/usr/share/doc/gcc-4.8-base/gccint.html

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<!-- Copyright (C) 1988-2013 Free Software Foundation, Inc.

Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with the
Invariant Sections being "Funding Free Software", the Front-Cover
Texts being (a) (see below), and with the Back-Cover Texts being (b)
(see below).  A copy of the license is included in the section entitled
"GNU Free Documentation License".

(a) The FSF's Front-Cover Text is:

A GNU Manual

(b) The FSF's Back-Cover Text is:

You have freedom to copy and modify this GNU Manual, like GNU
     software.  Copies published by the Free Software Foundation raise
     funds for GNU development. -->
<!-- Created by GNU Texinfo 5.2, http://www.gnu.org/software/texinfo/ -->
<head>
<title>GNU Compiler Collection (GCC) Internals</title>

<meta name="description" content="GNU Compiler Collection (GCC) Internals">
<meta name="keywords" content="GNU Compiler Collection (GCC) Internals">
<meta name="resource-type" content="document">
<meta name="distribution" content="global">
<meta name="Generator" content="makeinfo">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<link href="#Top" rel="start" title="Top">
<link href="#Option-Index" rel="index" title="Option Index">
<link href="#SEC_Contents" rel="contents" title="Table of Contents">
<link href="DIR.html#Top" rel="up" title="(DIR)">
<style type="text/css">
<!--
a.summary-letter {text-decoration: none}
blockquote.smallquotation {font-size: smaller}
div.display {margin-left: 3.2em}
div.example {margin-left: 3.2em}
div.indentedblock {margin-left: 3.2em}
div.lisp {margin-left: 3.2em}
div.smalldisplay {margin-left: 3.2em}
div.smallexample {margin-left: 3.2em}
div.smallindentedblock {margin-left: 3.2em; font-size: smaller}
div.smalllisp {margin-left: 3.2em}
kbd {font-style:oblique}
pre.display {font-family: inherit}
pre.format {font-family: inherit}
pre.menu-comment {font-family: serif}
pre.menu-preformatted {font-family: serif}
pre.smalldisplay {font-family: inherit; font-size: smaller}
pre.smallexample {font-size: smaller}
pre.smallformat {font-family: inherit; font-size: smaller}
pre.smalllisp {font-size: smaller}
span.nocodebreak {white-space:nowrap}
span.nolinebreak {white-space:nowrap}
span.roman {font-family:serif; font-weight:normal}
span.sansserif {font-family:sans-serif; font-weight:normal}
ul.no-bullet {list-style: none}
-->
</style>


</head>

<body lang="en" bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#800080" alink="#FF0000">
<h1 class="settitle" align="center">GNU Compiler Collection (GCC) Internals</h1>













<p>This file documents the internals of the GNU compilers.
</p><br>
<p>Copyright &copy; 1988-2013 Free Software Foundation, Inc.
</p>
<p>Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3 or
any later version published by the Free Software Foundation; with the
Invariant Sections being &ldquo;Funding Free Software&rdquo;, the Front-Cover
Texts being (a) (see below), and with the Back-Cover Texts being (b)
(see below).  A copy of the license is included in the section entitled
&ldquo;GNU Free Documentation License&rdquo;.
</p>
<p>(a) The FSF&rsquo;s Front-Cover Text is:
</p>
<p>A GNU Manual
</p>
<p>(b) The FSF&rsquo;s Back-Cover Text is:
</p>
<p>You have freedom to copy and modify this GNU Manual, like GNU
     software.  Copies published by the Free Software Foundation raise
     funds for GNU development.
</p>
<br>

<a name="SEC_Overview"></a>
<h2 class="shortcontents-heading">Short Table of Contents</h2>

<div class="shortcontents">
<ul class="no-bullet">
<li><a name="stoc-Contributing-to-GCC-Development" href="#toc-Contributing-to-GCC-Development">1 Contributing to GCC Development</a></li>
<li><a name="stoc-GCC-and-Portability" href="#toc-GCC-and-Portability">2 GCC and Portability</a></li>
<li><a name="stoc-Interfacing-to-GCC-Output" href="#toc-Interfacing-to-GCC-Output">3 Interfacing to GCC Output</a></li>
<li><a name="stoc-The-GCC-low_002dlevel-runtime-library" href="#toc-The-GCC-low_002dlevel-runtime-library">4 The GCC low-level runtime library</a></li>
<li><a name="stoc-Language-Front-Ends-in-GCC" href="#toc-Language-Front-Ends-in-GCC">5 Language Front Ends in GCC</a></li>
<li><a name="stoc-Source-Tree-Structure-and-Build-System" href="#toc-Source-Tree-Structure-and-Build-System">6 Source Tree Structure and Build System</a></li>
<li><a name="stoc-Testsuites-1" href="#toc-Testsuites-1">7 Testsuites</a></li>
<li><a name="stoc-Option-specification-files" href="#toc-Option-specification-files">8 Option specification files</a></li>
<li><a name="stoc-Passes-and-Files-of-the-Compiler" href="#toc-Passes-and-Files-of-the-Compiler">9 Passes and Files of the Compiler</a></li>
<li><a name="stoc-RTL-Representation" href="#toc-RTL-Representation">10 RTL Representation</a></li>
<li><a name="stoc-GENERIC-1" href="#toc-GENERIC-1">11 GENERIC</a></li>
<li><a name="stoc-GIMPLE-1" href="#toc-GIMPLE-1">12 GIMPLE</a></li>
<li><a name="stoc-Analysis-and-Optimization-of-GIMPLE-tuples" href="#toc-Analysis-and-Optimization-of-GIMPLE-tuples">13 Analysis and Optimization of GIMPLE tuples</a></li>
<li><a name="stoc-Analysis-and-Representation-of-Loops" href="#toc-Analysis-and-Representation-of-Loops">14 Analysis and Representation of Loops</a></li>
<li><a name="stoc-Control-Flow-Graph" href="#toc-Control-Flow-Graph">15 Control Flow Graph</a></li>
<li><a name="stoc-Machine-Descriptions" href="#toc-Machine-Descriptions">16 Machine Descriptions</a></li>
<li><a name="stoc-Target-Description-Macros-and-Functions" href="#toc-Target-Description-Macros-and-Functions">17 Target Description Macros and Functions</a></li>
<li><a name="stoc-Host-Configuration" href="#toc-Host-Configuration">18 Host Configuration</a></li>
<li><a name="stoc-Makefile-Fragments" href="#toc-Makefile-Fragments">19 Makefile Fragments</a></li>
<li><a name="stoc-collect2" href="#toc-collect2">20 <code>collect2</code></a></li>
<li><a name="stoc-Standard-Header-File-Directories" href="#toc-Standard-Header-File-Directories">21 Standard Header File Directories</a></li>
<li><a name="stoc-Memory-Management-and-Type-Information" href="#toc-Memory-Management-and-Type-Information">22 Memory Management and Type Information</a></li>
<li><a name="stoc-Plugins-1" href="#toc-Plugins-1">23 Plugins</a></li>
<li><a name="stoc-Link-Time-Optimization" href="#toc-Link-Time-Optimization">24 Link Time Optimization</a></li>
<li><a name="stoc-Funding-Free-Software" href="#toc-Funding-Free-Software">Funding Free Software</a></li>
<li><a name="stoc-The-GNU-Project-and-GNU_002fLinux" href="#toc-The-GNU-Project-and-GNU_002fLinux">The GNU Project and GNU/Linux</a></li>
<li><a name="stoc-GNU-General-Public-License" href="#toc-GNU-General-Public-License">GNU General Public License</a></li>
<li><a name="stoc-GNU-Free-Documentation-License-1" href="#toc-GNU-Free-Documentation-License-1">GNU Free Documentation License</a></li>
<li><a name="stoc-Contributors-to-GCC" href="#toc-Contributors-to-GCC">Contributors to GCC</a></li>
<li><a name="stoc-Option-Index-1" href="#toc-Option-Index-1">Option Index</a></li>
<li><a name="stoc-Concept-Index-1" href="#toc-Concept-Index-1">Concept Index</a></li>
</ul>
</div>

<a name="SEC_Contents"></a>
<h2 class="contents-heading">Table of Contents</h2>

<div class="contents">

<ul class="no-bullet">
  <li><a name="toc-Contributing-to-GCC-Development" href="#Contributing">1 Contributing to GCC Development</a></li>
  <li><a name="toc-GCC-and-Portability" href="#Portability">2 GCC and Portability</a></li>
  <li><a name="toc-Interfacing-to-GCC-Output" href="#Interface">3 Interfacing to GCC Output</a></li>
  <li><a name="toc-The-GCC-low_002dlevel-runtime-library" href="#Libgcc">4 The GCC low-level runtime library</a>
  <ul class="no-bullet">
    <li><a name="toc-Routines-for-integer-arithmetic" href="#Integer-library-routines">4.1 Routines for integer arithmetic</a>
    <ul class="no-bullet">
      <li><a name="toc-Arithmetic-functions" href="#Arithmetic-functions">4.1.1 Arithmetic functions</a></li>
      <li><a name="toc-Comparison-functions" href="#Comparison-functions">4.1.2 Comparison functions</a></li>
      <li><a name="toc-Trapping-arithmetic-functions" href="#Trapping-arithmetic-functions">4.1.3 Trapping arithmetic functions</a></li>
      <li><a name="toc-Bit-operations" href="#Bit-operations">4.1.4 Bit operations</a></li>
    </ul></li>
    <li><a name="toc-Routines-for-floating-point-emulation" href="#Soft-float-library-routines">4.2 Routines for floating point emulation</a>
    <ul class="no-bullet">
      <li><a name="toc-Arithmetic-functions-1" href="#Arithmetic-functions-1">4.2.1 Arithmetic functions</a></li>
      <li><a name="toc-Conversion-functions" href="#Conversion-functions">4.2.2 Conversion functions</a></li>
      <li><a name="toc-Comparison-functions-1" href="#Comparison-functions-1">4.2.3 Comparison functions</a></li>
      <li><a name="toc-Other-floating_002dpoint-functions" href="#Other-floating_002dpoint-functions">4.2.4 Other floating-point functions</a></li>
    </ul></li>
    <li><a name="toc-Routines-for-decimal-floating-point-emulation" href="#Decimal-float-library-routines">4.3 Routines for decimal floating point emulation</a>
    <ul class="no-bullet">
      <li><a name="toc-Arithmetic-functions-2" href="#Arithmetic-functions-2">4.3.1 Arithmetic functions</a></li>
      <li><a name="toc-Conversion-functions-1" href="#Conversion-functions-1">4.3.2 Conversion functions</a></li>
      <li><a name="toc-Comparison-functions-2" href="#Comparison-functions-2">4.3.3 Comparison functions</a></li>
    </ul></li>
    <li><a name="toc-Routines-for-fixed_002dpoint-fractional-emulation" href="#Fixed_002dpoint-fractional-library-routines">4.4 Routines for fixed-point fractional emulation</a>
    <ul class="no-bullet">
      <li><a name="toc-Arithmetic-functions-3" href="#Arithmetic-functions-3">4.4.1 Arithmetic functions</a></li>
      <li><a name="toc-Comparison-functions-3" href="#Comparison-functions-3">4.4.2 Comparison functions</a></li>
      <li><a name="toc-Conversion-functions-2" href="#Conversion-functions-2">4.4.3 Conversion functions</a></li>
    </ul></li>
    <li><a name="toc-Language_002dindependent-routines-for-exception-handling" href="#Exception-handling-routines">4.5 Language-independent routines for exception handling</a></li>
    <li><a name="toc-Miscellaneous-runtime-library-routines" href="#Miscellaneous-routines">4.6 Miscellaneous runtime library routines</a>
    <ul class="no-bullet">
      <li><a name="toc-Cache-control-functions" href="#Cache-control-functions">4.6.1 Cache control functions</a></li>
      <li><a name="toc-Split-stack-functions-and-variables" href="#Split-stack-functions-and-variables">4.6.2 Split stack functions and variables</a></li>
    </ul></li>
  </ul></li>
  <li><a name="toc-Language-Front-Ends-in-GCC" href="#Languages">5 Language Front Ends in GCC</a></li>
  <li><a name="toc-Source-Tree-Structure-and-Build-System" href="#Source-Tree">6 Source Tree Structure and Build System</a>
  <ul class="no-bullet">
    <li><a name="toc-Configure-Terms-and-History" href="#Configure-Terms">6.1 Configure Terms and History</a></li>
    <li><a name="toc-Top-Level-Source-Directory" href="#Top-Level">6.2 Top Level Source Directory</a></li>
    <li><a name="toc-The-gcc-Subdirectory" href="#gcc-Directory">6.3 The <samp>gcc</samp> Subdirectory</a>
    <ul class="no-bullet">
      <li><a name="toc-Subdirectories-of-gcc" href="#Subdirectories">6.3.1 Subdirectories of <samp>gcc</samp></a></li>
      <li><a name="toc-Configuration-in-the-gcc-Directory" href="#Configuration">6.3.2 Configuration in the <samp>gcc</samp> Directory</a>
      <ul class="no-bullet">
        <li><a name="toc-Scripts-Used-by-configure" href="#Config-Fragments">6.3.2.1 Scripts Used by <samp>configure</samp></a></li>
        <li><a name="toc-The-config_002ebuild_003b-config_002ehost_003b-and-config_002egcc-Files" href="#System-Config">6.3.2.2 The <samp>config.build</samp>; <samp>config.host</samp>; and <samp>config.gcc</samp> Files</a></li>
        <li><a name="toc-Files-Created-by-configure" href="#Configuration-Files">6.3.2.3 Files Created by <code>configure</code></a></li>
      </ul></li>
      <li><a name="toc-Build-System-in-the-gcc-Directory" href="#Build">6.3.3 Build System in the <samp>gcc</samp> Directory</a></li>
      <li><a name="toc-Makefile-Targets" href="#Makefile">6.3.4 Makefile Targets</a></li>
      <li><a name="toc-Library-Source-Files-and-Headers-under-the-gcc-Directory" href="#Library-Files">6.3.5 Library Source Files and Headers under the <samp>gcc</samp> Directory</a></li>
      <li><a name="toc-Headers-Installed-by-GCC" href="#Headers">6.3.6 Headers Installed by GCC</a></li>
      <li><a name="toc-Building-Documentation" href="#Documentation">6.3.7 Building Documentation</a>
      <ul class="no-bullet">
        <li><a name="toc-Texinfo-Manuals-1" href="#Texinfo-Manuals">6.3.7.1 Texinfo Manuals</a></li>
        <li><a name="toc-Man-Page-Generation-1" href="#Man-Page-Generation">6.3.7.2 Man Page Generation</a></li>
        <li><a name="toc-Miscellaneous-Documentation" href="#Miscellaneous-Docs">6.3.7.3 Miscellaneous Documentation</a></li>
      </ul></li>
      <li><a name="toc-Anatomy-of-a-Language-Front-End" href="#Front-End">6.3.8 Anatomy of a Language Front End</a>
      <ul class="no-bullet">
        <li><a name="toc-The-Front-End-language-Directory" href="#Front-End-Directory">6.3.8.1 The Front End <samp><var>language</var></samp> Directory</a></li>
        <li><a name="toc-The-Front-End-config_002dlang_002ein-File" href="#Front-End-Config">6.3.8.2 The Front End <samp>config-lang.in</samp> File</a></li>
        <li><a name="toc-The-Front-End-Make_002dlang_002ein-File" href="#Front-End-Makefile">6.3.8.3 The Front End <samp>Make-lang.in</samp> File</a></li>
      </ul></li>
      <li><a name="toc-Anatomy-of-a-Target-Back-End" href="#Back-End">6.3.9 Anatomy of a Target Back End</a></li>
    </ul></li>
  </ul></li>
  <li><a name="toc-Testsuites-1" href="#Testsuites">7 Testsuites</a>
  <ul class="no-bullet">
    <li><a name="toc-Idioms-Used-in-Testsuite-Code" href="#Test-Idioms">7.1 Idioms Used in Testsuite Code</a></li>
    <li><a name="toc-Directives-used-within-DejaGnu-tests" href="#Test-Directives">7.2 Directives used within DejaGnu tests</a>
    <ul class="no-bullet">
      <li><a name="toc-Syntax-and-Descriptions-of-test-directives" href="#Directives">7.2.1 Syntax and Descriptions of test directives</a>
      <ul class="no-bullet">
        <li><a name="toc-Specify-how-to-build-the-test" href="#Specify-how-to-build-the-test">7.2.1.1 Specify how to build the test</a></li>
        <li><a name="toc-Specify-additional-compiler-options" href="#Specify-additional-compiler-options">7.2.1.2 Specify additional compiler options</a></li>
        <li><a name="toc-Modify-the-test-timeout-value" href="#Modify-the-test-timeout-value">7.2.1.3 Modify the test timeout value</a></li>
        <li><a name="toc-Skip-a-test-for-some-targets" href="#Skip-a-test-for-some-targets">7.2.1.4 Skip a test for some targets</a></li>
        <li><a name="toc-Expect-a-test-to-fail-for-some-targets" href="#Expect-a-test-to-fail-for-some-targets">7.2.1.5 Expect a test to fail for some targets</a></li>
        <li><a name="toc-Expect-the-test-executable-to-fail" href="#Expect-the-test-executable-to-fail">7.2.1.6 Expect the test executable to fail</a></li>
        <li><a name="toc-Verify-compiler-messages" href="#Verify-compiler-messages">7.2.1.7 Verify compiler messages</a></li>
        <li><a name="toc-Verify-output-of-the-test-executable" href="#Verify-output-of-the-test-executable">7.2.1.8 Verify output of the test executable</a></li>
        <li><a name="toc-Specify-additional-files-for-a-test" href="#Specify-additional-files-for-a-test">7.2.1.9 Specify additional files for a test</a></li>
        <li><a name="toc-Add-checks-at-the-end-of-a-test" href="#Add-checks-at-the-end-of-a-test">7.2.1.10 Add checks at the end of a test</a></li>
      </ul></li>
      <li><a name="toc-Selecting-targets-to-which-a-test-applies" href="#Selectors">7.2.2 Selecting targets to which a test applies</a></li>
      <li><a name="toc-Keywords-describing-target-attributes" href="#Effective_002dTarget-Keywords">7.2.3 Keywords describing target attributes</a>
      <ul class="no-bullet">
        <li><a name="toc-Data-type-sizes" href="#Data-type-sizes">7.2.3.1 Data type sizes</a></li>
        <li><a name="toc-Fortran_002dspecific-attributes" href="#Fortran_002dspecific-attributes">7.2.3.2 Fortran-specific attributes</a></li>
        <li><a name="toc-Vector_002dspecific-attributes" href="#Vector_002dspecific-attributes">7.2.3.3 Vector-specific attributes</a></li>
        <li><a name="toc-Thread-Local-Storage-attributes" href="#Thread-Local-Storage-attributes">7.2.3.4 Thread Local Storage attributes</a></li>
        <li><a name="toc-Decimal-floating-point-attributes" href="#Decimal-floating-point-attributes">7.2.3.5 Decimal floating point attributes</a></li>
        <li><a name="toc-ARM_002dspecific-attributes" href="#ARM_002dspecific-attributes">7.2.3.6 ARM-specific attributes</a></li>
        <li><a name="toc-MIPS_002dspecific-attributes" href="#MIPS_002dspecific-attributes">7.2.3.7 MIPS-specific attributes</a></li>
        <li><a name="toc-PowerPC_002dspecific-attributes" href="#PowerPC_002dspecific-attributes">7.2.3.8 PowerPC-specific attributes</a></li>
        <li><a name="toc-Other-hardware-attributes" href="#Other-hardware-attributes">7.2.3.9 Other hardware attributes</a></li>
        <li><a name="toc-Environment-attributes" href="#Environment-attributes">7.2.3.10 Environment attributes</a></li>
        <li><a name="toc-Other-attributes" href="#Other-attributes">7.2.3.11 Other attributes</a></li>
        <li><a name="toc-Local-to-tests-in-gcc_002etarget_002fi386" href="#Local-to-tests-in-gcc_002etarget_002fi386">7.2.3.12 Local to tests in <code>gcc.target/i386</code></a></li>
        <li><a name="toc-Local-to-tests-in-gcc_002etarget_002fspu_002fea" href="#Local-to-tests-in-gcc_002etarget_002fspu_002fea">7.2.3.13 Local to tests in <code>gcc.target/spu/ea</code></a></li>
        <li><a name="toc-Local-to-tests-in-gcc_002etest_002dframework" href="#Local-to-tests-in-gcc_002etest_002dframework">7.2.3.14 Local to tests in <code>gcc.test-framework</code></a></li>
      </ul></li>
      <li><a name="toc-Features-for-dg_002dadd_002doptions" href="#Add-Options">7.2.4 Features for <code>dg-add-options</code></a></li>
      <li><a name="toc-Variants-of-dg_002drequire_002dsupport" href="#Require-Support">7.2.5 Variants of <code>dg-require-<var>support</var></code></a></li>
      <li><a name="toc-Commands-for-use-in-dg_002dfinal" href="#Final-Actions">7.2.6 Commands for use in <code>dg-final</code></a>
      <ul class="no-bullet">
        <li><a name="toc-Scan-a-particular-file" href="#Scan-a-particular-file">7.2.6.1 Scan a particular file</a></li>
        <li><a name="toc-Scan-the-assembly-output" href="#Scan-the-assembly-output">7.2.6.2 Scan the assembly output</a></li>
        <li><a name="toc-Scan-optimization-dump-files" href="#Scan-optimization-dump-files">7.2.6.3 Scan optimization dump files</a></li>
        <li><a name="toc-Verify-that-an-output-files-exists-or-not" href="#Verify-that-an-output-files-exists-or-not">7.2.6.4 Verify that an output files exists or not</a></li>
        <li><a name="toc-Check-for-LTO-tests" href="#Check-for-LTO-tests">7.2.6.5 Check for LTO tests</a></li>
        <li><a name="toc-Checks-for-gcov-tests" href="#Checks-for-gcov-tests">7.2.6.6 Checks for <code>gcov</code> tests</a></li>
        <li><a name="toc-Clean-up-generated-test-files" href="#Clean-up-generated-test-files">7.2.6.7 Clean up generated test files</a></li>
      </ul></li>
    </ul></li>
    <li><a name="toc-Ada-Language-Testsuites" href="#Ada-Tests">7.3 Ada Language Testsuites</a></li>
    <li><a name="toc-C-Language-Testsuites" href="#C-Tests">7.4 C Language Testsuites</a></li>
    <li><a name="toc-The-Java-library-testsuites_002e" href="#libgcj-Tests">7.5 The Java library testsuites.</a></li>
    <li><a name="toc-Support-for-testing-link_002dtime-optimizations" href="#LTO-Testing">7.6 Support for testing link-time optimizations</a></li>
    <li><a name="toc-Support-for-testing-gcov" href="#gcov-Testing">7.7 Support for testing <code>gcov</code></a></li>
    <li><a name="toc-Support-for-testing-profile_002ddirected-optimizations" href="#profopt-Testing">7.8 Support for testing profile-directed optimizations</a></li>
    <li><a name="toc-Support-for-testing-binary-compatibility" href="#compat-Testing">7.9 Support for testing binary compatibility</a></li>
    <li><a name="toc-Support-for-torture-testing-using-multiple-options" href="#Torture-Tests">7.10 Support for torture testing using multiple options</a></li>
  </ul></li>
  <li><a name="toc-Option-specification-files" href="#Options">8 Option specification files</a>
  <ul class="no-bullet">
    <li><a name="toc-Option-file-format-1" href="#Option-file-format">8.1 Option file format</a></li>
    <li><a name="toc-Option-properties-1" href="#Option-properties">8.2 Option properties</a></li>
  </ul></li>
  <li><a name="toc-Passes-and-Files-of-the-Compiler" href="#Passes">9 Passes and Files of the Compiler</a>
  <ul class="no-bullet">
    <li><a name="toc-Parsing-pass-1" href="#Parsing-pass">9.1 Parsing pass</a></li>
    <li><a name="toc-Gimplification-pass-1" href="#Gimplification-pass">9.2 Gimplification pass</a></li>
    <li><a name="toc-Pass-manager-1" href="#Pass-manager">9.3 Pass manager</a></li>
    <li><a name="toc-Tree-SSA-passes-1" href="#Tree-SSA-passes">9.4 Tree SSA passes</a></li>
    <li><a name="toc-RTL-passes-1" href="#RTL-passes">9.5 RTL passes</a></li>
  </ul></li>
  <li><a name="toc-RTL-Representation" href="#RTL">10 RTL Representation</a>
  <ul class="no-bullet">
    <li><a name="toc-RTL-Object-Types" href="#RTL-Objects">10.1 RTL Object Types</a></li>
    <li><a name="toc-RTL-Classes-and-Formats" href="#RTL-Classes">10.2 RTL Classes and Formats</a></li>
    <li><a name="toc-Access-to-Operands" href="#Accessors">10.3 Access to Operands</a></li>
    <li><a name="toc-Access-to-Special-Operands" href="#Special-Accessors">10.4 Access to Special Operands</a></li>
    <li><a name="toc-Flags-in-an-RTL-Expression" href="#Flags">10.5 Flags in an RTL Expression</a></li>
    <li><a name="toc-Machine-Modes-1" href="#Machine-Modes">10.6 Machine Modes</a></li>
    <li><a name="toc-Constant-Expression-Types" href="#Constants">10.7 Constant Expression Types</a></li>
    <li><a name="toc-Registers-and-Memory" href="#Regs-and-Memory">10.8 Registers and Memory</a></li>
    <li><a name="toc-RTL-Expressions-for-Arithmetic" href="#Arithmetic">10.9 RTL Expressions for Arithmetic</a></li>
    <li><a name="toc-Comparison-Operations" href="#Comparisons">10.10 Comparison Operations</a></li>
    <li><a name="toc-Bit_002dFields-1" href="#Bit_002dFields">10.11 Bit-Fields</a></li>
    <li><a name="toc-Vector-Operations-1" href="#Vector-Operations">10.12 Vector Operations</a></li>
    <li><a name="toc-Conversions-1" href="#Conversions">10.13 Conversions</a></li>
    <li><a name="toc-Declarations-1" href="#RTL-Declarations">10.14 Declarations</a></li>
    <li><a name="toc-Side-Effect-Expressions" href="#Side-Effects">10.15 Side Effect Expressions</a></li>
    <li><a name="toc-Embedded-Side_002dEffects-on-Addresses" href="#Incdec">10.16 Embedded Side-Effects on Addresses</a></li>
    <li><a name="toc-Assembler-Instructions-as-Expressions" href="#Assembler">10.17 Assembler Instructions as Expressions</a></li>
    <li><a name="toc-Variable-Location-Debug-Information-in-RTL" href="#Debug-Information">10.18 Variable Location Debug Information in RTL</a></li>
    <li><a name="toc-Insns-1" href="#Insns">10.19 Insns</a></li>
    <li><a name="toc-RTL-Representation-of-Function_002dCall-Insns" href="#Calls">10.20 RTL Representation of Function-Call Insns</a></li>
    <li><a name="toc-Structure-Sharing-Assumptions" href="#Sharing">10.21 Structure Sharing Assumptions</a></li>
    <li><a name="toc-Reading-RTL-1" href="#Reading-RTL">10.22 Reading RTL</a></li>
  </ul></li>
  <li><a name="toc-GENERIC-1" href="#GENERIC">11 GENERIC</a>
  <ul class="no-bullet">
    <li><a name="toc-Deficiencies-1" href="#Deficiencies">11.1 Deficiencies</a></li>
    <li><a name="toc-Overview-1" href="#Tree-overview">11.2 Overview</a>
    <ul class="no-bullet">
      <li><a name="toc-Trees" href="#Macros-and-Functions">11.2.1 Trees</a></li>
      <li><a name="toc-Identifiers-1" href="#Identifiers">11.2.2 Identifiers</a></li>
      <li><a name="toc-Containers-1" href="#Containers">11.2.3 Containers</a></li>
    </ul></li>
    <li><a name="toc-Types-1" href="#Types">11.3 Types</a></li>
    <li><a name="toc-Declarations-2" href="#Declarations">11.4 Declarations</a>
    <ul class="no-bullet">
      <li><a name="toc-Working-with-declarations-1" href="#Working-with-declarations">11.4.1 Working with declarations</a></li>
      <li><a name="toc-Internal-structure-1" href="#Internal-structure">11.4.2 Internal structure</a>
      <ul class="no-bullet">
        <li><a name="toc-Current-structure-hierarchy-1" href="#Current-structure-hierarchy">11.4.2.1 Current structure hierarchy</a></li>
        <li><a name="toc-Adding-new-DECL-node-types-1" href="#Adding-new-DECL-node-types">11.4.2.2 Adding new DECL node types</a></li>
      </ul></li>
    </ul></li>
    <li><a name="toc-Attributes-in-trees" href="#Attributes">11.5 Attributes in trees</a></li>
    <li><a name="toc-Expressions-1" href="#Expression-trees">11.6 Expressions</a>
    <ul class="no-bullet">
      <li><a name="toc-Constant-expressions-1" href="#Constant-expressions">11.6.1 Constant expressions</a></li>
      <li><a name="toc-References-to-storage" href="#Storage-References">11.6.2 References to storage</a></li>
      <li><a name="toc-Unary-and-Binary-Expressions-1" href="#Unary-and-Binary-Expressions">11.6.3 Unary and Binary Expressions</a></li>
      <li><a name="toc-Vectors-1" href="#Vectors">11.6.4 Vectors</a></li>
    </ul></li>
    <li><a name="toc-Statements-1" href="#Statements">11.7 Statements</a>
    <ul class="no-bullet">
      <li><a name="toc-Basic-Statements-1" href="#Basic-Statements">11.7.1 Basic Statements</a></li>
      <li><a name="toc-Blocks-1" href="#Blocks">11.7.2 Blocks</a></li>
      <li><a name="toc-Statement-Sequences-1" href="#Statement-Sequences">11.7.3 Statement Sequences</a></li>
      <li><a name="toc-Empty-Statements-1" href="#Empty-Statements">11.7.4 Empty Statements</a></li>
      <li><a name="toc-Jumps-1" href="#Jumps">11.7.5 Jumps</a></li>
      <li><a name="toc-Cleanups-1" href="#Cleanups">11.7.6 Cleanups</a></li>
      <li><a name="toc-OpenMP-1" href="#OpenMP">11.7.7 OpenMP</a></li>
    </ul></li>
    <li><a name="toc-Functions-1" href="#Functions">11.8 Functions</a>
    <ul class="no-bullet">
      <li><a name="toc-Function-Basics-1" href="#Function-Basics">11.8.1 Function Basics</a></li>
      <li><a name="toc-Function-Properties-1" href="#Function-Properties">11.8.2 Function Properties</a></li>
    </ul></li>
    <li><a name="toc-Language_002ddependent-trees-1" href="#Language_002ddependent-trees">11.9 Language-dependent trees</a></li>
    <li><a name="toc-C-and-C_002b_002b-Trees-1" href="#C-and-C_002b_002b-Trees">11.10 C and C++ Trees</a>
    <ul class="no-bullet">
      <li><a name="toc-Types-for-C_002b_002b-1" href="#Types-for-C_002b_002b">11.10.1 Types for C++</a></li>
      <li><a name="toc-Namespaces-1" href="#Namespaces">11.10.2 Namespaces</a></li>
      <li><a name="toc-Classes-1" href="#Classes">11.10.3 Classes</a></li>
      <li><a name="toc-Functions-for-C_002b_002b-1" href="#Functions-for-C_002b_002b">11.10.4 Functions for C++</a></li>
      <li><a name="toc-Statements-for-C_002b_002b-1" href="#Statements-for-C_002b_002b">11.10.5 Statements for C++</a>
      <ul class="no-bullet">
        <li><a name="toc-Statements-2" href="#Statements-2">11.10.5.1 Statements</a></li>
      </ul></li>
      <li><a name="toc-C_002b_002b-Expressions-1" href="#C_002b_002b-Expressions">11.10.6 C++ Expressions</a></li>
    </ul></li>
    <li><a name="toc-Java-Trees-1" href="#Java-Trees">11.11 Java Trees</a></li>
  </ul></li>
  <li><a name="toc-GIMPLE-1" href="#GIMPLE">12 GIMPLE</a>
  <ul class="no-bullet">
    <li><a name="toc-Tuple-representation-1" href="#Tuple-representation">12.1 Tuple representation</a>
    <ul class="no-bullet">
      <li><a name="toc-gimple_005fstatement_005fbase-_0028gsbase_0029" href="#gimple_005fstatement_005fbase-_0028gsbase_0029">12.1.1 <code>gimple_statement_base</code> (gsbase)</a></li>
      <li><a name="toc-gimple_005fstatement_005fwith_005fops" href="#gimple_005fstatement_005fwith_005fops">12.1.2 <code>gimple_statement_with_ops</code></a></li>
      <li><a name="toc-gimple_005fstatement_005fwith_005fmemory_005fops" href="#gimple_005fstatement_005fwith_005fmemory_005fops">12.1.3 <code>gimple_statement_with_memory_ops</code></a></li>
    </ul></li>
    <li><a name="toc-GIMPLE-instruction-set-1" href="#GIMPLE-instruction-set">12.2 GIMPLE instruction set</a></li>
    <li><a name="toc-Exception-Handling-1" href="#GIMPLE-Exception-Handling">12.3 Exception Handling</a></li>
    <li><a name="toc-Temporaries-1" href="#Temporaries">12.4 Temporaries</a></li>
    <li><a name="toc-Operands-1" href="#Operands">12.5 Operands</a>
    <ul class="no-bullet">
      <li><a name="toc-Compound-Expressions-1" href="#Compound-Expressions">12.5.1 Compound Expressions</a></li>
      <li><a name="toc-Compound-Lvalues-1" href="#Compound-Lvalues">12.5.2 Compound Lvalues</a></li>
      <li><a name="toc-Conditional-Expressions-1" href="#Conditional-Expressions">12.5.3 Conditional Expressions</a></li>
      <li><a name="toc-Logical-Operators-1" href="#Logical-Operators">12.5.4 Logical Operators</a></li>
      <li><a name="toc-Manipulating-operands" href="#Manipulating-operands">12.5.5 Manipulating operands</a></li>
      <li><a name="toc-Operand-vector-allocation" href="#Operand-vector-allocation">12.5.6 Operand vector allocation</a></li>
      <li><a name="toc-Operand-validation" href="#Operand-validation">12.5.7 Operand validation</a></li>
      <li><a name="toc-Statement-validation" href="#Statement-validation">12.5.8 Statement validation</a></li>
    </ul></li>
    <li><a name="toc-Manipulating-GIMPLE-statements-1" href="#Manipulating-GIMPLE-statements">12.6 Manipulating GIMPLE statements</a>
    <ul class="no-bullet">
      <li><a name="toc-Common-accessors" href="#Common-accessors">12.6.1 Common accessors</a></li>
    </ul></li>
    <li><a name="toc-Tuple-specific-accessors-1" href="#Tuple-specific-accessors">12.7 Tuple specific accessors</a>
    <ul class="no-bullet">
      <li><a name="toc-GIMPLE_005fASM-1" href="#GIMPLE_005fASM">12.7.1 <code>GIMPLE_ASM</code></a></li>
      <li><a name="toc-GIMPLE_005fASSIGN-1" href="#GIMPLE_005fASSIGN">12.7.2 <code>GIMPLE_ASSIGN</code></a></li>
      <li><a name="toc-GIMPLE_005fBIND-1" href="#GIMPLE_005fBIND">12.7.3 <code>GIMPLE_BIND</code></a></li>
      <li><a name="toc-GIMPLE_005fCALL-1" href="#GIMPLE_005fCALL">12.7.4 <code>GIMPLE_CALL</code></a></li>
      <li><a name="toc-GIMPLE_005fCATCH-1" href="#GIMPLE_005fCATCH">12.7.5 <code>GIMPLE_CATCH</code></a></li>
      <li><a name="toc-GIMPLE_005fCOND-1" href="#GIMPLE_005fCOND">12.7.6 <code>GIMPLE_COND</code></a></li>
      <li><a name="toc-GIMPLE_005fDEBUG-1" href="#GIMPLE_005fDEBUG">12.7.7 <code>GIMPLE_DEBUG</code></a></li>
      <li><a name="toc-GIMPLE_005fEH_005fFILTER-1" href="#GIMPLE_005fEH_005fFILTER">12.7.8 <code>GIMPLE_EH_FILTER</code></a></li>
      <li><a name="toc-GIMPLE_005fLABEL-1" href="#GIMPLE_005fLABEL">12.7.9 <code>GIMPLE_LABEL</code></a></li>
      <li><a name="toc-GIMPLE_005fNOP-1" href="#GIMPLE_005fNOP">12.7.10 <code>GIMPLE_NOP</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fATOMIC_005fLOAD-1" href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD">12.7.11 <code>GIMPLE_OMP_ATOMIC_LOAD</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fATOMIC_005fSTORE-1" href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE">12.7.12 <code>GIMPLE_OMP_ATOMIC_STORE</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fCONTINUE-1" href="#GIMPLE_005fOMP_005fCONTINUE">12.7.13 <code>GIMPLE_OMP_CONTINUE</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fCRITICAL-1" href="#GIMPLE_005fOMP_005fCRITICAL">12.7.14 <code>GIMPLE_OMP_CRITICAL</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fFOR-1" href="#GIMPLE_005fOMP_005fFOR">12.7.15 <code>GIMPLE_OMP_FOR</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fMASTER-1" href="#GIMPLE_005fOMP_005fMASTER">12.7.16 <code>GIMPLE_OMP_MASTER</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fORDERED-1" href="#GIMPLE_005fOMP_005fORDERED">12.7.17 <code>GIMPLE_OMP_ORDERED</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fPARALLEL-1" href="#GIMPLE_005fOMP_005fPARALLEL">12.7.18 <code>GIMPLE_OMP_PARALLEL</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fRETURN-1" href="#GIMPLE_005fOMP_005fRETURN">12.7.19 <code>GIMPLE_OMP_RETURN</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fSECTION-1" href="#GIMPLE_005fOMP_005fSECTION">12.7.20 <code>GIMPLE_OMP_SECTION</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fSECTIONS-1" href="#GIMPLE_005fOMP_005fSECTIONS">12.7.21 <code>GIMPLE_OMP_SECTIONS</code></a></li>
      <li><a name="toc-GIMPLE_005fOMP_005fSINGLE-1" href="#GIMPLE_005fOMP_005fSINGLE">12.7.22 <code>GIMPLE_OMP_SINGLE</code></a></li>
      <li><a name="toc-GIMPLE_005fPHI-1" href="#GIMPLE_005fPHI">12.7.23 <code>GIMPLE_PHI</code></a></li>
      <li><a name="toc-GIMPLE_005fRESX-1" href="#GIMPLE_005fRESX">12.7.24 <code>GIMPLE_RESX</code></a></li>
      <li><a name="toc-GIMPLE_005fRETURN-1" href="#GIMPLE_005fRETURN">12.7.25 <code>GIMPLE_RETURN</code></a></li>
      <li><a name="toc-GIMPLE_005fSWITCH-1" href="#GIMPLE_005fSWITCH">12.7.26 <code>GIMPLE_SWITCH</code></a></li>
      <li><a name="toc-GIMPLE_005fTRY-1" href="#GIMPLE_005fTRY">12.7.27 <code>GIMPLE_TRY</code></a></li>
      <li><a name="toc-GIMPLE_005fWITH_005fCLEANUP_005fEXPR-1" href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR">12.7.28 <code>GIMPLE_WITH_CLEANUP_EXPR</code></a></li>
    </ul></li>
    <li><a name="toc-GIMPLE-sequences-1" href="#GIMPLE-sequences">12.8 GIMPLE sequences</a></li>
    <li><a name="toc-Sequence-iterators-1" href="#Sequence-iterators">12.9 Sequence iterators</a></li>
    <li><a name="toc-Adding-a-new-GIMPLE-statement-code-1" href="#Adding-a-new-GIMPLE-statement-code">12.10 Adding a new GIMPLE statement code</a></li>
    <li><a name="toc-Statement-and-operand-traversals-1" href="#Statement-and-operand-traversals">12.11 Statement and operand traversals</a></li>
  </ul></li>
  <li><a name="toc-Analysis-and-Optimization-of-GIMPLE-tuples" href="#Tree-SSA">13 Analysis and Optimization of GIMPLE tuples</a>
  <ul class="no-bullet">
    <li><a name="toc-Annotations-1" href="#Annotations">13.1 Annotations</a></li>
    <li><a name="toc-SSA-Operands-1" href="#SSA-Operands">13.2 SSA Operands</a>
    <ul class="no-bullet">
      <li><a name="toc-Operand-Iterators-And-Access-Routines" href="#Operand-Iterators-And-Access-Routines">13.2.1 Operand Iterators And Access Routines</a></li>
      <li><a name="toc-Immediate-Uses" href="#Immediate-Uses">13.2.2 Immediate Uses</a></li>
    </ul></li>
    <li><a name="toc-Static-Single-Assignment" href="#SSA">13.3 Static Single Assignment</a>
    <ul class="no-bullet">
      <li><a name="toc-Preserving-the-SSA-form" href="#Preserving-the-SSA-form">13.3.1 Preserving the SSA form</a></li>
      <li><a name="toc-Preserving-the-virtual-SSA-form" href="#Preserving-the-virtual-SSA-form">13.3.2 Preserving the virtual SSA form</a></li>
      <li><a name="toc-Examining-SSA_005fNAME-nodes" href="#Examining-SSA_005fNAME-nodes">13.3.3 Examining <code>SSA_NAME</code> nodes</a></li>
      <li><a name="toc-Walking-use_002ddef-chains" href="#Walking-use_002ddef-chains">13.3.4 Walking use-def chains</a></li>
      <li><a name="toc-Walking-the-dominator-tree" href="#Walking-the-dominator-tree">13.3.5 Walking the dominator tree</a></li>
    </ul></li>
    <li><a name="toc-Alias-analysis-1" href="#Alias-analysis">13.4 Alias analysis</a></li>
    <li><a name="toc-Memory-model-1" href="#Memory-model">13.5 Memory model</a></li>
  </ul></li>
  <li><a name="toc-Analysis-and-Representation-of-Loops" href="#Loop-Analysis-and-Representation">14 Analysis and Representation of Loops</a>
  <ul class="no-bullet">
    <li><a name="toc-Loop-representation-1" href="#Loop-representation">14.1 Loop representation</a></li>
    <li><a name="toc-Loop-querying-1" href="#Loop-querying">14.2 Loop querying</a></li>
    <li><a name="toc-Loop-manipulation-1" href="#Loop-manipulation">14.3 Loop manipulation</a></li>
    <li><a name="toc-Loop_002dclosed-SSA-form" href="#LCSSA">14.4 Loop-closed SSA form</a></li>
    <li><a name="toc-Scalar-evolutions-1" href="#Scalar-evolutions">14.5 Scalar evolutions</a></li>
    <li><a name="toc-IV-analysis-on-RTL" href="#loop_002div">14.6 IV analysis on RTL</a></li>
    <li><a name="toc-Number-of-iterations-analysis" href="#Number-of-iterations">14.7 Number of iterations analysis</a></li>
    <li><a name="toc-Data-Dependency-Analysis" href="#Dependency-analysis">14.8 Data Dependency Analysis</a></li>
    <li><a name="toc-Linear-loop-transformations-framework" href="#Lambda">14.9 Linear loop transformations framework</a></li>
    <li><a name="toc-Omega-a-solver-for-linear-programming-problems" href="#Omega">14.10 Omega a solver for linear programming problems</a></li>
  </ul></li>
  <li><a name="toc-Control-Flow-Graph" href="#Control-Flow">15 Control Flow Graph</a>
  <ul class="no-bullet">
    <li><a name="toc-Basic-Blocks-1" href="#Basic-Blocks">15.1 Basic Blocks</a></li>
    <li><a name="toc-Edges-1" href="#Edges">15.2 Edges</a></li>
    <li><a name="toc-Profile-information-1" href="#Profile-information">15.3 Profile information</a></li>
    <li><a name="toc-Maintaining-the-CFG-1" href="#Maintaining-the-CFG">15.4 Maintaining the CFG</a></li>
    <li><a name="toc-Liveness-information-1" href="#Liveness-information">15.5 Liveness information</a></li>
  </ul></li>
  <li><a name="toc-Machine-Descriptions" href="#Machine-Desc">16 Machine Descriptions</a>
  <ul class="no-bullet">
    <li><a name="toc-Overview-of-How-the-Machine-Description-is-Used" href="#Overview">16.1 Overview of How the Machine Description is Used</a></li>
    <li><a name="toc-Everything-about-Instruction-Patterns" href="#Patterns">16.2 Everything about Instruction Patterns</a></li>
    <li><a name="toc-Example-of-define_005finsn" href="#Example">16.3 Example of <code>define_insn</code></a></li>
    <li><a name="toc-RTL-Template-1" href="#RTL-Template">16.4 RTL Template</a></li>
    <li><a name="toc-Output-Templates-and-Operand-Substitution" href="#Output-Template">16.5 Output Templates and Operand Substitution</a></li>
    <li><a name="toc-C-Statements-for-Assembler-Output" href="#Output-Statement">16.6 C Statements for Assembler Output</a></li>
    <li><a name="toc-Predicates-1" href="#Predicates">16.7 Predicates</a>
    <ul class="no-bullet">
      <li><a name="toc-Machine_002dIndependent-Predicates-1" href="#Machine_002dIndependent-Predicates">16.7.1 Machine-Independent Predicates</a></li>
      <li><a name="toc-Defining-Machine_002dSpecific-Predicates" href="#Defining-Predicates">16.7.2 Defining Machine-Specific Predicates</a></li>
    </ul></li>
    <li><a name="toc-Operand-Constraints" href="#Constraints">16.8 Operand Constraints</a>
    <ul class="no-bullet">
      <li><a name="toc-Simple-Constraints-1" href="#Simple-Constraints">16.8.1 Simple Constraints</a></li>
      <li><a name="toc-Multiple-Alternative-Constraints" href="#Multi_002dAlternative">16.8.2 Multiple Alternative Constraints</a></li>
      <li><a name="toc-Register-Class-Preferences" href="#Class-Preferences">16.8.3 Register Class Preferences</a></li>
      <li><a name="toc-Constraint-Modifier-Characters" href="#Modifiers">16.8.4 Constraint Modifier Characters</a></li>
      <li><a name="toc-Constraints-for-Particular-Machines" href="#Machine-Constraints">16.8.5 Constraints for Particular Machines</a></li>
      <li><a name="toc-Disable-insn-alternatives-using-the-enabled-attribute" href="#Disable-Insn-Alternatives">16.8.6 Disable insn alternatives using the <code>enabled</code> attribute</a></li>
      <li><a name="toc-Defining-Machine_002dSpecific-Constraints" href="#Define-Constraints">16.8.7 Defining Machine-Specific Constraints</a></li>
      <li><a name="toc-Testing-constraints-from-C" href="#C-Constraint-Interface">16.8.8 Testing constraints from C</a></li>
    </ul></li>
    <li><a name="toc-Standard-Pattern-Names-For-Generation" href="#Standard-Names">16.9 Standard Pattern Names For Generation</a></li>
    <li><a name="toc-When-the-Order-of-Patterns-Matters" href="#Pattern-Ordering">16.10 When the Order of Patterns Matters</a></li>
    <li><a name="toc-Interdependence-of-Patterns" href="#Dependent-Patterns">16.11 Interdependence of Patterns</a></li>
    <li><a name="toc-Defining-Jump-Instruction-Patterns" href="#Jump-Patterns">16.12 Defining Jump Instruction Patterns</a></li>
    <li><a name="toc-Defining-Looping-Instruction-Patterns" href="#Looping-Patterns">16.13 Defining Looping Instruction Patterns</a></li>
    <li><a name="toc-Canonicalization-of-Instructions" href="#Insn-Canonicalizations">16.14 Canonicalization of Instructions</a></li>
    <li><a name="toc-Defining-RTL-Sequences-for-Code-Generation" href="#Expander-Definitions">16.15 Defining RTL Sequences for Code Generation</a></li>
    <li><a name="toc-Defining-How-to-Split-Instructions" href="#Insn-Splitting">16.16 Defining How to Split Instructions</a></li>
    <li><a name="toc-Including-Patterns-in-Machine-Descriptions_002e" href="#Including-Patterns">16.17 Including Patterns in Machine Descriptions.</a>
    <ul class="no-bullet">
      <li><a name="toc-RTL-Generation-Tool-Options-for-Directory-Search" href="#RTL-Generation-Tool-Options-for-Directory-Search">16.17.1 RTL Generation Tool Options for Directory Search</a></li>
    </ul></li>
    <li><a name="toc-Machine_002dSpecific-Peephole-Optimizers" href="#Peephole-Definitions">16.18 Machine-Specific Peephole Optimizers</a>
    <ul class="no-bullet">
      <li><a name="toc-RTL-to-Text-Peephole-Optimizers" href="#define_005fpeephole">16.18.1 RTL to Text Peephole Optimizers</a></li>
      <li><a name="toc-RTL-to-RTL-Peephole-Optimizers" href="#define_005fpeephole2">16.18.2 RTL to RTL Peephole Optimizers</a></li>
    </ul></li>
    <li><a name="toc-Instruction-Attributes" href="#Insn-Attributes">16.19 Instruction Attributes</a>
    <ul class="no-bullet">
      <li><a name="toc-Defining-Attributes-and-their-Values" href="#Defining-Attributes">16.19.1 Defining Attributes and their Values</a></li>
      <li><a name="toc-Attribute-Expressions" href="#Expressions">16.19.2 Attribute Expressions</a></li>
      <li><a name="toc-Assigning-Attribute-Values-to-Insns" href="#Tagging-Insns">16.19.3 Assigning Attribute Values to Insns</a></li>
      <li><a name="toc-Example-of-Attribute-Specifications" href="#Attr-Example">16.19.4 Example of Attribute Specifications</a></li>
      <li><a name="toc-Computing-the-Length-of-an-Insn" href="#Insn-Lengths">16.19.5 Computing the Length of an Insn</a></li>
      <li><a name="toc-Constant-Attributes-1" href="#Constant-Attributes">16.19.6 Constant Attributes</a></li>
      <li><a name="toc-Delay-Slot-Scheduling" href="#Delay-Slots">16.19.7 Delay Slot Scheduling</a></li>
      <li><a name="toc-Specifying-processor-pipeline-description" href="#Processor-pipeline-description">16.19.8 Specifying processor pipeline description</a></li>
    </ul></li>
    <li><a name="toc-Conditional-Execution-1" href="#Conditional-Execution">16.20 Conditional Execution</a></li>
    <li><a name="toc-RTL-Templates-Transformations" href="#Define-Subst">16.21 RTL Templates Transformations</a>
    <ul class="no-bullet">
      <li><a name="toc-define_005fsubst-Example" href="#Define-Subst-Example">16.21.1 <code>define_subst</code> Example</a></li>
      <li><a name="toc-Pattern-Matching-in-define_005fsubst" href="#Define-Subst-Pattern-Matching">16.21.2 Pattern Matching in <code>define_subst</code></a></li>
      <li><a name="toc-Generation-of-output-template-in-define_005fsubst" href="#Define-Subst-Output-Template">16.21.3 Generation of output template in <code>define_subst</code></a></li>
    </ul></li>
    <li><a name="toc-Constant-Definitions-1" href="#Constant-Definitions">16.22 Constant Definitions</a></li>
    <li><a name="toc-Iterators-1" href="#Iterators">16.23 Iterators</a>
    <ul class="no-bullet">
      <li><a name="toc-Mode-Iterators-1" href="#Mode-Iterators">16.23.1 Mode Iterators</a>
      <ul class="no-bullet">
        <li><a name="toc-Defining-Mode-Iterators-1" href="#Defining-Mode-Iterators">16.23.1.1 Defining Mode Iterators</a></li>
        <li><a name="toc-Substitution-in-Mode-Iterators" href="#Substitutions">16.23.1.2 Substitution in Mode Iterators</a></li>
        <li><a name="toc-Mode-Iterator-Examples" href="#Examples">16.23.1.3 Mode Iterator Examples</a></li>
      </ul></li>
      <li><a name="toc-Code-Iterators-1" href="#Code-Iterators">16.23.2 Code Iterators</a></li>
      <li><a name="toc-Int-Iterators-1" href="#Int-Iterators">16.23.3 Int Iterators</a></li>
      <li><a name="toc-Subst-Iterators-1" href="#Subst-Iterators">16.23.4 Subst Iterators</a></li>
    </ul></li>
  </ul></li>
  <li><a name="toc-Target-Description-Macros-and-Functions" href="#Target-Macros">17 Target Description Macros and Functions</a>
  <ul class="no-bullet">
    <li><a name="toc-The-Global-targetm-Variable" href="#Target-Structure">17.1 The Global <code>targetm</code> Variable</a></li>
    <li><a name="toc-Controlling-the-Compilation-Driver_002c-gcc" href="#Driver">17.2 Controlling the Compilation Driver, <samp>gcc</samp></a></li>
    <li><a name="toc-Run_002dtime-Target-Specification" href="#Run_002dtime-Target">17.3 Run-time Target Specification</a></li>
    <li><a name="toc-Defining-data-structures-for-per_002dfunction-information_002e" href="#Per_002dFunction-Data">17.4 Defining data structures for per-function information.</a></li>
    <li><a name="toc-Storage-Layout-1" href="#Storage-Layout">17.5 Storage Layout</a></li>
    <li><a name="toc-Layout-of-Source-Language-Data-Types" href="#Type-Layout">17.6 Layout of Source Language Data Types</a></li>
    <li><a name="toc-Register-Usage" href="#Registers">17.7 Register Usage</a>
    <ul class="no-bullet">
      <li><a name="toc-Basic-Characteristics-of-Registers" href="#Register-Basics">17.7.1 Basic Characteristics of Registers</a></li>
      <li><a name="toc-Order-of-Allocation-of-Registers" href="#Allocation-Order">17.7.2 Order of Allocation of Registers</a></li>
      <li><a name="toc-How-Values-Fit-in-Registers" href="#Values-in-Registers">17.7.3 How Values Fit in Registers</a></li>
      <li><a name="toc-Handling-Leaf-Functions" href="#Leaf-Functions">17.7.4 Handling Leaf Functions</a></li>
      <li><a name="toc-Registers-That-Form-a-Stack" href="#Stack-Registers">17.7.5 Registers That Form a Stack</a></li>
    </ul></li>
    <li><a name="toc-Register-Classes-1" href="#Register-Classes">17.8 Register Classes</a></li>
    <li><a name="toc-Obsolete-Macros-for-Defining-Constraints" href="#Old-Constraints">17.9 Obsolete Macros for Defining Constraints</a></li>
    <li><a name="toc-Stack-Layout-and-Calling-Conventions" href="#Stack-and-Calling">17.10 Stack Layout and Calling Conventions</a>
    <ul class="no-bullet">
      <li><a name="toc-Basic-Stack-Layout" href="#Frame-Layout">17.10.1 Basic Stack Layout</a></li>
      <li><a name="toc-Exception-Handling-Support" href="#Exception-Handling">17.10.2 Exception Handling Support</a></li>
      <li><a name="toc-Specifying-How-Stack-Checking-is-Done" href="#Stack-Checking">17.10.3 Specifying How Stack Checking is Done</a></li>
      <li><a name="toc-Registers-That-Address-the-Stack-Frame" href="#Frame-Registers">17.10.4 Registers That Address the Stack Frame</a></li>
      <li><a name="toc-Eliminating-Frame-Pointer-and-Arg-Pointer" href="#Elimination">17.10.5 Eliminating Frame Pointer and Arg Pointer</a></li>
      <li><a name="toc-Passing-Function-Arguments-on-the-Stack" href="#Stack-Arguments">17.10.6 Passing Function Arguments on the Stack</a></li>
      <li><a name="toc-Passing-Arguments-in-Registers" href="#Register-Arguments">17.10.7 Passing Arguments in Registers</a></li>
      <li><a name="toc-How-Scalar-Function-Values-Are-Returned" href="#Scalar-Return">17.10.8 How Scalar Function Values Are Returned</a></li>
      <li><a name="toc-How-Large-Values-Are-Returned" href="#Aggregate-Return">17.10.9 How Large Values Are Returned</a></li>
      <li><a name="toc-Caller_002dSaves-Register-Allocation" href="#Caller-Saves">17.10.10 Caller-Saves Register Allocation</a></li>
      <li><a name="toc-Function-Entry-and-Exit" href="#Function-Entry">17.10.11 Function Entry and Exit</a></li>
      <li><a name="toc-Generating-Code-for-Profiling" href="#Profiling">17.10.12 Generating Code for Profiling</a></li>
      <li><a name="toc-Permitting-tail-calls" href="#Tail-Calls">17.10.13 Permitting tail calls</a></li>
      <li><a name="toc-Stack-smashing-protection" href="#Stack-Smashing-Protection">17.10.14 Stack smashing protection</a></li>
    </ul></li>
    <li><a name="toc-Implementing-the-Varargs-Macros" href="#Varargs">17.11 Implementing the Varargs Macros</a></li>
    <li><a name="toc-Trampolines-for-Nested-Functions" href="#Trampolines">17.12 Trampolines for Nested Functions</a></li>
    <li><a name="toc-Implicit-Calls-to-Library-Routines" href="#Library-Calls">17.13 Implicit Calls to Library Routines</a></li>
    <li><a name="toc-Addressing-Modes-1" href="#Addressing-Modes">17.14 Addressing Modes</a></li>
    <li><a name="toc-Anchored-Addresses-1" href="#Anchored-Addresses">17.15 Anchored Addresses</a></li>
    <li><a name="toc-Condition-Code-Status" href="#Condition-Code">17.16 Condition Code Status</a>
    <ul class="no-bullet">
      <li><a name="toc-Representation-of-condition-codes-using-_0028cc0_0029" href="#CC0-Condition-Codes">17.16.1 Representation of condition codes using <code>(cc0)</code></a></li>
      <li><a name="toc-Representation-of-condition-codes-using-registers" href="#MODE_005fCC-Condition-Codes">17.16.2 Representation of condition codes using registers</a></li>
    </ul></li>
    <li><a name="toc-Describing-Relative-Costs-of-Operations" href="#Costs">17.17 Describing Relative Costs of Operations</a></li>
    <li><a name="toc-Adjusting-the-Instruction-Scheduler" href="#Scheduling">17.18 Adjusting the Instruction Scheduler</a></li>
    <li><a name="toc-Dividing-the-Output-into-Sections-_0028Texts_002c-Data_002c-_2026_0029" href="#Sections">17.19 Dividing the Output into Sections (Texts, Data, &hellip;)</a></li>
    <li><a name="toc-Position-Independent-Code" href="#PIC">17.20 Position Independent Code</a></li>
    <li><a name="toc-Defining-the-Output-Assembler-Language" href="#Assembler-Format">17.21 Defining the Output Assembler Language</a>
    <ul class="no-bullet">
      <li><a name="toc-The-Overall-Framework-of-an-Assembler-File" href="#File-Framework">17.21.1 The Overall Framework of an Assembler File</a></li>
      <li><a name="toc-Output-of-Data" href="#Data-Output">17.21.2 Output of Data</a></li>
      <li><a name="toc-Output-of-Uninitialized-Variables" href="#Uninitialized-Data">17.21.3 Output of Uninitialized Variables</a></li>
      <li><a name="toc-Output-and-Generation-of-Labels" href="#Label-Output">17.21.4 Output and Generation of Labels</a></li>
      <li><a name="toc-How-Initialization-Functions-Are-Handled" href="#Initialization">17.21.5 How Initialization Functions Are Handled</a></li>
      <li><a name="toc-Macros-Controlling-Initialization-Routines" href="#Macros-for-Initialization">17.21.6 Macros Controlling Initialization Routines</a></li>
      <li><a name="toc-Output-of-Assembler-Instructions" href="#Instruction-Output">17.21.7 Output of Assembler Instructions</a></li>
      <li><a name="toc-Output-of-Dispatch-Tables" href="#Dispatch-Tables">17.21.8 Output of Dispatch Tables</a></li>
      <li><a name="toc-Assembler-Commands-for-Exception-Regions" href="#Exception-Region-Output">17.21.9 Assembler Commands for Exception Regions</a></li>
      <li><a name="toc-Assembler-Commands-for-Alignment" href="#Alignment-Output">17.21.10 Assembler Commands for Alignment</a></li>
    </ul></li>
    <li><a name="toc-Controlling-Debugging-Information-Format" href="#Debugging-Info">17.22 Controlling Debugging Information Format</a>
    <ul class="no-bullet">
      <li><a name="toc-Macros-Affecting-All-Debugging-Formats" href="#All-Debuggers">17.22.1 Macros Affecting All Debugging Formats</a></li>
      <li><a name="toc-Specific-Options-for-DBX-Output" href="#DBX-Options">17.22.2 Specific Options for DBX Output</a></li>
      <li><a name="toc-Open_002dEnded-Hooks-for-DBX-Format" href="#DBX-Hooks">17.22.3 Open-Ended Hooks for DBX Format</a></li>
      <li><a name="toc-File-Names-in-DBX-Format" href="#File-Names-and-DBX">17.22.4 File Names in DBX Format</a></li>
      <li><a name="toc-Macros-for-SDB-and-DWARF-Output" href="#SDB-and-DWARF">17.22.5 Macros for SDB and DWARF Output</a></li>
      <li><a name="toc-Macros-for-VMS-Debug-Format" href="#VMS-Debug">17.22.6 Macros for VMS Debug Format</a></li>
    </ul></li>
    <li><a name="toc-Cross-Compilation-and-Floating-Point" href="#Floating-Point">17.23 Cross Compilation and Floating Point</a></li>
    <li><a name="toc-Mode-Switching-Instructions" href="#Mode-Switching">17.24 Mode Switching Instructions</a></li>
    <li><a name="toc-Defining-target_002dspecific-uses-of-_005f_005fattribute_005f_005f" href="#Target-Attributes">17.25 Defining target-specific uses of <code>__attribute__</code></a></li>
    <li><a name="toc-Emulating-TLS" href="#Emulated-TLS">17.26 Emulating TLS</a></li>
    <li><a name="toc-Defining-coprocessor-specifics-for-MIPS-targets_002e" href="#MIPS-Coprocessors">17.27 Defining coprocessor specifics for MIPS targets.</a></li>
    <li><a name="toc-Parameters-for-Precompiled-Header-Validity-Checking" href="#PCH-Target">17.28 Parameters for Precompiled Header Validity Checking</a></li>
    <li><a name="toc-C_002b_002b-ABI-parameters" href="#C_002b_002b-ABI">17.29 C++ ABI parameters</a></li>
    <li><a name="toc-Adding-support-for-named-address-spaces" href="#Named-Address-Spaces">17.30 Adding support for named address spaces</a></li>
    <li><a name="toc-Miscellaneous-Parameters" href="#Misc">17.31 Miscellaneous Parameters</a></li>
  </ul></li>
  <li><a name="toc-Host-Configuration" href="#Host-Config">18 Host Configuration</a>
  <ul class="no-bullet">
    <li><a name="toc-Host-Common-1" href="#Host-Common">18.1 Host Common</a></li>
    <li><a name="toc-Host-Filesystem" href="#Filesystem">18.2 Host Filesystem</a></li>
    <li><a name="toc-Host-Misc-1" href="#Host-Misc">18.3 Host Misc</a></li>
  </ul></li>
  <li><a name="toc-Makefile-Fragments" href="#Fragments">19 Makefile Fragments</a>
  <ul class="no-bullet">
    <li><a name="toc-Target-Makefile-Fragments" href="#Target-Fragment">19.1 Target Makefile Fragments</a></li>
    <li><a name="toc-Host-Makefile-Fragments" href="#Host-Fragment">19.2 Host Makefile Fragments</a></li>
  </ul></li>
  <li><a name="toc-collect2" href="#Collect2">20 <code>collect2</code></a></li>
  <li><a name="toc-Standard-Header-File-Directories" href="#Header-Dirs">21 Standard Header File Directories</a></li>
  <li><a name="toc-Memory-Management-and-Type-Information" href="#Type-Information">22 Memory Management and Type Information</a>
  <ul class="no-bullet">
    <li><a name="toc-The-Inside-of-a-GTY_0028_0028_0029_0029" href="#GTY-Options">22.1 The Inside of a <code>GTY(())</code></a></li>
    <li><a name="toc-Support-for-user_002dprovided-GC-marking-routines" href="#User-GC">22.2 Support for user-provided GC marking routines</a>
    <ul class="no-bullet">
      <li><a name="toc-User_002dprovided-marking-routines-for-template-types" href="#User_002dprovided-marking-routines-for-template-types">22.2.1 User-provided marking routines for template types</a></li>
    </ul></li>
    <li><a name="toc-Marking-Roots-for-the-Garbage-Collector" href="#GGC-Roots">22.3 Marking Roots for the Garbage Collector</a></li>
    <li><a name="toc-Source-Files-Containing-Type-Information" href="#Files">22.4 Source Files Containing Type Information</a></li>
    <li><a name="toc-How-to-invoke-the-garbage-collector" href="#Invoking-the-garbage-collector">22.5 How to invoke the garbage collector</a></li>
    <li><a name="toc-Troubleshooting-the-garbage-collector" href="#Troubleshooting">22.6 Troubleshooting the garbage collector</a></li>
  </ul></li>
  <li><a name="toc-Plugins-1" href="#Plugins">23 Plugins</a>
  <ul class="no-bullet">
    <li><a name="toc-Loading-Plugins" href="#Plugins-loading">23.1 Loading Plugins</a></li>
    <li><a name="toc-Plugin-API-1" href="#Plugin-API">23.2 Plugin API</a>
    <ul class="no-bullet">
      <li><a name="toc-Plugin-license-check" href="#Plugin-license-check">23.2.1 Plugin license check</a></li>
      <li><a name="toc-Plugin-initialization" href="#Plugin-initialization">23.2.2 Plugin initialization</a></li>
      <li><a name="toc-Plugin-callbacks" href="#Plugin-callbacks">23.2.3 Plugin callbacks</a></li>
    </ul></li>
    <li><a name="toc-Interacting-with-the-pass-manager" href="#Plugins-pass">23.3 Interacting with the pass manager</a></li>
    <li><a name="toc-Interacting-with-the-GCC-Garbage-Collector" href="#Plugins-GC">23.4 Interacting with the GCC Garbage Collector</a></li>
    <li><a name="toc-Giving-information-about-a-plugin" href="#Plugins-description">23.5 Giving information about a plugin</a></li>
    <li><a name="toc-Registering-custom-attributes-or-pragmas" href="#Plugins-attr">23.6 Registering custom attributes or pragmas</a></li>
    <li><a name="toc-Recording-information-about-pass-execution" href="#Plugins-recording">23.7 Recording information about pass execution</a></li>
    <li><a name="toc-Controlling-which-passes-are-being-run" href="#Plugins-gate">23.8 Controlling which passes are being run</a></li>
    <li><a name="toc-Keeping-track-of-available-passes" href="#Plugins-tracking">23.9 Keeping track of available passes</a></li>
    <li><a name="toc-Building-GCC-plugins" href="#Plugins-building">23.10 Building GCC plugins</a></li>
  </ul></li>
  <li><a name="toc-Link-Time-Optimization" href="#LTO">24 Link Time Optimization</a>
  <ul class="no-bullet">
    <li><a name="toc-Design-Overview" href="#LTO-Overview">24.1 Design Overview</a>
    <ul class="no-bullet">
      <li><a name="toc-LTO-modes-of-operation" href="#LTO-modes-of-operation">24.1.1 LTO modes of operation</a></li>
    </ul></li>
    <li><a name="toc-LTO-file-sections" href="#LTO-object-file-layout">24.2 LTO file sections</a></li>
    <li><a name="toc-Using-summary-information-in-IPA-passes" href="#IPA">24.3 Using summary information in IPA passes</a>
    <ul class="no-bullet">
      <li><a name="toc-Virtual-clones" href="#Virtual-clones">24.3.1 Virtual clones</a></li>
      <li><a name="toc-IPA-references" href="#IPA-references">24.3.2 IPA references</a></li>
      <li><a name="toc-Jump-functions" href="#Jump-functions">24.3.3 Jump functions</a></li>
    </ul></li>
    <li><a name="toc-Whole-program-assumptions_002c-linker-plugin-and-symbol-visibilities" href="#WHOPR">24.4 Whole program assumptions, linker plugin and symbol visibilities</a></li>
    <li><a name="toc-Internal-flags-controlling-lto1" href="#Internal-flags">24.5 Internal flags controlling <code>lto1</code></a></li>
  </ul></li>
  <li><a name="toc-Funding-Free-Software" href="#Funding">Funding Free Software</a></li>
  <li><a name="toc-The-GNU-Project-and-GNU_002fLinux" href="#GNU-Project">The GNU Project and GNU/Linux</a></li>
  <li><a name="toc-GNU-General-Public-License" href="#Copying">GNU General Public License</a></li>
  <li><a name="toc-GNU-Free-Documentation-License-1" href="#GNU-Free-Documentation-License">GNU Free Documentation License</a>
  <ul class="no-bullet">
    <li><a name="toc-ADDENDUM_003a-How-to-use-this-License-for-your-documents" href="#ADDENDUM_003a-How-to-use-this-License-for-your-documents">ADDENDUM: How to use this License for your documents</a></li>
  </ul></li>
  <li><a name="toc-Contributors-to-GCC" href="#Contributors">Contributors to GCC</a></li>
  <li><a name="toc-Option-Index-1" href="#Option-Index">Option Index</a></li>
  <li><a name="toc-Concept-Index-1" href="#Concept-Index">Concept Index</a></li>
</ul>
</div>


<a name="Top"></a>
<div class="header">
<p>
Next: <a href="#Contributing" accesskey="n" rel="next">Contributing</a>, Up: <a href="DIR.html#Top" accesskey="u" rel="up">(DIR)</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Introduction"></a>
<h1 class="top">Introduction</h1>
<a name="index-introduction"></a>

<p>This manual documents the internals of the GNU compilers, including
how to port them to new targets and some information about how to
write front ends for new languages.  It corresponds to the compilers
(Ubuntu 4.8.4-2ubuntu1~14.04.4)
version 4.8.4.  The use of the GNU compilers is documented in a
separate manual.  See <a href="gcc-4.html#Top">Introduction</a> in <cite>Using the GNU
Compiler Collection (GCC)</cite>.
</p>
<p>This manual is mainly a reference manual rather than a tutorial.  It
discusses how to contribute to GCC (see <a href="#Contributing">Contributing</a>), the
characteristics of the machines supported by GCC as hosts and targets
(see <a href="#Portability">Portability</a>), how GCC relates to the ABIs on such systems
(see <a href="#Interface">Interface</a>), and the characteristics of the languages for
which GCC front ends are written (see <a href="#Languages">Languages</a>).  It then
describes the GCC source tree structure and build system, some of the
interfaces to GCC front ends, and how support for a target system is
implemented in GCC.
</p>
<p>Additional tutorial information is linked to from
<a href="http://gcc.gnu.org/readings.html">http://gcc.gnu.org/readings.html</a>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Contributing" accesskey="1">Contributing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to contribute to testing and developing GCC.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Portability" accesskey="2">Portability</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Goals of GCC&rsquo;s portability features.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Interface" accesskey="3">Interface</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Function-call interface of GCC output.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Libgcc" accesskey="4">Libgcc</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Low-level runtime library used by GCC.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Languages" accesskey="5">Languages</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Languages for which GCC front ends are written.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Source-Tree" accesskey="6">Source Tree</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">GCC source tree structure and build system.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Testsuites" accesskey="7">Testsuites</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">GCC testsuites.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Options" accesskey="8">Options</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Option specification files.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Passes" accesskey="9">Passes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Order of passes, what they do, and what each file is for.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GENERIC">GENERIC</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Language-independent representation generated by Front Ends
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE">GIMPLE</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Tuple representation used by Tree SSA optimizers
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tree-SSA">Tree SSA</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Analysis and optimization of GIMPLE
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#RTL">RTL</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Machine-dependent low-level intermediate representation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Control-Flow">Control Flow</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Maintaining and manipulating the control flow graph.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Loop-Analysis-and-Representation">Loop Analysis and Representation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Analysis and representation of loops
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Machine-Desc">Machine Desc</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to write machine description instruction patterns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Target-Macros">Target Macros</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to write the machine description C macros and functions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Host-Config">Host Config</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Writing the <samp>xm-<var>machine</var>.h</samp> file.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Fragments">Fragments</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Writing the <samp>t-<var>target</var></samp> and <samp>x-<var>host</var></samp> files.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Collect2">Collect2</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How <code>collect2</code> works; how it finds <code>ld</code>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Header-Dirs">Header Dirs</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Understanding the standard header file directories.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Type-Information">Type Information</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">GCC&rsquo;s memory management; generating type information.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins">Plugins</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Extending the compiler with plugins.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#LTO">LTO</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Using Link-Time Optimization.
</td></tr>
<tr><th colspan="3" align="left" valign="top"><pre class="menu-comment">
</pre></th></tr><tr><td align="left" valign="top">&bull; <a href="#Funding">Funding</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to help assure funding for free software.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GNU-Project">GNU Project</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The GNU Project and GNU/Linux.
</td></tr>
<tr><th colspan="3" align="left" valign="top"><pre class="menu-comment">
</pre></th></tr><tr><td align="left" valign="top">&bull; <a href="#Copying">Copying</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">GNU General Public License says
                    how you can copy and share GCC.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GNU-Free-Documentation-License">GNU Free Documentation License</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How you can copy and share this manual.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Contributors">Contributors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">People who have contributed to GCC.
</td></tr>
<tr><th colspan="3" align="left" valign="top"><pre class="menu-comment">
</pre></th></tr><tr><td align="left" valign="top">&bull; <a href="#Option-Index">Option Index</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Index to command line options.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Concept-Index">Concept Index</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Index of concepts and symbol names.
</td></tr>
</table>


<hr>
<a name="Contributing"></a>
<div class="header">
<p>
Next: <a href="#Portability" accesskey="n" rel="next">Portability</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Contributing-to-GCC-Development"></a>
<h2 class="chapter">1 Contributing to GCC Development</h2>

<p>If you would like to help pretest GCC releases to assure they work well,
current development sources are available by SVN (see
<a href="http://gcc.gnu.org/svn.html">http://gcc.gnu.org/svn.html</a>).  Source and binary snapshots are
also available for FTP; see <a href="http://gcc.gnu.org/snapshots.html">http://gcc.gnu.org/snapshots.html</a>.
</p>
<p>If you would like to work on improvements to GCC, please read the
advice at these URLs:
</p>
<div class="smallexample">
<pre class="smallexample"><a href="http://gcc.gnu.org/contribute.html">http://gcc.gnu.org/contribute.html</a>
<a href="http://gcc.gnu.org/contributewhy.html">http://gcc.gnu.org/contributewhy.html</a>
</pre></div>

<p>for information on how to make useful contributions and avoid
duplication of effort.  Suggested projects are listed at
<a href="http://gcc.gnu.org/projects/">http://gcc.gnu.org/projects/</a>.
</p>
<hr>
<a name="Portability"></a>
<div class="header">
<p>
Next: <a href="#Interface" accesskey="n" rel="next">Interface</a>, Previous: <a href="#Contributing" accesskey="p" rel="prev">Contributing</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GCC-and-Portability"></a>
<h2 class="chapter">2 GCC and Portability</h2>
<a name="index-portability"></a>
<a name="index-GCC-and-portability"></a>

<p>GCC itself aims to be portable to any machine where <code>int</code> is at least
a 32-bit type.  It aims to target machines with a flat (non-segmented) byte
addressed data address space (the code address space can be separate).
Target ABIs may have 8, 16, 32 or 64-bit <code>int</code> type.  <code>char</code>
can be wider than 8 bits.
</p>
<p>GCC gets most of the information about the target machine from a machine
description which gives an algebraic formula for each of the machine&rsquo;s
instructions.  This is a very clean way to describe the target.  But when
the compiler needs information that is difficult to express in this
fashion, ad-hoc parameters have been defined for machine descriptions.
The purpose of portability is to reduce the total work needed on the
compiler; it was not of interest for its own sake.
</p>
<a name="index-endianness"></a>
<a name="index-autoincrement-addressing_002c-availability"></a>
<a name="index-abort"></a>
<p>GCC does not contain machine dependent code, but it does contain code
that depends on machine parameters such as endianness (whether the most
significant byte has the highest or lowest address of the bytes in a word)
and the availability of autoincrement addressing.  In the RTL-generation
pass, it is often necessary to have multiple strategies for generating code
for a particular kind of syntax tree, strategies that are usable for different
combinations of parameters.  Often, not all possible cases have been
addressed, but only the common ones or only the ones that have been
encountered.  As a result, a new target may require additional
strategies.  You will know
if this happens because the compiler will call <code>abort</code>.  Fortunately,
the new strategies can be added in a machine-independent fashion, and will
affect only the target machines that need them.
</p>
<hr>
<a name="Interface"></a>
<div class="header">
<p>
Next: <a href="#Libgcc" accesskey="n" rel="next">Libgcc</a>, Previous: <a href="#Portability" accesskey="p" rel="prev">Portability</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Interfacing-to-GCC-Output"></a>
<h2 class="chapter">3 Interfacing to GCC Output</h2>
<a name="index-interfacing-to-GCC-output"></a>
<a name="index-run_002dtime-conventions"></a>
<a name="index-function-call-conventions"></a>
<a name="index-conventions_002c-run_002dtime"></a>

<p>GCC is normally configured to use the same function calling convention
normally in use on the target system.  This is done with the
machine-description macros described (see <a href="#Target-Macros">Target Macros</a>).
</p>
<a name="index-unions_002c-returning"></a>
<a name="index-structures_002c-returning"></a>
<a name="index-returning-structures-and-unions"></a>
<p>However, returning of structure and union values is done differently on
some target machines.  As a result, functions compiled with PCC
returning such types cannot be called from code compiled with GCC,
and vice versa.  This does not cause trouble often because few Unix
library routines return structures or unions.
</p>
<p>GCC code returns structures and unions that are 1, 2, 4 or 8 bytes
long in the same registers used for <code>int</code> or <code>double</code> return
values.  (GCC typically allocates variables of such types in
registers also.)  Structures and unions of other sizes are returned by
storing them into an address passed by the caller (usually in a
register).  The target hook <code>TARGET_STRUCT_VALUE_RTX</code>
tells GCC where to pass this address.
</p>
<p>By contrast, PCC on most target machines returns structures and unions
of any size by copying the data into an area of static storage, and then
returning the address of that storage as if it were a pointer value.
The caller must copy the data from that memory area to the place where
the value is wanted.  This is slower than the method used by GCC, and
fails to be reentrant.
</p>
<p>On some target machines, such as RISC machines and the 80386, the
standard system convention is to pass to the subroutine the address of
where to return the value.  On these machines, GCC has been
configured to be compatible with the standard compiler, when this method
is used.  It may not be compatible for structures of 1, 2, 4 or 8 bytes.
</p>
<a name="index-argument-passing"></a>
<a name="index-passing-arguments"></a>
<p>GCC uses the system&rsquo;s standard convention for passing arguments.  On
some machines, the first few arguments are passed in registers; in
others, all are passed on the stack.  It would be possible to use
registers for argument passing on any machine, and this would probably
result in a significant speedup.  But the result would be complete
incompatibility with code that follows the standard convention.  So this
change is practical only if you are switching to GCC as the sole C
compiler for the system.  We may implement register argument passing on
certain machines once we have a complete GNU system so that we can
compile the libraries with GCC.
</p>
<p>On some machines (particularly the SPARC), certain types of arguments
are passed &ldquo;by invisible reference&rdquo;.  This means that the value is
stored in memory, and the address of the memory location is passed to
the subroutine.
</p>
<a name="index-longjmp-and-automatic-variables"></a>
<p>If you use <code>longjmp</code>, beware of automatic variables.  ISO C says that
automatic variables that are not declared <code>volatile</code> have undefined
values after a <code>longjmp</code>.  And this is all GCC promises to do,
because it is very difficult to restore register variables correctly, and
one of GCC&rsquo;s features is that it can put variables in registers without
your asking it to.
</p>
<hr>
<a name="Libgcc"></a>
<div class="header">
<p>
Next: <a href="#Languages" accesskey="n" rel="next">Languages</a>, Previous: <a href="#Interface" accesskey="p" rel="prev">Interface</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-GCC-low_002dlevel-runtime-library"></a>
<h2 class="chapter">4 The GCC low-level runtime library</h2>

<p>GCC provides a low-level runtime library, <samp>libgcc.a</samp> or
<samp>libgcc_s.so.1</samp> on some platforms.  GCC generates calls to
routines in this library automatically, whenever it needs to perform
some operation that is too complicated to emit inline code for.
</p>
<p>Most of the routines in <code>libgcc</code> handle arithmetic operations
that the target processor cannot perform directly.  This includes
integer multiply and divide on some machines, and all floating-point
and fixed-point operations on other machines.  <code>libgcc</code> also includes
routines for exception handling, and a handful of miscellaneous operations.
</p>
<p>Some of these routines can be defined in mostly machine-independent C.
Others must be hand-written in assembly language for each processor
that needs them.
</p>
<p>GCC will also generate calls to C library routines, such as
<code>memcpy</code> and <code>memset</code>, in some cases.  The set of routines
that GCC may possibly use is documented in <a href="gcc-4.html#Other-Builtins">Other
Builtins</a> in <cite>Using the GNU Compiler Collection (GCC)</cite>.
</p>
<p>These routines take arguments and return values of a specific machine
mode, not a specific C type.  See <a href="#Machine-Modes">Machine Modes</a>, for an explanation
of this concept.  For illustrative purposes, in this chapter the
floating point type <code>float</code> is assumed to correspond to <code>SFmode</code>;
<code>double</code> to <code>DFmode</code>; and <code>long&nbsp;double<!-- /@w --></code> to both
<code>TFmode</code> and <code>XFmode</code>.  Similarly, the integer types <code>int</code>
and <code>unsigned&nbsp;int<!-- /@w --></code> correspond to <code>SImode</code>; <code>long</code> and
<code>unsigned&nbsp;long<!-- /@w --></code> to <code>DImode</code>; and <code>long&nbsp;long<!-- /@w --></code> and
<code>unsigned&nbsp;long&nbsp;long<!-- /@w --></code> to <code>TImode</code>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Integer-library-routines" accesskey="1">Integer library routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Soft-float-library-routines" accesskey="2">Soft float library routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Decimal-float-library-routines" accesskey="3">Decimal float library routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Fixed_002dpoint-fractional-library-routines" accesskey="4">Fixed-point fractional library routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Exception-handling-routines" accesskey="5">Exception handling routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Miscellaneous-routines" accesskey="6">Miscellaneous routines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Integer-library-routines"></a>
<div class="header">
<p>
Next: <a href="#Soft-float-library-routines" accesskey="n" rel="next">Soft float library routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Routines-for-integer-arithmetic"></a>
<h3 class="section">4.1 Routines for integer arithmetic</h3>

<p>The integer arithmetic routines are used on platforms that don&rsquo;t provide
hardware support for arithmetic operations on some modes.
</p>
<a name="Arithmetic-functions"></a>
<h4 class="subsection">4.1.1 Arithmetic functions</h4>

<dl>
<dt><a name="index-_005f_005fashlsi3"></a>Runtime Function: <em>int</em> <strong>__ashlsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashldi3"></a>Runtime Function: <em>long</em> <strong>__ashldi3</strong> <em>(long <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlti3"></a>Runtime Function: <em>long long</em> <strong>__ashlti3</strong> <em>(long long <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of shifting <var>a</var> left by <var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fashrsi3"></a>Runtime Function: <em>int</em> <strong>__ashrsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrdi3"></a>Runtime Function: <em>long</em> <strong>__ashrdi3</strong> <em>(long <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrti3"></a>Runtime Function: <em>long long</em> <strong>__ashrti3</strong> <em>(long long <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of arithmetically shifting <var>a</var> right
by <var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdivsi3"></a>Runtime Function: <em>int</em> <strong>__divsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivdi3"></a>Runtime Function: <em>long</em> <strong>__divdi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivti3"></a>Runtime Function: <em>long long</em> <strong>__divti3</strong> <em>(long long <var>a</var>, long long <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the signed division of <var>a</var> and
<var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005flshrsi3"></a>Runtime Function: <em>int</em> <strong>__lshrsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshrdi3"></a>Runtime Function: <em>long</em> <strong>__lshrdi3</strong> <em>(long <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshrti3"></a>Runtime Function: <em>long long</em> <strong>__lshrti3</strong> <em>(long long <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of logically shifting <var>a</var> right by
<var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmodsi3"></a>Runtime Function: <em>int</em> <strong>__modsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmoddi3"></a>Runtime Function: <em>long</em> <strong>__moddi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmodti3"></a>Runtime Function: <em>long long</em> <strong>__modti3</strong> <em>(long long <var>a</var>, long long <var>b</var>)</em></dt>
<dd><p>These functions return the remainder of the signed division of <var>a</var>
and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmulsi3"></a>Runtime Function: <em>int</em> <strong>__mulsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuldi3"></a>Runtime Function: <em>long</em> <strong>__muldi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulti3"></a>Runtime Function: <em>long long</em> <strong>__multi3</strong> <em>(long long <var>a</var>, long long <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fnegdi2"></a>Runtime Function: <em>long</em> <strong>__negdi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegti2"></a>Runtime Function: <em>long long</em> <strong>__negti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fudivsi3"></a>Runtime Function: <em>unsigned int</em> <strong>__udivsi3</strong> <em>(unsigned int <var>a</var>, unsigned int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivdi3"></a>Runtime Function: <em>unsigned long</em> <strong>__udivdi3</strong> <em>(unsigned long <var>a</var>, unsigned long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivti3"></a>Runtime Function: <em>unsigned long long</em> <strong>__udivti3</strong> <em>(unsigned long long <var>a</var>, unsigned long long <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the unsigned division of <var>a</var>
and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fudivmoddi4"></a>Runtime Function: <em>unsigned long</em> <strong>__udivmoddi4</strong> <em>(unsigned long <var>a</var>, unsigned long <var>b</var>, unsigned long *<var>c</var>)</em></dt>
<dt><a name="index-_005f_005fudivmodti4"></a>Runtime Function: <em>unsigned long long</em> <strong>__udivmodti4</strong> <em>(unsigned long long <var>a</var>, unsigned long long <var>b</var>, unsigned long long *<var>c</var>)</em></dt>
<dd><p>These functions calculate both the quotient and remainder of the unsigned
division of <var>a</var> and <var>b</var>.  The return value is the quotient, and
the remainder is placed in variable pointed to by <var>c</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fumodsi3"></a>Runtime Function: <em>unsigned int</em> <strong>__umodsi3</strong> <em>(unsigned int <var>a</var>, unsigned int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fumoddi3"></a>Runtime Function: <em>unsigned long</em> <strong>__umoddi3</strong> <em>(unsigned long <var>a</var>, unsigned long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fumodti3"></a>Runtime Function: <em>unsigned long long</em> <strong>__umodti3</strong> <em>(unsigned long long <var>a</var>, unsigned long long <var>b</var>)</em></dt>
<dd><p>These functions return the remainder of the unsigned division of <var>a</var>
and <var>b</var>.
</p></dd></dl>

<a name="Comparison-functions"></a>
<h4 class="subsection">4.1.2 Comparison functions</h4>

<p>The following functions implement integral comparisons.  These functions
implement a low-level compare, upon which the higher level comparison
operators (such as less than and greater than or equal to) can be
constructed.  The returned values lie in the range zero to two, to allow
the high-level operators to be implemented by testing the returned
result using either signed or unsigned comparison.
</p>
<dl>
<dt><a name="index-_005f_005fcmpdi2"></a>Runtime Function: <em>int</em> <strong>__cmpdi2</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpti2"></a>Runtime Function: <em>int</em> <strong>__cmpti2</strong> <em>(long long <var>a</var>, long long <var>b</var>)</em></dt>
<dd><p>These functions perform a signed comparison of <var>a</var> and <var>b</var>.  If
<var>a</var> is less than <var>b</var>, they return 0; if <var>a</var> is greater than
<var>b</var>, they return 2; and if <var>a</var> and <var>b</var> are equal they return 1.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fucmpdi2"></a>Runtime Function: <em>int</em> <strong>__ucmpdi2</strong> <em>(unsigned long <var>a</var>, unsigned long <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fucmpti2"></a>Runtime Function: <em>int</em> <strong>__ucmpti2</strong> <em>(unsigned long long <var>a</var>, unsigned long long <var>b</var>)</em></dt>
<dd><p>These functions perform an unsigned comparison of <var>a</var> and <var>b</var>.
If <var>a</var> is less than <var>b</var>, they return 0; if <var>a</var> is greater than
<var>b</var>, they return 2; and if <var>a</var> and <var>b</var> are equal they return 1.
</p></dd></dl>

<a name="Trapping-arithmetic-functions"></a>
<h4 class="subsection">4.1.3 Trapping arithmetic functions</h4>

<p>The following functions implement trapping arithmetic.  These functions
call the libc function <code>abort</code> upon signed arithmetic overflow.
</p>
<dl>
<dt><a name="index-_005f_005fabsvsi2"></a>Runtime Function: <em>int</em> <strong>__absvsi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fabsvdi2"></a>Runtime Function: <em>long</em> <strong>__absvdi2</strong> <em>(long <var>a</var>)</em></dt>
<dd><p>These functions return the absolute value of <var>a</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005faddvsi3"></a>Runtime Function: <em>int</em> <strong>__addvsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddvdi3"></a>Runtime Function: <em>long</em> <strong>__addvdi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var>; that is
<code><var>a</var> + <var>b</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmulvsi3"></a>Runtime Function: <em>int</em> <strong>__mulvsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulvdi3"></a>Runtime Function: <em>long</em> <strong>__mulvdi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dd><p>The functions return the product of <var>a</var> and <var>b</var>; that is
<code><var>a</var> * <var>b</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fnegvsi2"></a>Runtime Function: <em>int</em> <strong>__negvsi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegvdi2"></a>Runtime Function: <em>long</em> <strong>__negvdi2</strong> <em>(long <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var>; that is <code>-<var>a</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsubvsi3"></a>Runtime Function: <em>int</em> <strong>__subvsi3</strong> <em>(int <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubvdi3"></a>Runtime Function: <em>long</em> <strong>__subvdi3</strong> <em>(long <var>a</var>, long <var>b</var>)</em></dt>
<dd><p>These functions return the difference between <var>b</var> and <var>a</var>;
that is <code><var>a</var> - <var>b</var></code>.
</p></dd></dl>

<a name="Bit-operations"></a>
<h4 class="subsection">4.1.4 Bit operations</h4>

<dl>
<dt><a name="index-_005f_005fclzsi2"></a>Runtime Function: <em>int</em> <strong>__clzsi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fclzdi2"></a>Runtime Function: <em>int</em> <strong>__clzdi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fclzti2"></a>Runtime Function: <em>int</em> <strong>__clzti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the number of leading 0-bits in <var>a</var>, starting
at the most significant bit position.  If <var>a</var> is zero, the result is
undefined.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fctzsi2"></a>Runtime Function: <em>int</em> <strong>__ctzsi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fctzdi2"></a>Runtime Function: <em>int</em> <strong>__ctzdi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fctzti2"></a>Runtime Function: <em>int</em> <strong>__ctzti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the number of trailing 0-bits in <var>a</var>, starting
at the least significant bit position.  If <var>a</var> is zero, the result is
undefined.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fffsdi2"></a>Runtime Function: <em>int</em> <strong>__ffsdi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fffsti2"></a>Runtime Function: <em>int</em> <strong>__ffsti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the index of the least significant 1-bit in <var>a</var>,
or the value zero if <var>a</var> is zero.  The least significant bit is index
one.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fparitysi2"></a>Runtime Function: <em>int</em> <strong>__paritysi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fparitydi2"></a>Runtime Function: <em>int</em> <strong>__paritydi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fparityti2"></a>Runtime Function: <em>int</em> <strong>__parityti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the value zero if the number of bits set in
<var>a</var> is even, and the value one otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fpopcountsi2"></a>Runtime Function: <em>int</em> <strong>__popcountsi2</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fpopcountdi2"></a>Runtime Function: <em>int</em> <strong>__popcountdi2</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fpopcountti2"></a>Runtime Function: <em>int</em> <strong>__popcountti2</strong> <em>(long long <var>a</var>)</em></dt>
<dd><p>These functions return the number of bits set in <var>a</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fbswapsi2"></a>Runtime Function: <em>int32_t</em> <strong>__bswapsi2</strong> <em>(int32_t <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbswapdi2"></a>Runtime Function: <em>int64_t</em> <strong>__bswapdi2</strong> <em>(int64_t <var>a</var>)</em></dt>
<dd><p>These functions return the <var>a</var> byteswapped.
</p></dd></dl>

<hr>
<a name="Soft-float-library-routines"></a>
<div class="header">
<p>
Next: <a href="#Decimal-float-library-routines" accesskey="n" rel="next">Decimal float library routines</a>, Previous: <a href="#Integer-library-routines" accesskey="p" rel="prev">Integer library routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Routines-for-floating-point-emulation"></a>
<h3 class="section">4.2 Routines for floating point emulation</h3>
<a name="index-soft-float-library"></a>
<a name="index-arithmetic-library"></a>
<a name="index-math-library"></a>
<a name="index-msoft_002dfloat"></a>

<p>The software floating point library is used on machines which do not
have hardware support for floating point.  It is also used whenever
<samp>-msoft-float</samp> is used to disable generation of floating point
instructions.  (Not all targets support this switch.)
</p>
<p>For compatibility with other compilers, the floating point emulation
routines can be renamed with the <code>DECLARE_LIBRARY_RENAMES</code> macro
(see <a href="#Library-Calls">Library Calls</a>).  In this section, the default names are used.
</p>
<p>Presently the library does not support <code>XFmode</code>, which is used
for <code>long double</code> on some architectures.
</p>
<a name="Arithmetic-functions-1"></a>
<h4 class="subsection">4.2.1 Arithmetic functions</h4>

<dl>
<dt><a name="index-_005f_005faddsf3"></a>Runtime Function: <em>float</em> <strong>__addsf3</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadddf3"></a>Runtime Function: <em>double</em> <strong>__adddf3</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddtf3"></a>Runtime Function: <em>long double</em> <strong>__addtf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddxf3"></a>Runtime Function: <em>long double</em> <strong>__addxf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsubsf3"></a>Runtime Function: <em>float</em> <strong>__subsf3</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubdf3"></a>Runtime Function: <em>double</em> <strong>__subdf3</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubtf3"></a>Runtime Function: <em>long double</em> <strong>__subtf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubxf3"></a>Runtime Function: <em>long double</em> <strong>__subxf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return the difference between <var>b</var> and <var>a</var>;
that is, <em><var>a</var> - <var>b</var></em><!-- /@w -->.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmulsf3"></a>Runtime Function: <em>float</em> <strong>__mulsf3</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuldf3"></a>Runtime Function: <em>double</em> <strong>__muldf3</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmultf3"></a>Runtime Function: <em>long double</em> <strong>__multf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulxf3"></a>Runtime Function: <em>long double</em> <strong>__mulxf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdivsf3"></a>Runtime Function: <em>float</em> <strong>__divsf3</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivdf3"></a>Runtime Function: <em>double</em> <strong>__divdf3</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivtf3"></a>Runtime Function: <em>long double</em> <strong>__divtf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivxf3"></a>Runtime Function: <em>long double</em> <strong>__divxf3</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of <var>a</var> and <var>b</var>; that is,
<em><var>a</var> / <var>b</var></em><!-- /@w -->.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fnegsf2"></a>Runtime Function: <em>float</em> <strong>__negsf2</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegdf2"></a>Runtime Function: <em>double</em> <strong>__negdf2</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegtf2"></a>Runtime Function: <em>long double</em> <strong>__negtf2</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegxf2"></a>Runtime Function: <em>long double</em> <strong>__negxf2</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var>.  They simply flip the
sign bit, so they can produce negative zero and negative NaN.
</p></dd></dl>

<a name="Conversion-functions"></a>
<h4 class="subsection">4.2.2 Conversion functions</h4>

<dl>
<dt><a name="index-_005f_005fextendsfdf2"></a>Runtime Function: <em>double</em> <strong>__extendsfdf2</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fextendsftf2"></a>Runtime Function: <em>long double</em> <strong>__extendsftf2</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fextendsfxf2"></a>Runtime Function: <em>long double</em> <strong>__extendsfxf2</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fextenddftf2"></a>Runtime Function: <em>long double</em> <strong>__extenddftf2</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fextenddfxf2"></a>Runtime Function: <em>long double</em> <strong>__extenddfxf2</strong> <em>(double <var>a</var>)</em></dt>
<dd><p>These functions extend <var>a</var> to the wider mode of their return
type.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ftruncxfdf2"></a>Runtime Function: <em>double</em> <strong>__truncxfdf2</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ftrunctfdf2"></a>Runtime Function: <em>double</em> <strong>__trunctfdf2</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ftruncxfsf2"></a>Runtime Function: <em>float</em> <strong>__truncxfsf2</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ftrunctfsf2"></a>Runtime Function: <em>float</em> <strong>__trunctfsf2</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ftruncdfsf2"></a>Runtime Function: <em>float</em> <strong>__truncdfsf2</strong> <em>(double <var>a</var>)</em></dt>
<dd><p>These functions truncate <var>a</var> to the narrower mode of their return
type, rounding toward zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixsfsi"></a>Runtime Function: <em>int</em> <strong>__fixsfsi</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixdfsi"></a>Runtime Function: <em>int</em> <strong>__fixdfsi</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixtfsi"></a>Runtime Function: <em>int</em> <strong>__fixtfsi</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixxfsi"></a>Runtime Function: <em>int</em> <strong>__fixxfsi</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to a signed integer, rounding toward zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixsfdi"></a>Runtime Function: <em>long</em> <strong>__fixsfdi</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixdfdi"></a>Runtime Function: <em>long</em> <strong>__fixdfdi</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixtfdi"></a>Runtime Function: <em>long</em> <strong>__fixtfdi</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixxfdi"></a>Runtime Function: <em>long</em> <strong>__fixxfdi</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to a signed long, rounding toward zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixsfti"></a>Runtime Function: <em>long long</em> <strong>__fixsfti</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixdfti"></a>Runtime Function: <em>long long</em> <strong>__fixdfti</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixtfti"></a>Runtime Function: <em>long long</em> <strong>__fixtfti</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixxfti"></a>Runtime Function: <em>long long</em> <strong>__fixxfti</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to a signed long long, rounding toward zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixunssfsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fixunssfsi</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsdfsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fixunsdfsi</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunstfsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fixunstfsi</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsxfsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fixunsxfsi</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to an unsigned integer, rounding
toward zero.  Negative values all become zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixunssfdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fixunssfdi</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsdfdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fixunsdfdi</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunstfdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fixunstfdi</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsxfdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fixunsxfdi</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to an unsigned long, rounding
toward zero.  Negative values all become zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffixunssfti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fixunssfti</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsdfti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fixunsdfti</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunstfti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fixunstfti</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffixunsxfti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fixunsxfti</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to an unsigned long long, rounding
toward zero.  Negative values all become zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloatsisf"></a>Runtime Function: <em>float</em> <strong>__floatsisf</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatsidf"></a>Runtime Function: <em>double</em> <strong>__floatsidf</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatsitf"></a>Runtime Function: <em>long double</em> <strong>__floatsitf</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatsixf"></a>Runtime Function: <em>long double</em> <strong>__floatsixf</strong> <em>(int <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, a signed integer, to floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloatdisf"></a>Runtime Function: <em>float</em> <strong>__floatdisf</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatdidf"></a>Runtime Function: <em>double</em> <strong>__floatdidf</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatditf"></a>Runtime Function: <em>long double</em> <strong>__floatditf</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatdixf"></a>Runtime Function: <em>long double</em> <strong>__floatdixf</strong> <em>(long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, a signed long, to floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloattisf"></a>Runtime Function: <em>float</em> <strong>__floattisf</strong> <em>(long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloattidf"></a>Runtime Function: <em>double</em> <strong>__floattidf</strong> <em>(long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloattitf"></a>Runtime Function: <em>long double</em> <strong>__floattitf</strong> <em>(long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloattixf"></a>Runtime Function: <em>long double</em> <strong>__floattixf</strong> <em>(long long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, a signed long long, to floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloatunsisf"></a>Runtime Function: <em>float</em> <strong>__floatunsisf</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatunsidf"></a>Runtime Function: <em>double</em> <strong>__floatunsidf</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatunsitf"></a>Runtime Function: <em>long double</em> <strong>__floatunsitf</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatunsixf"></a>Runtime Function: <em>long double</em> <strong>__floatunsixf</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, an unsigned integer, to floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloatundisf"></a>Runtime Function: <em>float</em> <strong>__floatundisf</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatundidf"></a>Runtime Function: <em>double</em> <strong>__floatundidf</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatunditf"></a>Runtime Function: <em>long double</em> <strong>__floatunditf</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatundixf"></a>Runtime Function: <em>long double</em> <strong>__floatundixf</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, an unsigned long, to floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffloatuntisf"></a>Runtime Function: <em>float</em> <strong>__floatuntisf</strong> <em>(unsigned long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatuntidf"></a>Runtime Function: <em>double</em> <strong>__floatuntidf</strong> <em>(unsigned long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatuntitf"></a>Runtime Function: <em>long double</em> <strong>__floatuntitf</strong> <em>(unsigned long long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005ffloatuntixf"></a>Runtime Function: <em>long double</em> <strong>__floatuntixf</strong> <em>(unsigned long long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, an unsigned long long, to floating point.
</p></dd></dl>

<a name="Comparison-functions-1"></a>
<h4 class="subsection">4.2.3 Comparison functions</h4>

<p>There are two sets of basic comparison functions.
</p>
<dl>
<dt><a name="index-_005f_005fcmpsf2"></a>Runtime Function: <em>int</em> <strong>__cmpsf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpdf2"></a>Runtime Function: <em>int</em> <strong>__cmpdf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmptf2"></a>Runtime Function: <em>int</em> <strong>__cmptf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions calculate <em>a &lt;=&gt; b</em>.  That is, if <var>a</var> is less
than <var>b</var>, they return -1; if <var>a</var> is greater than <var>b</var>, they
return 1; and if <var>a</var> and <var>b</var> are equal they return 0.  If
either argument is NaN they return 1, but you should not rely on this;
if NaN is a possibility, use one of the higher-level comparison
functions.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005funordsf2"></a>Runtime Function: <em>int</em> <strong>__unordsf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005funorddf2"></a>Runtime Function: <em>int</em> <strong>__unorddf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005funordtf2"></a>Runtime Function: <em>int</em> <strong>__unordtf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a nonzero value if either argument is NaN, otherwise 0.
</p></dd></dl>

<p>There is also a complete group of higher level functions which
correspond directly to comparison operators.  They implement the ISO C
semantics for floating-point comparisons, taking NaN into account.
Pay careful attention to the return values defined for each set.
Under the hood, all of these routines are implemented as
</p>
<div class="smallexample">
<pre class="smallexample">  if (__unord<var>X</var>f2 (a, b))
    return <var>E</var>;
  return __cmp<var>X</var>f2 (a, b);
</pre></div>

<p>where <var>E</var> is a constant chosen to give the proper behavior for
NaN.  Thus, the meaning of the return value is different for each set.
Do not rely on this implementation; only the semantics documented
below are guaranteed.
</p>
<dl>
<dt><a name="index-_005f_005feqsf2"></a>Runtime Function: <em>int</em> <strong>__eqsf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005feqdf2"></a>Runtime Function: <em>int</em> <strong>__eqdf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005feqtf2"></a>Runtime Function: <em>int</em> <strong>__eqtf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return zero if neither argument is NaN, and <var>a</var> and
<var>b</var> are equal.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fnesf2"></a>Runtime Function: <em>int</em> <strong>__nesf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fnedf2"></a>Runtime Function: <em>int</em> <strong>__nedf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fnetf2"></a>Runtime Function: <em>int</em> <strong>__netf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a nonzero value if either argument is NaN, or
if <var>a</var> and <var>b</var> are unequal.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fgesf2"></a>Runtime Function: <em>int</em> <strong>__gesf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fgedf2"></a>Runtime Function: <em>int</em> <strong>__gedf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fgetf2"></a>Runtime Function: <em>int</em> <strong>__getf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a value greater than or equal to zero if
neither argument is NaN, and <var>a</var> is greater than or equal to
<var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fltsf2"></a>Runtime Function: <em>int</em> <strong>__ltsf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fltdf2"></a>Runtime Function: <em>int</em> <strong>__ltdf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flttf2"></a>Runtime Function: <em>int</em> <strong>__lttf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a value less than zero if neither argument is
NaN, and <var>a</var> is strictly less than <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005flesf2"></a>Runtime Function: <em>int</em> <strong>__lesf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fledf2"></a>Runtime Function: <em>int</em> <strong>__ledf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fletf2"></a>Runtime Function: <em>int</em> <strong>__letf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a value less than or equal to zero if neither
argument is NaN, and <var>a</var> is less than or equal to <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fgtsf2"></a>Runtime Function: <em>int</em> <strong>__gtsf2</strong> <em>(float <var>a</var>, float <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fgtdf2"></a>Runtime Function: <em>int</em> <strong>__gtdf2</strong> <em>(double <var>a</var>, double <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fgttf2"></a>Runtime Function: <em>int</em> <strong>__gttf2</strong> <em>(long double <var>a</var>, long double <var>b</var>)</em></dt>
<dd><p>These functions return a value greater than zero if neither argument
is NaN, and <var>a</var> is strictly greater than <var>b</var>.
</p></dd></dl>

<a name="Other-floating_002dpoint-functions"></a>
<h4 class="subsection">4.2.4 Other floating-point functions</h4>

<dl>
<dt><a name="index-_005f_005fpowisf2"></a>Runtime Function: <em>float</em> <strong>__powisf2</strong> <em>(float <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fpowidf2"></a>Runtime Function: <em>double</em> <strong>__powidf2</strong> <em>(double <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fpowitf2"></a>Runtime Function: <em>long double</em> <strong>__powitf2</strong> <em>(long double <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fpowixf2"></a>Runtime Function: <em>long double</em> <strong>__powixf2</strong> <em>(long double <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions convert raise <var>a</var> to the power <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmulsc3"></a>Runtime Function: <em>complex float</em> <strong>__mulsc3</strong> <em>(float <var>a</var>, float <var>b</var>, float <var>c</var>, float <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fmuldc3"></a>Runtime Function: <em>complex double</em> <strong>__muldc3</strong> <em>(double <var>a</var>, double <var>b</var>, double <var>c</var>, double <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fmultc3"></a>Runtime Function: <em>complex long double</em> <strong>__multc3</strong> <em>(long double <var>a</var>, long double <var>b</var>, long double <var>c</var>, long double <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fmulxc3"></a>Runtime Function: <em>complex long double</em> <strong>__mulxc3</strong> <em>(long double <var>a</var>, long double <var>b</var>, long double <var>c</var>, long double <var>d</var>)</em></dt>
<dd><p>These functions return the product of <em><var>a</var> + i<var>b</var></em> and
<em><var>c</var> + i<var>d</var></em>, following the rules of C99 Annex G.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdivsc3"></a>Runtime Function: <em>complex float</em> <strong>__divsc3</strong> <em>(float <var>a</var>, float <var>b</var>, float <var>c</var>, float <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fdivdc3"></a>Runtime Function: <em>complex double</em> <strong>__divdc3</strong> <em>(double <var>a</var>, double <var>b</var>, double <var>c</var>, double <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fdivtc3"></a>Runtime Function: <em>complex long double</em> <strong>__divtc3</strong> <em>(long double <var>a</var>, long double <var>b</var>, long double <var>c</var>, long double <var>d</var>)</em></dt>
<dt><a name="index-_005f_005fdivxc3"></a>Runtime Function: <em>complex long double</em> <strong>__divxc3</strong> <em>(long double <var>a</var>, long double <var>b</var>, long double <var>c</var>, long double <var>d</var>)</em></dt>
<dd><p>These functions return the quotient of <em><var>a</var> + i<var>b</var></em> and
<em><var>c</var> + i<var>d</var></em> (i.e., <em>(<var>a</var> + i<var>b</var>) / (<var>c</var>
+ i<var>d</var>)</em>), following the rules of C99 Annex G.
</p></dd></dl>

<hr>
<a name="Decimal-float-library-routines"></a>
<div class="header">
<p>
Next: <a href="#Fixed_002dpoint-fractional-library-routines" accesskey="n" rel="next">Fixed-point fractional library routines</a>, Previous: <a href="#Soft-float-library-routines" accesskey="p" rel="prev">Soft float library routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Routines-for-decimal-floating-point-emulation"></a>
<h3 class="section">4.3 Routines for decimal floating point emulation</h3>
<a name="index-decimal-float-library"></a>
<a name="index-IEEE-754_002d2008"></a>

<p>The software decimal floating point library implements IEEE 754-2008
decimal floating point arithmetic and is only activated on selected
targets.
</p>
<p>The software decimal floating point library supports either DPD
(Densely Packed Decimal) or BID (Binary Integer Decimal) encoding
as selected at configure time.
</p>

<a name="Arithmetic-functions-2"></a>
<h4 class="subsection">4.3.1 Arithmetic functions</h4>

<dl>
<dt><a name="index-_005f_005fdpd_005faddsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_addsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005faddsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_addsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fadddd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_adddd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fadddd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_adddd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005faddtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_addtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005faddtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_addtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fsubsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_subsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fsubsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_subsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fsubdd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_subdd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fsubdd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_subdd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fsubtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_subtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fsubtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_subtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return the difference between <var>b</var> and <var>a</var>;
that is, <em><var>a</var> - <var>b</var></em><!-- /@w -->.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fmulsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_mulsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fmulsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_mulsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fmuldd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_muldd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fmuldd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_muldd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fmultd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_multd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fmultd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_multd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fdivsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_divsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fdivsd3"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_divsd3</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fdivdd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_divdd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fdivdd3"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_divdd3</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fdivtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_divtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fdivtd3"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_divtd3</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of <var>a</var> and <var>b</var>; that is,
<em><var>a</var> / <var>b</var></em><!-- /@w -->.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fnegsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_negsd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnegsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_negsd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fnegdd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_negdd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnegdd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_negdd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fnegtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_negtd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnegtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_negtd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var>.  They simply flip the
sign bit, so they can produce negative zero and negative NaN.
</p></dd></dl>

<a name="Conversion-functions-1"></a>
<h4 class="subsection">4.3.2 Conversion functions</h4>

<dl>
<dt><a name="index-_005f_005fdpd_005fextendsddd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_extendsddd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsddd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_extendsddd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendsdtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extendsdtd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsdtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extendsdtd2</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendddtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extendddtd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendddtd2"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extendddtd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncddsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_truncddsd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncddsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_truncddsd2</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctdsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_trunctdsd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctdsd2"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_trunctdsd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctddd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_trunctddd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctddd2"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_trunctddd2</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert the value <var>a</var> from one decimal floating type
to another.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fextendsfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_extendsfdd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_extendsfdd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendsftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extendsftd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extendsftd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextenddftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extenddftd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextenddftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extenddftd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendxftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extendxftd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendxftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extendxftd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncdfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_truncdfsd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncdfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_truncdfsd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncxfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_truncxfsd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncxfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_truncxfsd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_trunctfsd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_trunctfsd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncxfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_truncxfdd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncxfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_truncxfdd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_trunctfdd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_trunctfdd</strong> <em>(long double <var>a</var>)</em></dt>
<dd><p>These functions convert the value of <var>a</var> from a binary floating type
to a decimal floating type of a different size.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ftruncddsf"></a>Runtime Function: <em>float</em> <strong>__dpd_truncddsf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncddsf"></a>Runtime Function: <em>float</em> <strong>__bid_truncddsf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctdsf"></a>Runtime Function: <em>float</em> <strong>__dpd_trunctdsf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctdsf"></a>Runtime Function: <em>float</em> <strong>__bid_trunctdsf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendsddf"></a>Runtime Function: <em>double</em> <strong>__dpd_extendsddf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsddf"></a>Runtime Function: <em>double</em> <strong>__bid_extendsddf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctddf"></a>Runtime Function: <em>double</em> <strong>__dpd_trunctddf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctddf"></a>Runtime Function: <em>double</em> <strong>__bid_trunctddf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendsdxf"></a>Runtime Function: <em>long double</em> <strong>__dpd_extendsdxf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsdxf"></a>Runtime Function: <em>long double</em> <strong>__bid_extendsdxf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendddxf"></a>Runtime Function: <em>long double</em> <strong>__dpd_extendddxf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendddxf"></a>Runtime Function: <em>long double</em> <strong>__bid_extendddxf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctdxf"></a>Runtime Function: <em>long double</em> <strong>__dpd_trunctdxf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctdxf"></a>Runtime Function: <em>long double</em> <strong>__bid_trunctdxf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendsdtf"></a>Runtime Function: <em>long double</em> <strong>__dpd_extendsdtf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsdtf"></a>Runtime Function: <em>long double</em> <strong>__bid_extendsdtf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendddtf"></a>Runtime Function: <em>long double</em> <strong>__dpd_extendddtf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendddtf"></a>Runtime Function: <em>long double</em> <strong>__bid_extendddtf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dd><p>These functions convert the value of <var>a</var> from a decimal floating type
to a binary floating type of a different size.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fextendsfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_extendsfsd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendsfsd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_extendsfsd</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextenddfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_extenddfdd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextenddfdd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_extenddfdd</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fextendtftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_extendtftd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fextendtftd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_extendtftd</strong> <em>(long double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncsdsf"></a>Runtime Function: <em>float</em> <strong>__dpd_truncsdsf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncsdsf"></a>Runtime Function: <em>float</em> <strong>__bid_truncsdsf</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftruncdddf"></a>Runtime Function: <em>double</em> <strong>__dpd_truncdddf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftruncdddf"></a>Runtime Function: <em>double</em> <strong>__bid_truncdddf</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ftrunctdtf"></a>Runtime Function: <em>long double</em> <strong>__dpd_trunctdtf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ftrunctdtf"></a>Runtime Function: <em>long double</em> <strong>__bid_trunctdtf</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert the value of <var>a</var> between decimal and
binary floating types of the same size.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffixsdsi"></a>Runtime Function: <em>int</em> <strong>__dpd_fixsdsi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixsdsi"></a>Runtime Function: <em>int</em> <strong>__bid_fixsdsi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixddsi"></a>Runtime Function: <em>int</em> <strong>__dpd_fixddsi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixddsi"></a>Runtime Function: <em>int</em> <strong>__bid_fixddsi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixtdsi"></a>Runtime Function: <em>int</em> <strong>__dpd_fixtdsi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixtdsi"></a>Runtime Function: <em>int</em> <strong>__bid_fixtdsi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to a signed integer.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffixsddi"></a>Runtime Function: <em>long</em> <strong>__dpd_fixsddi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixsddi"></a>Runtime Function: <em>long</em> <strong>__bid_fixsddi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixdddi"></a>Runtime Function: <em>long</em> <strong>__dpd_fixdddi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixdddi"></a>Runtime Function: <em>long</em> <strong>__bid_fixdddi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixtddi"></a>Runtime Function: <em>long</em> <strong>__dpd_fixtddi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixtddi"></a>Runtime Function: <em>long</em> <strong>__bid_fixtddi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to a signed long.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffixunssdsi"></a>Runtime Function: <em>unsigned int</em> <strong>__dpd_fixunssdsi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunssdsi"></a>Runtime Function: <em>unsigned int</em> <strong>__bid_fixunssdsi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixunsddsi"></a>Runtime Function: <em>unsigned int</em> <strong>__dpd_fixunsddsi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunsddsi"></a>Runtime Function: <em>unsigned int</em> <strong>__bid_fixunsddsi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixunstdsi"></a>Runtime Function: <em>unsigned int</em> <strong>__dpd_fixunstdsi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunstdsi"></a>Runtime Function: <em>unsigned int</em> <strong>__bid_fixunstdsi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to an unsigned integer.  Negative values all become zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffixunssddi"></a>Runtime Function: <em>unsigned long</em> <strong>__dpd_fixunssddi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunssddi"></a>Runtime Function: <em>unsigned long</em> <strong>__bid_fixunssddi</strong> <em>(_Decimal32 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixunsdddi"></a>Runtime Function: <em>unsigned long</em> <strong>__dpd_fixunsdddi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunsdddi"></a>Runtime Function: <em>unsigned long</em> <strong>__bid_fixunsdddi</strong> <em>(_Decimal64 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffixunstddi"></a>Runtime Function: <em>unsigned long</em> <strong>__dpd_fixunstddi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffixunstddi"></a>Runtime Function: <em>unsigned long</em> <strong>__bid_fixunstddi</strong> <em>(_Decimal128 <var>a</var>)</em></dt>
<dd><p>These functions convert <var>a</var> to an unsigned long.  Negative values
all become zero.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffloatsisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_floatsisd</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatsisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_floatsisd</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatsidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_floatsidd</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatsidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_floatsidd</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatsitd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_floatsitd</strong> <em>(int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatsitd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_floatsitd</strong> <em>(int <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, a signed integer, to decimal floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffloatdisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_floatdisd</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatdisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_floatdisd</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatdidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_floatdidd</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatdidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_floatdidd</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatditd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_floatditd</strong> <em>(long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatditd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_floatditd</strong> <em>(long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, a signed long, to decimal floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffloatunssisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_floatunssisd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunssisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_floatunssisd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatunssidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_floatunssidd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunssidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_floatunssidd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatunssitd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_floatunssitd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunssitd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_floatunssitd</strong> <em>(unsigned int <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, an unsigned integer, to decimal floating point.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005ffloatunsdisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__dpd_floatunsdisd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunsdisd"></a>Runtime Function: <em>_Decimal32</em> <strong>__bid_floatunsdisd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatunsdidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__dpd_floatunsdidd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunsdidd"></a>Runtime Function: <em>_Decimal64</em> <strong>__bid_floatunsdidd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005ffloatunsditd"></a>Runtime Function: <em>_Decimal128</em> <strong>__dpd_floatunsditd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005ffloatunsditd"></a>Runtime Function: <em>_Decimal128</em> <strong>__bid_floatunsditd</strong> <em>(unsigned long <var>i</var>)</em></dt>
<dd><p>These functions convert <var>i</var>, an unsigned long, to decimal floating point.
</p></dd></dl>

<a name="Comparison-functions-2"></a>
<h4 class="subsection">4.3.3 Comparison functions</h4>

<dl>
<dt><a name="index-_005f_005fdpd_005funordsd2"></a>Runtime Function: <em>int</em> <strong>__dpd_unordsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005funordsd2"></a>Runtime Function: <em>int</em> <strong>__bid_unordsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005funorddd2"></a>Runtime Function: <em>int</em> <strong>__dpd_unorddd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005funorddd2"></a>Runtime Function: <em>int</em> <strong>__bid_unorddd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005funordtd2"></a>Runtime Function: <em>int</em> <strong>__dpd_unordtd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005funordtd2"></a>Runtime Function: <em>int</em> <strong>__bid_unordtd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a nonzero value if either argument is NaN, otherwise 0.
</p></dd></dl>

<p>There is also a complete group of higher level functions which
correspond directly to comparison operators.  They implement the ISO C
semantics for floating-point comparisons, taking NaN into account.
Pay careful attention to the return values defined for each set.
Under the hood, all of these routines are implemented as
</p>
<div class="smallexample">
<pre class="smallexample">  if (__bid_unord<var>X</var>d2 (a, b))
    return <var>E</var>;
  return __bid_cmp<var>X</var>d2 (a, b);
</pre></div>

<p>where <var>E</var> is a constant chosen to give the proper behavior for
NaN.  Thus, the meaning of the return value is different for each set.
Do not rely on this implementation; only the semantics documented
below are guaranteed.
</p>
<dl>
<dt><a name="index-_005f_005fdpd_005feqsd2"></a>Runtime Function: <em>int</em> <strong>__dpd_eqsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005feqsd2"></a>Runtime Function: <em>int</em> <strong>__bid_eqsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005feqdd2"></a>Runtime Function: <em>int</em> <strong>__dpd_eqdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005feqdd2"></a>Runtime Function: <em>int</em> <strong>__bid_eqdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005feqtd2"></a>Runtime Function: <em>int</em> <strong>__dpd_eqtd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005feqtd2"></a>Runtime Function: <em>int</em> <strong>__bid_eqtd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return zero if neither argument is NaN, and <var>a</var> and
<var>b</var> are equal.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fnesd2"></a>Runtime Function: <em>int</em> <strong>__dpd_nesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnesd2"></a>Runtime Function: <em>int</em> <strong>__bid_nesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fnedd2"></a>Runtime Function: <em>int</em> <strong>__dpd_nedd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnedd2"></a>Runtime Function: <em>int</em> <strong>__bid_nedd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fnetd2"></a>Runtime Function: <em>int</em> <strong>__dpd_netd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fnetd2"></a>Runtime Function: <em>int</em> <strong>__bid_netd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a nonzero value if either argument is NaN, or
if <var>a</var> and <var>b</var> are unequal.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fgesd2"></a>Runtime Function: <em>int</em> <strong>__dpd_gesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgesd2"></a>Runtime Function: <em>int</em> <strong>__bid_gesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fgedd2"></a>Runtime Function: <em>int</em> <strong>__dpd_gedd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgedd2"></a>Runtime Function: <em>int</em> <strong>__bid_gedd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fgetd2"></a>Runtime Function: <em>int</em> <strong>__dpd_getd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgetd2"></a>Runtime Function: <em>int</em> <strong>__bid_getd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a value greater than or equal to zero if
neither argument is NaN, and <var>a</var> is greater than or equal to
<var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fltsd2"></a>Runtime Function: <em>int</em> <strong>__dpd_ltsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fltsd2"></a>Runtime Function: <em>int</em> <strong>__bid_ltsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fltdd2"></a>Runtime Function: <em>int</em> <strong>__dpd_ltdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fltdd2"></a>Runtime Function: <em>int</em> <strong>__bid_ltdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005flttd2"></a>Runtime Function: <em>int</em> <strong>__dpd_lttd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005flttd2"></a>Runtime Function: <em>int</em> <strong>__bid_lttd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a value less than zero if neither argument is
NaN, and <var>a</var> is strictly less than <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005flesd2"></a>Runtime Function: <em>int</em> <strong>__dpd_lesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005flesd2"></a>Runtime Function: <em>int</em> <strong>__bid_lesd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fledd2"></a>Runtime Function: <em>int</em> <strong>__dpd_ledd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fledd2"></a>Runtime Function: <em>int</em> <strong>__bid_ledd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fletd2"></a>Runtime Function: <em>int</em> <strong>__dpd_letd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fletd2"></a>Runtime Function: <em>int</em> <strong>__bid_letd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a value less than or equal to zero if neither
argument is NaN, and <var>a</var> is less than or equal to <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdpd_005fgtsd2"></a>Runtime Function: <em>int</em> <strong>__dpd_gtsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgtsd2"></a>Runtime Function: <em>int</em> <strong>__bid_gtsd2</strong> <em>(_Decimal32 <var>a</var>, _Decimal32 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fgtdd2"></a>Runtime Function: <em>int</em> <strong>__dpd_gtdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgtdd2"></a>Runtime Function: <em>int</em> <strong>__bid_gtdd2</strong> <em>(_Decimal64 <var>a</var>, _Decimal64 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdpd_005fgttd2"></a>Runtime Function: <em>int</em> <strong>__dpd_gttd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fbid_005fgttd2"></a>Runtime Function: <em>int</em> <strong>__bid_gttd2</strong> <em>(_Decimal128 <var>a</var>, _Decimal128 <var>b</var>)</em></dt>
<dd><p>These functions return a value greater than zero if neither argument
is NaN, and <var>a</var> is strictly greater than <var>b</var>.
</p></dd></dl>

<hr>
<a name="Fixed_002dpoint-fractional-library-routines"></a>
<div class="header">
<p>
Next: <a href="#Exception-handling-routines" accesskey="n" rel="next">Exception handling routines</a>, Previous: <a href="#Decimal-float-library-routines" accesskey="p" rel="prev">Decimal float library routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Routines-for-fixed_002dpoint-fractional-emulation"></a>
<h3 class="section">4.4 Routines for fixed-point fractional emulation</h3>
<a name="index-fixed_002dpoint-fractional-library"></a>
<a name="index-fractional-types"></a>
<a name="index-Embedded-C"></a>

<p>The software fixed-point library implements fixed-point fractional
arithmetic, and is only activated on selected targets.
</p>
<p>For ease of comprehension <code>fract</code> is an alias for the
<code>_Fract</code> type, <code>accum</code> an alias for <code>_Accum</code>, and
<code>sat</code> an alias for <code>_Sat</code>.
</p>
<p>For illustrative purposes, in this section the fixed-point fractional type
<code>short&nbsp;fract<!-- /@w --></code> is assumed to correspond to machine mode <code>QQmode</code>;
<code>unsigned&nbsp;short&nbsp;fract<!-- /@w --></code> to <code>UQQmode</code>;
<code>fract</code> to <code>HQmode</code>;
<code>unsigned&nbsp;fract<!-- /@w --></code> to <code>UHQmode</code>;
<code>long&nbsp;fract<!-- /@w --></code> to <code>SQmode</code>;
<code>unsigned&nbsp;long&nbsp;fract<!-- /@w --></code> to <code>USQmode</code>;
<code>long&nbsp;long&nbsp;fract<!-- /@w --></code> to <code>DQmode</code>;
and <code>unsigned&nbsp;long&nbsp;long&nbsp;fract<!-- /@w --></code> to <code>UDQmode</code>.
Similarly the fixed-point accumulator type
<code>short&nbsp;accum<!-- /@w --></code> corresponds to <code>HAmode</code>;
<code>unsigned&nbsp;short&nbsp;accum<!-- /@w --></code> to <code>UHAmode</code>;
<code>accum</code> to <code>SAmode</code>;
<code>unsigned&nbsp;accum<!-- /@w --></code> to <code>USAmode</code>;
<code>long&nbsp;accum<!-- /@w --></code> to <code>DAmode</code>;
<code>unsigned&nbsp;long&nbsp;accum<!-- /@w --></code> to <code>UDAmode</code>;
<code>long&nbsp;long&nbsp;accum<!-- /@w --></code> to <code>TAmode</code>;
and <code>unsigned&nbsp;long&nbsp;long&nbsp;accum<!-- /@w --></code> to <code>UTAmode</code>.
</p>
<a name="Arithmetic-functions-3"></a>
<h4 class="subsection">4.4.1 Arithmetic functions</h4>

<dl>
<dt><a name="index-_005f_005faddqq3"></a>Runtime Function: <em>short fract</em> <strong>__addqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddhq3"></a>Runtime Function: <em>fract</em> <strong>__addhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddsq3"></a>Runtime Function: <em>long fract</em> <strong>__addsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadddq3"></a>Runtime Function: <em>long long fract</em> <strong>__adddq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadduqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__adduqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadduhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__adduhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__addusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__addudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddha3"></a>Runtime Function: <em>short accum</em> <strong>__addha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddsa3"></a>Runtime Function: <em>accum</em> <strong>__addsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddda3"></a>Runtime Function: <em>long accum</em> <strong>__addda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddta3"></a>Runtime Function: <em>long long accum</em> <strong>__addta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadduha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__adduha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005faddusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__addusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadduda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__adduda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fadduta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__adduta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fssaddqq3"></a>Runtime Function: <em>short fract</em> <strong>__ssaddqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddhq3"></a>Runtime Function: <em>fract</em> <strong>__ssaddhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddsq3"></a>Runtime Function: <em>long fract</em> <strong>__ssaddsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssadddq3"></a>Runtime Function: <em>long long fract</em> <strong>__ssadddq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddha3"></a>Runtime Function: <em>short accum</em> <strong>__ssaddha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddsa3"></a>Runtime Function: <em>accum</em> <strong>__ssaddsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddda3"></a>Runtime Function: <em>long accum</em> <strong>__ssaddda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssaddta3"></a>Runtime Function: <em>long long accum</em> <strong>__ssaddta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var> with signed saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fusadduqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__usadduqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusadduhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__usadduhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusaddusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__usaddusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusaddudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__usaddudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusadduha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__usadduha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusaddusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__usaddusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusadduda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__usadduda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusadduta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__usadduta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the sum of <var>a</var> and <var>b</var> with unsigned saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsubqq3"></a>Runtime Function: <em>short fract</em> <strong>__subqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubhq3"></a>Runtime Function: <em>fract</em> <strong>__subhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubsq3"></a>Runtime Function: <em>long fract</em> <strong>__subsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubdq3"></a>Runtime Function: <em>long long fract</em> <strong>__subdq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubuqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__subuqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubuhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__subuhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__subusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__subudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubha3"></a>Runtime Function: <em>short accum</em> <strong>__subha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubsa3"></a>Runtime Function: <em>accum</em> <strong>__subsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubda3"></a>Runtime Function: <em>long accum</em> <strong>__subda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubta3"></a>Runtime Function: <em>long long accum</em> <strong>__subta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubuha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__subuha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__subusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubuda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__subuda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsubuta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__subuta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the difference of <var>a</var> and <var>b</var>;
that is, <code><var>a</var> - <var>b</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsssubqq3"></a>Runtime Function: <em>short fract</em> <strong>__sssubqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubhq3"></a>Runtime Function: <em>fract</em> <strong>__sssubhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubsq3"></a>Runtime Function: <em>long fract</em> <strong>__sssubsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubdq3"></a>Runtime Function: <em>long long fract</em> <strong>__sssubdq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubha3"></a>Runtime Function: <em>short accum</em> <strong>__sssubha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubsa3"></a>Runtime Function: <em>accum</em> <strong>__sssubsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubda3"></a>Runtime Function: <em>long accum</em> <strong>__sssubda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fsssubta3"></a>Runtime Function: <em>long long accum</em> <strong>__sssubta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the difference of <var>a</var> and <var>b</var> with signed
saturation;  that is, <code><var>a</var> - <var>b</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fussubuqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__ussubuqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubuhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__ussubuhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__ussubusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__ussubudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubuha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__ussubuha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__ussubusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubuda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__ussubuda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fussubuta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__ussubuta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the difference of <var>a</var> and <var>b</var> with unsigned
saturation;  that is, <code><var>a</var> - <var>b</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmulqq3"></a>Runtime Function: <em>short fract</em> <strong>__mulqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulhq3"></a>Runtime Function: <em>fract</em> <strong>__mulhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulsq3"></a>Runtime Function: <em>long fract</em> <strong>__mulsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuldq3"></a>Runtime Function: <em>long long fract</em> <strong>__muldq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuluqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__muluqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuluhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__muluhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__mulusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuludq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__muludq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulha3"></a>Runtime Function: <em>short accum</em> <strong>__mulha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulsa3"></a>Runtime Function: <em>accum</em> <strong>__mulsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulda3"></a>Runtime Function: <em>long accum</em> <strong>__mulda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulta3"></a>Runtime Function: <em>long long accum</em> <strong>__multa3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuluha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__muluha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmulusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__mulusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuluda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__muluda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fmuluta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__muluta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fssmulqq3"></a>Runtime Function: <em>short fract</em> <strong>__ssmulqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulhq3"></a>Runtime Function: <em>fract</em> <strong>__ssmulhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulsq3"></a>Runtime Function: <em>long fract</em> <strong>__ssmulsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmuldq3"></a>Runtime Function: <em>long long fract</em> <strong>__ssmuldq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulha3"></a>Runtime Function: <em>short accum</em> <strong>__ssmulha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulsa3"></a>Runtime Function: <em>accum</em> <strong>__ssmulsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulda3"></a>Runtime Function: <em>long accum</em> <strong>__ssmulda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssmulta3"></a>Runtime Function: <em>long long accum</em> <strong>__ssmulta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var> with signed
saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fusmuluqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__usmuluqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmuluhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__usmuluhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmulusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__usmulusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmuludq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__usmuludq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmuluha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__usmuluha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmulusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__usmulusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmuluda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__usmuluda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusmuluta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__usmuluta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the product of <var>a</var> and <var>b</var> with unsigned
saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fdivqq3"></a>Runtime Function: <em>short fract</em> <strong>__divqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivhq3"></a>Runtime Function: <em>fract</em> <strong>__divhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivsq3"></a>Runtime Function: <em>long fract</em> <strong>__divsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivdq3"></a>Runtime Function: <em>long long fract</em> <strong>__divdq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivha3"></a>Runtime Function: <em>short accum</em> <strong>__divha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivsa3"></a>Runtime Function: <em>accum</em> <strong>__divsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivda3"></a>Runtime Function: <em>long accum</em> <strong>__divda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fdivta3"></a>Runtime Function: <em>long long accum</em> <strong>__divta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the signed division of <var>a</var>
and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fudivuqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__udivuqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivuhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__udivuhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__udivusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__udivudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivuha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__udivuha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__udivusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivuda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__udivuda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fudivuta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__udivuta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the unsigned division of <var>a</var>
and <var>b</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fssdivqq3"></a>Runtime Function: <em>short fract</em> <strong>__ssdivqq3</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivhq3"></a>Runtime Function: <em>fract</em> <strong>__ssdivhq3</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivsq3"></a>Runtime Function: <em>long fract</em> <strong>__ssdivsq3</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivdq3"></a>Runtime Function: <em>long long fract</em> <strong>__ssdivdq3</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivha3"></a>Runtime Function: <em>short accum</em> <strong>__ssdivha3</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivsa3"></a>Runtime Function: <em>accum</em> <strong>__ssdivsa3</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivda3"></a>Runtime Function: <em>long accum</em> <strong>__ssdivda3</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssdivta3"></a>Runtime Function: <em>long long accum</em> <strong>__ssdivta3</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the signed division of <var>a</var>
and <var>b</var> with signed saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fusdivuqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__usdivuqq3</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivuhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__usdivuhq3</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__usdivusq3</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__usdivudq3</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivuha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__usdivuha3</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__usdivusa3</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivuda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__usdivuda3</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusdivuta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__usdivuta3</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions return the quotient of the unsigned division of <var>a</var>
and <var>b</var> with unsigned saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fnegqq2"></a>Runtime Function: <em>short fract</em> <strong>__negqq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneghq2"></a>Runtime Function: <em>fract</em> <strong>__neghq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegsq2"></a>Runtime Function: <em>long fract</em> <strong>__negsq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegdq2"></a>Runtime Function: <em>long long fract</em> <strong>__negdq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneguqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__neguqq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneguhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__neguhq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__negusq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__negudq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegha2"></a>Runtime Function: <em>short accum</em> <strong>__negha2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegsa2"></a>Runtime Function: <em>accum</em> <strong>__negsa2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegda2"></a>Runtime Function: <em>long accum</em> <strong>__negda2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegta2"></a>Runtime Function: <em>long long accum</em> <strong>__negta2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneguha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__neguha2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fnegusa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__negusa2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneguda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__neguda2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fneguta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__neguta2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var>.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fssnegqq2"></a>Runtime Function: <em>short fract</em> <strong>__ssnegqq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssneghq2"></a>Runtime Function: <em>fract</em> <strong>__ssneghq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegsq2"></a>Runtime Function: <em>long fract</em> <strong>__ssnegsq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegdq2"></a>Runtime Function: <em>long long fract</em> <strong>__ssnegdq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegha2"></a>Runtime Function: <em>short accum</em> <strong>__ssnegha2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegsa2"></a>Runtime Function: <em>accum</em> <strong>__ssnegsa2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegda2"></a>Runtime Function: <em>long accum</em> <strong>__ssnegda2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fssnegta2"></a>Runtime Function: <em>long long accum</em> <strong>__ssnegta2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var> with signed saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fusneguqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__usneguqq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusneguhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__usneguhq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusnegusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__usnegusq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusnegudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__usnegudq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusneguha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__usneguha2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusnegusa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__usnegusa2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusneguda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__usneguda2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fusneguta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__usneguta2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dd><p>These functions return the negation of <var>a</var> with unsigned saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fashlqq3"></a>Runtime Function: <em>short fract</em> <strong>__ashlqq3</strong> <em>(short fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlhq3"></a>Runtime Function: <em>fract</em> <strong>__ashlhq3</strong> <em>(fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlsq3"></a>Runtime Function: <em>long fract</em> <strong>__ashlsq3</strong> <em>(long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashldq3"></a>Runtime Function: <em>long long fract</em> <strong>__ashldq3</strong> <em>(long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashluqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__ashluqq3</strong> <em>(unsigned short fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashluhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__ashluhq3</strong> <em>(unsigned fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__ashlusq3</strong> <em>(unsigned long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashludq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__ashludq3</strong> <em>(unsigned long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlha3"></a>Runtime Function: <em>short accum</em> <strong>__ashlha3</strong> <em>(short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlsa3"></a>Runtime Function: <em>accum</em> <strong>__ashlsa3</strong> <em>(accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlda3"></a>Runtime Function: <em>long accum</em> <strong>__ashlda3</strong> <em>(long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlta3"></a>Runtime Function: <em>long long accum</em> <strong>__ashlta3</strong> <em>(long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashluha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__ashluha3</strong> <em>(unsigned short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashlusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__ashlusa3</strong> <em>(unsigned accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashluda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__ashluda3</strong> <em>(unsigned long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashluta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__ashluta3</strong> <em>(unsigned long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of shifting <var>a</var> left by <var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fashrqq3"></a>Runtime Function: <em>short fract</em> <strong>__ashrqq3</strong> <em>(short fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrhq3"></a>Runtime Function: <em>fract</em> <strong>__ashrhq3</strong> <em>(fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrsq3"></a>Runtime Function: <em>long fract</em> <strong>__ashrsq3</strong> <em>(long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrdq3"></a>Runtime Function: <em>long long fract</em> <strong>__ashrdq3</strong> <em>(long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrha3"></a>Runtime Function: <em>short accum</em> <strong>__ashrha3</strong> <em>(short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrsa3"></a>Runtime Function: <em>accum</em> <strong>__ashrsa3</strong> <em>(accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrda3"></a>Runtime Function: <em>long accum</em> <strong>__ashrda3</strong> <em>(long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fashrta3"></a>Runtime Function: <em>long long accum</em> <strong>__ashrta3</strong> <em>(long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of arithmetically shifting <var>a</var> right
by <var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005flshruqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__lshruqq3</strong> <em>(unsigned short fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshruhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__lshruhq3</strong> <em>(unsigned fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshrusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__lshrusq3</strong> <em>(unsigned long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshrudq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__lshrudq3</strong> <em>(unsigned long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshruha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__lshruha3</strong> <em>(unsigned short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshrusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__lshrusa3</strong> <em>(unsigned accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshruda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__lshruda3</strong> <em>(unsigned long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005flshruta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__lshruta3</strong> <em>(unsigned long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of logically shifting <var>a</var> right
by <var>b</var> bits.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fssashlhq3"></a>Runtime Function: <em>fract</em> <strong>__ssashlhq3</strong> <em>(fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashlsq3"></a>Runtime Function: <em>long fract</em> <strong>__ssashlsq3</strong> <em>(long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashldq3"></a>Runtime Function: <em>long long fract</em> <strong>__ssashldq3</strong> <em>(long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashlha3"></a>Runtime Function: <em>short accum</em> <strong>__ssashlha3</strong> <em>(short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashlsa3"></a>Runtime Function: <em>accum</em> <strong>__ssashlsa3</strong> <em>(accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashlda3"></a>Runtime Function: <em>long accum</em> <strong>__ssashlda3</strong> <em>(long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fssashlta3"></a>Runtime Function: <em>long long accum</em> <strong>__ssashlta3</strong> <em>(long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of shifting <var>a</var> left by <var>b</var> bits
with signed saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fusashluqq3"></a>Runtime Function: <em>unsigned short fract</em> <strong>__usashluqq3</strong> <em>(unsigned short fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashluhq3"></a>Runtime Function: <em>unsigned fract</em> <strong>__usashluhq3</strong> <em>(unsigned fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashlusq3"></a>Runtime Function: <em>unsigned long fract</em> <strong>__usashlusq3</strong> <em>(unsigned long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashludq3"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__usashludq3</strong> <em>(unsigned long long fract <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashluha3"></a>Runtime Function: <em>unsigned short accum</em> <strong>__usashluha3</strong> <em>(unsigned short accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashlusa3"></a>Runtime Function: <em>unsigned accum</em> <strong>__usashlusa3</strong> <em>(unsigned accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashluda3"></a>Runtime Function: <em>unsigned long accum</em> <strong>__usashluda3</strong> <em>(unsigned long accum <var>a</var>, int <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fusashluta3"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__usashluta3</strong> <em>(unsigned long long accum <var>a</var>, int <var>b</var>)</em></dt>
<dd><p>These functions return the result of shifting <var>a</var> left by <var>b</var> bits
with unsigned saturation.
</p></dd></dl>

<a name="Comparison-functions-3"></a>
<h4 class="subsection">4.4.2 Comparison functions</h4>

<p>The following functions implement fixed-point comparisons.  These functions
implement a low-level compare, upon which the higher level comparison
operators (such as less than and greater than or equal to) can be
constructed.  The returned values lie in the range zero to two, to allow
the high-level operators to be implemented by testing the returned
result using either signed or unsigned comparison.
</p>
<dl>
<dt><a name="index-_005f_005fcmpqq2"></a>Runtime Function: <em>int</em> <strong>__cmpqq2</strong> <em>(short fract <var>a</var>, short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmphq2"></a>Runtime Function: <em>int</em> <strong>__cmphq2</strong> <em>(fract <var>a</var>, fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpsq2"></a>Runtime Function: <em>int</em> <strong>__cmpsq2</strong> <em>(long fract <var>a</var>, long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpdq2"></a>Runtime Function: <em>int</em> <strong>__cmpdq2</strong> <em>(long long fract <var>a</var>, long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpuqq2"></a>Runtime Function: <em>int</em> <strong>__cmpuqq2</strong> <em>(unsigned short fract <var>a</var>, unsigned short fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpuhq2"></a>Runtime Function: <em>int</em> <strong>__cmpuhq2</strong> <em>(unsigned fract <var>a</var>, unsigned fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpusq2"></a>Runtime Function: <em>int</em> <strong>__cmpusq2</strong> <em>(unsigned long fract <var>a</var>, unsigned long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpudq2"></a>Runtime Function: <em>int</em> <strong>__cmpudq2</strong> <em>(unsigned long long fract <var>a</var>, unsigned long long fract <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpha2"></a>Runtime Function: <em>int</em> <strong>__cmpha2</strong> <em>(short accum <var>a</var>, short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpsa2"></a>Runtime Function: <em>int</em> <strong>__cmpsa2</strong> <em>(accum <var>a</var>, accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpda2"></a>Runtime Function: <em>int</em> <strong>__cmpda2</strong> <em>(long accum <var>a</var>, long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpta2"></a>Runtime Function: <em>int</em> <strong>__cmpta2</strong> <em>(long long accum <var>a</var>, long long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpuha2"></a>Runtime Function: <em>int</em> <strong>__cmpuha2</strong> <em>(unsigned short accum <var>a</var>, unsigned short accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpusa2"></a>Runtime Function: <em>int</em> <strong>__cmpusa2</strong> <em>(unsigned accum <var>a</var>, unsigned accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmpuda2"></a>Runtime Function: <em>int</em> <strong>__cmpuda2</strong> <em>(unsigned long accum <var>a</var>, unsigned long accum <var>b</var>)</em></dt>
<dt><a name="index-_005f_005fcmputa2"></a>Runtime Function: <em>int</em> <strong>__cmputa2</strong> <em>(unsigned long long accum <var>a</var>, unsigned long long accum <var>b</var>)</em></dt>
<dd><p>These functions perform a signed or unsigned comparison of <var>a</var> and
<var>b</var> (depending on the selected machine mode).  If <var>a</var> is less
than <var>b</var>, they return 0; if <var>a</var> is greater than <var>b</var>, they
return 2; and if <var>a</var> and <var>b</var> are equal they return 1.
</p></dd></dl>

<a name="Conversion-functions-2"></a>
<h4 class="subsection">4.4.3 Conversion functions</h4>

<dl>
<dt><a name="index-_005f_005ffractqqhq2"></a>Runtime Function: <em>fract</em> <strong>__fractqqhq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqsq2"></a>Runtime Function: <em>long fract</em> <strong>__fractqqsq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__fractqqdq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqha"></a>Runtime Function: <em>short accum</em> <strong>__fractqqha</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqsa"></a>Runtime Function: <em>accum</em> <strong>__fractqqsa</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqda"></a>Runtime Function: <em>long accum</em> <strong>__fractqqda</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractqqta</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractqquqq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractqquhq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractqqusq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractqqudq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractqquha</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractqqusa</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractqquda</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractqquta</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractqqqi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqhi"></a>Runtime Function: <em>short</em> <strong>__fractqqhi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqsi"></a>Runtime Function: <em>int</em> <strong>__fractqqsi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqdi"></a>Runtime Function: <em>long</em> <strong>__fractqqdi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqti"></a>Runtime Function: <em>long long</em> <strong>__fractqqti</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqsf"></a>Runtime Function: <em>float</em> <strong>__fractqqsf</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqqdf"></a>Runtime Function: <em>double</em> <strong>__fractqqdf</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqqq2"></a>Runtime Function: <em>short fract</em> <strong>__fracthqqq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqsq2"></a>Runtime Function: <em>long fract</em> <strong>__fracthqsq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__fracthqdq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqha"></a>Runtime Function: <em>short accum</em> <strong>__fracthqha</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqsa"></a>Runtime Function: <em>accum</em> <strong>__fracthqsa</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqda"></a>Runtime Function: <em>long accum</em> <strong>__fracthqda</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqta"></a>Runtime Function: <em>long long accum</em> <strong>__fracthqta</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fracthquqq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fracthquhq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fracthqusq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fracthqudq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fracthquha</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fracthqusa</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fracthquda</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fracthquta</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqqi"></a>Runtime Function: <em>signed char</em> <strong>__fracthqqi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqhi"></a>Runtime Function: <em>short</em> <strong>__fracthqhi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqsi"></a>Runtime Function: <em>int</em> <strong>__fracthqsi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqdi"></a>Runtime Function: <em>long</em> <strong>__fracthqdi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqti"></a>Runtime Function: <em>long long</em> <strong>__fracthqti</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqsf"></a>Runtime Function: <em>float</em> <strong>__fracthqsf</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthqdf"></a>Runtime Function: <em>double</em> <strong>__fracthqdf</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqqq2"></a>Runtime Function: <em>short fract</em> <strong>__fractsqqq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqhq2"></a>Runtime Function: <em>fract</em> <strong>__fractsqhq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__fractsqdq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqha"></a>Runtime Function: <em>short accum</em> <strong>__fractsqha</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqsa"></a>Runtime Function: <em>accum</em> <strong>__fractsqsa</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqda"></a>Runtime Function: <em>long accum</em> <strong>__fractsqda</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractsqta</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractsquqq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractsquhq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractsqusq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractsqudq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractsquha</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractsqusa</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractsquda</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractsquta</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractsqqi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqhi"></a>Runtime Function: <em>short</em> <strong>__fractsqhi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqsi"></a>Runtime Function: <em>int</em> <strong>__fractsqsi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqdi"></a>Runtime Function: <em>long</em> <strong>__fractsqdi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqti"></a>Runtime Function: <em>long long</em> <strong>__fractsqti</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqsf"></a>Runtime Function: <em>float</em> <strong>__fractsqsf</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsqdf"></a>Runtime Function: <em>double</em> <strong>__fractsqdf</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqqq2"></a>Runtime Function: <em>short fract</em> <strong>__fractdqqq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqhq2"></a>Runtime Function: <em>fract</em> <strong>__fractdqhq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqsq2"></a>Runtime Function: <em>long fract</em> <strong>__fractdqsq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqha"></a>Runtime Function: <em>short accum</em> <strong>__fractdqha</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqsa"></a>Runtime Function: <em>accum</em> <strong>__fractdqsa</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqda"></a>Runtime Function: <em>long accum</em> <strong>__fractdqda</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractdqta</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractdquqq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractdquhq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractdqusq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractdqudq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractdquha</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractdqusa</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractdquda</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractdquta</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractdqqi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqhi"></a>Runtime Function: <em>short</em> <strong>__fractdqhi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqsi"></a>Runtime Function: <em>int</em> <strong>__fractdqsi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqdi"></a>Runtime Function: <em>long</em> <strong>__fractdqdi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqti"></a>Runtime Function: <em>long long</em> <strong>__fractdqti</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqsf"></a>Runtime Function: <em>float</em> <strong>__fractdqsf</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdqdf"></a>Runtime Function: <em>double</em> <strong>__fractdqdf</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthaqq"></a>Runtime Function: <em>short fract</em> <strong>__fracthaqq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthahq"></a>Runtime Function: <em>fract</em> <strong>__fracthahq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthasq"></a>Runtime Function: <em>long fract</em> <strong>__fracthasq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthadq"></a>Runtime Function: <em>long long fract</em> <strong>__fracthadq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthasa2"></a>Runtime Function: <em>accum</em> <strong>__fracthasa2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthada2"></a>Runtime Function: <em>long accum</em> <strong>__fracthada2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthata2"></a>Runtime Function: <em>long long accum</em> <strong>__fracthata2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fracthauqq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fracthauhq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fracthausq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fracthaudq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fracthauha</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fracthausa</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fracthauda</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fracthauta</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthaqi"></a>Runtime Function: <em>signed char</em> <strong>__fracthaqi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthahi"></a>Runtime Function: <em>short</em> <strong>__fracthahi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthasi"></a>Runtime Function: <em>int</em> <strong>__fracthasi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthadi"></a>Runtime Function: <em>long</em> <strong>__fracthadi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthati"></a>Runtime Function: <em>long long</em> <strong>__fracthati</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthasf"></a>Runtime Function: <em>float</em> <strong>__fracthasf</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthadf"></a>Runtime Function: <em>double</em> <strong>__fracthadf</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractsaqq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsahq"></a>Runtime Function: <em>fract</em> <strong>__fractsahq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsasq"></a>Runtime Function: <em>long fract</em> <strong>__fractsasq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractsadq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsaha2"></a>Runtime Function: <em>short accum</em> <strong>__fractsaha2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsada2"></a>Runtime Function: <em>long accum</em> <strong>__fractsada2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsata2"></a>Runtime Function: <em>long long accum</em> <strong>__fractsata2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractsauqq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractsauhq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractsausq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractsaudq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractsauha</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractsausa</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractsauda</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractsauta</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractsaqi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsahi"></a>Runtime Function: <em>short</em> <strong>__fractsahi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsasi"></a>Runtime Function: <em>int</em> <strong>__fractsasi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsadi"></a>Runtime Function: <em>long</em> <strong>__fractsadi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsati"></a>Runtime Function: <em>long long</em> <strong>__fractsati</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsasf"></a>Runtime Function: <em>float</em> <strong>__fractsasf</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsadf"></a>Runtime Function: <em>double</em> <strong>__fractsadf</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractdaqq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdahq"></a>Runtime Function: <em>fract</em> <strong>__fractdahq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdasq"></a>Runtime Function: <em>long fract</em> <strong>__fractdasq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractdadq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdaha2"></a>Runtime Function: <em>short accum</em> <strong>__fractdaha2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdasa2"></a>Runtime Function: <em>accum</em> <strong>__fractdasa2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdata2"></a>Runtime Function: <em>long long accum</em> <strong>__fractdata2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractdauqq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractdauhq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractdausq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractdaudq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractdauha</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractdausa</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractdauda</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractdauta</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractdaqi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdahi"></a>Runtime Function: <em>short</em> <strong>__fractdahi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdasi"></a>Runtime Function: <em>int</em> <strong>__fractdasi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdadi"></a>Runtime Function: <em>long</em> <strong>__fractdadi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdati"></a>Runtime Function: <em>long long</em> <strong>__fractdati</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdasf"></a>Runtime Function: <em>float</em> <strong>__fractdasf</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdadf"></a>Runtime Function: <em>double</em> <strong>__fractdadf</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttaqq"></a>Runtime Function: <em>short fract</em> <strong>__fracttaqq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttahq"></a>Runtime Function: <em>fract</em> <strong>__fracttahq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttasq"></a>Runtime Function: <em>long fract</em> <strong>__fracttasq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttadq"></a>Runtime Function: <em>long long fract</em> <strong>__fracttadq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttaha2"></a>Runtime Function: <em>short accum</em> <strong>__fracttaha2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttasa2"></a>Runtime Function: <em>accum</em> <strong>__fracttasa2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttada2"></a>Runtime Function: <em>long accum</em> <strong>__fracttada2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fracttauqq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fracttauhq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fracttausq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fracttaudq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fracttauha</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fracttausa</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fracttauda</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fracttauta</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttaqi"></a>Runtime Function: <em>signed char</em> <strong>__fracttaqi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttahi"></a>Runtime Function: <em>short</em> <strong>__fracttahi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttasi"></a>Runtime Function: <em>int</em> <strong>__fracttasi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttadi"></a>Runtime Function: <em>long</em> <strong>__fracttadi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttati"></a>Runtime Function: <em>long long</em> <strong>__fracttati</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttasf"></a>Runtime Function: <em>float</em> <strong>__fracttasf</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttadf"></a>Runtime Function: <em>double</em> <strong>__fracttadf</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqqq"></a>Runtime Function: <em>short fract</em> <strong>__fractuqqqq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqhq"></a>Runtime Function: <em>fract</em> <strong>__fractuqqhq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqsq"></a>Runtime Function: <em>long fract</em> <strong>__fractuqqsq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractuqqdq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqha"></a>Runtime Function: <em>short accum</em> <strong>__fractuqqha</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqsa"></a>Runtime Function: <em>accum</em> <strong>__fractuqqsa</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqda"></a>Runtime Function: <em>long accum</em> <strong>__fractuqqda</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractuqqta</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractuqquhq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractuqqusq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractuqqudq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractuqquha</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractuqqusa</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractuqquda</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractuqquta</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractuqqqi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqhi"></a>Runtime Function: <em>short</em> <strong>__fractuqqhi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqsi"></a>Runtime Function: <em>int</em> <strong>__fractuqqsi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqdi"></a>Runtime Function: <em>long</em> <strong>__fractuqqdi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqti"></a>Runtime Function: <em>long long</em> <strong>__fractuqqti</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqsf"></a>Runtime Function: <em>float</em> <strong>__fractuqqsf</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuqqdf"></a>Runtime Function: <em>double</em> <strong>__fractuqqdf</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqqq"></a>Runtime Function: <em>short fract</em> <strong>__fractuhqqq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqhq"></a>Runtime Function: <em>fract</em> <strong>__fractuhqhq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqsq"></a>Runtime Function: <em>long fract</em> <strong>__fractuhqsq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractuhqdq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqha"></a>Runtime Function: <em>short accum</em> <strong>__fractuhqha</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqsa"></a>Runtime Function: <em>accum</em> <strong>__fractuhqsa</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqda"></a>Runtime Function: <em>long accum</em> <strong>__fractuhqda</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractuhqta</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractuhquqq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractuhqusq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractuhqudq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractuhquha</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractuhqusa</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractuhquda</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractuhquta</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractuhqqi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqhi"></a>Runtime Function: <em>short</em> <strong>__fractuhqhi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqsi"></a>Runtime Function: <em>int</em> <strong>__fractuhqsi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqdi"></a>Runtime Function: <em>long</em> <strong>__fractuhqdi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqti"></a>Runtime Function: <em>long long</em> <strong>__fractuhqti</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqsf"></a>Runtime Function: <em>float</em> <strong>__fractuhqsf</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhqdf"></a>Runtime Function: <em>double</em> <strong>__fractuhqdf</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqqq"></a>Runtime Function: <em>short fract</em> <strong>__fractusqqq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqhq"></a>Runtime Function: <em>fract</em> <strong>__fractusqhq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqsq"></a>Runtime Function: <em>long fract</em> <strong>__fractusqsq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractusqdq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqha"></a>Runtime Function: <em>short accum</em> <strong>__fractusqha</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqsa"></a>Runtime Function: <em>accum</em> <strong>__fractusqsa</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqda"></a>Runtime Function: <em>long accum</em> <strong>__fractusqda</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractusqta</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractusquqq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractusquhq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractusqudq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractusquha</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractusqusa</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractusquda</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractusquta</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractusqqi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqhi"></a>Runtime Function: <em>short</em> <strong>__fractusqhi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqsi"></a>Runtime Function: <em>int</em> <strong>__fractusqsi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqdi"></a>Runtime Function: <em>long</em> <strong>__fractusqdi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqti"></a>Runtime Function: <em>long long</em> <strong>__fractusqti</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqsf"></a>Runtime Function: <em>float</em> <strong>__fractusqsf</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusqdf"></a>Runtime Function: <em>double</em> <strong>__fractusqdf</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqqq"></a>Runtime Function: <em>short fract</em> <strong>__fractudqqq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqhq"></a>Runtime Function: <em>fract</em> <strong>__fractudqhq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqsq"></a>Runtime Function: <em>long fract</em> <strong>__fractudqsq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractudqdq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqha"></a>Runtime Function: <em>short accum</em> <strong>__fractudqha</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqsa"></a>Runtime Function: <em>accum</em> <strong>__fractudqsa</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqda"></a>Runtime Function: <em>long accum</em> <strong>__fractudqda</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqta"></a>Runtime Function: <em>long long accum</em> <strong>__fractudqta</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractudquqq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractudquhq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractudqusq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractudquha</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractudqusa</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractudquda</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractudquta</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqqi"></a>Runtime Function: <em>signed char</em> <strong>__fractudqqi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqhi"></a>Runtime Function: <em>short</em> <strong>__fractudqhi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqsi"></a>Runtime Function: <em>int</em> <strong>__fractudqsi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqdi"></a>Runtime Function: <em>long</em> <strong>__fractudqdi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqti"></a>Runtime Function: <em>long long</em> <strong>__fractudqti</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqsf"></a>Runtime Function: <em>float</em> <strong>__fractudqsf</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudqdf"></a>Runtime Function: <em>double</em> <strong>__fractudqdf</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractuhaqq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhahq"></a>Runtime Function: <em>fract</em> <strong>__fractuhahq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhasq"></a>Runtime Function: <em>long fract</em> <strong>__fractuhasq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractuhadq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhaha"></a>Runtime Function: <em>short accum</em> <strong>__fractuhaha</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhasa"></a>Runtime Function: <em>accum</em> <strong>__fractuhasa</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhada"></a>Runtime Function: <em>long accum</em> <strong>__fractuhada</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhata"></a>Runtime Function: <em>long long accum</em> <strong>__fractuhata</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractuhauqq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractuhauhq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractuhausq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractuhaudq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractuhausa2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractuhauda2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractuhauta2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractuhaqi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhahi"></a>Runtime Function: <em>short</em> <strong>__fractuhahi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhasi"></a>Runtime Function: <em>int</em> <strong>__fractuhasi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhadi"></a>Runtime Function: <em>long</em> <strong>__fractuhadi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhati"></a>Runtime Function: <em>long long</em> <strong>__fractuhati</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhasf"></a>Runtime Function: <em>float</em> <strong>__fractuhasf</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractuhadf"></a>Runtime Function: <em>double</em> <strong>__fractuhadf</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractusaqq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusahq"></a>Runtime Function: <em>fract</em> <strong>__fractusahq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusasq"></a>Runtime Function: <em>long fract</em> <strong>__fractusasq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractusadq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusaha"></a>Runtime Function: <em>short accum</em> <strong>__fractusaha</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusasa"></a>Runtime Function: <em>accum</em> <strong>__fractusasa</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusada"></a>Runtime Function: <em>long accum</em> <strong>__fractusada</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusata"></a>Runtime Function: <em>long long accum</em> <strong>__fractusata</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractusauqq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractusauhq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractusausq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractusaudq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractusauha2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractusauda2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractusauta2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractusaqi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusahi"></a>Runtime Function: <em>short</em> <strong>__fractusahi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusasi"></a>Runtime Function: <em>int</em> <strong>__fractusasi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusadi"></a>Runtime Function: <em>long</em> <strong>__fractusadi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusati"></a>Runtime Function: <em>long long</em> <strong>__fractusati</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusasf"></a>Runtime Function: <em>float</em> <strong>__fractusasf</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractusadf"></a>Runtime Function: <em>double</em> <strong>__fractusadf</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractudaqq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudahq"></a>Runtime Function: <em>fract</em> <strong>__fractudahq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudasq"></a>Runtime Function: <em>long fract</em> <strong>__fractudasq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractudadq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudaha"></a>Runtime Function: <em>short accum</em> <strong>__fractudaha</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudasa"></a>Runtime Function: <em>accum</em> <strong>__fractudasa</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudada"></a>Runtime Function: <em>long accum</em> <strong>__fractudada</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudata"></a>Runtime Function: <em>long long accum</em> <strong>__fractudata</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractudauqq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractudauhq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractudausq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractudaudq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractudauha2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractudausa2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractudauta2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractudaqi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudahi"></a>Runtime Function: <em>short</em> <strong>__fractudahi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudasi"></a>Runtime Function: <em>int</em> <strong>__fractudasi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudadi"></a>Runtime Function: <em>long</em> <strong>__fractudadi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudati"></a>Runtime Function: <em>long long</em> <strong>__fractudati</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudasf"></a>Runtime Function: <em>float</em> <strong>__fractudasf</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractudadf"></a>Runtime Function: <em>double</em> <strong>__fractudadf</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutaqq"></a>Runtime Function: <em>short fract</em> <strong>__fractutaqq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutahq"></a>Runtime Function: <em>fract</em> <strong>__fractutahq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutasq"></a>Runtime Function: <em>long fract</em> <strong>__fractutasq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutadq"></a>Runtime Function: <em>long long fract</em> <strong>__fractutadq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutaha"></a>Runtime Function: <em>short accum</em> <strong>__fractutaha</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutasa"></a>Runtime Function: <em>accum</em> <strong>__fractutasa</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutada"></a>Runtime Function: <em>long accum</em> <strong>__fractutada</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutata"></a>Runtime Function: <em>long long accum</em> <strong>__fractutata</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractutauqq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractutauhq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractutausq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractutaudq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractutauha2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractutausa2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractutauda2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutaqi"></a>Runtime Function: <em>signed char</em> <strong>__fractutaqi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutahi"></a>Runtime Function: <em>short</em> <strong>__fractutahi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutasi"></a>Runtime Function: <em>int</em> <strong>__fractutasi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutadi"></a>Runtime Function: <em>long</em> <strong>__fractutadi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutati"></a>Runtime Function: <em>long long</em> <strong>__fractutati</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutasf"></a>Runtime Function: <em>float</em> <strong>__fractutasf</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractutadf"></a>Runtime Function: <em>double</em> <strong>__fractutadf</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractqiqq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqihq"></a>Runtime Function: <em>fract</em> <strong>__fractqihq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqisq"></a>Runtime Function: <em>long fract</em> <strong>__fractqisq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractqidq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiha"></a>Runtime Function: <em>short accum</em> <strong>__fractqiha</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqisa"></a>Runtime Function: <em>accum</em> <strong>__fractqisa</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqida"></a>Runtime Function: <em>long accum</em> <strong>__fractqida</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqita"></a>Runtime Function: <em>long long accum</em> <strong>__fractqita</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractqiuqq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractqiuhq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractqiusq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractqiudq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractqiuha</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractqiusa</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractqiuda</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractqiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractqiuta</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiqq"></a>Runtime Function: <em>short fract</em> <strong>__fracthiqq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthihq"></a>Runtime Function: <em>fract</em> <strong>__fracthihq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthisq"></a>Runtime Function: <em>long fract</em> <strong>__fracthisq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthidq"></a>Runtime Function: <em>long long fract</em> <strong>__fracthidq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiha"></a>Runtime Function: <em>short accum</em> <strong>__fracthiha</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthisa"></a>Runtime Function: <em>accum</em> <strong>__fracthisa</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthida"></a>Runtime Function: <em>long accum</em> <strong>__fracthida</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthita"></a>Runtime Function: <em>long long accum</em> <strong>__fracthita</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fracthiuqq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fracthiuhq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fracthiusq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fracthiudq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fracthiuha</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fracthiusa</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fracthiuda</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracthiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fracthiuta</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractsiqq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsihq"></a>Runtime Function: <em>fract</em> <strong>__fractsihq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsisq"></a>Runtime Function: <em>long fract</em> <strong>__fractsisq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractsidq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiha"></a>Runtime Function: <em>short accum</em> <strong>__fractsiha</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsisa"></a>Runtime Function: <em>accum</em> <strong>__fractsisa</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsida"></a>Runtime Function: <em>long accum</em> <strong>__fractsida</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsita"></a>Runtime Function: <em>long long accum</em> <strong>__fractsita</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractsiuqq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractsiuhq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractsiusq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractsiudq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractsiuha</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractsiusa</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractsiuda</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractsiuta</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractdiqq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdihq"></a>Runtime Function: <em>fract</em> <strong>__fractdihq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdisq"></a>Runtime Function: <em>long fract</em> <strong>__fractdisq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractdidq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiha"></a>Runtime Function: <em>short accum</em> <strong>__fractdiha</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdisa"></a>Runtime Function: <em>accum</em> <strong>__fractdisa</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdida"></a>Runtime Function: <em>long accum</em> <strong>__fractdida</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdita"></a>Runtime Function: <em>long long accum</em> <strong>__fractdita</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractdiuqq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractdiuhq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractdiusq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractdiudq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractdiuha</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractdiusa</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractdiuda</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractdiuta</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiqq"></a>Runtime Function: <em>short fract</em> <strong>__fracttiqq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttihq"></a>Runtime Function: <em>fract</em> <strong>__fracttihq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttisq"></a>Runtime Function: <em>long fract</em> <strong>__fracttisq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttidq"></a>Runtime Function: <em>long long fract</em> <strong>__fracttidq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiha"></a>Runtime Function: <em>short accum</em> <strong>__fracttiha</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttisa"></a>Runtime Function: <em>accum</em> <strong>__fracttisa</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttida"></a>Runtime Function: <em>long accum</em> <strong>__fracttida</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttita"></a>Runtime Function: <em>long long accum</em> <strong>__fracttita</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fracttiuqq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fracttiuhq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fracttiusq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fracttiudq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fracttiuha</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fracttiusa</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fracttiuda</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffracttiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fracttiuta</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfqq"></a>Runtime Function: <em>short fract</em> <strong>__fractsfqq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfhq"></a>Runtime Function: <em>fract</em> <strong>__fractsfhq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfsq"></a>Runtime Function: <em>long fract</em> <strong>__fractsfsq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractsfdq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfha"></a>Runtime Function: <em>short accum</em> <strong>__fractsfha</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfsa"></a>Runtime Function: <em>accum</em> <strong>__fractsfsa</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfda"></a>Runtime Function: <em>long accum</em> <strong>__fractsfda</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfta"></a>Runtime Function: <em>long long accum</em> <strong>__fractsfta</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractsfuqq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractsfuhq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractsfusq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractsfudq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractsfuha</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractsfusa</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractsfuda</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractsfuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractsfuta</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfqq"></a>Runtime Function: <em>short fract</em> <strong>__fractdfqq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfhq"></a>Runtime Function: <em>fract</em> <strong>__fractdfhq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfsq"></a>Runtime Function: <em>long fract</em> <strong>__fractdfsq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfdq"></a>Runtime Function: <em>long long fract</em> <strong>__fractdfdq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfha"></a>Runtime Function: <em>short accum</em> <strong>__fractdfha</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfsa"></a>Runtime Function: <em>accum</em> <strong>__fractdfsa</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfda"></a>Runtime Function: <em>long accum</em> <strong>__fractdfda</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfta"></a>Runtime Function: <em>long long accum</em> <strong>__fractdfta</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractdfuqq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractdfuhq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractdfusq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractdfudq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractdfuha</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractdfusa</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractdfuda</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractdfuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractdfuta</strong> <em>(double <var>a</var>)</em></dt>
<dd><p>These functions convert from fractional and signed non-fractionals to
fractionals and signed non-fractionals, without saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsatfractqqhq2"></a>Runtime Function: <em>fract</em> <strong>__satfractqqhq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqsq2"></a>Runtime Function: <em>long fract</em> <strong>__satfractqqsq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__satfractqqdq2</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractqqha</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractqqsa</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractqqda</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractqqta</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractqquqq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractqquhq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractqqusq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractqqudq</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractqquha</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractqqusa</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractqquda</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractqquta</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqqq2"></a>Runtime Function: <em>short fract</em> <strong>__satfracthqqq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqsq2"></a>Runtime Function: <em>long fract</em> <strong>__satfracthqsq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__satfracthqdq2</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqha"></a>Runtime Function: <em>short accum</em> <strong>__satfracthqha</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqsa"></a>Runtime Function: <em>accum</em> <strong>__satfracthqsa</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqda"></a>Runtime Function: <em>long accum</em> <strong>__satfracthqda</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfracthqta</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfracthquqq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfracthquhq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfracthqusq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfracthqudq</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfracthquha</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfracthqusa</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfracthquda</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfracthquta</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqqq2"></a>Runtime Function: <em>short fract</em> <strong>__satfractsqqq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqhq2"></a>Runtime Function: <em>fract</em> <strong>__satfractsqhq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqdq2"></a>Runtime Function: <em>long long fract</em> <strong>__satfractsqdq2</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractsqha</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractsqsa</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractsqda</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractsqta</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractsquqq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractsquhq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractsqusq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractsqudq</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractsquha</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractsqusa</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractsquda</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractsquta</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqqq2"></a>Runtime Function: <em>short fract</em> <strong>__satfractdqqq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqhq2"></a>Runtime Function: <em>fract</em> <strong>__satfractdqhq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqsq2"></a>Runtime Function: <em>long fract</em> <strong>__satfractdqsq2</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractdqha</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractdqsa</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractdqda</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractdqta</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdquqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractdquqq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdquhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractdquhq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractdqusq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractdqudq</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractdquha</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractdqusa</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractdquda</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractdquta</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfracthaqq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthahq"></a>Runtime Function: <em>fract</em> <strong>__satfracthahq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthasq"></a>Runtime Function: <em>long fract</em> <strong>__satfracthasq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfracthadq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthasa2"></a>Runtime Function: <em>accum</em> <strong>__satfracthasa2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthada2"></a>Runtime Function: <em>long accum</em> <strong>__satfracthada2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthata2"></a>Runtime Function: <em>long long accum</em> <strong>__satfracthata2</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfracthauqq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfracthauhq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfracthausq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfracthaudq</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfracthauha</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfracthausa</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfracthauda</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfracthauta</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractsaqq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsahq"></a>Runtime Function: <em>fract</em> <strong>__satfractsahq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractsasq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractsadq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsaha2"></a>Runtime Function: <em>short accum</em> <strong>__satfractsaha2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsada2"></a>Runtime Function: <em>long accum</em> <strong>__satfractsada2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsata2"></a>Runtime Function: <em>long long accum</em> <strong>__satfractsata2</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractsauqq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractsauhq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractsausq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractsaudq</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractsauha</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractsausa</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractsauda</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractsauta</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractdaqq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdahq"></a>Runtime Function: <em>fract</em> <strong>__satfractdahq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractdasq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractdadq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdaha2"></a>Runtime Function: <em>short accum</em> <strong>__satfractdaha2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdasa2"></a>Runtime Function: <em>accum</em> <strong>__satfractdasa2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdata2"></a>Runtime Function: <em>long long accum</em> <strong>__satfractdata2</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractdauqq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractdauhq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractdausq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractdaudq</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractdauha</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractdausa</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractdauda</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractdauta</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfracttaqq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttahq"></a>Runtime Function: <em>fract</em> <strong>__satfracttahq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttasq"></a>Runtime Function: <em>long fract</em> <strong>__satfracttasq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfracttadq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttaha2"></a>Runtime Function: <em>short accum</em> <strong>__satfracttaha2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttasa2"></a>Runtime Function: <em>accum</em> <strong>__satfracttasa2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttada2"></a>Runtime Function: <em>long accum</em> <strong>__satfracttada2</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfracttauqq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfracttauhq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfracttausq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfracttaudq</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttauha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfracttauha</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttausa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfracttausa</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttauda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfracttauda</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttauta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfracttauta</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractuqqqq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqhq"></a>Runtime Function: <em>fract</em> <strong>__satfractuqqhq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractuqqsq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractuqqdq</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractuqqha</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractuqqsa</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractuqqda</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractuqqta</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractuqquhq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractuqqusq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractuqqudq2</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractuqquha</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractuqqusa</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractuqquda</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuqquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractuqquta</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractuhqqq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqhq"></a>Runtime Function: <em>fract</em> <strong>__satfractuhqhq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractuhqsq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractuhqdq</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractuhqha</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractuhqsa</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractuhqda</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractuhqta</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractuhquqq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractuhqusq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractuhqudq2</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractuhquha</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractuhqusa</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractuhquda</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractuhquta</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractusqqq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqhq"></a>Runtime Function: <em>fract</em> <strong>__satfractusqhq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractusqsq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractusqdq</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractusqha</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractusqsa</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractusqda</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractusqta</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractusquqq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractusquhq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqudq2"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractusqudq2</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractusquha</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractusqusa</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractusquda</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractusquta</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractudqqq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqhq"></a>Runtime Function: <em>fract</em> <strong>__satfractudqhq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractudqsq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractudqdq</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqha"></a>Runtime Function: <em>short accum</em> <strong>__satfractudqha</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqsa"></a>Runtime Function: <em>accum</em> <strong>__satfractudqsa</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqda"></a>Runtime Function: <em>long accum</em> <strong>__satfractudqda</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractudqta</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudquqq2"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractudquqq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudquhq2"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractudquhq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqusq2"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractudqusq2</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudquha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractudquha</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudqusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractudqusa</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudquda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractudquda</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudquta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractudquta</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractuhaqq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhahq"></a>Runtime Function: <em>fract</em> <strong>__satfractuhahq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractuhasq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractuhadq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhaha"></a>Runtime Function: <em>short accum</em> <strong>__satfractuhaha</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhasa"></a>Runtime Function: <em>accum</em> <strong>__satfractuhasa</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhada"></a>Runtime Function: <em>long accum</em> <strong>__satfractuhada</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhata"></a>Runtime Function: <em>long long accum</em> <strong>__satfractuhata</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractuhauqq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractuhauhq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractuhausq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractuhaudq</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractuhausa2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractuhauda2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractuhauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractuhauta2</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractusaqq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusahq"></a>Runtime Function: <em>fract</em> <strong>__satfractusahq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractusasq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractusadq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusaha"></a>Runtime Function: <em>short accum</em> <strong>__satfractusaha</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusasa"></a>Runtime Function: <em>accum</em> <strong>__satfractusasa</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusada"></a>Runtime Function: <em>long accum</em> <strong>__satfractusada</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusata"></a>Runtime Function: <em>long long accum</em> <strong>__satfractusata</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractusauqq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractusauhq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractusausq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractusaudq</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractusauha2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractusauda2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractusauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractusauta2</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractudaqq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudahq"></a>Runtime Function: <em>fract</em> <strong>__satfractudahq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractudasq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractudadq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudaha"></a>Runtime Function: <em>short accum</em> <strong>__satfractudaha</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudasa"></a>Runtime Function: <em>accum</em> <strong>__satfractudasa</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudada"></a>Runtime Function: <em>long accum</em> <strong>__satfractudada</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudata"></a>Runtime Function: <em>long long accum</em> <strong>__satfractudata</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractudauqq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractudauhq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractudausq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractudaudq</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractudauha2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractudausa2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractudauta2"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractudauta2</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutaqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractutaqq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutahq"></a>Runtime Function: <em>fract</em> <strong>__satfractutahq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutasq"></a>Runtime Function: <em>long fract</em> <strong>__satfractutasq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutadq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractutadq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutaha"></a>Runtime Function: <em>short accum</em> <strong>__satfractutaha</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutasa"></a>Runtime Function: <em>accum</em> <strong>__satfractutasa</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutada"></a>Runtime Function: <em>long accum</em> <strong>__satfractutada</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutata"></a>Runtime Function: <em>long long accum</em> <strong>__satfractutata</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutauqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractutauqq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutauhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractutauhq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutausq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractutausq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutaudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractutaudq</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutauha2"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractutauha2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutausa2"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractutausa2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractutauda2"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractutauda2</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractqiqq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqihq"></a>Runtime Function: <em>fract</em> <strong>__satfractqihq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractqisq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractqidq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractqiha</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqisa"></a>Runtime Function: <em>accum</em> <strong>__satfractqisa</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqida"></a>Runtime Function: <em>long accum</em> <strong>__satfractqida</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractqita</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractqiuqq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractqiuhq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractqiusq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractqiudq</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractqiuha</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractqiusa</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractqiuda</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractqiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractqiuta</strong> <em>(signed char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfracthiqq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthihq"></a>Runtime Function: <em>fract</em> <strong>__satfracthihq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthisq"></a>Runtime Function: <em>long fract</em> <strong>__satfracthisq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfracthidq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiha"></a>Runtime Function: <em>short accum</em> <strong>__satfracthiha</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthisa"></a>Runtime Function: <em>accum</em> <strong>__satfracthisa</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthida"></a>Runtime Function: <em>long accum</em> <strong>__satfracthida</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthita"></a>Runtime Function: <em>long long accum</em> <strong>__satfracthita</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfracthiuqq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfracthiuhq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfracthiusq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfracthiudq</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfracthiuha</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfracthiusa</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfracthiuda</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracthiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfracthiuta</strong> <em>(short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractsiqq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsihq"></a>Runtime Function: <em>fract</em> <strong>__satfractsihq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractsisq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractsidq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractsiha</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsisa"></a>Runtime Function: <em>accum</em> <strong>__satfractsisa</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsida"></a>Runtime Function: <em>long accum</em> <strong>__satfractsida</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractsita</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractsiuqq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractsiuhq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractsiusq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractsiudq</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractsiuha</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractsiusa</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractsiuda</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractsiuta</strong> <em>(int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractdiqq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdihq"></a>Runtime Function: <em>fract</em> <strong>__satfractdihq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractdisq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractdidq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractdiha</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdisa"></a>Runtime Function: <em>accum</em> <strong>__satfractdisa</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdida"></a>Runtime Function: <em>long accum</em> <strong>__satfractdida</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractdita</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractdiuqq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractdiuhq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractdiusq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractdiudq</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractdiuha</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractdiusa</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractdiuda</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractdiuta</strong> <em>(long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfracttiqq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttihq"></a>Runtime Function: <em>fract</em> <strong>__satfracttihq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttisq"></a>Runtime Function: <em>long fract</em> <strong>__satfracttisq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfracttidq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiha"></a>Runtime Function: <em>short accum</em> <strong>__satfracttiha</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttisa"></a>Runtime Function: <em>accum</em> <strong>__satfracttisa</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttida"></a>Runtime Function: <em>long accum</em> <strong>__satfracttida</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttita"></a>Runtime Function: <em>long long accum</em> <strong>__satfracttita</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfracttiuqq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfracttiuhq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfracttiusq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfracttiudq</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfracttiuha</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfracttiusa</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfracttiuda</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfracttiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfracttiuta</strong> <em>(long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractsfqq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfhq"></a>Runtime Function: <em>fract</em> <strong>__satfractsfhq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractsfsq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractsfdq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfha"></a>Runtime Function: <em>short accum</em> <strong>__satfractsfha</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfsa"></a>Runtime Function: <em>accum</em> <strong>__satfractsfsa</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfda"></a>Runtime Function: <em>long accum</em> <strong>__satfractsfda</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractsfta</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractsfuqq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractsfuhq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractsfusq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractsfudq</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractsfuha</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractsfusa</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractsfuda</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractsfuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractsfuta</strong> <em>(float <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractdfqq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfhq"></a>Runtime Function: <em>fract</em> <strong>__satfractdfhq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfsq"></a>Runtime Function: <em>long fract</em> <strong>__satfractdfsq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfdq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractdfdq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfha"></a>Runtime Function: <em>short accum</em> <strong>__satfractdfha</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfsa"></a>Runtime Function: <em>accum</em> <strong>__satfractdfsa</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfda"></a>Runtime Function: <em>long accum</em> <strong>__satfractdfda</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfta"></a>Runtime Function: <em>long long accum</em> <strong>__satfractdfta</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractdfuqq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractdfuhq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractdfusq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractdfudq</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractdfuha</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractdfusa</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractdfuda</strong> <em>(double <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractdfuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractdfuta</strong> <em>(double <var>a</var>)</em></dt>
<dd><p>The functions convert from fractional and signed non-fractionals to
fractionals, with saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005ffractunsqqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsqqqi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsqqhi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsqqsi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsqqdi</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsqqti</strong> <em>(short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunshqqi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunshqhi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunshqsi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunshqdi</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunshqti</strong> <em>(fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunssqqi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunssqhi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunssqsi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunssqdi</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunssqti</strong> <em>(long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsdqqi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsdqhi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsdqsi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsdqdi</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsdqti</strong> <em>(long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunshaqi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunshahi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunshasi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunshadi</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunshati</strong> <em>(short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunssaqi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunssahi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunssasi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunssadi</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunssati</strong> <em>(accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsdaqi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsdahi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsdasi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsdadi</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsdati</strong> <em>(long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunstaqi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunstahi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunstasi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunstadi</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunstati</strong> <em>(long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuqqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsuqqqi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuqqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsuqqhi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuqqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsuqqsi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuqqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsuqqdi</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuqqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsuqqti</strong> <em>(unsigned short fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsuhqqi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsuhqhi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsuhqsi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsuhqdi</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsuhqti</strong> <em>(unsigned fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsusqqi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsusqhi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsusqsi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsusqdi</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsusqti</strong> <em>(unsigned long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudqqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsudqqi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudqhi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsudqhi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudqsi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsudqsi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudqdi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsudqdi</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudqti"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsudqti</strong> <em>(unsigned long long fract <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsuhaqi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsuhahi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsuhasi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsuhadi</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsuhati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsuhati</strong> <em>(unsigned short accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsusaqi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsusahi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsusasi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsusadi</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsusati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsusati</strong> <em>(unsigned accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsudaqi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsudahi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsudasi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsudadi</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsudati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsudati</strong> <em>(unsigned long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsutaqi"></a>Runtime Function: <em>unsigned char</em> <strong>__fractunsutaqi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsutahi"></a>Runtime Function: <em>unsigned short</em> <strong>__fractunsutahi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsutasi"></a>Runtime Function: <em>unsigned int</em> <strong>__fractunsutasi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsutadi"></a>Runtime Function: <em>unsigned long</em> <strong>__fractunsutadi</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsutati"></a>Runtime Function: <em>unsigned long long</em> <strong>__fractunsutati</strong> <em>(unsigned long long accum <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractunsqiqq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqihq"></a>Runtime Function: <em>fract</em> <strong>__fractunsqihq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqisq"></a>Runtime Function: <em>long fract</em> <strong>__fractunsqisq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractunsqidq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiha"></a>Runtime Function: <em>short accum</em> <strong>__fractunsqiha</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqisa"></a>Runtime Function: <em>accum</em> <strong>__fractunsqisa</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqida"></a>Runtime Function: <em>long accum</em> <strong>__fractunsqida</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqita"></a>Runtime Function: <em>long long accum</em> <strong>__fractunsqita</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractunsqiuqq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractunsqiuhq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractunsqiusq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractunsqiudq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractunsqiuha</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractunsqiusa</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractunsqiuda</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsqiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractunsqiuta</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractunshiqq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshihq"></a>Runtime Function: <em>fract</em> <strong>__fractunshihq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshisq"></a>Runtime Function: <em>long fract</em> <strong>__fractunshisq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractunshidq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiha"></a>Runtime Function: <em>short accum</em> <strong>__fractunshiha</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshisa"></a>Runtime Function: <em>accum</em> <strong>__fractunshisa</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshida"></a>Runtime Function: <em>long accum</em> <strong>__fractunshida</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshita"></a>Runtime Function: <em>long long accum</em> <strong>__fractunshita</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractunshiuqq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractunshiuhq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractunshiusq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractunshiudq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractunshiuha</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractunshiusa</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractunshiuda</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunshiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractunshiuta</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractunssiqq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssihq"></a>Runtime Function: <em>fract</em> <strong>__fractunssihq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssisq"></a>Runtime Function: <em>long fract</em> <strong>__fractunssisq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractunssidq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiha"></a>Runtime Function: <em>short accum</em> <strong>__fractunssiha</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssisa"></a>Runtime Function: <em>accum</em> <strong>__fractunssisa</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssida"></a>Runtime Function: <em>long accum</em> <strong>__fractunssida</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssita"></a>Runtime Function: <em>long long accum</em> <strong>__fractunssita</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractunssiuqq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractunssiuhq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractunssiusq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractunssiudq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractunssiuha</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractunssiusa</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractunssiuda</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunssiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractunssiuta</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractunsdiqq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdihq"></a>Runtime Function: <em>fract</em> <strong>__fractunsdihq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdisq"></a>Runtime Function: <em>long fract</em> <strong>__fractunsdisq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractunsdidq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiha"></a>Runtime Function: <em>short accum</em> <strong>__fractunsdiha</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdisa"></a>Runtime Function: <em>accum</em> <strong>__fractunsdisa</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdida"></a>Runtime Function: <em>long accum</em> <strong>__fractunsdida</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdita"></a>Runtime Function: <em>long long accum</em> <strong>__fractunsdita</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractunsdiuqq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractunsdiuhq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractunsdiusq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractunsdiudq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractunsdiuha</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractunsdiusa</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractunsdiuda</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunsdiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractunsdiuta</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiqq"></a>Runtime Function: <em>short fract</em> <strong>__fractunstiqq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstihq"></a>Runtime Function: <em>fract</em> <strong>__fractunstihq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstisq"></a>Runtime Function: <em>long fract</em> <strong>__fractunstisq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstidq"></a>Runtime Function: <em>long long fract</em> <strong>__fractunstidq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiha"></a>Runtime Function: <em>short accum</em> <strong>__fractunstiha</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstisa"></a>Runtime Function: <em>accum</em> <strong>__fractunstisa</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstida"></a>Runtime Function: <em>long accum</em> <strong>__fractunstida</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstita"></a>Runtime Function: <em>long long accum</em> <strong>__fractunstita</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__fractunstiuqq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__fractunstiuhq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__fractunstiusq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__fractunstiudq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__fractunstiuha</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__fractunstiusa</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__fractunstiuda</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005ffractunstiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__fractunstiuta</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dd><p>These functions convert from fractionals to unsigned non-fractionals;
and from unsigned non-fractionals to fractionals, without saturation.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fsatfractunsqiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractunsqiqq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqihq"></a>Runtime Function: <em>fract</em> <strong>__satfractunsqihq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractunsqisq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractunsqidq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractunsqiha</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqisa"></a>Runtime Function: <em>accum</em> <strong>__satfractunsqisa</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqida"></a>Runtime Function: <em>long accum</em> <strong>__satfractunsqida</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractunsqita</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractunsqiuqq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractunsqiuhq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractunsqiusq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractunsqiudq</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractunsqiuha</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractunsqiusa</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractunsqiuda</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsqiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractunsqiuta</strong> <em>(unsigned char <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractunshiqq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshihq"></a>Runtime Function: <em>fract</em> <strong>__satfractunshihq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractunshisq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractunshidq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractunshiha</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshisa"></a>Runtime Function: <em>accum</em> <strong>__satfractunshisa</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshida"></a>Runtime Function: <em>long accum</em> <strong>__satfractunshida</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractunshita</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractunshiuqq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractunshiuhq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractunshiusq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractunshiudq</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractunshiuha</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractunshiusa</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractunshiuda</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunshiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractunshiuta</strong> <em>(unsigned short <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractunssiqq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssihq"></a>Runtime Function: <em>fract</em> <strong>__satfractunssihq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractunssisq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractunssidq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractunssiha</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssisa"></a>Runtime Function: <em>accum</em> <strong>__satfractunssisa</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssida"></a>Runtime Function: <em>long accum</em> <strong>__satfractunssida</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractunssita</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractunssiuqq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractunssiuhq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractunssiusq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractunssiudq</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractunssiuha</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractunssiusa</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractunssiuda</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunssiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractunssiuta</strong> <em>(unsigned int <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractunsdiqq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdihq"></a>Runtime Function: <em>fract</em> <strong>__satfractunsdihq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractunsdisq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractunsdidq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractunsdiha</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdisa"></a>Runtime Function: <em>accum</em> <strong>__satfractunsdisa</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdida"></a>Runtime Function: <em>long accum</em> <strong>__satfractunsdida</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractunsdita</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractunsdiuqq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractunsdiuhq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractunsdiusq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractunsdiudq</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractunsdiuha</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractunsdiusa</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractunsdiuda</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunsdiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractunsdiuta</strong> <em>(unsigned long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiqq"></a>Runtime Function: <em>short fract</em> <strong>__satfractunstiqq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstihq"></a>Runtime Function: <em>fract</em> <strong>__satfractunstihq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstisq"></a>Runtime Function: <em>long fract</em> <strong>__satfractunstisq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstidq"></a>Runtime Function: <em>long long fract</em> <strong>__satfractunstidq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiha"></a>Runtime Function: <em>short accum</em> <strong>__satfractunstiha</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstisa"></a>Runtime Function: <em>accum</em> <strong>__satfractunstisa</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstida"></a>Runtime Function: <em>long accum</em> <strong>__satfractunstida</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstita"></a>Runtime Function: <em>long long accum</em> <strong>__satfractunstita</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiuqq"></a>Runtime Function: <em>unsigned short fract</em> <strong>__satfractunstiuqq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiuhq"></a>Runtime Function: <em>unsigned fract</em> <strong>__satfractunstiuhq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiusq"></a>Runtime Function: <em>unsigned long fract</em> <strong>__satfractunstiusq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiudq"></a>Runtime Function: <em>unsigned long long fract</em> <strong>__satfractunstiudq</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiuha"></a>Runtime Function: <em>unsigned short accum</em> <strong>__satfractunstiuha</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiusa"></a>Runtime Function: <em>unsigned accum</em> <strong>__satfractunstiusa</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiuda"></a>Runtime Function: <em>unsigned long accum</em> <strong>__satfractunstiuda</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dt><a name="index-_005f_005fsatfractunstiuta"></a>Runtime Function: <em>unsigned long long accum</em> <strong>__satfractunstiuta</strong> <em>(unsigned long long <var>a</var>)</em></dt>
<dd><p>These functions convert from unsigned non-fractionals to fractionals,
with saturation.
</p></dd></dl>

<hr>
<a name="Exception-handling-routines"></a>
<div class="header">
<p>
Next: <a href="#Miscellaneous-routines" accesskey="n" rel="next">Miscellaneous routines</a>, Previous: <a href="#Fixed_002dpoint-fractional-library-routines" accesskey="p" rel="prev">Fixed-point fractional library routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Language_002dindependent-routines-for-exception-handling"></a>
<h3 class="section">4.5 Language-independent routines for exception handling</h3>

<p>document me!
</p>
<div class="smallexample">
<pre class="smallexample">  _Unwind_DeleteException
  _Unwind_Find_FDE
  _Unwind_ForcedUnwind
  _Unwind_GetGR
  _Unwind_GetIP
  _Unwind_GetLanguageSpecificData
  _Unwind_GetRegionStart
  _Unwind_GetTextRelBase
  _Unwind_GetDataRelBase
  _Unwind_RaiseException
  _Unwind_Resume
  _Unwind_SetGR
  _Unwind_SetIP
  _Unwind_FindEnclosingFunction
  _Unwind_SjLj_Register
  _Unwind_SjLj_Unregister
  _Unwind_SjLj_RaiseException
  _Unwind_SjLj_ForcedUnwind
  _Unwind_SjLj_Resume
  __deregister_frame
  __deregister_frame_info
  __deregister_frame_info_bases
  __register_frame
  __register_frame_info
  __register_frame_info_bases
  __register_frame_info_table
  __register_frame_info_table_bases
  __register_frame_table
</pre></div>

<hr>
<a name="Miscellaneous-routines"></a>
<div class="header">
<p>
Previous: <a href="#Exception-handling-routines" accesskey="p" rel="prev">Exception handling routines</a>, Up: <a href="#Libgcc" accesskey="u" rel="up">Libgcc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Miscellaneous-runtime-library-routines"></a>
<h3 class="section">4.6 Miscellaneous runtime library routines</h3>

<a name="Cache-control-functions"></a>
<h4 class="subsection">4.6.1 Cache control functions</h4>
<dl>
<dt><a name="index-_005f_005fclear_005fcache"></a>Runtime Function: <em>void</em> <strong>__clear_cache</strong> <em>(char *<var>beg</var>, char *<var>end</var>)</em></dt>
<dd><p>This function clears the instruction cache between <var>beg</var> and <var>end</var>.
</p></dd></dl>

<a name="Split-stack-functions-and-variables"></a>
<h4 class="subsection">4.6.2 Split stack functions and variables</h4>
<dl>
<dt><a name="index-_005f_005fsplitstack_005ffind"></a>Runtime Function: <em>void *</em> <strong>__splitstack_find</strong> <em>(void *<var>segment_arg</var>, void *<var>sp</var>, size_t <var>len</var>, void **<var>next_segment</var>, void **<var>next_sp</var>, void **<var>initial_sp</var>)</em></dt>
<dd><p>When using <samp>-fsplit-stack</samp>, this call may be used to iterate
over the stack segments.  It may be called like this:
</p><div class="smallexample">
<pre class="smallexample">  void *next_segment = NULL;
  void *next_sp = NULL;
  void *initial_sp = NULL;
  void *stack;
  size_t stack_size;
  while ((stack = __splitstack_find (next_segment, next_sp,
                                     &amp;stack_size, &amp;next_segment,
                                     &amp;next_sp, &amp;initial_sp))
         != NULL)
    {
      /* Stack segment starts at stack and is
         stack_size bytes long.  */
    }
</pre></div>

<p>There is no way to iterate over the stack segments of a different
thread.  However, what is permitted is for one thread to call this
with the <var>segment_arg</var> and <var>sp</var> arguments NULL, to pass
<var>next_segment</var>, <var>next_sp</var>, and <var>initial_sp</var> to a different
thread, and then to suspend one way or another.  A different thread
may run the subsequent <code>__splitstack_find</code> iterations.  Of
course, this will only work if the first thread is suspended while the
second thread is calling <code>__splitstack_find</code>.  If not, the second
thread could be looking at the stack while it is changing, and
anything could happen.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fmorestack_005fsegments"></a>Variable: <strong>__morestack_segments</strong></dt>
<dt><a name="index-_005f_005fmorestack_005fcurrent_005fsegment"></a>Variable: <strong>__morestack_current_segment</strong></dt>
<dt><a name="index-_005f_005fmorestack_005finitial_005fsp"></a>Variable: <strong>__morestack_initial_sp</strong></dt>
<dd><p>Internal variables used by the <samp>-fsplit-stack</samp> implementation.
</p></dd></dl>

<hr>
<a name="Languages"></a>
<div class="header">
<p>
Next: <a href="#Source-Tree" accesskey="n" rel="next">Source Tree</a>, Previous: <a href="#Libgcc" accesskey="p" rel="prev">Libgcc</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Language-Front-Ends-in-GCC"></a>
<h2 class="chapter">5 Language Front Ends in GCC</h2>

<p>The interface to front ends for languages in GCC, and in particular
the <code>tree</code> structure (see <a href="#GENERIC">GENERIC</a>), was initially designed for
C, and many aspects of it are still somewhat biased towards C and
C-like languages.  It is, however, reasonably well suited to other
procedural languages, and front ends for many such languages have been
written for GCC.
</p>
<p>Writing a compiler as a front end for GCC, rather than compiling
directly to assembler or generating C code which is then compiled by
GCC, has several advantages:
</p>
<ul>
<li> GCC front ends benefit from the support for many different
target machines already present in GCC.
</li><li> GCC front ends benefit from all the optimizations in GCC.  Some
of these, such as alias analysis, may work better when GCC is
compiling directly from source code then when it is compiling from
generated C code.
</li><li> Better debugging information is generated when compiling
directly from source code than when going via intermediate generated C
code.
</li></ul>

<p>Because of the advantages of writing a compiler as a GCC front end,
GCC front ends have also been created for languages very different
from those for which GCC was designed, such as the declarative
logic/functional language Mercury.  For these reasons, it may also be
useful to implement compilers created for specialized purposes (for
example, as part of a research project) as GCC front ends.
</p>
<hr>
<a name="Source-Tree"></a>
<div class="header">
<p>
Next: <a href="#Testsuites" accesskey="n" rel="next">Testsuites</a>, Previous: <a href="#Languages" accesskey="p" rel="prev">Languages</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Source-Tree-Structure-and-Build-System"></a>
<h2 class="chapter">6 Source Tree Structure and Build System</h2>

<p>This chapter describes the structure of the GCC source tree, and how
GCC is built.  The user documentation for building and installing GCC
is in a separate manual (<a href="http://gcc.gnu.org/install/">http://gcc.gnu.org/install/</a>), with
which it is presumed that you are familiar.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Configure-Terms" accesskey="1">Configure Terms</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Configuration terminology and history.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Top-Level" accesskey="2">Top Level</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The top level source directory.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#gcc-Directory" accesskey="3">gcc Directory</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The <samp>gcc</samp> subdirectory.
</td></tr>
</table>


<hr>
<a name="Configure-Terms"></a>
<div class="header">
<p>
Next: <a href="#Top-Level" accesskey="n" rel="next">Top Level</a>, Up: <a href="#Source-Tree" accesskey="u" rel="up">Source Tree</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Configure-Terms-and-History"></a>
<h3 class="section">6.1 Configure Terms and History</h3>
<a name="index-configure-terms"></a>
<a name="index-canadian"></a>

<p>The configure and build process has a long and colorful history, and can
be confusing to anyone who doesn&rsquo;t know why things are the way they are.
While there are other documents which describe the configuration process
in detail, here are a few things that everyone working on GCC should
know.
</p>
<p>There are three system names that the build knows about: the machine you
are building on (<em>build</em>), the machine that you are building for
(<em>host</em>), and the machine that GCC will produce code for
(<em>target</em>).  When you configure GCC, you specify these with
<samp>--build=</samp>, <samp>--host=</samp>, and <samp>--target=</samp>.
</p>
<p>Specifying the host without specifying the build should be avoided, as
<code>configure</code> may (and once did) assume that the host you specify
is also the build, which may not be true.
</p>
<p>If build, host, and target are all the same, this is called a
<em>native</em>.  If build and host are the same but target is different,
this is called a <em>cross</em>.  If build, host, and target are all
different this is called a <em>canadian</em> (for obscure reasons dealing
with Canada&rsquo;s political party and the background of the person working
on the build at that time).  If host and target are the same, but build
is different, you are using a cross-compiler to build a native for a
different system.  Some people call this a <em>host-x-host</em>,
<em>crossed native</em>, or <em>cross-built native</em>.  If build and target
are the same, but host is different, you are using a cross compiler to
build a cross compiler that produces code for the machine you&rsquo;re
building on.  This is rare, so there is no common way of describing it.
There is a proposal to call this a <em>crossback</em>.
</p>
<p>If build and host are the same, the GCC you are building will also be
used to build the target libraries (like <code>libstdc++</code>).  If build and host
are different, you must have already built and installed a cross
compiler that will be used to build the target libraries (if you
configured with <samp>--target=foo-bar</samp>, this compiler will be called
<code>foo-bar-gcc</code>).
</p>
<p>In the case of target libraries, the machine you&rsquo;re building for is the
machine you specified with <samp>--target</samp>.  So, build is the machine
you&rsquo;re building on (no change there), host is the machine you&rsquo;re
building for (the target libraries are built for the target, so host is
the target you specified), and target doesn&rsquo;t apply (because you&rsquo;re not
building a compiler, you&rsquo;re building libraries).  The configure/make
process will adjust these variables as needed.  It also sets
<code>$with_cross_host</code> to the original <samp>--host</samp> value in case you
need it.
</p>
<p>The <code>libiberty</code> support library is built up to three times: once
for the host, once for the target (even if they are the same), and once
for the build if build and host are different.  This allows it to be
used by all programs which are generated in the course of the build
process.
</p>
<hr>
<a name="Top-Level"></a>
<div class="header">
<p>
Next: <a href="#gcc-Directory" accesskey="n" rel="next">gcc Directory</a>, Previous: <a href="#Configure-Terms" accesskey="p" rel="prev">Configure Terms</a>, Up: <a href="#Source-Tree" accesskey="u" rel="up">Source Tree</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Top-Level-Source-Directory"></a>
<h3 class="section">6.2 Top Level Source Directory</h3>

<p>The top level source directory in a GCC distribution contains several
files and directories that are shared with other software
distributions such as that of GNU Binutils.  It also contains several
subdirectories that contain parts of GCC and its runtime libraries:
</p>
<dl compact="compact">
<dt><samp>boehm-gc</samp></dt>
<dd><p>The Boehm conservative garbage collector, used as part of the Java
runtime library.
</p>
</dd>
<dt><samp>config</samp></dt>
<dd><p>Autoconf macros and Makefile fragments used throughout the tree.
</p>
</dd>
<dt><samp>contrib</samp></dt>
<dd><p>Contributed scripts that may be found useful in conjunction with GCC.
One of these, <samp>contrib/texi2pod.pl</samp>, is used to generate man
pages from Texinfo manuals as part of the GCC build process.
</p>
</dd>
<dt><samp>fixincludes</samp></dt>
<dd><p>The support for fixing system headers to work with GCC.  See
<samp>fixincludes/README</samp> for more information.  The headers fixed by
this mechanism are installed in <samp><var>libsubdir</var>/include-fixed</samp>.
Along with those headers, <samp>README-fixinc</samp> is also installed, as
<samp><var>libsubdir</var>/include-fixed/README</samp>.
</p>
</dd>
<dt><samp>gcc</samp></dt>
<dd><p>The main sources of GCC itself (except for runtime libraries),
including optimizers, support for different target architectures,
language front ends, and testsuites.  See <a href="#gcc-Directory">The
<samp>gcc</samp> Subdirectory</a>, for details.
</p>
</dd>
<dt><samp>gnattools</samp></dt>
<dd><p>Support tools for GNAT.
</p>
</dd>
<dt><samp>include</samp></dt>
<dd><p>Headers for the <code>libiberty</code> library.
</p>
</dd>
<dt><samp>intl</samp></dt>
<dd><p>GNU <code>libintl</code>, from GNU <code>gettext</code>, for systems which do not
include it in <code>libc</code>.
</p>
</dd>
<dt><samp>libada</samp></dt>
<dd><p>The Ada runtime library.
</p>
</dd>
<dt><samp>libatomic</samp></dt>
<dd><p>The runtime support library for atomic operations (e.g. for <code>__sync</code>
and <code>__atomic</code>).
</p>
</dd>
<dt><samp>libcpp</samp></dt>
<dd><p>The C preprocessor library.
</p>
</dd>
<dt><samp>libdecnumber</samp></dt>
<dd><p>The Decimal Float support library.
</p>
</dd>
<dt><samp>libffi</samp></dt>
<dd><p>The <code>libffi</code> library, used as part of the Java runtime library.
</p>
</dd>
<dt><samp>libgcc</samp></dt>
<dd><p>The GCC runtime library.
</p>
</dd>
<dt><samp>libgfortran</samp></dt>
<dd><p>The Fortran runtime library.
</p>
</dd>
<dt><samp>libgo</samp></dt>
<dd><p>The Go runtime library.  The bulk of this library is mirrored from the
<a href="http://code.google.com/p/go/">master Go repository</a>.
</p>
</dd>
<dt><samp>libgomp</samp></dt>
<dd><p>The GNU OpenMP runtime library.
</p>
</dd>
<dt><samp>libiberty</samp></dt>
<dd><p>The <code>libiberty</code> library, used for portability and for some
generally useful data structures and algorithms.  See <a href="http://gcc.gnu.org/onlinedocs/libiberty/index.html#Top">Introduction</a> in <cite><small>GNU</small> libiberty</cite>, for more information
about this library.
</p>
</dd>
<dt><samp>libitm</samp></dt>
<dd><p>The runtime support library for transactional memory.
</p>
</dd>
<dt><samp>libjava</samp></dt>
<dd><p>The Java runtime library.
</p>
</dd>
<dt><samp>libmudflap</samp></dt>
<dd><p>The <code>libmudflap</code> library, used for instrumenting pointer and array
dereferencing operations.
</p>
</dd>
<dt><samp>libobjc</samp></dt>
<dd><p>The Objective-C and Objective-C++ runtime library.
</p>
</dd>
<dt><samp>libquadmath</samp></dt>
<dd><p>The runtime support library for quad-precision math operations.
</p>
</dd>
<dt><samp>libphobos</samp></dt>
<dd><p>The D standard runtime library.
</p>
</dd>
<dt><samp>libssp</samp></dt>
<dd><p>The Stack protector runtime library.
</p>
</dd>
<dt><samp>libstdc++-v3</samp></dt>
<dd><p>The C++ runtime library.
</p>
</dd>
<dt><samp>lto-plugin</samp></dt>
<dd><p>Plugin used by <code>gold</code> if link-time optimizations are enabled.
</p>
</dd>
<dt><samp>maintainer-scripts</samp></dt>
<dd><p>Scripts used by the <code>gccadmin</code> account on <code>gcc.gnu.org</code>.
</p>
</dd>
<dt><samp>zlib</samp></dt>
<dd><p>The <code>zlib</code> compression library, used by the Java front end, as
part of the Java runtime library, and for compressing and uncompressing
GCC&rsquo;s intermediate language in LTO object files.
</p></dd>
</dl>

<p>The build system in the top level directory, including how recursion
into subdirectories works and how building runtime libraries for
multilibs is handled, is documented in a separate manual, included
with GNU Binutils.  See <a href="configure.html#Top">GNU configure and build system</a> in <cite>The GNU configure and build system</cite>, for details.
</p>
<hr>
<a name="gcc-Directory"></a>
<div class="header">
<p>
Previous: <a href="#Top-Level" accesskey="p" rel="prev">Top Level</a>, Up: <a href="#Source-Tree" accesskey="u" rel="up">Source Tree</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-gcc-Subdirectory"></a>
<h3 class="section">6.3 The <samp>gcc</samp> Subdirectory</h3>

<p>The <samp>gcc</samp> directory contains many files that are part of the C
sources of GCC, other files used as part of the configuration and
build process, and subdirectories including documentation and a
testsuite.  The files that are sources of GCC are documented in a
separate chapter.  See <a href="#Passes">Passes and Files of the Compiler</a>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Subdirectories" accesskey="1">Subdirectories</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Subdirectories of <samp>gcc</samp>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Configuration" accesskey="2">Configuration</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The configuration process, and the files it uses.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Build" accesskey="3">Build</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The build system in the <samp>gcc</samp> directory.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Makefile" accesskey="4">Makefile</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Targets in <samp>gcc/Makefile</samp>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Library-Files" accesskey="5">Library Files</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Library source files and headers under <samp>gcc/</samp>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Headers" accesskey="6">Headers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Headers installed by GCC.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Documentation" accesskey="7">Documentation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Building documentation in GCC.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Front-End" accesskey="8">Front End</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Anatomy of a language front end.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Back-End" accesskey="9">Back End</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Anatomy of a target back end.
</td></tr>
</table>

<hr>
<a name="Subdirectories"></a>
<div class="header">
<p>
Next: <a href="#Configuration" accesskey="n" rel="next">Configuration</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Subdirectories-of-gcc"></a>
<h4 class="subsection">6.3.1 Subdirectories of <samp>gcc</samp></h4>

<p>The <samp>gcc</samp> directory contains the following subdirectories:
</p>
<dl compact="compact">
<dt><samp><var>language</var></samp></dt>
<dd><p>Subdirectories for various languages.  Directories containing a file
<samp>config-lang.in</samp> are language subdirectories.  The contents of
the subdirectories <samp>c</samp> (for C), <samp>cp</samp> (for C++),
<samp>objc</samp> (for Objective-C), <samp>objcp</samp> (for Objective-C++),
and <samp>lto</samp> (for LTO) are documented in this
manual (see <a href="#Passes">Passes and Files of the Compiler</a>);
those for other languages are not.  See <a href="#Front-End">Anatomy of a Language Front End</a>, for details of the files in these
directories.
</p>
</dd>
<dt><samp>common</samp></dt>
<dd><p>Source files shared between the compiler drivers (such as
<code>gcc</code>) and the compilers proper (such as <samp>cc1</samp>).  If an
architecture defines target hooks shared between those places, it also
has a subdirectory in <samp>common/config</samp>.  See <a href="#Target-Structure">Target Structure</a>.
</p>
</dd>
<dt><samp>config</samp></dt>
<dd><p>Configuration files for supported architectures and operating
systems.  See <a href="#Back-End">Anatomy of a Target Back End</a>, for
details of the files in this directory.
</p>
</dd>
<dt><samp>doc</samp></dt>
<dd><p>Texinfo documentation for GCC, together with automatically generated
man pages and support for converting the installation manual to
HTML.  See <a href="#Documentation">Documentation</a>.
</p>
</dd>
<dt><samp>ginclude</samp></dt>
<dd><p>System headers installed by GCC, mainly those required by the C
standard of freestanding implementations.  See <a href="#Headers">Headers
Installed by GCC</a>, for details of when these and other headers are
installed.
</p>
</dd>
<dt><samp>po</samp></dt>
<dd><p>Message catalogs with translations of messages produced by GCC into
various languages, <samp><var>language</var>.po</samp>.  This directory also
contains <samp>gcc.pot</samp>, the template for these message catalogues,
<samp>exgettext</samp>, a wrapper around <code>gettext</code> to extract the
messages from the GCC sources and create <samp>gcc.pot</samp>, which is run
by &lsquo;<samp>make gcc.pot</samp>&rsquo;, and <samp>EXCLUDES</samp>, a list of files from
which messages should not be extracted.
</p>
</dd>
<dt><samp>testsuite</samp></dt>
<dd><p>The GCC testsuites (except for those for runtime libraries).
See <a href="#Testsuites">Testsuites</a>.
</p></dd>
</dl>

<hr>
<a name="Configuration"></a>
<div class="header">
<p>
Next: <a href="#Build" accesskey="n" rel="next">Build</a>, Previous: <a href="#Subdirectories" accesskey="p" rel="prev">Subdirectories</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Configuration-in-the-gcc-Directory"></a>
<h4 class="subsection">6.3.2 Configuration in the <samp>gcc</samp> Directory</h4>

<p>The <samp>gcc</samp> directory is configured with an Autoconf-generated
script <samp>configure</samp>.  The <samp>configure</samp> script is generated
from <samp>configure.ac</samp> and <samp>aclocal.m4</samp>.  From the files
<samp>configure.ac</samp> and <samp>acconfig.h</samp>, Autoheader generates the
file <samp>config.in</samp>.  The file <samp>cstamp-h.in</samp> is used as a
timestamp.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Config-Fragments" accesskey="1">Config Fragments</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Scripts used by <samp>configure</samp>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#System-Config" accesskey="2">System Config</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The <samp>config.build</samp>, <samp>config.host</samp>, and
                         <samp>config.gcc</samp> files.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Configuration-Files" accesskey="3">Configuration Files</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Files created by running <samp>configure</samp>.
</td></tr>
</table>

<hr>
<a name="Config-Fragments"></a>
<div class="header">
<p>
Next: <a href="#System-Config" accesskey="n" rel="next">System Config</a>, Up: <a href="#Configuration" accesskey="u" rel="up">Configuration</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Scripts-Used-by-configure"></a>
<h4 class="subsubsection">6.3.2.1 Scripts Used by <samp>configure</samp></h4>

<p><samp>configure</samp> uses some other scripts to help in its work:
</p>
<ul>
<li> The standard GNU <samp>config.sub</samp> and <samp>config.guess</samp>
files, kept in the top level directory, are used.

</li><li> The file <samp>config.gcc</samp> is used to handle configuration
specific to the particular target machine.  The file
<samp>config.build</samp> is used to handle configuration specific to the
particular build machine.  The file <samp>config.host</samp> is used to handle
configuration specific to the particular host machine.  (In general,
these should only be used for features that cannot reasonably be tested in
Autoconf feature tests.)
See <a href="#System-Config">The <samp>config.build</samp>; <samp>config.host</samp>;
and <samp>config.gcc</samp> Files</a>, for details of the contents of these files.

</li><li> Each language subdirectory has a file
<samp><var>language</var>/config-lang.in</samp> that is used for
front-end-specific configuration.  See <a href="#Front-End-Config">The Front
End <samp>config-lang.in</samp> File</a>, for details of this file.

</li><li> A helper script <samp>configure.frag</samp> is used as part of
creating the output of <samp>configure</samp>.
</li></ul>

<hr>
<a name="System-Config"></a>
<div class="header">
<p>
Next: <a href="#Configuration-Files" accesskey="n" rel="next">Configuration Files</a>, Previous: <a href="#Config-Fragments" accesskey="p" rel="prev">Config Fragments</a>, Up: <a href="#Configuration" accesskey="u" rel="up">Configuration</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-config_002ebuild_003b-config_002ehost_003b-and-config_002egcc-Files"></a>
<h4 class="subsubsection">6.3.2.2 The <samp>config.build</samp>; <samp>config.host</samp>; and <samp>config.gcc</samp> Files</h4>

<p>The <samp>config.build</samp> file contains specific rules for particular systems
which GCC is built on.  This should be used as rarely as possible, as the
behavior of the build system can always be detected by autoconf.
</p>
<p>The <samp>config.host</samp> file contains specific rules for particular systems
which GCC will run on.  This is rarely needed.
</p>
<p>The <samp>config.gcc</samp> file contains specific rules for particular systems
which GCC will generate code for.  This is usually needed.
</p>
<p>Each file has a list of the shell variables it sets, with descriptions, at the
top of the file.
</p>
<p>FIXME: document the contents of these files, and what variables should
be set to control build, host and target configuration.
</p>

<hr>
<a name="Configuration-Files"></a>
<div class="header">
<p>
Previous: <a href="#System-Config" accesskey="p" rel="prev">System Config</a>, Up: <a href="#Configuration" accesskey="u" rel="up">Configuration</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Files-Created-by-configure"></a>
<h4 class="subsubsection">6.3.2.3 Files Created by <code>configure</code></h4>

<p>Here we spell out what files will be set up by <samp>configure</samp> in the
<samp>gcc</samp> directory.  Some other files are created as temporary files
in the configuration process, and are not used in the subsequent
build; these are not documented.
</p>
<ul>
<li> <samp>Makefile</samp> is constructed from <samp>Makefile.in</samp>, together with
the host and target fragments (see <a href="#Fragments">Makefile
Fragments</a>) <samp>t-<var>target</var></samp> and <samp>x-<var>host</var></samp> from
<samp>config</samp>, if any, and language Makefile fragments
<samp><var>language</var>/Make-lang.in</samp>.
</li><li> <samp>auto-host.h</samp> contains information about the host machine
determined by <samp>configure</samp>.  If the host machine is different from
the build machine, then <samp>auto-build.h</samp> is also created,
containing such information about the build machine.
</li><li> <samp>config.status</samp> is a script that may be run to recreate the
current configuration.
</li><li> <samp>configargs.h</samp> is a header containing details of the arguments
passed to <samp>configure</samp> to configure GCC, and of the thread model
used.
</li><li> <samp>cstamp-h</samp> is used as a timestamp.
</li><li> If a language <samp>config-lang.in</samp> file (see <a href="#Front-End-Config">The Front End <samp>config-lang.in</samp> File</a>) sets <code>outputs</code>, then
the files listed in <code>outputs</code> there are also generated.
</li></ul>

<p>The following configuration headers are created from the Makefile,
using <samp>mkconfig.sh</samp>, rather than directly by <samp>configure</samp>.
<samp>config.h</samp>, <samp>bconfig.h</samp> and <samp>tconfig.h</samp> all contain the
<samp>xm-<var>machine</var>.h</samp> header, if any, appropriate to the host,
build and target machines respectively, the configuration headers for
the target, and some definitions; for the host and build machines,
these include the autoconfigured headers generated by
<samp>configure</samp>.  The other configuration headers are determined by
<samp>config.gcc</samp>.  They also contain the typedefs for <code>rtx</code>,
<code>rtvec</code> and <code>tree</code>.
</p>
<ul>
<li> <samp>config.h</samp>, for use in programs that run on the host machine.
</li><li> <samp>bconfig.h</samp>, for use in programs that run on the build machine.
</li><li> <samp>tconfig.h</samp>, for use in programs and libraries for the target
machine.
</li><li> <samp>tm_p.h</samp>, which includes the header <samp><var>machine</var>-protos.h</samp>
that contains prototypes for functions in the target
<samp><var>machine</var>.c</samp> file.  The header <samp><var>machine</var>-protos.h</samp>
can include prototypes of functions that use rtl and tree data
structures inside appropriate <code>#ifdef RTX_CODE</code> and <code>#ifdef
TREE_CODE</code> conditional code segements.  The
<samp><var>machine</var>-protos.h</samp> is included after the <samp>rtl.h</samp>
and/or <samp>tree.h</samp> would have been included.  The <samp>tm_p.h</samp> also
includes the header <samp>tm-preds.h</samp> which is generated by
<samp>genpreds</samp> program during the build to define the declarations
and inline functions for the predicate functions.
</li></ul>

<hr>
<a name="Build"></a>
<div class="header">
<p>
Next: <a href="#Makefile" accesskey="n" rel="next">Makefile</a>, Previous: <a href="#Configuration" accesskey="p" rel="prev">Configuration</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Build-System-in-the-gcc-Directory"></a>
<h4 class="subsection">6.3.3 Build System in the <samp>gcc</samp> Directory</h4>

<p>FIXME: describe the build system, including what is built in what
stages.  Also list the various source files that are used in the build
process but aren&rsquo;t source files of GCC itself and so aren&rsquo;t documented
below (see <a href="#Passes">Passes</a>).
</p>

<hr>
<a name="Makefile"></a>
<div class="header">
<p>
Next: <a href="#Library-Files" accesskey="n" rel="next">Library Files</a>, Previous: <a href="#Build" accesskey="p" rel="prev">Build</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Makefile-Targets"></a>
<h4 class="subsection">6.3.4 Makefile Targets</h4>
<a name="index-makefile-targets"></a>
<a name="index-targets_002c-makefile"></a>

<p>These targets are available from the &lsquo;<samp>gcc</samp>&rsquo; directory:
</p>
<dl compact="compact">
<dt><code>all</code></dt>
<dd><p>This is the default target.  Depending on what your build/host/target
configuration is, it coordinates all the things that need to be built.
</p>
</dd>
<dt><code>doc</code></dt>
<dd><p>Produce info-formatted documentation and man pages.  Essentially it
calls &lsquo;<samp>make man</samp>&rsquo; and &lsquo;<samp>make info</samp>&rsquo;.
</p>
</dd>
<dt><code>dvi</code></dt>
<dd><p>Produce DVI-formatted documentation.
</p>
</dd>
<dt><code>pdf</code></dt>
<dd><p>Produce PDF-formatted documentation.
</p>
</dd>
<dt><code>html</code></dt>
<dd><p>Produce HTML-formatted documentation.
</p>
</dd>
<dt><code>man</code></dt>
<dd><p>Generate man pages.
</p>
</dd>
<dt><code>info</code></dt>
<dd><p>Generate info-formatted pages.
</p>
</dd>
<dt><code>mostlyclean</code></dt>
<dd><p>Delete the files made while building the compiler.
</p>
</dd>
<dt><code>clean</code></dt>
<dd><p>That, and all the other files built by &lsquo;<samp>make all</samp>&rsquo;.
</p>
</dd>
<dt><code>distclean</code></dt>
<dd><p>That, and all the files created by <code>configure</code>.
</p>
</dd>
<dt><code>maintainer-clean</code></dt>
<dd><p>Distclean plus any file that can be generated from other files.  Note
that additional tools may be required beyond what is normally needed to
build GCC.
</p>
</dd>
<dt><code>srcextra</code></dt>
<dd><p>Generates files in the source directory that are not version-controlled but
should go into a release tarball.
</p>
</dd>
<dt><code>srcinfo</code></dt>
<dt><code>srcman</code></dt>
<dd><p>Copies the info-formatted and manpage documentation into the source
directory usually for the purpose of generating a release tarball.
</p>
</dd>
<dt><code>install</code></dt>
<dd><p>Installs GCC.
</p>
</dd>
<dt><code>uninstall</code></dt>
<dd><p>Deletes installed files, though this is not supported.
</p>
</dd>
<dt><code>check</code></dt>
<dd><p>Run the testsuite.  This creates a <samp>testsuite</samp> subdirectory that
has various <samp>.sum</samp> and <samp>.log</samp> files containing the results of
the testing.  You can run subsets with, for example, &lsquo;<samp>make check-gcc</samp>&rsquo;.
You can specify specific tests by setting <code>RUNTESTFLAGS</code> to be the name
of the <samp>.exp</samp> file, optionally followed by (for some tests) an equals
and a file wildcard, like:
</p>
<div class="smallexample">
<pre class="smallexample">make check-gcc RUNTESTFLAGS=&quot;execute.exp=19980413-*&quot;
</pre></div>

<p>Note that running the testsuite may require additional tools be
installed, such as Tcl or DejaGnu.
</p></dd>
</dl>

<p>The toplevel tree from which you start GCC compilation is not
the GCC directory, but rather a complex Makefile that coordinates
the various steps of the build, including bootstrapping the compiler
and using the new compiler to build target libraries.
</p>
<p>When GCC is configured for a native configuration, the default action
for <code>make</code> is to do a full three-stage bootstrap.  This means
that GCC is built three times&mdash;once with the native compiler, once with
the native-built compiler it just built, and once with the compiler it
built the second time.  In theory, the last two should produce the same
results, which &lsquo;<samp>make compare</samp>&rsquo; can check.  Each stage is configured
separately and compiled into a separate directory, to minimize problems
due to ABI incompatibilities between the native compiler and GCC.
</p>
<p>If you do a change, rebuilding will also start from the first stage
and &ldquo;bubble&rdquo; up the change through the three stages.  Each stage
is taken from its build directory (if it had been built previously),
rebuilt, and copied to its subdirectory.  This will allow you to, for
example, continue a bootstrap after fixing a bug which causes the
stage2 build to crash.  It does not provide as good coverage of the
compiler as bootstrapping from scratch, but it ensures that the new
code is syntactically correct (e.g., that you did not use GCC extensions
by mistake), and avoids spurious bootstrap comparison
failures<a name="DOCF1" href="#FOOT1"><sup>1</sup></a>.
</p>
<p>Other targets available from the top level include:
</p>
<dl compact="compact">
<dt><code>bootstrap-lean</code></dt>
<dd><p>Like <code>bootstrap</code>, except that the various stages are removed once
they&rsquo;re no longer needed.  This saves disk space.
</p>
</dd>
<dt><code>bootstrap2</code></dt>
<dt><code>bootstrap2-lean</code></dt>
<dd><p>Performs only the first two stages of bootstrap.  Unlike a three-stage
bootstrap, this does not perform a comparison to test that the compiler
is running properly.  Note that the disk space required by a &ldquo;lean&rdquo;
bootstrap is approximately independent of the number of stages.
</p>
</dd>
<dt><code>stage<var>N</var>-bubble (<var>N</var> = 1&hellip;4, profile, feedback)</code></dt>
<dd><p>Rebuild all the stages up to <var>N</var>, with the appropriate flags,
&ldquo;bubbling&rdquo; the changes as described above.
</p>
</dd>
<dt><code>all-stage<var>N</var> (<var>N</var> = 1&hellip;4, profile, feedback)</code></dt>
<dd><p>Assuming that stage <var>N</var> has already been built, rebuild it with the
appropriate flags.  This is rarely needed.
</p>
</dd>
<dt><code>cleanstrap</code></dt>
<dd><p>Remove everything (&lsquo;<samp>make clean</samp>&rsquo;) and rebuilds (&lsquo;<samp>make bootstrap</samp>&rsquo;).
</p>
</dd>
<dt><code>compare</code></dt>
<dd><p>Compares the results of stages 2 and 3.  This ensures that the compiler
is running properly, since it should produce the same object files
regardless of how it itself was compiled.
</p>
</dd>
<dt><code>profiledbootstrap</code></dt>
<dd><p>Builds a compiler with profiling feedback information.  In this case,
the second and third stages are named &lsquo;<samp>profile</samp>&rsquo; and &lsquo;<samp>feedback</samp>&rsquo;,
respectively.  For more information, see
<a href="gccinstall-4.html#Building">Building with profile feedback</a> in <cite>Installing GCC</cite>.
</p>
</dd>
<dt><code>restrap</code></dt>
<dd><p>Restart a bootstrap, so that everything that was not built with
the system compiler is rebuilt.
</p>
</dd>
<dt><code>stage<var>N</var>-start (<var>N</var> = 1&hellip;4, profile, feedback)</code></dt>
<dd><p>For each package that is bootstrapped, rename directories so that,
for example, <samp>gcc</samp> points to the stage<var>N</var> GCC, compiled
with the stage<var>N-1</var> GCC<a name="DOCF2" href="#FOOT2"><sup>2</sup></a>.
</p>
<p>You will invoke this target if you need to test or debug the
stage<var>N</var> GCC.  If you only need to execute GCC (but you need
not run &lsquo;<samp>make</samp>&rsquo; either to rebuild it or to run test suites),
you should be able to work directly in the <samp>stage<var>N</var>-gcc</samp>
directory.  This makes it easier to debug multiple stages in
parallel.
</p>
</dd>
<dt><code>stage</code></dt>
<dd><p>For each package that is bootstrapped, relocate its build directory
to indicate its stage.  For example, if the <samp>gcc</samp> directory
points to the stage2 GCC, after invoking this target it will be
renamed to <samp>stage2-gcc</samp>.
</p>
</dd>
</dl>

<p>If you wish to use non-default GCC flags when compiling the stage2 and
stage3 compilers, set <code>BOOT_CFLAGS</code> on the command line when doing
&lsquo;<samp>make</samp>&rsquo;.
</p>
<p>Usually, the first stage only builds the languages that the compiler
is written in: typically, C and maybe Ada.  If you are debugging a
miscompilation of a different stage2 front-end (for example, of the
Fortran front-end), you may want to have front-ends for other languages
in the first stage as well.  To do so, set <code>STAGE1_LANGUAGES</code>
on the command line when doing &lsquo;<samp>make</samp>&rsquo;.
</p>
<p>For example, in the aforementioned scenario of debugging a Fortran
front-end miscompilation caused by the stage1 compiler, you may need a
command like
</p>
<div class="example">
<pre class="example">make stage2-bubble STAGE1_LANGUAGES=c,fortran
</pre></div>

<p>Alternatively, you can use per-language targets to build and test
languages that are not enabled by default in stage1.  For example,
<code>make f951</code> will build a Fortran compiler even in the stage1
build directory.
</p>

<hr>
<a name="Library-Files"></a>
<div class="header">
<p>
Next: <a href="#Headers" accesskey="n" rel="next">Headers</a>, Previous: <a href="#Makefile" accesskey="p" rel="prev">Makefile</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Library-Source-Files-and-Headers-under-the-gcc-Directory"></a>
<h4 class="subsection">6.3.5 Library Source Files and Headers under the <samp>gcc</samp> Directory</h4>

<p>FIXME: list here, with explanation, all the C source files and headers
under the <samp>gcc</samp> directory that aren&rsquo;t built into the GCC
executable but rather are part of runtime libraries and object files,
such as <samp>crtstuff.c</samp> and <samp>unwind-dw2.c</samp>.  See <a href="#Headers">Headers Installed by GCC</a>, for more information about the
<samp>ginclude</samp> directory.
</p>
<hr>
<a name="Headers"></a>
<div class="header">
<p>
Next: <a href="#Documentation" accesskey="n" rel="next">Documentation</a>, Previous: <a href="#Library-Files" accesskey="p" rel="prev">Library Files</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Headers-Installed-by-GCC"></a>
<h4 class="subsection">6.3.6 Headers Installed by GCC</h4>

<p>In general, GCC expects the system C library to provide most of the
headers to be used with it.  However, GCC will fix those headers if
necessary to make them work with GCC, and will install some headers
required of freestanding implementations.  These headers are installed
in <samp><var>libsubdir</var>/include</samp>.  Headers for non-C runtime
libraries are also installed by GCC; these are not documented here.
(FIXME: document them somewhere.)
</p>
<p>Several of the headers GCC installs are in the <samp>ginclude</samp>
directory.  These headers, <samp>iso646.h</samp>,
<samp>stdarg.h</samp>, <samp>stdbool.h</samp>, and <samp>stddef.h</samp>,
are installed in <samp><var>libsubdir</var>/include</samp>,
unless the target Makefile fragment (see <a href="#Target-Fragment">Target Fragment</a>)
overrides this by setting <code>USER_H</code>.
</p>
<p>In addition to these headers and those generated by fixing system
headers to work with GCC, some other headers may also be installed in
<samp><var>libsubdir</var>/include</samp>.  <samp>config.gcc</samp> may set
<code>extra_headers</code>; this specifies additional headers under
<samp>config</samp> to be installed on some systems.
</p>
<p>GCC installs its own version of <code>&lt;float.h&gt;</code>, from <samp>ginclude/float.h</samp>.
This is done to cope with command-line options that change the
representation of floating point numbers.
</p>
<p>GCC also installs its own version of <code>&lt;limits.h&gt;</code>; this is generated
from <samp>glimits.h</samp>, together with <samp>limitx.h</samp> and
<samp>limity.h</samp> if the system also has its own version of
<code>&lt;limits.h&gt;</code>.  (GCC provides its own header because it is
required of ISO C freestanding implementations, but needs to include
the system header from its own header as well because other standards
such as POSIX specify additional values to be defined in
<code>&lt;limits.h&gt;</code>.)  The system&rsquo;s <code>&lt;limits.h&gt;</code> header is used via
<samp><var>libsubdir</var>/include/syslimits.h</samp>, which is copied from
<samp>gsyslimits.h</samp> if it does not need fixing to work with GCC; if it
needs fixing, <samp>syslimits.h</samp> is the fixed copy.
</p>
<p>GCC can also install <code>&lt;tgmath.h&gt;</code>.  It will do this when
<samp>config.gcc</samp> sets <code>use_gcc_tgmath</code> to <code>yes</code>.
</p>
<hr>
<a name="Documentation"></a>
<div class="header">
<p>
Next: <a href="#Front-End" accesskey="n" rel="next">Front End</a>, Previous: <a href="#Headers" accesskey="p" rel="prev">Headers</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Building-Documentation"></a>
<h4 class="subsection">6.3.7 Building Documentation</h4>

<p>The main GCC documentation is in the form of manuals in Texinfo
format.  These are installed in Info format; DVI versions may be
generated by &lsquo;<samp>make dvi</samp>&rsquo;, PDF versions by &lsquo;<samp>make pdf</samp>&rsquo;, and
HTML versions by &lsquo;<samp>make html</samp>&rsquo;.  In addition, some man pages are
generated from the Texinfo manuals, there are some other text files
with miscellaneous documentation, and runtime libraries have their own
documentation outside the <samp>gcc</samp> directory.  FIXME: document the
documentation for runtime libraries somewhere.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Texinfo-Manuals" accesskey="1">Texinfo Manuals</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">GCC manuals in Texinfo format.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Man-Page-Generation" accesskey="2">Man Page Generation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating man pages from Texinfo manuals.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Miscellaneous-Docs" accesskey="3">Miscellaneous Docs</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Miscellaneous text files with documentation.
</td></tr>
</table>

<hr>
<a name="Texinfo-Manuals"></a>
<div class="header">
<p>
Next: <a href="#Man-Page-Generation" accesskey="n" rel="next">Man Page Generation</a>, Up: <a href="#Documentation" accesskey="u" rel="up">Documentation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Texinfo-Manuals-1"></a>
<h4 class="subsubsection">6.3.7.1 Texinfo Manuals</h4>

<p>The manuals for GCC as a whole, and the C and C++ front ends, are in
files <samp>doc/*.texi</samp>.  Other front ends have their own manuals in
files <samp><var>language</var>/*.texi</samp>.  Common files
<samp>doc/include/*.texi</samp> are provided which may be included in
multiple manuals; the following files are in <samp>doc/include</samp>:
</p>
<dl compact="compact">
<dt><samp>fdl.texi</samp></dt>
<dd><p>The GNU Free Documentation License.
</p></dd>
<dt><samp>funding.texi</samp></dt>
<dd><p>The section &ldquo;Funding Free Software&rdquo;.
</p></dd>
<dt><samp>gcc-common.texi</samp></dt>
<dd><p>Common definitions for manuals.
</p></dd>
<dt><samp>gpl_v3.texi</samp></dt>
<dd><p>The GNU General Public License.
</p></dd>
<dt><samp>texinfo.tex</samp></dt>
<dd><p>A copy of <samp>texinfo.tex</samp> known to work with the GCC manuals.
</p></dd>
</dl>

<p>DVI-formatted manuals are generated by &lsquo;<samp>make dvi</samp>&rsquo;, which uses
<code>texi2dvi</code> (via the Makefile macro <code>$(TEXI2DVI)</code>).
PDF-formatted manuals are generated by &lsquo;<samp>make pdf</samp>&rsquo;, which uses
<code>texi2pdf</code> (via the Makefile macro <code>$(TEXI2PDF)</code>).  HTML
formatted manuals are generated by &lsquo;<samp>make html</samp>&rsquo;.  Info
manuals are generated by &lsquo;<samp>make info</samp>&rsquo; (which is run as part of
a bootstrap); this generates the manuals in the source directory,
using <code>makeinfo</code> via the Makefile macro <code>$(MAKEINFO)</code>,
and they are included in release distributions.
</p>
<p>Manuals are also provided on the GCC web site, in both HTML and
PostScript forms.  This is done via the script
<samp>maintainer-scripts/update_web_docs_svn</samp>.  Each manual to be
provided online must be listed in the definition of <code>MANUALS</code> in
that file; a file <samp><var>name</var>.texi</samp> must only appear once in the
source tree, and the output manual must have the same name as the
source file.  (However, other Texinfo files, included in manuals but
not themselves the root files of manuals, may have names that appear
more than once in the source tree.)  The manual file
<samp><var>name</var>.texi</samp> should only include other files in its own
directory or in <samp>doc/include</samp>.  HTML manuals will be generated by
&lsquo;<samp>makeinfo --html</samp>&rsquo;, PostScript manuals by <code>texi2dvi</code>
and <code>dvips</code>, and PDF manuals by <code>texi2pdf</code>.
All Texinfo files that are parts of manuals must
be version-controlled, even if they are generated files, for the
generation of online manuals to work.
</p>
<p>The installation manual, <samp>doc/install.texi</samp>, is also provided on
the GCC web site.  The HTML version is generated by the script
<samp>doc/install.texi2html</samp>.
</p>
<hr>
<a name="Man-Page-Generation"></a>
<div class="header">
<p>
Next: <a href="#Miscellaneous-Docs" accesskey="n" rel="next">Miscellaneous Docs</a>, Previous: <a href="#Texinfo-Manuals" accesskey="p" rel="prev">Texinfo Manuals</a>, Up: <a href="#Documentation" accesskey="u" rel="up">Documentation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Man-Page-Generation-1"></a>
<h4 class="subsubsection">6.3.7.2 Man Page Generation</h4>

<p>Because of user demand, in addition to full Texinfo manuals, man pages
are provided which contain extracts from those manuals.  These man
pages are generated from the Texinfo manuals using
<samp>contrib/texi2pod.pl</samp> and <code>pod2man</code>.  (The man page for
<code>g++</code>, <samp>cp/g++.1</samp>, just contains a &lsquo;<samp>.so</samp>&rsquo; reference
to <samp>gcc.1</samp>, but all the other man pages are generated from
Texinfo manuals.)
</p>
<p>Because many systems may not have the necessary tools installed to
generate the man pages, they are only generated if the
<samp>configure</samp> script detects that recent enough tools are
installed, and the Makefiles allow generating man pages to fail
without aborting the build.  Man pages are also included in release
distributions.  They are generated in the source directory.
</p>
<p>Magic comments in Texinfo files starting &lsquo;<samp>@c man</samp>&rsquo; control what
parts of a Texinfo file go into a man page.  Only a subset of Texinfo
is supported by <samp>texi2pod.pl</samp>, and it may be necessary to add
support for more Texinfo features to this script when generating new
man pages.  To improve the man page output, some special Texinfo
macros are provided in <samp>doc/include/gcc-common.texi</samp> which
<samp>texi2pod.pl</samp> understands:
</p>
<dl compact="compact">
<dt><code>@gcctabopt</code></dt>
<dd><p>Use in the form &lsquo;<samp>@table @gcctabopt</samp>&rsquo; for tables of options,
where for printed output the effect of &lsquo;<samp>@code</samp>&rsquo; is better than
that of &lsquo;<samp>@option</samp>&rsquo; but for man page output a different effect is
wanted.
</p></dd>
<dt><code>@gccoptlist</code></dt>
<dd><p>Use for summary lists of options in manuals.
</p></dd>
<dt><code>@gol</code></dt>
<dd><p>Use at the end of each line inside &lsquo;<samp>@gccoptlist</samp>&rsquo;.  This is
necessary to avoid problems with differences in how the
&lsquo;<samp>@gccoptlist</samp>&rsquo; macro is handled by different Texinfo formatters.
</p></dd>
</dl>

<p>FIXME: describe the <samp>texi2pod.pl</samp> input language and magic
comments in more detail.
</p>
<hr>
<a name="Miscellaneous-Docs"></a>
<div class="header">
<p>
Previous: <a href="#Man-Page-Generation" accesskey="p" rel="prev">Man Page Generation</a>, Up: <a href="#Documentation" accesskey="u" rel="up">Documentation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Miscellaneous-Documentation"></a>
<h4 class="subsubsection">6.3.7.3 Miscellaneous Documentation</h4>

<p>In addition to the formal documentation that is installed by GCC,
there are several other text files in the <samp>gcc</samp> subdirectory
with miscellaneous documentation:
</p>
<dl compact="compact">
<dt><samp>ABOUT-GCC-NLS</samp></dt>
<dd><p>Notes on GCC&rsquo;s Native Language Support.  FIXME: this should be part of
this manual rather than a separate file.
</p></dd>
<dt><samp>ABOUT-NLS</samp></dt>
<dd><p>Notes on the Free Translation Project.
</p></dd>
<dt><samp>COPYING</samp></dt>
<dt><samp>COPYING3</samp></dt>
<dd><p>The GNU General Public License, Versions 2 and 3.
</p></dd>
<dt><samp>COPYING.LIB</samp></dt>
<dt><samp>COPYING3.LIB</samp></dt>
<dd><p>The GNU Lesser General Public License, Versions 2.1 and 3.
</p></dd>
<dt><samp>*ChangeLog*</samp></dt>
<dt><samp>*/ChangeLog*</samp></dt>
<dd><p>Change log files for various parts of GCC.
</p></dd>
<dt><samp>LANGUAGES</samp></dt>
<dd><p>Details of a few changes to the GCC front-end interface.  FIXME: the
information in this file should be part of general documentation of
the front-end interface in this manual.
</p></dd>
<dt><samp>ONEWS</samp></dt>
<dd><p>Information about new features in old versions of GCC.  (For recent
versions, the information is on the GCC web site.)
</p></dd>
<dt><samp>README.Portability</samp></dt>
<dd><p>Information about portability issues when writing code in GCC.  FIXME:
why isn&rsquo;t this part of this manual or of the GCC Coding Conventions?
</p></dd>
</dl>

<p>FIXME: document such files in subdirectories, at least <samp>config</samp>,
<samp>c</samp>, <samp>cp</samp>, <samp>objc</samp>, <samp>testsuite</samp>.
</p>
<hr>
<a name="Front-End"></a>
<div class="header">
<p>
Next: <a href="#Back-End" accesskey="n" rel="next">Back End</a>, Previous: <a href="#Documentation" accesskey="p" rel="prev">Documentation</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Anatomy-of-a-Language-Front-End"></a>
<h4 class="subsection">6.3.8 Anatomy of a Language Front End</h4>

<p>A front end for a language in GCC has the following parts:
</p>
<ul>
<li> A directory <samp><var>language</var></samp> under <samp>gcc</samp> containing source
files for that front end.  See <a href="#Front-End-Directory">The Front End
<samp><var>language</var></samp> Directory</a>, for details.
</li><li> A mention of the language in the list of supported languages in
<samp>gcc/doc/install.texi</samp>.
</li><li> A mention of the name under which the language&rsquo;s runtime library is
recognized by <samp>--enable-shared=<var>package</var></samp> in the
documentation of that option in <samp>gcc/doc/install.texi</samp>.
</li><li> A mention of any special prerequisites for building the front end in
the documentation of prerequisites in <samp>gcc/doc/install.texi</samp>.
</li><li> Details of contributors to that front end in
<samp>gcc/doc/contrib.texi</samp>.  If the details are in that front end&rsquo;s
own manual then there should be a link to that manual&rsquo;s list in
<samp>contrib.texi</samp>.
</li><li> Information about support for that language in
<samp>gcc/doc/frontends.texi</samp>.
</li><li> Information about standards for that language, and the front end&rsquo;s
support for them, in <samp>gcc/doc/standards.texi</samp>.  This may be a
link to such information in the front end&rsquo;s own manual.
</li><li> Details of source file suffixes for that language and <samp>-x
<var>lang</var></samp> options supported, in <samp>gcc/doc/invoke.texi</samp>.
</li><li> Entries in <code>default_compilers</code> in <samp>gcc.c</samp> for source file
suffixes for that language.
</li><li> Preferably testsuites, which may be under <samp>gcc/testsuite</samp> or
runtime library directories.  FIXME: document somewhere how to write
testsuite harnesses.
</li><li> Probably a runtime library for the language, outside the <samp>gcc</samp>
directory.  FIXME: document this further.
</li><li> Details of the directories of any runtime libraries in
<samp>gcc/doc/sourcebuild.texi</samp>.
</li><li> Check targets in <samp>Makefile.def</samp> for the top-level <samp>Makefile</samp>
to check just the compiler or the compiler and runtime library for the
language.
</li></ul>

<p>If the front end is added to the official GCC source repository, the
following are also necessary:
</p>
<ul>
<li> At least one Bugzilla component for bugs in that front end and runtime
libraries.  This category needs to be added to the Bugzilla database.
</li><li> Normally, one or more maintainers of that front end listed in
<samp>MAINTAINERS</samp>.
</li><li> Mentions on the GCC web site in <samp>index.html</samp> and
<samp>frontends.html</samp>, with any relevant links on
<samp>readings.html</samp>.  (Front ends that are not an official part of
GCC may also be listed on <samp>frontends.html</samp>, with relevant links.)
</li><li> A news item on <samp>index.html</samp>, and possibly an announcement on the
<a href="mailto:gcc-announce@gcc.gnu.org">gcc-announce@gcc.gnu.org</a> mailing list.
</li><li> The front end&rsquo;s manuals should be mentioned in
<samp>maintainer-scripts/update_web_docs_svn</samp> (see <a href="#Texinfo-Manuals">Texinfo Manuals</a>)
and the online manuals should be linked to from
<samp>onlinedocs/index.html</samp>.
</li><li> Any old releases or CVS repositories of the front end, before its
inclusion in GCC, should be made available on the GCC FTP site
<a href="ftp://gcc.gnu.org/pub/gcc/old-releases/">ftp://gcc.gnu.org/pub/gcc/old-releases/</a>.
</li><li> The release and snapshot script <samp>maintainer-scripts/gcc_release</samp>
should be updated to generate appropriate tarballs for this front end.
</li><li> If this front end includes its own version files that include the
current date, <samp>maintainer-scripts/update_version</samp> should be
updated accordingly.
</li></ul>

<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Front-End-Directory" accesskey="1">Front End Directory</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The front end <samp><var>language</var></samp> directory.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Front-End-Config" accesskey="2">Front End Config</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The front end <samp>config-lang.in</samp> file.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Front-End-Makefile" accesskey="3">Front End Makefile</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The front end <samp>Make-lang.in</samp> file.
</td></tr>
</table>

<hr>
<a name="Front-End-Directory"></a>
<div class="header">
<p>
Next: <a href="#Front-End-Config" accesskey="n" rel="next">Front End Config</a>, Up: <a href="#Front-End" accesskey="u" rel="up">Front End</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Front-End-language-Directory"></a>
<h4 class="subsubsection">6.3.8.1 The Front End <samp><var>language</var></samp> Directory</h4>

<p>A front end <samp><var>language</var></samp> directory contains the source files
of that front end (but not of any runtime libraries, which should be
outside the <samp>gcc</samp> directory).  This includes documentation, and
possibly some subsidiary programs built alongside the front end.
Certain files are special and other parts of the compiler depend on
their names:
</p>
<dl compact="compact">
<dt><samp>config-lang.in</samp></dt>
<dd><p>This file is required in all language subdirectories.  See <a href="#Front-End-Config">The Front End <samp>config-lang.in</samp> File</a>, for details of
its contents
</p></dd>
<dt><samp>Make-lang.in</samp></dt>
<dd><p>This file is required in all language subdirectories.  See <a href="#Front-End-Makefile">The Front End <samp>Make-lang.in</samp> File</a>, for details of its
contents.
</p></dd>
<dt><samp>lang.opt</samp></dt>
<dd><p>This file registers the set of switches that the front end accepts on
the command line, and their <samp>--help</samp> text.  See <a href="#Options">Options</a>.
</p></dd>
<dt><samp>lang-specs.h</samp></dt>
<dd><p>This file provides entries for <code>default_compilers</code> in
<samp>gcc.c</samp> which override the default of giving an error that a
compiler for that language is not installed.
</p></dd>
<dt><samp><var>language</var>-tree.def</samp></dt>
<dd><p>This file, which need not exist, defines any language-specific tree
codes.
</p></dd>
</dl>

<hr>
<a name="Front-End-Config"></a>
<div class="header">
<p>
Next: <a href="#Front-End-Makefile" accesskey="n" rel="next">Front End Makefile</a>, Previous: <a href="#Front-End-Directory" accesskey="p" rel="prev">Front End Directory</a>, Up: <a href="#Front-End" accesskey="u" rel="up">Front End</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Front-End-config_002dlang_002ein-File"></a>
<h4 class="subsubsection">6.3.8.2 The Front End <samp>config-lang.in</samp> File</h4>

<p>Each language subdirectory contains a <samp>config-lang.in</samp> file.
This file is a shell script that may define some variables describing
the language:
</p>
<dl compact="compact">
<dt><code>language</code></dt>
<dd><p>This definition must be present, and gives the name of the language
for some purposes such as arguments to <samp>--enable-languages</samp>.
</p></dd>
<dt><code>lang_requires</code></dt>
<dd><p>If defined, this variable lists (space-separated) language front ends
other than C that this front end requires to be enabled (with the
names given being their <code>language</code> settings).  For example, the
Java front end depends on the C++ front end, so sets
&lsquo;<samp>lang_requires=c++</samp>&rsquo;.
</p></dd>
<dt><code>subdir_requires</code></dt>
<dd><p>If defined, this variable lists (space-separated) front end directories
other than C that this front end requires to be present.  For example,
the Objective-C++ front end uses source files from the C++ and
Objective-C front ends, so sets &lsquo;<samp>subdir_requires=&quot;cp objc&quot;</samp>&rsquo;.
</p></dd>
<dt><code>target_libs</code></dt>
<dd><p>If defined, this variable lists (space-separated) targets in the top
level <samp>Makefile</samp> to build the runtime libraries for this
language, such as <code>target-libobjc</code>.
</p></dd>
<dt><code>lang_dirs</code></dt>
<dd><p>If defined, this variable lists (space-separated) top level
directories (parallel to <samp>gcc</samp>), apart from the runtime libraries,
that should not be configured if this front end is not built.
</p></dd>
<dt><code>build_by_default</code></dt>
<dd><p>If defined to &lsquo;<samp>no</samp>&rsquo;, this language front end is not built unless
enabled in a <samp>--enable-languages</samp> argument.  Otherwise, front
ends are built by default, subject to any special logic in
<samp>configure.ac</samp> (as is present to disable the Ada front end if the
Ada compiler is not already installed).
</p></dd>
<dt><code>boot_language</code></dt>
<dd><p>If defined to &lsquo;<samp>yes</samp>&rsquo;, this front end is built in stage1 of the
bootstrap.  This is only relevant to front ends written in their own
languages.
</p></dd>
<dt><code>compilers</code></dt>
<dd><p>If defined, a space-separated list of compiler executables that will
be run by the driver.  The names here will each end
with &lsquo;<samp>\$(exeext)</samp>&rsquo;.
</p></dd>
<dt><code>outputs</code></dt>
<dd><p>If defined, a space-separated list of files that should be generated
by <samp>configure</samp> substituting values in them.  This mechanism can
be used to create a file <samp><var>language</var>/Makefile</samp> from
<samp><var>language</var>/Makefile.in</samp>, but this is deprecated, building
everything from the single <samp>gcc/Makefile</samp> is preferred.
</p></dd>
<dt><code>gtfiles</code></dt>
<dd><p>If defined, a space-separated list of files that should be scanned by
<samp>gengtype.c</samp> to generate the garbage collection tables and routines for
this language.  This excludes the files that are common to all front
ends.  See <a href="#Type-Information">Type Information</a>.
</p>
</dd>
</dl>

<hr>
<a name="Front-End-Makefile"></a>
<div class="header">
<p>
Previous: <a href="#Front-End-Config" accesskey="p" rel="prev">Front End Config</a>, Up: <a href="#Front-End" accesskey="u" rel="up">Front End</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Front-End-Make_002dlang_002ein-File"></a>
<h4 class="subsubsection">6.3.8.3 The Front End <samp>Make-lang.in</samp> File</h4>

<p>Each language subdirectory contains a <samp>Make-lang.in</samp> file.  It contains
targets <code><var>lang</var>.<var>hook</var></code> (where <code><var>lang</var></code> is the
setting of <code>language</code> in <samp>config-lang.in</samp>) for the following
values of <code><var>hook</var></code>, and any other Makefile rules required to
build those targets (which may if necessary use other Makefiles
specified in <code>outputs</code> in <samp>config-lang.in</samp>, although this is
deprecated).  It also adds any testsuite targets that can use the
standard rule in <samp>gcc/Makefile.in</samp> to the variable
<code>lang_checks</code>.
</p>
<dl compact="compact">
<dt><code>all.cross</code></dt>
<dt><code>start.encap</code></dt>
<dt><code>rest.encap</code></dt>
<dd><p>FIXME: exactly what goes in each of these targets?
</p></dd>
<dt><code>tags</code></dt>
<dd><p>Build an <code>etags</code> <samp>TAGS</samp> file in the language subdirectory
in the source tree.
</p></dd>
<dt><code>info</code></dt>
<dd><p>Build info documentation for the front end, in the build directory.
This target is only called by &lsquo;<samp>make bootstrap</samp>&rsquo; if a suitable
version of <code>makeinfo</code> is available, so does not need to check
for this, and should fail if an error occurs.
</p></dd>
<dt><code>dvi</code></dt>
<dd><p>Build DVI documentation for the front end, in the build directory.
This should be done using <code>$(TEXI2DVI)</code>, with appropriate
<samp>-I</samp> arguments pointing to directories of included files.
</p></dd>
<dt><code>pdf</code></dt>
<dd><p>Build PDF documentation for the front end, in the build directory.
This should be done using <code>$(TEXI2PDF)</code>, with appropriate
<samp>-I</samp> arguments pointing to directories of included files.
</p></dd>
<dt><code>html</code></dt>
<dd><p>Build HTML documentation for the front end, in the build directory.
</p></dd>
<dt><code>man</code></dt>
<dd><p>Build generated man pages for the front end from Texinfo manuals
(see <a href="#Man-Page-Generation">Man Page Generation</a>), in the build directory.  This target
is only called if the necessary tools are available, but should ignore
errors so as not to stop the build if errors occur; man pages are
optional and the tools involved may be installed in a broken way.
</p></dd>
<dt><code>install-common</code></dt>
<dd><p>Install everything that is part of the front end, apart from the
compiler executables listed in <code>compilers</code> in
<samp>config-lang.in</samp>.
</p></dd>
<dt><code>install-info</code></dt>
<dd><p>Install info documentation for the front end, if it is present in the
source directory.  This target should have dependencies on info files
that should be installed.
</p></dd>
<dt><code>install-man</code></dt>
<dd><p>Install man pages for the front end.  This target should ignore
errors.
</p></dd>
<dt><code>install-plugin</code></dt>
<dd><p>Install headers needed for plugins.
</p></dd>
<dt><code>srcextra</code></dt>
<dd><p>Copies its dependencies into the source directory.  This generally should
be used for generated files such as Bison output files which are not
version-controlled, but should be included in any release tarballs.  This
target will be executed during a bootstrap if
&lsquo;<samp>--enable-generated-files-in-srcdir</samp>&rsquo; was specified as a
<samp>configure</samp> option.
</p></dd>
<dt><code>srcinfo</code></dt>
<dt><code>srcman</code></dt>
<dd><p>Copies its dependencies into the source directory.  These targets will be
executed during a bootstrap if &lsquo;<samp>--enable-generated-files-in-srcdir</samp>&rsquo;
was specified as a <samp>configure</samp> option.
</p></dd>
<dt><code>uninstall</code></dt>
<dd><p>Uninstall files installed by installing the compiler.  This is
currently documented not to be supported, so the hook need not do
anything.
</p></dd>
<dt><code>mostlyclean</code></dt>
<dt><code>clean</code></dt>
<dt><code>distclean</code></dt>
<dt><code>maintainer-clean</code></dt>
<dd><p>The language parts of the standard GNU
&lsquo;<samp>*clean</samp>&rsquo; targets.  See <a href="http://www.gnu.org/prep/standards/standards.html#Standard-Targets">Standard Targets for
Users</a> in <cite>GNU Coding Standards</cite>, for details of the standard
targets.  For GCC, <code>maintainer-clean</code> should delete
all generated files in the source directory that are not version-controlled,
but should not delete anything that is.
</p></dd>
</dl>

<p><samp>Make-lang.in</samp> must also define a variable <code><var>lang</var>_OBJS</code>
to a list of host object files that are used by that language.
</p>
<hr>
<a name="Back-End"></a>
<div class="header">
<p>
Previous: <a href="#Front-End" accesskey="p" rel="prev">Front End</a>, Up: <a href="#gcc-Directory" accesskey="u" rel="up">gcc Directory</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Anatomy-of-a-Target-Back-End"></a>
<h4 class="subsection">6.3.9 Anatomy of a Target Back End</h4>

<p>A back end for a target architecture in GCC has the following parts:
</p>
<ul>
<li> A directory <samp><var>machine</var></samp> under <samp>gcc/config</samp>, containing a
machine description <samp><var>machine</var>.md</samp> file (see <a href="#Machine-Desc">Machine Descriptions</a>), header files <samp><var>machine</var>.h</samp> and
<samp><var>machine</var>-protos.h</samp> and a source file <samp><var>machine</var>.c</samp>
(see <a href="#Target-Macros">Target Description Macros and Functions</a>),
possibly a target Makefile fragment <samp>t-<var>machine</var></samp>
(see <a href="#Target-Fragment">The Target Makefile Fragment</a>), and maybe
some other files.  The names of these files may be changed from the
defaults given by explicit specifications in <samp>config.gcc</samp>.
</li><li> If necessary, a file <samp><var>machine</var>-modes.def</samp> in the
<samp><var>machine</var></samp> directory, containing additional machine modes to
represent condition codes.  See <a href="#Condition-Code">Condition Code</a>, for further details.
</li><li> An optional <samp><var>machine</var>.opt</samp> file in the <samp><var>machine</var></samp>
directory, containing a list of target-specific options.  You can also
add other option files using the <code>extra_options</code> variable in
<samp>config.gcc</samp>.  See <a href="#Options">Options</a>.
</li><li> Entries in <samp>config.gcc</samp> (see <a href="#System-Config">The
<samp>config.gcc</samp> File</a>) for the systems with this target
architecture.
</li><li> Documentation in <samp>gcc/doc/invoke.texi</samp> for any command-line
options supported by this target (see <a href="#Run_002dtime-Target">Run-time
Target Specification</a>).  This means both entries in the summary table
of options and details of the individual options.
</li><li> Documentation in <samp>gcc/doc/extend.texi</samp> for any target-specific
attributes supported (see <a href="#Target-Attributes">Defining
target-specific uses of <code>__attribute__</code></a>), including where the
same attribute is already supported on some targets, which are
enumerated in the manual.
</li><li> Documentation in <samp>gcc/doc/extend.texi</samp> for any target-specific
pragmas supported.
</li><li> Documentation in <samp>gcc/doc/extend.texi</samp> of any target-specific
built-in functions supported.
</li><li> Documentation in <samp>gcc/doc/extend.texi</samp> of any target-specific
format checking styles supported.
</li><li> Documentation in <samp>gcc/doc/md.texi</samp> of any target-specific
constraint letters (see <a href="#Machine-Constraints">Constraints for
Particular Machines</a>).
</li><li> A note in <samp>gcc/doc/contrib.texi</samp> under the person or people who
contributed the target support.
</li><li> Entries in <samp>gcc/doc/install.texi</samp> for all target triplets
supported with this target architecture, giving details of any special
notes about installation for this target, or saying that there are no
special notes if there are none.
</li><li> Possibly other support outside the <samp>gcc</samp> directory for runtime
libraries.  FIXME: reference docs for this.  The <code>libstdc++</code> porting
manual needs to be installed as info for this to work, or to be a
chapter of this manual.
</li></ul>

<p>If the back end is added to the official GCC source repository, the
following are also necessary:
</p>
<ul>
<li> An entry for the target architecture in <samp>readings.html</samp> on the
GCC web site, with any relevant links.
</li><li> Details of the properties of the back end and target architecture in
<samp>backends.html</samp> on the GCC web site.
</li><li> A news item about the contribution of support for that target
architecture, in <samp>index.html</samp> on the GCC web site.
</li><li> Normally, one or more maintainers of that target listed in
<samp>MAINTAINERS</samp>.  Some existing architectures may be unmaintained,
but it would be unusual to add support for a target that does not have
a maintainer when support is added.
</li><li> Target triplets covering all <samp>config.gcc</samp> stanzas for the target,
in the list in <samp>contrib/config-list.mk</samp>.
</li></ul>

<hr>
<a name="Testsuites"></a>
<div class="header">
<p>
Next: <a href="#Options" accesskey="n" rel="next">Options</a>, Previous: <a href="#Source-Tree" accesskey="p" rel="prev">Source Tree</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Testsuites-1"></a>
<h2 class="chapter">7 Testsuites</h2>

<p>GCC contains several testsuites to help maintain compiler quality.
Most of the runtime libraries and language front ends in GCC have
testsuites.  Currently only the C language testsuites are documented
here; FIXME: document the others.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Test-Idioms" accesskey="1">Test Idioms</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Idioms used in testsuite code.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Test-Directives" accesskey="2">Test Directives</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Directives used within DejaGnu tests.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Ada-Tests" accesskey="3">Ada Tests</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The Ada language testsuites.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#C-Tests" accesskey="4">C Tests</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The C language testsuites.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#libgcj-Tests" accesskey="5">libgcj Tests</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The Java library testsuites.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#LTO-Testing" accesskey="6">LTO Testing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Support for testing link-time optimizations.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#gcov-Testing" accesskey="7">gcov Testing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Support for testing gcov.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#profopt-Testing" accesskey="8">profopt Testing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Support for testing profile-directed optimizations.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#compat-Testing" accesskey="9">compat Testing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Support for testing binary compatibility.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Torture-Tests">Torture Tests</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Support for torture testing using multiple options.
</td></tr>
</table>

<hr>
<a name="Test-Idioms"></a>
<div class="header">
<p>
Next: <a href="#Test-Directives" accesskey="n" rel="next">Test Directives</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Idioms-Used-in-Testsuite-Code"></a>
<h3 class="section">7.1 Idioms Used in Testsuite Code</h3>

<p>In general, C testcases have a trailing <samp>-<var>n</var>.c</samp>, starting
with <samp>-1.c</samp>, in case other testcases with similar names are added
later.  If the test is a test of some well-defined feature, it should
have a name referring to that feature such as
<samp><var>feature</var>-1.c</samp>.  If it does not test a well-defined feature
but just happens to exercise a bug somewhere in the compiler, and a
bug report has been filed for this bug in the GCC bug database,
<samp>pr<var>bug-number</var>-1.c</samp> is the appropriate form of name.
Otherwise (for miscellaneous bugs not filed in the GCC bug database),
and previously more generally, test cases are named after the date on
which they were added.  This allows people to tell at a glance whether
a test failure is because of a recently found bug that has not yet
been fixed, or whether it may be a regression, but does not give any
other information about the bug or where discussion of it may be
found.  Some other language testsuites follow similar conventions.
</p>
<p>In the <samp>gcc.dg</samp> testsuite, it is often necessary to test that an
error is indeed a hard error and not just a warning&mdash;for example,
where it is a constraint violation in the C standard, which must
become an error with <samp>-pedantic-errors</samp>.  The following idiom,
where the first line shown is line <var>line</var> of the file and the line
that generates the error, is used for this:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-bogus &quot;warning&quot; &quot;warning in place of error&quot; } */
/* { dg-error &quot;<var>regexp</var>&quot; &quot;<var>message</var>&quot; { target *-*-* } <var>line</var> } */
</pre></div>

<p>It may be necessary to check that an expression is an integer constant
expression and has a certain value.  To check that <code><var>E</var></code> has
value <code><var>V</var></code>, an idiom similar to the following is used:
</p>
<div class="smallexample">
<pre class="smallexample">char x[((E) == (V) ? 1 : -1)];
</pre></div>

<p>In <samp>gcc.dg</samp> tests, <code>__typeof__</code> is sometimes used to make
assertions about the types of expressions.  See, for example,
<samp>gcc.dg/c99-condexpr-1.c</samp>.  The more subtle uses depend on the
exact rules for the types of conditional expressions in the C
standard; see, for example, <samp>gcc.dg/c99-intconst-1.c</samp>.
</p>
<p>It is useful to be able to test that optimizations are being made
properly.  This cannot be done in all cases, but it can be done where
the optimization will lead to code being optimized away (for example,
where flow analysis or alias analysis should show that certain code
cannot be called) or to functions not being called because they have
been expanded as built-in functions.  Such tests go in
<samp>gcc.c-torture/execute</samp>.  Where code should be optimized away, a
call to a nonexistent function such as <code>link_failure ()</code> may be
inserted; a definition
</p>
<div class="smallexample">
<pre class="smallexample">#ifndef __OPTIMIZE__
void
link_failure (void)
{
  abort ();
}
#endif
</pre></div>

<p>will also be needed so that linking still succeeds when the test is
run without optimization.  When all calls to a built-in function
should have been optimized and no calls to the non-built-in version of
the function should remain, that function may be defined as
<code>static</code> to call <code>abort ()</code> (although redeclaring a function
as static may not work on all targets).
</p>
<p>All testcases must be portable.  Target-specific testcases must have
appropriate code to avoid causing failures on unsupported systems;
unfortunately, the mechanisms for this differ by directory.
</p>
<p>FIXME: discuss non-C testsuites here.
</p>
<hr>
<a name="Test-Directives"></a>
<div class="header">
<p>
Next: <a href="#Ada-Tests" accesskey="n" rel="next">Ada Tests</a>, Previous: <a href="#Test-Idioms" accesskey="p" rel="prev">Test Idioms</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Directives-used-within-DejaGnu-tests"></a>
<h3 class="section">7.2 Directives used within DejaGnu tests</h3>

<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Directives" accesskey="1">Directives</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Syntax and descriptions of test directives.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Selectors" accesskey="2">Selectors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Selecting targets to which a test applies.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Effective_002dTarget-Keywords" accesskey="3">Effective-Target Keywords</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Keywords describing target attributes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Add-Options" accesskey="4">Add Options</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Features for <code>dg-add-options</code>
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Require-Support" accesskey="5">Require Support</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Variants of <code>dg-require-<var>support</var></code>
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Final-Actions" accesskey="6">Final Actions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Commands for use in <code>dg-final</code>
</td></tr>
</table>

<hr>
<a name="Directives"></a>
<div class="header">
<p>
Next: <a href="#Selectors" accesskey="n" rel="next">Selectors</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Syntax-and-Descriptions-of-test-directives"></a>
<h4 class="subsection">7.2.1 Syntax and Descriptions of test directives</h4>

<p>Test directives appear within comments in a test source file and begin
with <code>dg-</code>.  Some of these are defined within DejaGnu and others
are local to the GCC testsuite.
</p>
<p>The order in which test directives appear in a test can be important:
directives local to GCC sometimes override information used by the
DejaGnu directives, which know nothing about the GCC directives, so the
DejaGnu directives must precede GCC directives.
</p>
<p>Several test directives include selectors (see <a href="#Selectors">Selectors</a>)
which are usually preceded by the keyword <code>target</code> or <code>xfail</code>.
</p>
<a name="Specify-how-to-build-the-test"></a>
<h4 class="subsubsection">7.2.1.1 Specify how to build the test</h4>

<dl compact="compact">
<dt><code>{ dg-do <var>do-what-keyword</var> [{ target/xfail <var>selector</var> }] }</code></dt>
<dd><p><var>do-what-keyword</var> specifies how the test is compiled and whether
it is executed.  It is one of:
</p>
<dl compact="compact">
<dt><code>preprocess</code></dt>
<dd><p>Compile with <samp>-E</samp> to run only the preprocessor.
</p></dd>
<dt><code>compile</code></dt>
<dd><p>Compile with <samp>-S</samp> to produce an assembly code file.
</p></dd>
<dt><code>assemble</code></dt>
<dd><p>Compile with <samp>-c</samp> to produce a relocatable object file.
</p></dd>
<dt><code>link</code></dt>
<dd><p>Compile, assemble, and link to produce an executable file.
</p></dd>
<dt><code>run</code></dt>
<dd><p>Produce and run an executable file, which is expected to return
an exit code of 0.
</p></dd>
</dl>

<p>The default is <code>compile</code>.  That can be overridden for a set of
tests by redefining <code>dg-do-what-default</code> within the <code>.exp</code>
file for those tests.
</p>
<p>If the directive includes the optional &lsquo;<samp>{ target <var>selector</var> }</samp>&rsquo;
then the test is skipped unless the target system matches the
<var>selector</var>.
</p>
<p>If <var>do-what-keyword</var> is <code>run</code> and the directive includes
the optional &lsquo;<samp>{ xfail <var>selector</var> }</samp>&rsquo; and the selector is met
then the test is expected to fail.  The <code>xfail</code> clause is ignored
for other values of <var>do-what-keyword</var>; those tests can use
directive <code>dg-xfail-if</code>.
</p></dd>
</dl>

<a name="Specify-additional-compiler-options"></a>
<h4 class="subsubsection">7.2.1.2 Specify additional compiler options</h4>

<dl compact="compact">
<dt><code>{ dg-options <var>options</var> [{ target <var>selector</var> }] }</code></dt>
<dd><p>This DejaGnu directive provides a list of compiler options, to be used
if the target system matches <var>selector</var>, that replace the default
options used for this set of tests.
</p>
</dd>
<dt><code>{ dg-add-options <var>feature</var> &hellip; }</code></dt>
<dd><p>Add any compiler options that are needed to access certain features.
This directive does nothing on targets that enable the features by
default, or that don&rsquo;t provide them at all.  It must come after
all <code>dg-options</code> directives.
For supported values of <var>feature</var> see <a href="#Add-Options">Add Options</a>.
</p>
</dd>
<dt><code>{ dg-additional-options <var>options</var> [{ target <var>selector</var> }] }</code></dt>
<dd><p>This directive provides a list of compiler options, to be used
if the target system matches <var>selector</var>, that are added to the default
options used for this set of tests.
</p></dd>
</dl>

<a name="Modify-the-test-timeout-value"></a>
<h4 class="subsubsection">7.2.1.3 Modify the test timeout value</h4>

<p>The normal timeout limit, in seconds, is found by searching the
following in order:
</p>
<ul>
<li> the value defined by an earlier <code>dg-timeout</code> directive in
the test

</li><li> variable <var>tool_timeout</var> defined by the set of tests

</li><li> <var>gcc</var>,<var>timeout</var> set in the target board

</li><li> 300
</li></ul>

<dl compact="compact">
<dt><code>{ dg-timeout <var>n</var> [{target <var>selector</var> }] }</code></dt>
<dd><p>Set the time limit for the compilation and for the execution of the test
to the specified number of seconds.
</p>
</dd>
<dt><code>{ dg-timeout-factor <var>x</var> [{ target <var>selector</var> }] }</code></dt>
<dd><p>Multiply the normal time limit for compilation and execution of the test
by the specified floating-point factor.
</p></dd>
</dl>

<a name="Skip-a-test-for-some-targets"></a>
<h4 class="subsubsection">7.2.1.4 Skip a test for some targets</h4>

<dl compact="compact">
<dt><code>{ dg-skip-if <var>comment</var> { <var>selector</var> } [{ <var>include-opts</var> } [{ <var>exclude-opts</var> }]] }</code></dt>
<dd><p>Arguments <var>include-opts</var> and <var>exclude-opts</var> are lists in which
each element is a string of zero or more GCC options.
Skip the test if all of the following conditions are met:
</p><ul>
<li> the test system is included in <var>selector</var>

</li><li> for at least one of the option strings in <var>include-opts</var>,
every option from that string is in the set of options with which
the test would be compiled; use &lsquo;<samp>&quot;*&quot;</samp>&rsquo; for an <var>include-opts</var> list
that matches any options; that is the default if <var>include-opts</var> is
not specified

</li><li> for each of the option strings in <var>exclude-opts</var>, at least one
option from that string is not in the set of options with which the test
would be compiled; use &lsquo;<samp>&quot;&quot;</samp>&rsquo; for an empty <var>exclude-opts</var> list;
that is the default if <var>exclude-opts</var> is not specified
</li></ul>

<p>For example, to skip a test if option <code>-Os</code> is present:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-skip-if &quot;&quot; { *-*-* }  { &quot;-Os&quot; } { &quot;&quot; } } */
</pre></div>

<p>To skip a test if both options <code>-O2</code> and <code>-g</code> are present:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-skip-if &quot;&quot; { *-*-* }  { &quot;-O2 -g&quot; } { &quot;&quot; } } */
</pre></div>

<p>To skip a test if either <code>-O2</code> or <code>-O3</code> is present:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-skip-if &quot;&quot; { *-*-* }  { &quot;-O2&quot; &quot;-O3&quot; } { &quot;&quot; } } */
</pre></div>

<p>To skip a test unless option <code>-Os</code> is present:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-skip-if &quot;&quot; { *-*-* }  { &quot;*&quot; } { &quot;-Os&quot; } } */
</pre></div>

<p>To skip a test if either <code>-O2</code> or <code>-O3</code> is used with <code>-g</code>
but not if <code>-fpic</code> is also present:
</p>
<div class="smallexample">
<pre class="smallexample">/* { dg-skip-if &quot;&quot; { *-*-* }  { &quot;-O2 -g&quot; &quot;-O3 -g&quot; } { &quot;-fpic&quot; } } */
</pre></div>

</dd>
<dt><code>{ dg-require-effective-target <var>keyword</var> [{ <var>selector</var> }] }</code></dt>
<dd><p>Skip the test if the test target, including current multilib flags,
is not covered by the effective-target keyword.
If the directive includes the optional &lsquo;<samp>{ <var>selector</var> }</samp>&rsquo;
then the effective-target test is only performed if the target system
matches the <var>selector</var>.
This directive must appear after any <code>dg-do</code> directive in the test
and before any <code>dg-additional-sources</code> directive.
See <a href="#Effective_002dTarget-Keywords">Effective-Target Keywords</a>.
</p>
</dd>
<dt><code>{ dg-require-<var>support</var> args }</code></dt>
<dd><p>Skip the test if the target does not provide the required support.
These directives must appear after any <code>dg-do</code> directive in the test
and before any <code>dg-additional-sources</code> directive.
They require at least one argument, which can be an empty string if the
specific procedure does not examine the argument.
See <a href="#Require-Support">Require Support</a>, for a complete list of these directives.
</p></dd>
</dl>

<a name="Expect-a-test-to-fail-for-some-targets"></a>
<h4 class="subsubsection">7.2.1.5 Expect a test to fail for some targets</h4>

<dl compact="compact">
<dt><code>{ dg-xfail-if <var>comment</var> { <var>selector</var> } [{ <var>include-opts</var> } [{ <var>exclude-opts</var> }]] }</code></dt>
<dd><p>Expect the test to fail if the conditions (which are the same as for
<code>dg-skip-if</code>) are met.  This does not affect the execute step.
</p>
</dd>
<dt><code>{ dg-xfail-run-if <var>comment</var> { <var>selector</var> } [{ <var>include-opts</var> } [{ <var>exclude-opts</var> }]] }</code></dt>
<dd><p>Expect the execute step of a test to fail if the conditions (which are
the same as for <code>dg-skip-if</code>) are met.
</p></dd>
</dl>

<a name="Expect-the-test-executable-to-fail"></a>
<h4 class="subsubsection">7.2.1.6 Expect the test executable to fail</h4>

<dl compact="compact">
<dt><code>{ dg-shouldfail <var>comment</var> [{ <var>selector</var> } [{ <var>include-opts</var> } [{ <var>exclude-opts</var> }]]] }</code></dt>
<dd><p>Expect the test executable to return a nonzero exit status if the
conditions (which are the same as for <code>dg-skip-if</code>) are met.
</p></dd>
</dl>

<a name="Verify-compiler-messages"></a>
<h4 class="subsubsection">7.2.1.7 Verify compiler messages</h4>

<dl compact="compact">
<dt><code>{ dg-error <var>regexp</var> [<var>comment</var> [{ target/xfail <var>selector</var> } [<var>line</var>] }]] }</code></dt>
<dd><p>This DejaGnu directive appears on a source line that is expected to get
an error message, or else specifies the source line associated with the
message.  If there is no message for that line or if the text of that
message is not matched by <var>regexp</var> then the check fails and
<var>comment</var> is included in the <code>FAIL</code> message.  The check does
not look for the string &lsquo;<samp>error</samp>&rsquo; unless it is part of <var>regexp</var>.
</p>
</dd>
<dt><code>{ dg-warning <var>regexp</var> [<var>comment</var> [{ target/xfail <var>selector</var> } [<var>line</var>] }]] }</code></dt>
<dd><p>This DejaGnu directive appears on a source line that is expected to get
a warning message, or else specifies the source line associated with the
message.  If there is no message for that line or if the text of that
message is not matched by <var>regexp</var> then the check fails and
<var>comment</var> is included in the <code>FAIL</code> message.  The check does
not look for the string &lsquo;<samp>warning</samp>&rsquo; unless it is part of <var>regexp</var>.
</p>
</dd>
<dt><code>{ dg-message <var>regexp</var> [<var>comment</var> [{ target/xfail <var>selector</var> } [<var>line</var>] }]] }</code></dt>
<dd><p>The line is expected to get a message other than an error or warning.
If there is no message for that line or if the text of that message is
not matched by <var>regexp</var> then the check fails and <var>comment</var> is
included in the <code>FAIL</code> message.
</p>
</dd>
<dt><code>{ dg-bogus <var>regexp</var> [<var>comment</var> [{ target/xfail <var>selector</var> } [<var>line</var>] }]] }</code></dt>
<dd><p>This DejaGnu directive appears on a source line that should not get a
message matching <var>regexp</var>, or else specifies the source line
associated with the bogus message.  It is usually used with &lsquo;<samp>xfail</samp>&rsquo;
to indicate that the message is a known problem for a particular set of
targets.
</p>
</dd>
<dt><code>{ dg-excess-errors <var>comment</var> [{ target/xfail <var>selector</var> }] }</code></dt>
<dd><p>This DejaGnu directive indicates that the test is expected to fail due
to compiler messages that are not handled by &lsquo;<samp>dg-error</samp>&rsquo;,
&lsquo;<samp>dg-warning</samp>&rsquo; or &lsquo;<samp>dg-bogus</samp>&rsquo;.  For this directive &lsquo;<samp>xfail</samp>&rsquo;
has the same effect as &lsquo;<samp>target</samp>&rsquo;.
</p>
</dd>
<dt><code>{ dg-prune-output <var>regexp</var> }</code></dt>
<dd><p>Prune messages matching <var>regexp</var> from the test output.
</p></dd>
</dl>

<a name="Verify-output-of-the-test-executable"></a>
<h4 class="subsubsection">7.2.1.8 Verify output of the test executable</h4>

<dl compact="compact">
<dt><code>{ dg-output <var>regexp</var> [{ target/xfail <var>selector</var> }] }</code></dt>
<dd><p>This DejaGnu directive compares <var>regexp</var> to the combined output
that the test executable writes to <samp>stdout</samp> and <samp>stderr</samp>.
</p></dd>
</dl>

<a name="Specify-additional-files-for-a-test"></a>
<h4 class="subsubsection">7.2.1.9 Specify additional files for a test</h4>

<dl compact="compact">
<dt><code>{ dg-additional-files &quot;<var>filelist</var>&quot; }</code></dt>
<dd><p>Specify additional files, other than source files, that must be copied
to the system where the compiler runs.
</p>
</dd>
<dt><code>{ dg-additional-sources &quot;<var>filelist</var>&quot; }</code></dt>
<dd><p>Specify additional source files to appear in the compile line
following the main test file.
</p></dd>
</dl>

<a name="Add-checks-at-the-end-of-a-test"></a>
<h4 class="subsubsection">7.2.1.10 Add checks at the end of a test</h4>

<dl compact="compact">
<dt><code>{ dg-final { <var>local-directive</var> } }</code></dt>
<dd><p>This DejaGnu directive is placed within a comment anywhere in the
source file and is processed after the test has been compiled and run.
Multiple &lsquo;<samp>dg-final</samp>&rsquo; commands are processed in the order in which
they appear in the source file.  See <a href="#Final-Actions">Final Actions</a>, for a list
of directives that can be used within <code>dg-final</code>.
</p></dd>
</dl>

<hr>
<a name="Selectors"></a>
<div class="header">
<p>
Next: <a href="#Effective_002dTarget-Keywords" accesskey="n" rel="next">Effective-Target Keywords</a>, Previous: <a href="#Directives" accesskey="p" rel="prev">Directives</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Selecting-targets-to-which-a-test-applies"></a>
<h4 class="subsection">7.2.2 Selecting targets to which a test applies</h4>

<p>Several test directives include <var>selector</var>s to limit the targets
for which a test is run or to declare that a test is expected to fail
on particular targets.
</p>
<p>A selector is:
</p><ul>
<li> one or more target triplets, possibly including wildcard characters;
use &lsquo;<samp>*-*-*</samp>&rsquo; to match any target
</li><li> a single effective-target keyword (see <a href="#Effective_002dTarget-Keywords">Effective-Target Keywords</a>)
</li><li> a logical expression
</li></ul>

<p>Depending on the context, the selector specifies whether a test is
skipped and reported as unsupported or is expected to fail.  A context
that allows either &lsquo;<samp>target</samp>&rsquo; or &lsquo;<samp>xfail</samp>&rsquo; also allows
&lsquo;<samp>{ target <var>selector1</var> xfail <var>selector2</var> }</samp>&rsquo;
to skip the test for targets that don&rsquo;t match <var>selector1</var> and the
test to fail for targets that match <var>selector2</var>.
</p>
<p>A selector expression appears within curly braces and uses a single
logical operator: one of &lsquo;<samp>!</samp>&rsquo;, &lsquo;<samp>&amp;&amp;</samp>&rsquo;, or &lsquo;<samp>||</samp>&rsquo;.  An
operand is another selector expression, an effective-target keyword,
a single target triplet, or a list of target triplets within quotes or
curly braces.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">{ target { ! &quot;hppa*-*-* ia64*-*-*&quot; } }
{ target { powerpc*-*-* &amp;&amp; lp64 } }
{ xfail { lp64 || vect_no_align } }
</pre></div>

<hr>
<a name="Effective_002dTarget-Keywords"></a>
<div class="header">
<p>
Next: <a href="#Add-Options" accesskey="n" rel="next">Add Options</a>, Previous: <a href="#Selectors" accesskey="p" rel="prev">Selectors</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Keywords-describing-target-attributes"></a>
<h4 class="subsection">7.2.3 Keywords describing target attributes</h4>

<p>Effective-target keywords identify sets of targets that support
particular functionality.  They are used to limit tests to be run only
for particular targets, or to specify that particular sets of targets
are expected to fail some tests.
</p>
<p>Effective-target keywords are defined in <samp>lib/target-supports.exp</samp> in
the GCC testsuite, with the exception of those that are documented as
being local to a particular test directory.
</p>
<p>The &lsquo;<samp>effective target</samp>&rsquo; takes into account all of the compiler options
with which the test will be compiled, including the multilib options.
By convention, keywords ending in <code>_nocache</code> can also include options
specified for the particular test in an earlier <code>dg-options</code> or
<code>dg-add-options</code> directive.
</p>
<a name="Data-type-sizes"></a>
<h4 class="subsubsection">7.2.3.1 Data type sizes</h4>

<dl compact="compact">
<dt><code>ilp32</code></dt>
<dd><p>Target has 32-bit <code>int</code>, <code>long</code>, and pointers.
</p>
</dd>
<dt><code>lp64</code></dt>
<dd><p>Target has 32-bit <code>int</code>, 64-bit <code>long</code> and pointers.
</p>
</dd>
<dt><code>llp64</code></dt>
<dd><p>Target has 32-bit <code>int</code> and <code>long</code>, 64-bit <code>long long</code>
and pointers.
</p>
</dd>
<dt><code>double64</code></dt>
<dd><p>Target has 64-bit <code>double</code>.
</p>
</dd>
<dt><code>double64plus</code></dt>
<dd><p>Target has <code>double</code> that is 64 bits or longer.
</p>
</dd>
<dt><code>longdouble128</code></dt>
<dd><p>Target has 128-bit <code>long double</code>.
</p>
</dd>
<dt><code>int32plus</code></dt>
<dd><p>Target has <code>int</code> that is at 32 bits or longer.
</p>
</dd>
<dt><code>int16</code></dt>
<dd><p>Target has <code>int</code> that is 16 bits or shorter.
</p>
</dd>
<dt><code>long_neq_int</code></dt>
<dd><p>Target has <code>int</code> and <code>long</code> with different sizes.
</p>
</dd>
<dt><code>large_double</code></dt>
<dd><p>Target supports <code>double</code> that is longer than <code>float</code>.
</p>
</dd>
<dt><code>large_long_double</code></dt>
<dd><p>Target supports <code>long double</code> that is longer than <code>double</code>.
</p>
</dd>
<dt><code>ptr32plus</code></dt>
<dd><p>Target has pointers that are 32 bits or longer.
</p>
</dd>
<dt><code>size32plus</code></dt>
<dd><p>Target supports array and structure sizes that are 32 bits or longer.
</p>
</dd>
<dt><code>4byte_wchar_t</code></dt>
<dd><p>Target has <code>wchar_t</code> that is at least 4 bytes.
</p></dd>
</dl>

<a name="Fortran_002dspecific-attributes"></a>
<h4 class="subsubsection">7.2.3.2 Fortran-specific attributes</h4>

<dl compact="compact">
<dt><code>fortran_integer_16</code></dt>
<dd><p>Target supports Fortran <code>integer</code> that is 16 bytes or longer.
</p>
</dd>
<dt><code>fortran_large_int</code></dt>
<dd><p>Target supports Fortran <code>integer</code> kinds larger than <code>integer(8)</code>.
</p>
</dd>
<dt><code>fortran_large_real</code></dt>
<dd><p>Target supports Fortran <code>real</code> kinds larger than <code>real(8)</code>.
</p></dd>
</dl>

<a name="Vector_002dspecific-attributes"></a>
<h4 class="subsubsection">7.2.3.3 Vector-specific attributes</h4>

<dl compact="compact">
<dt><code>vect_condition</code></dt>
<dd><p>Target supports vector conditional operations.
</p>
</dd>
<dt><code>vect_double</code></dt>
<dd><p>Target supports hardware vectors of <code>double</code>.
</p>
</dd>
<dt><code>vect_float</code></dt>
<dd><p>Target supports hardware vectors of <code>float</code>.
</p>
</dd>
<dt><code>vect_int</code></dt>
<dd><p>Target supports hardware vectors of <code>int</code>.
</p>
</dd>
<dt><code>vect_long</code></dt>
<dd><p>Target supports hardware vectors of <code>long</code>.
</p>
</dd>
<dt><code>vect_long_long</code></dt>
<dd><p>Target supports hardware vectors of <code>long long</code>.
</p>
</dd>
<dt><code>vect_aligned_arrays</code></dt>
<dd><p>Target aligns arrays to vector alignment boundary.
</p>
</dd>
<dt><code>vect_hw_misalign</code></dt>
<dd><p>Target supports a vector misalign access.
</p>
</dd>
<dt><code>vect_no_align</code></dt>
<dd><p>Target does not support a vector alignment mechanism.
</p>
</dd>
<dt><code>vect_no_int_max</code></dt>
<dd><p>Target does not support a vector max instruction on <code>int</code>.
</p>
</dd>
<dt><code>vect_no_int_add</code></dt>
<dd><p>Target does not support a vector add instruction on <code>int</code>.
</p>
</dd>
<dt><code>vect_no_bitwise</code></dt>
<dd><p>Target does not support vector bitwise instructions.
</p>
</dd>
<dt><code>vect_char_mult</code></dt>
<dd><p>Target supports <code>vector char</code> multiplication.
</p>
</dd>
<dt><code>vect_short_mult</code></dt>
<dd><p>Target supports <code>vector short</code> multiplication.
</p>
</dd>
<dt><code>vect_int_mult</code></dt>
<dd><p>Target supports <code>vector int</code> multiplication.
</p>
</dd>
<dt><code>vect_extract_even_odd</code></dt>
<dd><p>Target supports vector even/odd element extraction.
</p>
</dd>
<dt><code>vect_extract_even_odd_wide</code></dt>
<dd><p>Target supports vector even/odd element extraction of vectors with elements
<code>SImode</code> or larger.
</p>
</dd>
<dt><code>vect_interleave</code></dt>
<dd><p>Target supports vector interleaving.
</p>
</dd>
<dt><code>vect_strided</code></dt>
<dd><p>Target supports vector interleaving and extract even/odd.
</p>
</dd>
<dt><code>vect_strided_wide</code></dt>
<dd><p>Target supports vector interleaving and extract even/odd for wide
element types.
</p>
</dd>
<dt><code>vect_perm</code></dt>
<dd><p>Target supports vector permutation.
</p>
</dd>
<dt><code>vect_shift</code></dt>
<dd><p>Target supports a hardware vector shift operation.
</p>
</dd>
<dt><code>vect_widen_sum_hi_to_si</code></dt>
<dd><p>Target supports a vector widening summation of <code>short</code> operands
into <code>int</code> results, or can promote (unpack) from <code>short</code>
to <code>int</code>.
</p>
</dd>
<dt><code>vect_widen_sum_qi_to_hi</code></dt>
<dd><p>Target supports a vector widening summation of <code>char</code> operands
into <code>short</code> results, or can promote (unpack) from <code>char</code>
to <code>short</code>.
</p>
</dd>
<dt><code>vect_widen_sum_qi_to_si</code></dt>
<dd><p>Target supports a vector widening summation of <code>char</code> operands
into <code>int</code> results.
</p>
</dd>
<dt><code>vect_widen_mult_qi_to_hi</code></dt>
<dd><p>Target supports a vector widening multiplication of <code>char</code> operands
into <code>short</code> results, or can promote (unpack) from <code>char</code> to
<code>short</code> and perform non-widening multiplication of <code>short</code>.
</p>
</dd>
<dt><code>vect_widen_mult_hi_to_si</code></dt>
<dd><p>Target supports a vector widening multiplication of <code>short</code> operands
into <code>int</code> results, or can promote (unpack) from <code>short</code> to
<code>int</code> and perform non-widening multiplication of <code>int</code>.
</p>
</dd>
<dt><code>vect_sdot_qi</code></dt>
<dd><p>Target supports a vector dot-product of <code>signed char</code>.
</p>
</dd>
<dt><code>vect_udot_qi</code></dt>
<dd><p>Target supports a vector dot-product of <code>unsigned char</code>.
</p>
</dd>
<dt><code>vect_sdot_hi</code></dt>
<dd><p>Target supports a vector dot-product of <code>signed short</code>.
</p>
</dd>
<dt><code>vect_udot_hi</code></dt>
<dd><p>Target supports a vector dot-product of <code>unsigned short</code>.
</p>
</dd>
<dt><code>vect_pack_trunc</code></dt>
<dd><p>Target supports a vector demotion (packing) of <code>short</code> to <code>char</code>
and from <code>int</code> to <code>short</code> using modulo arithmetic.
</p>
</dd>
<dt><code>vect_unpack</code></dt>
<dd><p>Target supports a vector promotion (unpacking) of <code>char</code> to <code>short</code>
and from <code>char</code> to <code>int</code>.
</p>
</dd>
<dt><code>vect_intfloat_cvt</code></dt>
<dd><p>Target supports conversion from <code>signed int</code> to <code>float</code>.
</p>
</dd>
<dt><code>vect_uintfloat_cvt</code></dt>
<dd><p>Target supports conversion from <code>unsigned int</code> to <code>float</code>.
</p>
</dd>
<dt><code>vect_floatint_cvt</code></dt>
<dd><p>Target supports conversion from <code>float</code> to <code>signed int</code>.
</p>
</dd>
<dt><code>vect_floatuint_cvt</code></dt>
<dd><p>Target supports conversion from <code>float</code> to <code>unsigned int</code>.
</p></dd>
</dl>

<a name="Thread-Local-Storage-attributes"></a>
<h4 class="subsubsection">7.2.3.4 Thread Local Storage attributes</h4>

<dl compact="compact">
<dt><code>tls</code></dt>
<dd><p>Target supports thread-local storage.
</p>
</dd>
<dt><code>tls_native</code></dt>
<dd><p>Target supports native (rather than emulated) thread-local storage.
</p>
</dd>
<dt><code>tls_runtime</code></dt>
<dd><p>Test system supports executing TLS executables.
</p></dd>
</dl>

<a name="Decimal-floating-point-attributes"></a>
<h4 class="subsubsection">7.2.3.5 Decimal floating point attributes</h4>

<dl compact="compact">
<dt><code>dfp</code></dt>
<dd><p>Targets supports compiling decimal floating point extension to C.
</p>
</dd>
<dt><code>dfp_nocache</code></dt>
<dd><p>Including the options used to compile this particular test, the
target supports compiling decimal floating point extension to C.
</p>
</dd>
<dt><code>dfprt</code></dt>
<dd><p>Test system can execute decimal floating point tests.
</p>
</dd>
<dt><code>dfprt_nocache</code></dt>
<dd><p>Including the options used to compile this particular test, the
test system can execute decimal floating point tests.
</p>
</dd>
<dt><code>hard_dfp</code></dt>
<dd><p>Target generates decimal floating point instructions with current options.
</p></dd>
</dl>

<a name="ARM_002dspecific-attributes"></a>
<h4 class="subsubsection">7.2.3.6 ARM-specific attributes</h4>

<dl compact="compact">
<dt><code>arm32</code></dt>
<dd><p>ARM target generates 32-bit code.
</p>
</dd>
<dt><code>arm_eabi</code></dt>
<dd><p>ARM target adheres to the ABI for the ARM Architecture.
</p>
</dd>
<dt><code>arm_hf_eabi</code></dt>
<dd><p>ARM target adheres to the VFP and Advanced SIMD Register Arguments
variant of the ABI for the ARM Architecture (as selected with
<code>-mfloat-abi=hard</code>).
</p>
</dd>
<dt><code>arm_hard_vfp_ok</code></dt>
<dd><p>ARM target supports <code>-mfpu=vfp -mfloat-abi=hard</code>.
Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_iwmmxt_ok</code></dt>
<dd><p>ARM target supports <code>-mcpu=iwmmxt</code>.
Some multilibs may be incompatible with this option.
</p>
</dd>
<dt><code>arm_neon</code></dt>
<dd><p>ARM target supports generating NEON instructions.
</p>
</dd>
<dt><code>arm_neon_hw</code></dt>
<dd><p>Test system supports executing NEON instructions.
</p>
</dd>
<dt><code>arm_neonv2_hw</code></dt>
<dd><p>Test system supports executing NEON v2 instructions.
</p>
</dd>
<dt><code>arm_neon_ok</code></dt>
<dd><a name="arm_005fneon_005fok"></a><p>ARM Target supports <code>-mfpu=neon -mfloat-abi=softfp</code> or compatible
options.  Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_neonv2_ok</code></dt>
<dd><a name="arm_005fneonv2_005fok"></a><p>ARM Target supports <code>-mfpu=neon-vfpv4 -mfloat-abi=softfp</code> or compatible
options.  Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_neon_fp16_ok</code></dt>
<dd><a name="arm_005fneon_005ffp16_005fok"></a><p>ARM Target supports <code>-mfpu=neon-fp16 -mfloat-abi=softfp</code> or compatible
options.  Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_thumb1_ok</code></dt>
<dd><p>ARM target generates Thumb-1 code for <code>-mthumb</code>.
</p>
</dd>
<dt><code>arm_thumb2_ok</code></dt>
<dd><p>ARM target generates Thumb-2 code for <code>-mthumb</code>.
</p>
</dd>
<dt><code>arm_vfp_ok</code></dt>
<dd><p>ARM target supports <code>-mfpu=vfp -mfloat-abi=softfp</code>.
Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_v8_vfp_ok</code></dt>
<dd><p>ARM target supports <code>-mfpu=fp-armv8 -mfloat-abi=softfp</code>.
Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_v8_neon_ok</code></dt>
<dd><p>ARM target supports <code>-mfpu=neon-fp-armv8 -mfloat-abi=softfp</code>.
Some multilibs may be incompatible with these options.
</p>
</dd>
<dt><code>arm_prefer_ldrd_strd</code></dt>
<dd><p>ARM target prefers <code>LDRD</code> and <code>STRD</code> instructions over
<code>LDM</code> and <code>STM</code> instructions.
</p>
</dd>
</dl>

<a name="MIPS_002dspecific-attributes"></a>
<h4 class="subsubsection">7.2.3.7 MIPS-specific attributes</h4>

<dl compact="compact">
<dt><code>mips64</code></dt>
<dd><p>MIPS target supports 64-bit instructions.
</p>
</dd>
<dt><code>nomips16</code></dt>
<dd><p>MIPS target does not produce MIPS16 code.
</p>
</dd>
<dt><code>mips16_attribute</code></dt>
<dd><p>MIPS target can generate MIPS16 code.
</p>
</dd>
<dt><code>mips_loongson</code></dt>
<dd><p>MIPS target is a Loongson-2E or -2F target using an ABI that supports
the Loongson vector modes.
</p>
</dd>
<dt><code>mips_newabi_large_long_double</code></dt>
<dd><p>MIPS target supports <code>long double</code> larger than <code>double</code>
when using the new ABI.
</p>
</dd>
<dt><code>mpaired_single</code></dt>
<dd><p>MIPS target supports <code>-mpaired-single</code>.
</p></dd>
</dl>

<a name="PowerPC_002dspecific-attributes"></a>
<h4 class="subsubsection">7.2.3.8 PowerPC-specific attributes</h4>

<dl compact="compact">
<dt><code>dfp_hw</code></dt>
<dd><p>PowerPC target supports executing hardware DFP instructions.
</p>
</dd>
<dt><code>p8vector_hw</code></dt>
<dd><p>PowerPC target supports executing VSX instructions (ISA 2.07).
</p>
</dd>
<dt><code>powerpc64</code></dt>
<dd><p>Test system supports executing 64-bit instructions.
</p>
</dd>
<dt><code>powerpc_altivec</code></dt>
<dd><p>PowerPC target supports AltiVec.
</p>
</dd>
<dt><code>powerpc_altivec_ok</code></dt>
<dd><p>PowerPC target supports <code>-maltivec</code>.
</p>
</dd>
<dt><code>powerpc_eabi_ok</code></dt>
<dd><p>PowerPC target supports <code>-meabi</code>.
</p>
</dd>
<dt><code>powerpc_elfv2</code></dt>
<dd><p>PowerPC target supports <code>-mabi=elfv2</code>.
</p>
</dd>
<dt><code>powerpc_fprs</code></dt>
<dd><p>PowerPC target supports floating-point registers.
</p>
</dd>
<dt><code>powerpc_hard_double</code></dt>
<dd><p>PowerPC target supports hardware double-precision floating-point.
</p>
</dd>
<dt><code>powerpc_htm_ok</code></dt>
<dd><p>PowerPC target supports <code>-mhtm</code>
</p>
</dd>
<dt><code>powerpc_p8vector_ok</code></dt>
<dd><p>PowerPC target supports <code>-mpower8-vector</code>
</p>
</dd>
<dt><code>powerpc_ppu_ok</code></dt>
<dd><p>PowerPC target supports <code>-mcpu=cell</code>.
</p>
</dd>
<dt><code>powerpc_spe</code></dt>
<dd><p>PowerPC target supports PowerPC SPE.
</p>
</dd>
<dt><code>powerpc_spe_nocache</code></dt>
<dd><p>Including the options used to compile this particular test, the
PowerPC target supports PowerPC SPE.
</p>
</dd>
<dt><code>powerpc_spu</code></dt>
<dd><p>PowerPC target supports PowerPC SPU.
</p>
</dd>
<dt><code>powerpc_vsx_ok</code></dt>
<dd><p>PowerPC target supports <code>-mvsx</code>.
</p>
</dd>
<dt><code>powerpc_405_nocache</code></dt>
<dd><p>Including the options used to compile this particular test, the
PowerPC target supports PowerPC 405.
</p>
</dd>
<dt><code>ppc_recip_hw</code></dt>
<dd><p>PowerPC target supports executing reciprocal estimate instructions.
</p>
</dd>
<dt><code>spu_auto_overlay</code></dt>
<dd><p>SPU target has toolchain that supports automatic overlay generation.
</p>
</dd>
<dt><code>vmx_hw</code></dt>
<dd><p>PowerPC target supports executing AltiVec instructions.
</p>
</dd>
<dt><code>vsx_hw</code></dt>
<dd><p>PowerPC target supports executing VSX instructions (ISA 2.06).
</p></dd>
</dl>

<a name="Other-hardware-attributes"></a>
<h4 class="subsubsection">7.2.3.9 Other hardware attributes</h4>

<dl compact="compact">
<dt><code>avx</code></dt>
<dd><p>Target supports compiling <code>avx</code> instructions.
</p>
</dd>
<dt><code>avx_runtime</code></dt>
<dd><p>Target supports the execution of <code>avx</code> instructions.
</p>
</dd>
<dt><code>cell_hw</code></dt>
<dd><p>Test system can execute AltiVec and Cell PPU instructions.
</p>
</dd>
<dt><code>coldfire_fpu</code></dt>
<dd><p>Target uses a ColdFire FPU.
</p>
</dd>
<dt><code>hard_float</code></dt>
<dd><p>Target supports FPU instructions.
</p>
</dd>
<dt><code>sse</code></dt>
<dd><p>Target supports compiling <code>sse</code> instructions.
</p>
</dd>
<dt><code>sse_runtime</code></dt>
<dd><p>Target supports the execution of <code>sse</code> instructions.
</p>
</dd>
<dt><code>sse2</code></dt>
<dd><p>Target supports compiling <code>sse2</code> instructions.
</p>
</dd>
<dt><code>sse2_runtime</code></dt>
<dd><p>Target supports the execution of <code>sse2</code> instructions.
</p>
</dd>
<dt><code>sync_char_short</code></dt>
<dd><p>Target supports atomic operations on <code>char</code> and <code>short</code>.
</p>
</dd>
<dt><code>sync_int_long</code></dt>
<dd><p>Target supports atomic operations on <code>int</code> and <code>long</code>.
</p>
</dd>
<dt><code>ultrasparc_hw</code></dt>
<dd><p>Test environment appears to run executables on a simulator that
accepts only <code>EM_SPARC</code> executables and chokes on <code>EM_SPARC32PLUS</code>
or <code>EM_SPARCV9</code> executables.
</p>
</dd>
<dt><code>vect_cmdline_needed</code></dt>
<dd><p>Target requires a command line argument to enable a SIMD instruction set.
</p></dd>
</dl>

<a name="Environment-attributes"></a>
<h4 class="subsubsection">7.2.3.10 Environment attributes</h4>

<dl compact="compact">
<dt><code>c</code></dt>
<dd><p>The language for the compiler under test is C.
</p>
</dd>
<dt><code>c++</code></dt>
<dd><p>The language for the compiler under test is C++.
</p>
</dd>
<dt><code>c99_runtime</code></dt>
<dd><p>Target provides a full C99 runtime.
</p>
</dd>
<dt><code>correct_iso_cpp_string_wchar_protos</code></dt>
<dd><p>Target <code>string.h</code> and <code>wchar.h</code> headers provide C++ required
overloads for <code>strchr</code> etc. functions.
</p>
</dd>
<dt><code>dummy_wcsftime</code></dt>
<dd><p>Target uses a dummy <code>wcsftime</code> function that always returns zero.
</p>
</dd>
<dt><code>fd_truncate</code></dt>
<dd><p>Target can truncate a file from a file descriptor, as used by
<samp>libgfortran/io/unix.c:fd_truncate</samp>; i.e. <code>ftruncate</code> or
<code>chsize</code>.
</p>
</dd>
<dt><code>freestanding</code></dt>
<dd><p>Target is &lsquo;<samp>freestanding</samp>&rsquo; as defined in section 4 of the C99 standard.
Effectively, it is a target which supports no extra headers or libraries
other than what is considered essential.
</p>
</dd>
<dt><code>init_priority</code></dt>
<dd><p>Target supports constructors with initialization priority arguments.
</p>
</dd>
<dt><code>inttypes_types</code></dt>
<dd><p>Target has the basic signed and unsigned types in <code>inttypes.h</code>.
This is for tests that GCC&rsquo;s notions of these types agree with those
in the header, as some systems have only <code>inttypes.h</code>.
</p>
</dd>
<dt><code>lax_strtofp</code></dt>
<dd><p>Target might have errors of a few ULP in string to floating-point
conversion functions and overflow is not always detected correctly by
those functions.
</p>
</dd>
<dt><code>mmap</code></dt>
<dd><p>Target supports <code>mmap</code>.
</p>
</dd>
<dt><code>newlib</code></dt>
<dd><p>Target supports Newlib.
</p>
</dd>
<dt><code>pow10</code></dt>
<dd><p>Target provides <code>pow10</code> function.
</p>
</dd>
<dt><code>pthread</code></dt>
<dd><p>Target can compile using <code>pthread.h</code> with no errors or warnings.
</p>
</dd>
<dt><code>pthread_h</code></dt>
<dd><p>Target has <code>pthread.h</code>.
</p>
</dd>
<dt><code>run_expensive_tests</code></dt>
<dd><p>Expensive testcases (usually those that consume excessive amounts of CPU
time) should be run on this target.  This can be enabled by setting the
<code>GCC_TEST_RUN_EXPENSIVE</code> environment variable to a non-empty string.
</p>
</dd>
<dt><code>simulator</code></dt>
<dd><p>Test system runs executables on a simulator (i.e. slowly) rather than
hardware (i.e. fast).
</p>
</dd>
<dt><code>stdint_types</code></dt>
<dd><p>Target has the basic signed and unsigned C types in <code>stdint.h</code>.
This will be obsolete when GCC ensures a working <code>stdint.h</code> for
all targets.
</p>
</dd>
<dt><code>trampolines</code></dt>
<dd><p>Target supports trampolines.
</p>
</dd>
<dt><code>uclibc</code></dt>
<dd><p>Target supports uClibc.
</p>
</dd>
<dt><code>unwrapped</code></dt>
<dd><p>Target does not use a status wrapper.
</p>
</dd>
<dt><code>vxworks_kernel</code></dt>
<dd><p>Target is a VxWorks kernel.
</p>
</dd>
<dt><code>vxworks_rtp</code></dt>
<dd><p>Target is a VxWorks RTP.
</p>
</dd>
<dt><code>wchar</code></dt>
<dd><p>Target supports wide characters.
</p></dd>
</dl>

<a name="Other-attributes"></a>
<h4 class="subsubsection">7.2.3.11 Other attributes</h4>

<dl compact="compact">
<dt><code>automatic_stack_alignment</code></dt>
<dd><p>Target supports automatic stack alignment.
</p>
</dd>
<dt><code>cxa_atexit</code></dt>
<dd><p>Target uses <code>__cxa_atexit</code>.
</p>
</dd>
<dt><code>default_packed</code></dt>
<dd><p>Target has packed layout of structure members by default.
</p>
</dd>
<dt><code>fgraphite</code></dt>
<dd><p>Target supports Graphite optimizations.
</p>
</dd>
<dt><code>fixed_point</code></dt>
<dd><p>Target supports fixed-point extension to C.
</p>
</dd>
<dt><code>fopenmp</code></dt>
<dd><p>Target supports OpenMP via <samp>-fopenmp</samp>.
</p>
</dd>
<dt><code>fpic</code></dt>
<dd><p>Target supports <samp>-fpic</samp> and <samp>-fPIC</samp>.
</p>
</dd>
<dt><code>freorder</code></dt>
<dd><p>Target supports <samp>-freorder-blocks-and-partition</samp>.
</p>
</dd>
<dt><code>fstack_protector</code></dt>
<dd><p>Target supports <samp>-fstack-protector</samp>.
</p>
</dd>
<dt><code>gas</code></dt>
<dd><p>Target uses GNU <code>as</code>.
</p>
</dd>
<dt><code>gc_sections</code></dt>
<dd><p>Target supports <samp>--gc-sections</samp>.
</p>
</dd>
<dt><code>gld</code></dt>
<dd><p>Target uses GNU <code>ld</code>.
</p>
</dd>
<dt><code>keeps_null_pointer_checks</code></dt>
<dd><p>Target keeps null pointer checks, either due to the use of
<samp>-fno-delete-null-pointer-checks</samp> or hardwired into the target.
</p>
</dd>
<dt><code>lto</code></dt>
<dd><p>Compiler has been configured to support link-time optimization (LTO).
</p>
</dd>
<dt><code>naked_functions</code></dt>
<dd><p>Target supports the <code>naked</code> function attribute.
</p>
</dd>
<dt><code>named_sections</code></dt>
<dd><p>Target supports named sections.
</p>
</dd>
<dt><code>natural_alignment_32</code></dt>
<dd><p>Target uses natural alignment (aligned to type size) for types of
32 bits or less.
</p>
</dd>
<dt><code>target_natural_alignment_64</code></dt>
<dd><p>Target uses natural alignment (aligned to type size) for types of
64 bits or less.
</p>
</dd>
<dt><code>nonpic</code></dt>
<dd><p>Target does not generate PIC by default.
</p>
</dd>
<dt><code>pcc_bitfield_type_matters</code></dt>
<dd><p>Target defines <code>PCC_BITFIELD_TYPE_MATTERS</code>.
</p>
</dd>
<dt><code>pe_aligned_commons</code></dt>
<dd><p>Target supports <samp>-mpe-aligned-commons</samp>.
</p>
</dd>
<dt><code>pie</code></dt>
<dd><p>Target supports <samp>-pie</samp>, <samp>-fpie</samp> and <samp>-fPIE</samp>.
</p>
</dd>
<dt><code>section_anchors</code></dt>
<dd><p>Target supports section anchors.
</p>
</dd>
<dt><code>short_enums</code></dt>
<dd><p>Target defaults to short enums.
</p>
</dd>
<dt><code>static</code></dt>
<dd><p>Target supports <samp>-static</samp>.
</p>
</dd>
<dt><code>static_libgfortran</code></dt>
<dd><p>Target supports statically linking &lsquo;<samp>libgfortran</samp>&rsquo;.
</p>
</dd>
<dt><code>string_merging</code></dt>
<dd><p>Target supports merging string constants at link time.
</p>
</dd>
<dt><code>ucn</code></dt>
<dd><p>Target supports compiling and assembling UCN.
</p>
</dd>
<dt><code>ucn_nocache</code></dt>
<dd><p>Including the options used to compile this particular test, the
target supports compiling and assembling UCN.
</p>
</dd>
<dt><code>unaligned_stack</code></dt>
<dd><p>Target does not guarantee that its <code>STACK_BOUNDARY</code> is greater than
or equal to the required vector alignment.
</p>
</dd>
<dt><code>vector_alignment_reachable</code></dt>
<dd><p>Vector alignment is reachable for types of 32 bits or less.
</p>
</dd>
<dt><code>vector_alignment_reachable_for_64bit</code></dt>
<dd><p>Vector alignment is reachable for types of 64 bits or less.
</p>
</dd>
<dt><code>wchar_t_char16_t_compatible</code></dt>
<dd><p>Target supports <code>wchar_t</code> that is compatible with <code>char16_t</code>.
</p>
</dd>
<dt><code>wchar_t_char32_t_compatible</code></dt>
<dd><p>Target supports <code>wchar_t</code> that is compatible with <code>char32_t</code>.
</p></dd>
</dl>

<a name="Local-to-tests-in-gcc_002etarget_002fi386"></a>
<h4 class="subsubsection">7.2.3.12 Local to tests in <code>gcc.target/i386</code></h4>

<dl compact="compact">
<dt><code>3dnow</code></dt>
<dd><p>Target supports compiling <code>3dnow</code> instructions.
</p>
</dd>
<dt><code>aes</code></dt>
<dd><p>Target supports compiling <code>aes</code> instructions.
</p>
</dd>
<dt><code>fma4</code></dt>
<dd><p>Target supports compiling <code>fma4</code> instructions.
</p>
</dd>
<dt><code>ms_hook_prologue</code></dt>
<dd><p>Target supports attribute <code>ms_hook_prologue</code>.
</p>
</dd>
<dt><code>pclmul</code></dt>
<dd><p>Target supports compiling <code>pclmul</code> instructions.
</p>
</dd>
<dt><code>sse3</code></dt>
<dd><p>Target supports compiling <code>sse3</code> instructions.
</p>
</dd>
<dt><code>sse4</code></dt>
<dd><p>Target supports compiling <code>sse4</code> instructions.
</p>
</dd>
<dt><code>sse4a</code></dt>
<dd><p>Target supports compiling <code>sse4a</code> instructions.
</p>
</dd>
<dt><code>ssse3</code></dt>
<dd><p>Target supports compiling <code>ssse3</code> instructions.
</p>
</dd>
<dt><code>vaes</code></dt>
<dd><p>Target supports compiling <code>vaes</code> instructions.
</p>
</dd>
<dt><code>vpclmul</code></dt>
<dd><p>Target supports compiling <code>vpclmul</code> instructions.
</p>
</dd>
<dt><code>xop</code></dt>
<dd><p>Target supports compiling <code>xop</code> instructions.
</p></dd>
</dl>

<a name="Local-to-tests-in-gcc_002etarget_002fspu_002fea"></a>
<h4 class="subsubsection">7.2.3.13 Local to tests in <code>gcc.target/spu/ea</code></h4>

<dl compact="compact">
<dt><code>ealib</code></dt>
<dd><p>Target <code>__ea</code> library functions are available.
</p></dd>
</dl>

<a name="Local-to-tests-in-gcc_002etest_002dframework"></a>
<h4 class="subsubsection">7.2.3.14 Local to tests in <code>gcc.test-framework</code></h4>

<dl compact="compact">
<dt><code>no</code></dt>
<dd><p>Always returns 0.
</p>
</dd>
<dt><code>yes</code></dt>
<dd><p>Always returns 1.
</p></dd>
</dl>

<hr>
<a name="Add-Options"></a>
<div class="header">
<p>
Next: <a href="#Require-Support" accesskey="n" rel="next">Require Support</a>, Previous: <a href="#Effective_002dTarget-Keywords" accesskey="p" rel="prev">Effective-Target Keywords</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Features-for-dg_002dadd_002doptions"></a>
<h4 class="subsection">7.2.4 Features for <code>dg-add-options</code></h4>

<p>The supported values of <var>feature</var> for directive <code>dg-add-options</code>
are:
</p>
<dl compact="compact">
<dt><code>arm_neon</code></dt>
<dd><p>NEON support.  Only ARM targets support this feature, and only then
in certain modes; see the <a href="#arm_005fneon_005fok">arm_neon_ok effective target
keyword</a>.
</p>
</dd>
<dt><code>arm_neon_fp16</code></dt>
<dd><p>NEON and half-precision floating point support.  Only ARM targets
support this feature, and only then in certain modes; see
the <a href="#arm_005fneon_005fok">arm_neon_fp16_ok effective target keyword</a>.
</p>
</dd>
<dt><code>bind_pic_locally</code></dt>
<dd><p>Add the target-specific flags needed to enable functions to bind
locally when using pic/PIC passes in the testsuite.
</p>
</dd>
<dt><code>c99_runtime</code></dt>
<dd><p>Add the target-specific flags needed to access the C99 runtime.
</p>
</dd>
<dt><code>ieee</code></dt>
<dd><p>Add the target-specific flags needed to enable full IEEE
compliance mode.
</p>
</dd>
<dt><code>mips16_attribute</code></dt>
<dd><p><code>mips16</code> function attributes.
Only MIPS targets support this feature, and only then in certain modes.
</p>
</dd>
<dt><code>tls</code></dt>
<dd><p>Add the target-specific flags needed to use thread-local storage.
</p></dd>
</dl>

<hr>
<a name="Require-Support"></a>
<div class="header">
<p>
Next: <a href="#Final-Actions" accesskey="n" rel="next">Final Actions</a>, Previous: <a href="#Add-Options" accesskey="p" rel="prev">Add Options</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Variants-of-dg_002drequire_002dsupport"></a>
<h4 class="subsection">7.2.5 Variants of <code>dg-require-<var>support</var></code></h4>

<p>A few of the <code>dg-require</code> directives take arguments.
</p>
<dl compact="compact">
<dt><code>dg-require-iconv <var>codeset</var></code></dt>
<dd><p>Skip the test if the target does not support iconv.  <var>codeset</var> is
the codeset to convert to.
</p>
</dd>
<dt><code>dg-require-profiling <var>profopt</var></code></dt>
<dd><p>Skip the test if the target does not support profiling with option
<var>profopt</var>.
</p>
</dd>
<dt><code>dg-require-visibility <var>vis</var></code></dt>
<dd><p>Skip the test if the target does not support the <code>visibility</code> attribute.
If <var>vis</var> is <code>&quot;&quot;</code>, support for <code>visibility(&quot;hidden&quot;)</code> is
checked, for <code>visibility(&quot;<var>vis</var>&quot;)</code> otherwise.
</p></dd>
</dl>

<p>The original <code>dg-require</code> directives were defined before there
was support for effective-target keywords.  The directives that do not
take arguments could be replaced with effective-target keywords.
</p>
<dl compact="compact">
<dt><code>dg-require-alias &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support the &lsquo;<samp>alias</samp>&rsquo; attribute.
</p>
</dd>
<dt><code>dg-require-ascii-locale &quot;&quot;</code></dt>
<dd><p>Skip the test if the host does not support an ASCII locale.
</p>
</dd>
<dt><code>dg-require-compat-dfp &quot;&quot;</code></dt>
<dd><p>Skip this test unless both compilers in a <samp>compat</samp> testsuite
support decimal floating point.
</p>
</dd>
<dt><code>dg-require-cxa-atexit &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support <code>__cxa_atexit</code>.
This is equivalent to <code>dg-require-effective-target cxa_atexit</code>.
</p>
</dd>
<dt><code>dg-require-dll &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support DLL attributes.
</p>
</dd>
<dt><code>dg-require-fork &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support <code>fork</code>.
</p>
</dd>
<dt><code>dg-require-gc-sections &quot;&quot;</code></dt>
<dd><p>Skip the test if the target&rsquo;s linker does not support the
<code>--gc-sections</code> flags.
This is equivalent to <code>dg-require-effective-target gc-sections</code>.
</p>
</dd>
<dt><code>dg-require-host-local &quot;&quot;</code></dt>
<dd><p>Skip the test if the host is remote, rather than the same as the build
system.  Some tests are incompatible with DejaGnu&rsquo;s handling of remote
hosts, which involves copying the source file to the host and compiling
it with a relative path and &quot;<code>-o a.out</code>&quot;.
</p>
</dd>
<dt><code>dg-require-mkfifo &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support <code>mkfifo</code>.
</p>
</dd>
<dt><code>dg-require-named-sections &quot;&quot;</code></dt>
<dd><p>Skip the test is the target does not support named sections.
This is equivalent to <code>dg-require-effective-target named_sections</code>.
</p>
</dd>
<dt><code>dg-require-weak &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support weak symbols.
</p>
</dd>
<dt><code>dg-require-weak-override &quot;&quot;</code></dt>
<dd><p>Skip the test if the target does not support overriding weak symbols.
</p></dd>
</dl>

<hr>
<a name="Final-Actions"></a>
<div class="header">
<p>
Previous: <a href="#Require-Support" accesskey="p" rel="prev">Require Support</a>, Up: <a href="#Test-Directives" accesskey="u" rel="up">Test Directives</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Commands-for-use-in-dg_002dfinal"></a>
<h4 class="subsection">7.2.6 Commands for use in <code>dg-final</code></h4>

<p>The GCC testsuite defines the following directives to be used within
<code>dg-final</code>.
</p>
<a name="Scan-a-particular-file"></a>
<h4 class="subsubsection">7.2.6.1 Scan a particular file</h4>

<dl compact="compact">
<dt><code>scan-file <var>filename</var> <var>regexp</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regexp</var> matches text in <var>filename</var>.
</p></dd>
<dt><code>scan-file-not <var>filename</var> <var>regexp</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regexp</var> does not match text in <var>filename</var>.
</p></dd>
<dt><code>scan-module <var>module</var> <var>regexp</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regexp</var> matches in Fortran module <var>module</var>.
</p></dd>
</dl>

<a name="Scan-the-assembly-output"></a>
<h4 class="subsubsection">7.2.6.2 Scan the assembly output</h4>

<dl compact="compact">
<dt><code>scan-assembler <var>regex</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> matches text in the test&rsquo;s assembler output.
</p>
</dd>
<dt><code>scan-assembler-not <var>regex</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> does not match text in the test&rsquo;s assembler output.
</p>
</dd>
<dt><code>scan-assembler-times <var>regex</var> <var>num</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> is matched exactly <var>num</var> times in the test&rsquo;s
assembler output.
</p>
</dd>
<dt><code>scan-assembler-dem <var>regex</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> matches text in the test&rsquo;s demangled assembler output.
</p>
</dd>
<dt><code>scan-assembler-dem-not <var>regex</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> does not match text in the test&rsquo;s demangled assembler
output.
</p>
</dd>
<dt><code>scan-hidden <var>symbol</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>symbol</var> is defined as a hidden symbol in the test&rsquo;s
assembly output.
</p>
</dd>
<dt><code>scan-not-hidden <var>symbol</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>symbol</var> is not defined as a hidden symbol in the test&rsquo;s
assembly output.
</p></dd>
</dl>

<a name="Scan-optimization-dump-files"></a>
<h4 class="subsubsection">7.2.6.3 Scan optimization dump files</h4>

<p>These commands are available for <var>kind</var> of <code>tree</code>, <code>rtl</code>,
and <code>ipa</code>.
</p>
<dl compact="compact">
<dt><code>scan-<var>kind</var>-dump <var>regex</var> <var>suffix</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> matches text in the dump file with suffix <var>suffix</var>.
</p>
</dd>
<dt><code>scan-<var>kind</var>-dump-not <var>regex</var> <var>suffix</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> does not match text in the dump file with suffix
<var>suffix</var>.
</p>
</dd>
<dt><code>scan-<var>kind</var>-dump-times <var>regex</var> <var>num</var> <var>suffix</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> is found exactly <var>num</var> times in the dump file
with suffix <var>suffix</var>.
</p>
</dd>
<dt><code>scan-<var>kind</var>-dump-dem <var>regex</var> <var>suffix</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> matches demangled text in the dump file with
suffix <var>suffix</var>.
</p>
</dd>
<dt><code>scan-<var>kind</var>-dump-dem-not <var>regex</var> <var>suffix</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if <var>regex</var> does not match demangled text in the dump file with
suffix <var>suffix</var>.
</p></dd>
</dl>

<a name="Verify-that-an-output-files-exists-or-not"></a>
<h4 class="subsubsection">7.2.6.4 Verify that an output files exists or not</h4>

<dl compact="compact">
<dt><code>output-exists [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if compiler output file exists.
</p>
</dd>
<dt><code>output-exists-not [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if compiler output file does not exist.
</p></dd>
</dl>

<a name="Check-for-LTO-tests"></a>
<h4 class="subsubsection">7.2.6.5 Check for LTO tests</h4>

<dl compact="compact">
<dt><code>scan-symbol <var>regexp</var> [{ target/xfail <var>selector</var> }]</code></dt>
<dd><p>Passes if the pattern is present in the final executable.
</p></dd>
</dl>

<a name="Checks-for-gcov-tests"></a>
<h4 class="subsubsection">7.2.6.6 Checks for <code>gcov</code> tests</h4>

<dl compact="compact">
<dt><code>run-gcov <var>sourcefile</var></code></dt>
<dd><p>Check line counts in <code>gcov</code> tests.
</p>
</dd>
<dt><code>run-gcov [branches] [calls] { <var>opts</var> <var>sourcefile</var> }</code></dt>
<dd><p>Check branch and/or call counts, in addition to line counts, in
<code>gcov</code> tests.
</p></dd>
</dl>

<a name="Clean-up-generated-test-files"></a>
<h4 class="subsubsection">7.2.6.7 Clean up generated test files</h4>

<dl compact="compact">
<dt><code>cleanup-coverage-files</code></dt>
<dd><p>Removes coverage data files generated for this test.
</p>
</dd>
<dt><code>cleanup-ipa-dump <var>suffix</var></code></dt>
<dd><p>Removes IPA dump files generated for this test.
</p>
</dd>
<dt><code>cleanup-modules &quot;<var>list-of-extra-modules</var>&quot;</code></dt>
<dd><p>Removes Fortran module files generated for this test, excluding the
module names listed in keep-modules.
Cleaning up module files is usually done automatically by the testsuite
by looking at the source files and removing the modules after the test
has been executed.
</p><div class="smallexample">
<pre class="smallexample">module MoD1
end module MoD1
module Mod2
end module Mod2
module moD3
end module moD3
module mod4
end module mod4
! { dg-final { cleanup-modules &quot;mod1 mod2&quot; } } ! redundant
! { dg-final { keep-modules &quot;mod3 mod4&quot; } }
</pre></div>

</dd>
<dt><code>keep-modules &quot;<var>list-of-modules-not-to-delete</var>&quot;</code></dt>
<dd><p>Whitespace separated list of module names that should not be deleted by
cleanup-modules.
If the list of modules is empty, all modules defined in this file are kept.
</p><div class="smallexample">
<pre class="smallexample">module maybe_unneeded
end module maybe_unneeded
module keep1
end module keep1
module keep2
end module keep2
! { dg-final { keep-modules &quot;keep1 keep2&quot; } } ! just keep these two
! { dg-final { keep-modules &quot;&quot; } } ! keep all
</pre></div>

</dd>
<dt><code>cleanup-profile-file</code></dt>
<dd><p>Removes profiling files generated for this test.
</p>
</dd>
<dt><code>cleanup-repo-files</code></dt>
<dd><p>Removes files generated for this test for <samp>-frepo</samp>.
</p>
</dd>
<dt><code>cleanup-rtl-dump <var>suffix</var></code></dt>
<dd><p>Removes RTL dump files generated for this test.
</p>
</dd>
<dt><code>cleanup-saved-temps</code></dt>
<dd><p>Removes files for the current test which were kept for <samp>-save-temps</samp>.
</p>
</dd>
<dt><code>cleanup-tree-dump <var>suffix</var></code></dt>
<dd><p>Removes tree dump files matching <var>suffix</var> which were generated for
this test.
</p></dd>
</dl>

<hr>
<a name="Ada-Tests"></a>
<div class="header">
<p>
Next: <a href="#C-Tests" accesskey="n" rel="next">C Tests</a>, Previous: <a href="#Test-Directives" accesskey="p" rel="prev">Test Directives</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Ada-Language-Testsuites"></a>
<h3 class="section">7.3 Ada Language Testsuites</h3>

<p>The Ada testsuite includes executable tests from the ACATS
testsuite, publicly available at
<a href="http://www.ada-auth.org/acats.html">http://www.ada-auth.org/acats.html</a>.
</p>
<p>These tests are integrated in the GCC testsuite in the
<samp>ada/acats</samp> directory, and
enabled automatically when running <code>make check</code>, assuming
the Ada language has been enabled when configuring GCC.
</p>
<p>You can also run the Ada testsuite independently, using
<code>make check-ada</code>, or run a subset of the tests by specifying which
chapter to run, e.g.:
</p>
<div class="smallexample">
<pre class="smallexample">$ make check-ada CHAPTERS=&quot;c3 c9&quot;
</pre></div>

<p>The tests are organized by directory, each directory corresponding to
a chapter of the Ada Reference Manual.  So for example, <samp>c9</samp> corresponds
to chapter 9, which deals with tasking features of the language.
</p>
<p>There is also an extra chapter called <samp>gcc</samp> containing a template for
creating new executable tests, although this is deprecated in favor of
the <samp>gnat.dg</samp> testsuite.
</p>
<p>The tests are run using two <code>sh</code> scripts: <samp>run_acats</samp> and
<samp>run_all.sh</samp>.  To run the tests using a simulator or a cross
target, see the small
customization section at the top of <samp>run_all.sh</samp>.
</p>
<p>These tests are run using the build tree: they can be run without doing
a <code>make install</code>.
</p>
<hr>
<a name="C-Tests"></a>
<div class="header">
<p>
Next: <a href="#libgcj-Tests" accesskey="n" rel="next">libgcj Tests</a>, Previous: <a href="#Ada-Tests" accesskey="p" rel="prev">Ada Tests</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="C-Language-Testsuites"></a>
<h3 class="section">7.4 C Language Testsuites</h3>

<p>GCC contains the following C language testsuites, in the
<samp>gcc/testsuite</samp> directory:
</p>
<dl compact="compact">
<dt><samp>gcc.dg</samp></dt>
<dd><p>This contains tests of particular features of the C compiler, using the
more modern &lsquo;<samp>dg</samp>&rsquo; harness.  Correctness tests for various compiler
features should go here if possible.
</p>
<p>Magic comments determine whether the file
is preprocessed, compiled, linked or run.  In these tests, error and warning
message texts are compared against expected texts or regular expressions
given in comments.  These tests are run with the options &lsquo;<samp>-ansi -pedantic</samp>&rsquo;
unless other options are given in the test.  Except as noted below they
are not run with multiple optimization options.
</p></dd>
<dt><samp>gcc.dg/compat</samp></dt>
<dd><p>This subdirectory contains tests for binary compatibility using
<samp>lib/compat.exp</samp>, which in turn uses the language-independent support
(see <a href="#compat-Testing">Support for testing binary compatibility</a>).
</p></dd>
<dt><samp>gcc.dg/cpp</samp></dt>
<dd><p>This subdirectory contains tests of the preprocessor.
</p></dd>
<dt><samp>gcc.dg/debug</samp></dt>
<dd><p>This subdirectory contains tests for debug formats.  Tests in this
subdirectory are run for each debug format that the compiler supports.
</p></dd>
<dt><samp>gcc.dg/format</samp></dt>
<dd><p>This subdirectory contains tests of the <samp>-Wformat</samp> format
checking.  Tests in this directory are run with and without
<samp>-DWIDE</samp>.
</p></dd>
<dt><samp>gcc.dg/noncompile</samp></dt>
<dd><p>This subdirectory contains tests of code that should not compile and
does not need any special compilation options.  They are run with
multiple optimization options, since sometimes invalid code crashes
the compiler with optimization.
</p></dd>
<dt><samp>gcc.dg/special</samp></dt>
<dd><p>FIXME: describe this.
</p>
</dd>
<dt><samp>gcc.c-torture</samp></dt>
<dd><p>This contains particular code fragments which have historically broken easily.
These tests are run with multiple optimization options, so tests for features
which only break at some optimization levels belong here.  This also contains
tests to check that certain optimizations occur.  It might be worthwhile to
separate the correctness tests cleanly from the code quality tests, but
it hasn&rsquo;t been done yet.
</p>
</dd>
<dt><samp>gcc.c-torture/compat</samp></dt>
<dd><p>FIXME: describe this.
</p>
<p>This directory should probably not be used for new tests.
</p></dd>
<dt><samp>gcc.c-torture/compile</samp></dt>
<dd><p>This testsuite contains test cases that should compile, but do not
need to link or run.  These test cases are compiled with several
different combinations of optimization options.  All warnings are
disabled for these test cases, so this directory is not suitable if
you wish to test for the presence or absence of compiler warnings.
While special options can be set, and tests disabled on specific
platforms, by the use of <samp>.x</samp> files, mostly these test cases
should not contain platform dependencies.  FIXME: discuss how defines
such as <code>NO_LABEL_VALUES</code> and <code>STACK_SIZE</code> are used.
</p></dd>
<dt><samp>gcc.c-torture/execute</samp></dt>
<dd><p>This testsuite contains test cases that should compile, link and run;
otherwise the same comments as for <samp>gcc.c-torture/compile</samp> apply.
</p></dd>
<dt><samp>gcc.c-torture/execute/ieee</samp></dt>
<dd><p>This contains tests which are specific to IEEE floating point.
</p></dd>
<dt><samp>gcc.c-torture/unsorted</samp></dt>
<dd><p>FIXME: describe this.
</p>
<p>This directory should probably not be used for new tests.
</p></dd>
<dt><samp>gcc.misc-tests</samp></dt>
<dd><p>This directory contains C tests that require special handling.  Some
of these tests have individual expect files, and others share
special-purpose expect files:
</p>
<dl compact="compact">
<dt><samp><code>bprob*.c</code></samp></dt>
<dd><p>Test <samp>-fbranch-probabilities</samp> using
<samp>gcc.misc-tests/bprob.exp</samp>, which
in turn uses the generic, language-independent framework
(see <a href="#profopt-Testing">Support for testing profile-directed
optimizations</a>).
</p>
</dd>
<dt><samp><code>gcov*.c</code></samp></dt>
<dd><p>Test <code>gcov</code> output using <samp>gcov.exp</samp>, which in turn uses the
language-independent support (see <a href="#gcov-Testing">Support for testing gcov</a>).
</p>
</dd>
<dt><samp><code>i386-pf-*.c</code></samp></dt>
<dd><p>Test i386-specific support for data prefetch using <samp>i386-prefetch.exp</samp>.
</p></dd>
</dl>

</dd>
<dt><samp>gcc.test-framework</samp></dt>
<dd><dl compact="compact">
<dt><samp><code>dg-*.c</code></samp></dt>
<dd><p>Test the testsuite itself using <samp>gcc.test-framework/test-framework.exp</samp>.
</p></dd>
</dl>

</dd>
</dl>

<p>FIXME: merge in <samp>testsuite/README.gcc</samp> and discuss the format of
test cases and magic comments more.
</p>
<hr>
<a name="libgcj-Tests"></a>
<div class="header">
<p>
Next: <a href="#LTO-Testing" accesskey="n" rel="next">LTO Testing</a>, Previous: <a href="#C-Tests" accesskey="p" rel="prev">C Tests</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Java-library-testsuites_002e"></a>
<h3 class="section">7.5 The Java library testsuites.</h3>

<p>Runtime tests are executed via &lsquo;<samp>make check</samp>&rsquo; in the
<samp><var>target</var>/libjava/testsuite</samp> directory in the build
tree.  Additional runtime tests can be checked into this testsuite.
</p>
<p>Regression testing of the core packages in libgcj is also covered by the
Mauve testsuite.  The <a href="http://sourceware.org/mauve/">Mauve Project</a>
develops tests for the Java Class Libraries.  These tests are run as part
of libgcj testing by placing the Mauve tree within the libjava testsuite
sources at <samp>libjava/testsuite/libjava.mauve/mauve</samp>, or by specifying
the location of that tree when invoking &lsquo;<samp>make</samp>&rsquo;, as in
&lsquo;<samp>make MAUVEDIR=~/mauve check</samp>&rsquo;.
</p>
<p>To detect regressions, a mechanism in <samp>mauve.exp</samp> compares the
failures for a test run against the list of expected failures in
<samp>libjava/testsuite/libjava.mauve/xfails</samp> from the source hierarchy.
Update this file when adding new failing tests to Mauve, or when fixing
bugs in libgcj that had caused Mauve test failures.
</p>
<p>We encourage developers to contribute test cases to Mauve.
</p>
<hr>
<a name="LTO-Testing"></a>
<div class="header">
<p>
Next: <a href="#gcov-Testing" accesskey="n" rel="next">gcov Testing</a>, Previous: <a href="#libgcj-Tests" accesskey="p" rel="prev">libgcj Tests</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-testing-link_002dtime-optimizations"></a>
<h3 class="section">7.6 Support for testing link-time optimizations</h3>

<p>Tests for link-time optimizations usually require multiple source files
that are compiled separately, perhaps with different sets of options.
There are several special-purpose test directives used for these tests.
</p>
<dl compact="compact">
<dt><code>{ dg-lto-do <var>do-what-keyword</var> }</code></dt>
<dd><p><var>do-what-keyword</var> specifies how the test is compiled and whether
it is executed.  It is one of:
</p>
<dl compact="compact">
<dt><code>assemble</code></dt>
<dd><p>Compile with <samp>-c</samp> to produce a relocatable object file.
</p></dd>
<dt><code>link</code></dt>
<dd><p>Compile, assemble, and link to produce an executable file.
</p></dd>
<dt><code>run</code></dt>
<dd><p>Produce and run an executable file, which is expected to return
an exit code of 0.
</p></dd>
</dl>

<p>The default is <code>assemble</code>.  That can be overridden for a set of
tests by redefining <code>dg-do-what-default</code> within the <code>.exp</code>
file for those tests.
</p>
<p>Unlike <code>dg-do</code>, <code>dg-lto-do</code> does not support an optional
&lsquo;<samp>target</samp>&rsquo; or &lsquo;<samp>xfail</samp>&rsquo; list.  Use <code>dg-skip-if</code>,
<code>dg-xfail-if</code>, or <code>dg-xfail-run-if</code>.
</p>
</dd>
<dt><code>{ dg-lto-options { { <var>options</var> } [{ <var>options</var> }] } [{ target <var>selector</var> }]}</code></dt>
<dd><p>This directive provides a list of one or more sets of compiler options
to override <var>LTO_OPTIONS</var>.  Each test will be compiled and run with
each of these sets of options.
</p>
</dd>
<dt><code>{ dg-extra-ld-options <var>options</var> [{ target <var>selector</var> }]}</code></dt>
<dd><p>This directive adds <var>options</var> to the linker options used.
</p>
</dd>
<dt><code>{ dg-suppress-ld-options <var>options</var> [{ target <var>selector</var> }]}</code></dt>
<dd><p>This directive removes <var>options</var> from the set of linker options used.
</p></dd>
</dl>

<hr>
<a name="gcov-Testing"></a>
<div class="header">
<p>
Next: <a href="#profopt-Testing" accesskey="n" rel="next">profopt Testing</a>, Previous: <a href="#LTO-Testing" accesskey="p" rel="prev">LTO Testing</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-testing-gcov"></a>
<h3 class="section">7.7 Support for testing <code>gcov</code></h3>

<p>Language-independent support for testing <code>gcov</code>, and for checking
that branch profiling produces expected values, is provided by the
expect file <samp>lib/gcov.exp</samp>.  <code>gcov</code> tests also rely on procedures
in <samp>lib/gcc-dg.exp</samp> to compile and run the test program.  A typical
<code>gcov</code> test contains the following DejaGnu commands within comments:
</p>
<div class="smallexample">
<pre class="smallexample">{ dg-options &quot;-fprofile-arcs -ftest-coverage&quot; }
{ dg-do run { target native } }
{ dg-final { run-gcov sourcefile } }
</pre></div>

<p>Checks of <code>gcov</code> output can include line counts, branch percentages,
and call return percentages.  All of these checks are requested via
commands that appear in comments in the test&rsquo;s source file.
Commands to check line counts are processed by default.
Commands to check branch percentages and call return percentages are
processed if the <code>run-gcov</code> command has arguments <code>branches</code>
or <code>calls</code>, respectively.  For example, the following specifies
checking both, as well as passing <samp>-b</samp> to <code>gcov</code>:
</p>
<div class="smallexample">
<pre class="smallexample">{ dg-final { run-gcov branches calls { -b sourcefile } } }
</pre></div>

<p>A line count command appears within a comment on the source line
that is expected to get the specified count and has the form
<code>count(<var>cnt</var>)</code>.  A test should only check line counts for
lines that will get the same count for any architecture.
</p>
<p>Commands to check branch percentages (<code>branch</code>) and call
return percentages (<code>returns</code>) are very similar to each other.
A beginning command appears on or before the first of a range of
lines that will report the percentage, and the ending command
follows that range of lines.  The beginning command can include a
list of percentages, all of which are expected to be found within
the range.  A range is terminated by the next command of the same
kind.  A command <code>branch(end)</code> or <code>returns(end)</code> marks
the end of a range without starting a new one.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">if (i &gt; 10 &amp;&amp; j &gt; i &amp;&amp; j &lt; 20)  /* <span class="roman">branch(27 50 75)</span> */
                                /* <span class="roman">branch(end)</span> */
  foo (i, j);
</pre></div>

<p>For a call return percentage, the value specified is the
percentage of calls reported to return.  For a branch percentage,
the value is either the expected percentage or 100 minus that
value, since the direction of a branch can differ depending on the
target or the optimization level.
</p>
<p>Not all branches and calls need to be checked.  A test should not
check for branches that might be optimized away or replaced with
predicated instructions.  Don&rsquo;t check for calls inserted by the
compiler or ones that might be inlined or optimized away.
</p>
<p>A single test can check for combinations of line counts, branch
percentages, and call return percentages.  The command to check a
line count must appear on the line that will report that count, but
commands to check branch percentages and call return percentages can
bracket the lines that report them.
</p>
<hr>
<a name="profopt-Testing"></a>
<div class="header">
<p>
Next: <a href="#compat-Testing" accesskey="n" rel="next">compat Testing</a>, Previous: <a href="#gcov-Testing" accesskey="p" rel="prev">gcov Testing</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-testing-profile_002ddirected-optimizations"></a>
<h3 class="section">7.8 Support for testing profile-directed optimizations</h3>

<p>The file <samp>profopt.exp</samp> provides language-independent support for
checking correct execution of a test built with profile-directed
optimization.  This testing requires that a test program be built and
executed twice.  The first time it is compiled to generate profile
data, and the second time it is compiled to use the data that was
generated during the first execution.  The second execution is to
verify that the test produces the expected results.
</p>
<p>To check that the optimization actually generated better code, a
test can be built and run a third time with normal optimizations to
verify that the performance is better with the profile-directed
optimizations.  <samp>profopt.exp</samp> has the beginnings of this kind
of support.
</p>
<p><samp>profopt.exp</samp> provides generic support for profile-directed
optimizations.  Each set of tests that uses it provides information
about a specific optimization:
</p>
<dl compact="compact">
<dt><code>tool</code></dt>
<dd><p>tool being tested, e.g., <code>gcc</code>
</p>
</dd>
<dt><code>profile_option</code></dt>
<dd><p>options used to generate profile data
</p>
</dd>
<dt><code>feedback_option</code></dt>
<dd><p>options used to optimize using that profile data
</p>
</dd>
<dt><code>prof_ext</code></dt>
<dd><p>suffix of profile data files
</p>
</dd>
<dt><code>PROFOPT_OPTIONS</code></dt>
<dd><p>list of options with which to run each test, similar to the lists for
torture tests
</p>
</dd>
<dt><code>{ dg-final-generate { <var>local-directive</var> } }</code></dt>
<dd><p>This directive is similar to <code>dg-final</code>, but the
<var>local-directive</var> is run after the generation of profile data.
</p>
</dd>
<dt><code>{ dg-final-use { <var>local-directive</var> } }</code></dt>
<dd><p>The <var>local-directive</var> is run after the profile data have been
used.
</p></dd>
</dl>

<hr>
<a name="compat-Testing"></a>
<div class="header">
<p>
Next: <a href="#Torture-Tests" accesskey="n" rel="next">Torture Tests</a>, Previous: <a href="#profopt-Testing" accesskey="p" rel="prev">profopt Testing</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-testing-binary-compatibility"></a>
<h3 class="section">7.9 Support for testing binary compatibility</h3>

<p>The file <samp>compat.exp</samp> provides language-independent support for
binary compatibility testing.  It supports testing interoperability of
two compilers that follow the same ABI, or of multiple sets of
compiler options that should not affect binary compatibility.  It is
intended to be used for testsuites that complement ABI testsuites.
</p>
<p>A test supported by this framework has three parts, each in a
separate source file: a main program and two pieces that interact
with each other to split up the functionality being tested.
</p>
<dl compact="compact">
<dt><samp><var>testname</var>_main.<var>suffix</var></samp></dt>
<dd><p>Contains the main program, which calls a function in file
<samp><var>testname</var>_x.<var>suffix</var></samp>.
</p>
</dd>
<dt><samp><var>testname</var>_x.<var>suffix</var></samp></dt>
<dd><p>Contains at least one call to a function in
<samp><var>testname</var>_y.<var>suffix</var></samp>.
</p>
</dd>
<dt><samp><var>testname</var>_y.<var>suffix</var></samp></dt>
<dd><p>Shares data with, or gets arguments from,
<samp><var>testname</var>_x.<var>suffix</var></samp>.
</p></dd>
</dl>

<p>Within each test, the main program and one functional piece are
compiled by the GCC under test.  The other piece can be compiled by
an alternate compiler.  If no alternate compiler is specified,
then all three source files are all compiled by the GCC under test.
You can specify pairs of sets of compiler options.  The first element
of such a pair specifies options used with the GCC under test, and the
second element of the pair specifies options used with the alternate
compiler.  Each test is compiled with each pair of options.
</p>
<p><samp>compat.exp</samp> defines default pairs of compiler options.
These can be overridden by defining the environment variable
<code>COMPAT_OPTIONS</code> as:
</p>
<div class="smallexample">
<pre class="smallexample">COMPAT_OPTIONS=&quot;[list [list {<var>tst1</var>} {<var>alt1</var>}]
  &hellip;[list {<var>tstn</var>} {<var>altn</var>}]]&quot;
</pre></div>

<p>where <var>tsti</var> and <var>alti</var> are lists of options, with <var>tsti</var>
used by the compiler under test and <var>alti</var> used by the alternate
compiler.  For example, with
<code>[list [list {-g -O0} {-O3}] [list {-fpic} {-fPIC -O2}]]</code>,
the test is first built with <samp>-g -O0</samp> by the compiler under
test and with <samp>-O3</samp> by the alternate compiler.  The test is
built a second time using <samp>-fpic</samp> by the compiler under test
and <samp>-fPIC -O2</samp> by the alternate compiler.
</p>
<p>An alternate compiler is specified by defining an environment
variable to be the full pathname of an installed compiler; for C
define <code>ALT_CC_UNDER_TEST</code>, and for C++ define
<code>ALT_CXX_UNDER_TEST</code>.  These will be written to the
<samp>site.exp</samp> file used by DejaGnu.  The default is to build each
test with the compiler under test using the first of each pair of
compiler options from <code>COMPAT_OPTIONS</code>.  When
<code>ALT_CC_UNDER_TEST</code> or
<code>ALT_CXX_UNDER_TEST</code> is <code>same</code>, each test is built using
the compiler under test but with combinations of the options from
<code>COMPAT_OPTIONS</code>.
</p>
<p>To run only the C++ compatibility suite using the compiler under test
and another version of GCC using specific compiler options, do the
following from <samp><var>objdir</var>/gcc</samp>:
</p>
<div class="smallexample">
<pre class="smallexample">rm site.exp
make -k \
  ALT_CXX_UNDER_TEST=${alt_prefix}/bin/g++ \
  COMPAT_OPTIONS=&quot;<var>lists as shown above</var>&quot; \
  check-c++ \
  RUNTESTFLAGS=&quot;compat.exp&quot;
</pre></div>

<p>A test that fails when the source files are compiled with different
compilers, but passes when the files are compiled with the same
compiler, demonstrates incompatibility of the generated code or
runtime support.  A test that fails for the alternate compiler but
passes for the compiler under test probably tests for a bug that was
fixed in the compiler under test but is present in the alternate
compiler.
</p>
<p>The binary compatibility tests support a small number of test framework
commands that appear within comments in a test file.
</p>
<dl compact="compact">
<dt><code>dg-require-*</code></dt>
<dd><p>These commands can be used in <samp><var>testname</var>_main.<var>suffix</var></samp>
to skip the test if specific support is not available on the target.
</p>
</dd>
<dt><code>dg-options</code></dt>
<dd><p>The specified options are used for compiling this particular source
file, appended to the options from <code>COMPAT_OPTIONS</code>.  When this
command appears in <samp><var>testname</var>_main.<var>suffix</var></samp> the options
are also used to link the test program.
</p>
</dd>
<dt><code>dg-xfail-if</code></dt>
<dd><p>This command can be used in a secondary source file to specify that
compilation is expected to fail for particular options on particular
targets.
</p></dd>
</dl>

<hr>
<a name="Torture-Tests"></a>
<div class="header">
<p>
Previous: <a href="#compat-Testing" accesskey="p" rel="prev">compat Testing</a>, Up: <a href="#Testsuites" accesskey="u" rel="up">Testsuites</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-torture-testing-using-multiple-options"></a>
<h3 class="section">7.10 Support for torture testing using multiple options</h3>

<p>Throughout the compiler testsuite there are several directories whose
tests are run multiple times, each with a different set of options.
These are known as torture tests.
<samp>lib/torture-options.exp</samp> defines procedures to
set up these lists:
</p>
<dl compact="compact">
<dt><code>torture-init</code></dt>
<dd><p>Initialize use of torture lists.
</p></dd>
<dt><code>set-torture-options</code></dt>
<dd><p>Set lists of torture options to use for tests with and without loops.
Optionally combine a set of torture options with a set of other
options, as is done with Objective-C runtime options.
</p></dd>
<dt><code>torture-finish</code></dt>
<dd><p>Finalize use of torture lists.
</p></dd>
</dl>

<p>The <samp>.exp</samp> file for a set of tests that use torture options must
include calls to these three procedures if:
</p>
<ul>
<li> It calls <code>gcc-dg-runtest</code> and overrides <var>DG_TORTURE_OPTIONS</var>.

</li><li> It calls <var>${tool}</var><code>-torture</code> or
<var>${tool}</var><code>-torture-execute</code>, where <var>tool</var> is <code>c</code>,
<code>fortran</code>, or <code>objc</code>.

</li><li> It calls <code>dg-pch</code>.
</li></ul>

<p>It is not necessary for a <samp>.exp</samp> file that calls <code>gcc-dg-runtest</code>
to call the torture procedures if the tests should use the list in
<var>DG_TORTURE_OPTIONS</var> defined in <samp>gcc-dg.exp</samp>.
</p>
<p>Most uses of torture options can override the default lists by defining
<var>TORTURE_OPTIONS</var> or add to the default list by defining
<var>ADDITIONAL_TORTURE_OPTIONS</var>.  Define these in a <samp>.dejagnurc</samp>
file or add them to the <samp>site.exp</samp> file; for example
</p>
<div class="smallexample">
<pre class="smallexample">set ADDITIONAL_TORTURE_OPTIONS  [list \
  { -O2 -ftree-loop-linear } \
  { -O2 -fpeel-loops } ]
</pre></div>

<hr>
<a name="Options"></a>
<div class="header">
<p>
Next: <a href="#Passes" accesskey="n" rel="next">Passes</a>, Previous: <a href="#Testsuites" accesskey="p" rel="prev">Testsuites</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Option-specification-files"></a>
<h2 class="chapter">8 Option specification files</h2>
<a name="index-option-specification-files"></a>
<a name="index-optc_002dgen_002eawk"></a>

<p>Most GCC command-line options are described by special option
definition files, the names of which conventionally end in
<code>.opt</code>.  This chapter describes the format of these files.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Option-file-format" accesskey="1">Option file format</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The general layout of the files
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Option-properties" accesskey="2">Option properties</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Supported option properties
</td></tr>
</table>

<hr>
<a name="Option-file-format"></a>
<div class="header">
<p>
Next: <a href="#Option-properties" accesskey="n" rel="next">Option properties</a>, Up: <a href="#Options" accesskey="u" rel="up">Options</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Option-file-format-1"></a>
<h3 class="section">8.1 Option file format</h3>

<p>Option files are a simple list of records in which each field occupies
its own line and in which the records themselves are separated by
blank lines.  Comments may appear on their own line anywhere within
the file and are preceded by semicolons.  Whitespace is allowed before
the semicolon.
</p>
<p>The files can contain the following types of record:
</p>
<ul>
<li> A language definition record.  These records have two fields: the
string &lsquo;<samp>Language</samp>&rsquo; and the name of the language.  Once a language
has been declared in this way, it can be used as an option property.
See <a href="#Option-properties">Option properties</a>.

</li><li> A target specific save record to save additional information. These
records have two fields: the string &lsquo;<samp>TargetSave</samp>&rsquo;, and a
declaration type to go in the <code>cl_target_option</code> structure.

</li><li> A variable record to define a variable used to store option
information.  These records have two fields: the string
&lsquo;<samp>Variable</samp>&rsquo;, and a declaration of the type and name of the
variable, optionally with an initializer (but without any trailing
&lsquo;<samp>;</samp>&rsquo;).  These records may be used for variables used for many
options where declaring the initializer in a single option definition
record, or duplicating it in many records, would be inappropriate, or
for variables set in option handlers rather than referenced by
<code>Var</code> properties.

</li><li> A variable record to define a variable used to store option
information.  These records have two fields: the string
&lsquo;<samp>TargetVariable</samp>&rsquo;, and a declaration of the type and name of the
variable, optionally with an initializer (but without any trailing
&lsquo;<samp>;</samp>&rsquo;).  &lsquo;<samp>TargetVariable</samp>&rsquo; is a combination of &lsquo;<samp>Variable</samp>&rsquo;
and &lsquo;<samp>TargetSave</samp>&rsquo; records in that the variable is defined in the
<code>gcc_options</code> structure, but these variables are also stored in
the <code>cl_target_option</code> structure.  The variables are saved in the
target save code and restored in the target restore code.

</li><li> A variable record to record any additional files that the
<samp>options.h</samp> file should include.  This is useful to provide
enumeration or structure definitions needed for target variables.
These records have two fields: the string &lsquo;<samp>HeaderInclude</samp>&rsquo; and the
name of the include file.

</li><li> A variable record to record any additional files that the
<samp>options.c</samp> or <samp>options-save.c</samp> file should include.  This
is useful to provide
inline functions needed for target variables and/or <code>#ifdef</code>
sequences to properly set up the initialization.  These records have
two fields: the string &lsquo;<samp>SourceInclude</samp>&rsquo; and the name of the
include file.

</li><li> An enumeration record to define a set of strings that may be used as
arguments to an option or options.  These records have three fields:
the string &lsquo;<samp>Enum</samp>&rsquo;, a space-separated list of properties and help
text used to describe the set of strings in <samp>--help</samp> output.
Properties use the same format as option properties; the following are
valid:
<dl compact="compact">
<dt><code>Name(<var>name</var>)</code></dt>
<dd><p>This property is required; <var>name</var> must be a name (suitable for use
in C identifiers) used to identify the set of strings in <code>Enum</code>
option properties.
</p>
</dd>
<dt><code>Type(<var>type</var>)</code></dt>
<dd><p>This property is required; <var>type</var> is the C type for variables set
by options using this enumeration together with <code>Var</code>.
</p>
</dd>
<dt><code>UnknownError(<var>message</var>)</code></dt>
<dd><p>The message <var>message</var> will be used as an error message if the
argument is invalid; for enumerations without <code>UnknownError</code>, a
generic error message is used.  <var>message</var> should contain a single
&lsquo;<samp>%qs</samp>&rsquo; format, which will be used to format the invalid argument.
</p></dd>
</dl>

</li><li> An enumeration value record to define one of the strings in a set
given in an &lsquo;<samp>Enum</samp>&rsquo; record.  These records have two fields: the
string &lsquo;<samp>EnumValue</samp>&rsquo; and a space-separated list of properties.
Properties use the same format as option properties; the following are
valid:
<dl compact="compact">
<dt><code>Enum(<var>name</var>)</code></dt>
<dd><p>This property is required; <var>name</var> says which &lsquo;<samp>Enum</samp>&rsquo; record
this &lsquo;<samp>EnumValue</samp>&rsquo; record corresponds to.
</p>
</dd>
<dt><code>String(<var>string</var>)</code></dt>
<dd><p>This property is required; <var>string</var> is the string option argument
being described by this record.
</p>
</dd>
<dt><code>Value(<var>value</var>)</code></dt>
<dd><p>This property is required; it says what value (representable as
<code>int</code>) should be used for the given string.
</p>
</dd>
<dt><code>Canonical</code></dt>
<dd><p>This property is optional.  If present, it says the present string is
the canonical one among all those with the given value.  Other strings
yielding that value will be mapped to this one so specs do not need to
handle them.
</p>
</dd>
<dt><code>DriverOnly</code></dt>
<dd><p>This property is optional.  If present, the present string will only
be accepted by the driver.  This is used for cases such as
<samp>-march=native</samp> that are processed by the driver so that
&lsquo;<samp>gcc -v</samp>&rsquo; shows how the options chosen depended on the system on
which the compiler was run.
</p></dd>
</dl>

</li><li> An option definition record.  These records have the following fields:
<ol>
<li> the name of the option, with the leading &ldquo;-&rdquo; removed
</li><li> a space-separated list of option properties (see <a href="#Option-properties">Option properties</a>)
</li><li> the help text to use for <samp>--help</samp> (omitted if the second field
contains the <code>Undocumented</code> property).
</li></ol>

<p>By default, all options beginning with &ldquo;f&rdquo;, &ldquo;W&rdquo; or &ldquo;m&rdquo; are
implicitly assumed to take a &ldquo;no-&rdquo; form.  This form should not be
listed separately.  If an option beginning with one of these letters
does not have a &ldquo;no-&rdquo; form, you can use the <code>RejectNegative</code>
property to reject it.
</p>
<p>The help text is automatically line-wrapped before being displayed.
Normally the name of the option is printed on the left-hand side of
the output and the help text is printed on the right.  However, if the
help text contains a tab character, the text to the left of the tab is
used instead of the option&rsquo;s name and the text to the right of the
tab forms the help text.  This allows you to elaborate on what type
of argument the option takes.
</p>
</li><li> A target mask record.  These records have one field of the form
&lsquo;<samp>Mask(<var>x</var>)</samp>&rsquo;.  The options-processing script will automatically
allocate a bit in <code>target_flags</code> (see <a href="#Run_002dtime-Target">Run-time Target</a>) for
each mask name <var>x</var> and set the macro <code>MASK_<var>x</var></code> to the
appropriate bitmask.  It will also declare a <code>TARGET_<var>x</var></code>
macro that has the value 1 when bit <code>MASK_<var>x</var></code> is set and
0 otherwise.

<p>They are primarily intended to declare target masks that are not
associated with user options, either because these masks represent
internal switches or because the options are not available on all
configurations and yet the masks always need to be defined.
</p></li></ul>

<hr>
<a name="Option-properties"></a>
<div class="header">
<p>
Previous: <a href="#Option-file-format" accesskey="p" rel="prev">Option file format</a>, Up: <a href="#Options" accesskey="u" rel="up">Options</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Option-properties-1"></a>
<h3 class="section">8.2 Option properties</h3>

<p>The second field of an option record can specify any of the following
properties.  When an option takes an argument, it is enclosed in parentheses
following the option property name.  The parser that handles option files
is quite simplistic, and will be tricked by any nested parentheses within
the argument text itself; in this case, the entire option argument can
be wrapped in curly braces within the parentheses to demarcate it, e.g.:
</p>
<div class="smallexample">
<pre class="smallexample">Condition({defined (USE_CYGWIN_LIBSTDCXX_WRAPPERS)})
</pre></div>

<dl compact="compact">
<dt><code>Common</code></dt>
<dd><p>The option is available for all languages and targets.
</p>
</dd>
<dt><code>Target</code></dt>
<dd><p>The option is available for all languages but is target-specific.
</p>
</dd>
<dt><code>Driver</code></dt>
<dd><p>The option is handled by the compiler driver using code not shared
with the compilers proper (<samp>cc1</samp> etc.).
</p>
</dd>
<dt><code><var>language</var></code></dt>
<dd><p>The option is available when compiling for the given language.
</p>
<p>It is possible to specify several different languages for the same
option.  Each <var>language</var> must have been declared by an earlier
<code>Language</code> record.  See <a href="#Option-file-format">Option file format</a>.
</p>
</dd>
<dt><code>RejectDriver</code></dt>
<dd><p>The option is only handled by the compilers proper (<samp>cc1</samp> etc.)
and should not be accepted by the driver.
</p>
</dd>
<dt><code>RejectNegative</code></dt>
<dd><p>The option does not have a &ldquo;no-&rdquo; form.  All options beginning with
&ldquo;f&rdquo;, &ldquo;W&rdquo; or &ldquo;m&rdquo; are assumed to have a &ldquo;no-&rdquo; form unless this
property is used.
</p>
</dd>
<dt><code>Negative(<var>othername</var>)</code></dt>
<dd><p>The option will turn off another option <var>othername</var>, which is
the option name with the leading &ldquo;-&rdquo; removed.  This chain action will
propagate through the <code>Negative</code> property of the option to be
turned off.
</p>
<p>As a consequence, if you have a group of mutually-exclusive
options, their <code>Negative</code> properties should form a circular chain.
For example, if options <samp>-<var>a</var></samp>, <samp>-<var>b</var></samp> and
<samp>-<var>c</var></samp> are mutually exclusive, their respective <code>Negative</code>
properties should be &lsquo;<samp>Negative(<var>b</var>)</samp>&rsquo;, &lsquo;<samp>Negative(<var>c</var>)</samp>&rsquo;
and &lsquo;<samp>Negative(<var>a</var>)</samp>&rsquo;.
</p>
</dd>
<dt><code>Joined</code></dt>
<dt><code>Separate</code></dt>
<dd><p>The option takes a mandatory argument.  <code>Joined</code> indicates
that the option and argument can be included in the same <code>argv</code>
entry (as with <code>-mflush-func=<var>name</var></code>, for example).
<code>Separate</code> indicates that the option and argument can be
separate <code>argv</code> entries (as with <code>-o</code>).  An option is
allowed to have both of these properties.
</p>
</dd>
<dt><code>JoinedOrMissing</code></dt>
<dd><p>The option takes an optional argument.  If the argument is given,
it will be part of the same <code>argv</code> entry as the option itself.
</p>
<p>This property cannot be used alongside <code>Joined</code> or <code>Separate</code>.
</p>
</dd>
<dt><code>MissingArgError(<var>message</var>)</code></dt>
<dd><p>For an option marked <code>Joined</code> or <code>Separate</code>, the message
<var>message</var> will be used as an error message if the mandatory
argument is missing; for options without <code>MissingArgError</code>, a
generic error message is used.  <var>message</var> should contain a single
&lsquo;<samp>%qs</samp>&rsquo; format, which will be used to format the name of the option
passed.
</p>
</dd>
<dt><code>Args(<var>n</var>)</code></dt>
<dd><p>For an option marked <code>Separate</code>, indicate that it takes <var>n</var>
arguments.  The default is 1.
</p>
</dd>
<dt><code>UInteger</code></dt>
<dd><p>The option&rsquo;s argument is a non-negative integer.  The option parser
will check and convert the argument before passing it to the relevant
option handler.  <code>UInteger</code> should also be used on options like
<code>-falign-loops</code> where both <code>-falign-loops</code> and
<code>-falign-loops</code>=<var>n</var> are supported to make sure the saved
options are given a full integer.
</p>
</dd>
<dt><code>ToLower</code></dt>
<dd><p>The option&rsquo;s argument should be converted to lowercase as part of
putting it in canonical form, and before comparing with the strings
indicated by any <code>Enum</code> property.
</p>
</dd>
<dt><code>NoDriverArg</code></dt>
<dd><p>For an option marked <code>Separate</code>, the option only takes an
argument in the compiler proper, not in the driver.  This is for
compatibility with existing options that are used both directly and
via <samp>-Wp,</samp>; new options should not have this property.
</p>
</dd>
<dt><code>Var(<var>var</var>)</code></dt>
<dd><p>The state of this option should be stored in variable <var>var</var>
(actually a macro for <code>global_options.x_<var>var</var></code>).
The way that the state is stored depends on the type of option:
</p>
<ul>
<li> If the option uses the <code>Mask</code> or <code>InverseMask</code> properties,
<var>var</var> is the integer variable that contains the mask.

</li><li> If the option is a normal on/off switch, <var>var</var> is an integer
variable that is nonzero when the option is enabled.  The options
parser will set the variable to 1 when the positive form of the
option is used and 0 when the &ldquo;no-&rdquo; form is used.

</li><li> If the option takes an argument and has the <code>UInteger</code> property,
<var>var</var> is an integer variable that stores the value of the argument.

</li><li> If the option takes an argument and has the <code>Enum</code> property,
<var>var</var> is a variable (type given in the <code>Type</code> property of the
&lsquo;<samp>Enum</samp>&rsquo; record whose <code>Name</code> property has the same argument as
the <code>Enum</code> property of this option) that stores the value of the
argument.

</li><li> If the option has the <code>Defer</code> property, <var>var</var> is a pointer to
a <code>VEC(cl_deferred_option,heap)</code> that stores the option for later
processing.  (<var>var</var> is declared with type <code>void *</code> and needs
to be cast to <code>VEC(cl_deferred_option,heap)</code> before use.)

</li><li> Otherwise, if the option takes an argument, <var>var</var> is a pointer to
the argument string.  The pointer will be null if the argument is optional
and wasn&rsquo;t given.
</li></ul>

<p>The option-processing script will usually zero-initialize <var>var</var>.
You can modify this behavior using <code>Init</code>.
</p>
</dd>
<dt><code>Var(<var>var</var>, <var>set</var>)</code></dt>
<dd><p>The option controls an integer variable <var>var</var> and is active when
<var>var</var> equals <var>set</var>.  The option parser will set <var>var</var> to
<var>set</var> when the positive form of the option is used and <code>!<var>set</var></code>
when the &ldquo;no-&rdquo; form is used.
</p>
<p><var>var</var> is declared in the same way as for the single-argument form
described above.
</p>
</dd>
<dt><code>Init(<var>value</var>)</code></dt>
<dd><p>The variable specified by the <code>Var</code> property should be statically
initialized to <var>value</var>.  If more than one option using the same
variable specifies <code>Init</code>, all must specify the same initializer.
</p>
</dd>
<dt><code>Mask(<var>name</var>)</code></dt>
<dd><p>The option is associated with a bit in the <code>target_flags</code>
variable (see <a href="#Run_002dtime-Target">Run-time Target</a>) and is active when that bit is set.
You may also specify <code>Var</code> to select a variable other than
<code>target_flags</code>.
</p>
<p>The options-processing script will automatically allocate a unique bit
for the option.  If the option is attached to &lsquo;<samp>target_flags</samp>&rsquo;,
the script will set the macro <code>MASK_<var>name</var></code> to the appropriate
bitmask.  It will also declare a <code>TARGET_<var>name</var></code> macro that has
the value 1 when the option is active and 0 otherwise.  If you use <code>Var</code>
to attach the option to a different variable, the bitmask macro with be
called <code>OPTION_MASK_<var>name</var></code>.
</p>
</dd>
<dt><code>InverseMask(<var>othername</var>)</code></dt>
<dt><code>InverseMask(<var>othername</var>, <var>thisname</var>)</code></dt>
<dd><p>The option is the inverse of another option that has the
<code>Mask(<var>othername</var>)</code> property.  If <var>thisname</var> is given,
the options-processing script will declare a <code>TARGET_<var>thisname</var></code>
macro that is 1 when the option is active and 0 otherwise.
</p>
</dd>
<dt><code>Enum(<var>name</var>)</code></dt>
<dd><p>The option&rsquo;s argument is a string from the set of strings associated
with the corresponding &lsquo;<samp>Enum</samp>&rsquo; record.  The string is checked and
converted to the integer specified in the corresponding
&lsquo;<samp>EnumValue</samp>&rsquo; record before being passed to option handlers.
</p>
</dd>
<dt><code>Defer</code></dt>
<dd><p>The option should be stored in a vector, specified with <code>Var</code>,
for later processing.
</p>
</dd>
<dt><code>Alias(<var>opt</var>)</code></dt>
<dt><code>Alias(<var>opt</var>, <var>arg</var>)</code></dt>
<dt><code>Alias(<var>opt</var>, <var>posarg</var>, <var>negarg</var>)</code></dt>
<dd><p>The option is an alias for <samp>-<var>opt</var></samp> (or the negative form
of that option, depending on <code>NegativeAlias</code>).  In the first form,
any argument passed to the alias is considered to be passed to
<samp>-<var>opt</var></samp>, and <samp>-<var>opt</var></samp> is considered to be
negated if the alias is used in negated form.  In the second form, the
alias may not be negated or have an argument, and <var>posarg</var> is
considered to be passed as an argument to <samp>-<var>opt</var></samp>.  In the
third form, the alias may not have an argument, if the alias is used
in the positive form then <var>posarg</var> is considered to be passed to
<samp>-<var>opt</var></samp>, and if the alias is used in the negative form
then <var>negarg</var> is considered to be passed to <samp>-<var>opt</var></samp>.
</p>
<p>Aliases should not specify <code>Var</code> or <code>Mask</code> or
<code>UInteger</code>.  Aliases should normally specify the same languages
as the target of the alias; the flags on the target will be used to
determine any diagnostic for use of an option for the wrong language,
while those on the alias will be used to identify what command-line
text is the option and what text is any argument to that option.
</p>
<p>When an <code>Alias</code> definition is used for an option, driver specs do
not need to handle it and no &lsquo;<samp>OPT_</samp>&rsquo; enumeration value is defined
for it; only the canonical form of the option will be seen in those
places.
</p>
</dd>
<dt><code>NegativeAlias</code></dt>
<dd><p>For an option marked with <code>Alias(<var>opt</var>)</code>, the option is
considered to be an alias for the positive form of <samp>-<var>opt</var></samp>
if negated and for the negative form of <samp>-<var>opt</var></samp> if not
negated.  <code>NegativeAlias</code> may not be used with the forms of
<code>Alias</code> taking more than one argument.
</p>
</dd>
<dt><code>Ignore</code></dt>
<dd><p>This option is ignored apart from printing any warning specified using
<code>Warn</code>.  The option will not be seen by specs and no &lsquo;<samp>OPT_</samp>&rsquo;
enumeration value is defined for it.
</p>
</dd>
<dt><code>SeparateAlias</code></dt>
<dd><p>For an option marked with <code>Joined</code>, <code>Separate</code> and
<code>Alias</code>, the option only acts as an alias when passed a separate
argument; with a joined argument it acts as a normal option, with an
&lsquo;<samp>OPT_</samp>&rsquo; enumeration value.  This is for compatibility with the
Java <samp>-d</samp> option and should not be used for new options.
</p>
</dd>
<dt><code>Warn(<var>message</var>)</code></dt>
<dd><p>If this option is used, output the warning <var>message</var>.
<var>message</var> is a format string, either taking a single operand with
a &lsquo;<samp>%qs</samp>&rsquo; format which is the option name, or not taking any
operands, which is passed to the &lsquo;<samp>warning</samp>&rsquo; function.  If an alias
is marked <code>Warn</code>, the target of the alias must not also be marked
<code>Warn</code>.
</p>
</dd>
<dt><code>Report</code></dt>
<dd><p>The state of the option should be printed by <samp>-fverbose-asm</samp>.
</p>
</dd>
<dt><code>Warning</code></dt>
<dd><p>This is a warning option and should be shown as such in
<samp>--help</samp> output.  This flag does not currently affect anything
other than <samp>--help</samp>.
</p>
</dd>
<dt><code>Optimization</code></dt>
<dd><p>This is an optimization option.  It should be shown as such in
<samp>--help</samp> output, and any associated variable named using
<code>Var</code> should be saved and restored when the optimization level is
changed with <code>optimize</code> attributes.
</p>
</dd>
<dt><code>Undocumented</code></dt>
<dd><p>The option is deliberately missing documentation and should not
be included in the <samp>--help</samp> output.
</p>
</dd>
<dt><code>Condition(<var>cond</var>)</code></dt>
<dd><p>The option should only be accepted if preprocessor condition
<var>cond</var> is true.  Note that any C declarations associated with the
option will be present even if <var>cond</var> is false; <var>cond</var> simply
controls whether the option is accepted and whether it is printed in
the <samp>--help</samp> output.
</p>
</dd>
<dt><code>Save</code></dt>
<dd><p>Build the <code>cl_target_option</code> structure to hold a copy of the
option, add the functions <code>cl_target_option_save</code> and
<code>cl_target_option_restore</code> to save and restore the options.
</p>
</dd>
<dt><code>SetByCombined</code></dt>
<dd><p>The option may also be set by a combined option such as
<samp>-ffast-math</samp>.  This causes the <code>gcc_options</code> struct to
have a field <code>frontend_set_<var>name</var></code>, where <code><var>name</var></code>
is the name of the field holding the value of this option (without the
leading <code>x_</code>).  This gives the front end a way to indicate that
the value has been set explicitly and should not be changed by the
combined option.  For example, some front ends use this to prevent
<samp>-ffast-math</samp> and <samp>-fno-fast-math</samp> from changing the
value of <samp>-fmath-errno</samp> for languages that do not use
<code>errno</code>.
</p>
</dd>
<dt><code>EnabledBy(<var>opt</var>)</code></dt>
<dt><code>EnabledBy(<var>opt</var> &amp;&amp; <var>opt2</var>)</code></dt>
<dd><p>If not explicitly set, the option is set to the value of
<samp>-<var>opt</var></samp>.  The second form specifies that the option is
only set if both <var>opt</var> and <var>opt2</var> are set.
</p>
</dd>
<dt><code>LangEnabledBy(<var>language</var>, <var>opt</var>)</code></dt>
<dt><code>LangEnabledBy(<var>language</var>, <var>opt</var>, <var>posarg</var>, <var>negarg</var>)</code></dt>
<dd><p>When compiling for the given language, the option is set to the value
of <samp>-<var>opt</var></samp>, if not explicitly set. In the second form, if
<var>opt</var> is used in the positive form then <var>posarg</var> is considered
to be passed to the option, and if <var>opt</var> is used in the negative
form then <var>negarg</var> is considered to be passed to the option.  It
is possible to specify several different languages.  Each
<var>language</var> must have been declared by an earlier <code>Language</code>
record.  See <a href="#Option-file-format">Option file format</a>.
</p>
</dd>
<dt><code>NoDWARFRecord</code></dt>
<dd><p>The option is omitted from the producer string written by
<samp>-grecord-gcc-switches</samp>.
</p></dd>
</dl>


<hr>
<a name="Passes"></a>
<div class="header">
<p>
Next: <a href="#RTL" accesskey="n" rel="next">RTL</a>, Previous: <a href="#Options" accesskey="p" rel="prev">Options</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Passes-and-Files-of-the-Compiler"></a>
<h2 class="chapter">9 Passes and Files of the Compiler</h2>
<a name="index-passes-and-files-of-the-compiler"></a>
<a name="index-files-and-passes-of-the-compiler"></a>
<a name="index-compiler-passes-and-files"></a>

<p>This chapter is dedicated to giving an overview of the optimization and
code generation passes of the compiler.  In the process, it describes
some of the language front end interface, though this description is no
where near complete.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Parsing-pass" accesskey="1">Parsing pass</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The language front end turns text into bits.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Gimplification-pass" accesskey="2">Gimplification pass</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The bits are turned into something we can optimize.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Pass-manager" accesskey="3">Pass manager</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Sequencing the optimization passes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tree-SSA-passes" accesskey="4">Tree SSA passes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Optimizations on a high-level representation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#RTL-passes" accesskey="5">RTL passes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Optimizations on a low-level representation.
</td></tr>
</table>

<hr>
<a name="Parsing-pass"></a>
<div class="header">
<p>
Next: <a href="#Gimplification-pass" accesskey="n" rel="next">Gimplification pass</a>, Up: <a href="#Passes" accesskey="u" rel="up">Passes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Parsing-pass-1"></a>
<h3 class="section">9.1 Parsing pass</h3>
<a name="index-GENERIC"></a>
<a name="index-lang_005fhooks_002eparse_005ffile"></a>
<p>The language front end is invoked only once, via
<code>lang_hooks.parse_file</code>, to parse the entire input.  The language
front end may use any intermediate language representation deemed
appropriate.  The C front end uses GENERIC trees (see <a href="#GENERIC">GENERIC</a>), plus
a double handful of language specific tree codes defined in
<samp>c-common.def</samp>.  The Fortran front end uses a completely different
private representation.
</p>
<a name="index-GIMPLE"></a>
<a name="index-gimplification"></a>
<a name="index-gimplifier"></a>
<a name="index-language_002dindependent-intermediate-representation"></a>
<a name="index-intermediate-representation-lowering"></a>
<a name="index-lowering_002c-language_002ddependent-intermediate-representation"></a>
<p>At some point the front end must translate the representation used in the
front end to a representation understood by the language-independent
portions of the compiler.  Current practice takes one of two forms.
The C front end manually invokes the gimplifier (see <a href="#GIMPLE">GIMPLE</a>) on each function,
and uses the gimplifier callbacks to convert the language-specific tree
nodes directly to GIMPLE before passing the function off to be compiled.
The Fortran front end converts from a private representation to GENERIC,
which is later lowered to GIMPLE when the function is compiled.  Which
route to choose probably depends on how well GENERIC (plus extensions)
can be made to match up with the source language and necessary parsing
data structures.
</p>
<p>BUG: Gimplification must occur before nested function lowering,
and nested function lowering must be done by the front end before
passing the data off to cgraph.
</p>
<p>TODO: Cgraph should control nested function lowering.  It would
only be invoked when it is certain that the outer-most function
is used.
</p>
<p>TODO: Cgraph needs a gimplify_function callback.  It should be
invoked when (1) it is certain that the function is used, (2)
warning flags specified by the user require some amount of
compilation in order to honor, (3) the language indicates that
semantic analysis is not complete until gimplification occurs.
Hum&hellip; this sounds overly complicated.  Perhaps we should just
have the front end gimplify always; in most cases it&rsquo;s only one
function call.
</p>
<p>The front end needs to pass all function definitions and top level
declarations off to the middle-end so that they can be compiled and
emitted to the object file.  For a simple procedural language, it is
usually most convenient to do this as each top level declaration or
definition is seen.  There is also a distinction to be made between
generating functional code and generating complete debug information.
The only thing that is absolutely required for functional code is that
function and data <em>definitions</em> be passed to the middle-end.  For
complete debug information, function, data and type declarations
should all be passed as well.
</p>
<a name="index-rest_005fof_005fdecl_005fcompilation"></a>
<a name="index-rest_005fof_005ftype_005fcompilation"></a>
<a name="index-cgraph_005ffinalize_005ffunction"></a>
<p>In any case, the front end needs each complete top-level function or
data declaration, and each data definition should be passed to
<code>rest_of_decl_compilation</code>.  Each complete type definition should
be passed to <code>rest_of_type_compilation</code>.  Each function definition
should be passed to <code>cgraph_finalize_function</code>.
</p>
<p>TODO: I know rest_of_compilation currently has all sorts of
RTL generation semantics.  I plan to move all code generation
bits (both Tree and RTL) to compile_function.  Should we hide
cgraph from the front ends and move back to rest_of_compilation
as the official interface?  Possibly we should rename all three
interfaces such that the names match in some meaningful way and
that is more descriptive than &quot;rest_of&quot;.
</p>
<p>The middle-end will, at its option, emit the function and data
definitions immediately or queue them for later processing.
</p>
<hr>
<a name="Gimplification-pass"></a>
<div class="header">
<p>
Next: <a href="#Pass-manager" accesskey="n" rel="next">Pass manager</a>, Previous: <a href="#Parsing-pass" accesskey="p" rel="prev">Parsing pass</a>, Up: <a href="#Passes" accesskey="u" rel="up">Passes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Gimplification-pass-1"></a>
<h3 class="section">9.2 Gimplification pass</h3>

<a name="index-gimplification-1"></a>
<a name="index-GIMPLE-1"></a>
<p><em>Gimplification</em> is a whimsical term for the process of converting
the intermediate representation of a function into the GIMPLE language
(see <a href="#GIMPLE">GIMPLE</a>).  The term stuck, and so words like &ldquo;gimplification&rdquo;,
&ldquo;gimplify&rdquo;, &ldquo;gimplifier&rdquo; and the like are sprinkled throughout this
section of code.
</p>
<p>While a front end may certainly choose to generate GIMPLE directly if
it chooses, this can be a moderately complex process unless the
intermediate language used by the front end is already fairly simple.
Usually it is easier to generate GENERIC trees plus extensions
and let the language-independent gimplifier do most of the work.
</p>
<a name="index-gimplify_005ffunction_005ftree"></a>
<a name="index-gimplify_005fexpr"></a>
<a name="index-lang_005fhooks_002egimplify_005fexpr"></a>
<p>The main entry point to this pass is <code>gimplify_function_tree</code>
located in <samp>gimplify.c</samp>.  From here we process the entire
function gimplifying each statement in turn.  The main workhorse
for this pass is <code>gimplify_expr</code>.  Approximately everything
passes through here at least once, and it is from here that we
invoke the <code>lang_hooks.gimplify_expr</code> callback.
</p>
<p>The callback should examine the expression in question and return
<code>GS_UNHANDLED</code> if the expression is not a language specific
construct that requires attention.  Otherwise it should alter the
expression in some way to such that forward progress is made toward
producing valid GIMPLE.  If the callback is certain that the
transformation is complete and the expression is valid GIMPLE, it
should return <code>GS_ALL_DONE</code>.  Otherwise it should return
<code>GS_OK</code>, which will cause the expression to be processed again.
If the callback encounters an error during the transformation (because
the front end is relying on the gimplification process to finish
semantic checks), it should return <code>GS_ERROR</code>.
</p>
<hr>
<a name="Pass-manager"></a>
<div class="header">
<p>
Next: <a href="#Tree-SSA-passes" accesskey="n" rel="next">Tree SSA passes</a>, Previous: <a href="#Gimplification-pass" accesskey="p" rel="prev">Gimplification pass</a>, Up: <a href="#Passes" accesskey="u" rel="up">Passes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Pass-manager-1"></a>
<h3 class="section">9.3 Pass manager</h3>

<p>The pass manager is located in <samp>passes.c</samp>, <samp>tree-optimize.c</samp>
and <samp>tree-pass.h</samp>.
Its job is to run all of the individual passes in the correct order,
and take care of standard bookkeeping that applies to every pass.
</p>
<p>The theory of operation is that each pass defines a structure that
represents everything we need to know about that pass&mdash;when it
should be run, how it should be run, what intermediate language
form or on-the-side data structures it needs.  We register the pass
to be run in some particular order, and the pass manager arranges
for everything to happen in the correct order.
</p>
<p>The actuality doesn&rsquo;t completely live up to the theory at present.
Command-line switches and <code>timevar_id_t</code> enumerations must still
be defined elsewhere.  The pass manager validates constraints but does
not attempt to (re-)generate data structures or lower intermediate
language form based on the requirements of the next pass.  Nevertheless,
what is present is useful, and a far sight better than nothing at all.
</p>
<p>Each pass should have a unique name.
Each pass may have its own dump file (for GCC debugging purposes).
Passes with a name starting with a star do not dump anything.
Sometimes passes are supposed to share a dump file / option name.
To still give these unique names, you can use a prefix that is delimited
by a space from the part that is used for the dump file / option name.
E.g. When the pass name is &quot;ud dce&quot;, the name used for dump file/options
is &quot;dce&quot;.
</p>
<p>TODO: describe the global variables set up by the pass manager,
and a brief description of how a new pass should use it.
I need to look at what info RTL passes use first<small class="enddots">...</small>
</p>
<hr>
<a name="Tree-SSA-passes"></a>
<div class="header">
<p>
Next: <a href="#RTL-passes" accesskey="n" rel="next">RTL passes</a>, Previous: <a href="#Pass-manager" accesskey="p" rel="prev">Pass manager</a>, Up: <a href="#Passes" accesskey="u" rel="up">Passes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Tree-SSA-passes-1"></a>
<h3 class="section">9.4 Tree SSA passes</h3>

<p>The following briefly describes the Tree optimization passes that are
run after gimplification and what source files they are located in.
</p>
<ul>
<li> Remove useless statements

<p>This pass is an extremely simple sweep across the gimple code in which
we identify obviously dead code and remove it.  Here we do things like
simplify <code>if</code> statements with constant conditions, remove
exception handling constructs surrounding code that obviously cannot
throw, remove lexical bindings that contain no variables, and other
assorted simplistic cleanups.  The idea is to get rid of the obvious
stuff quickly rather than wait until later when it&rsquo;s more work to get
rid of it.  This pass is located in <samp>tree-cfg.c</samp> and described by
<code>pass_remove_useless_stmts</code>.
</p>
</li><li> Mudflap declaration registration

<p>If mudflap (see <a href="gcc-4.html#Optimize-Options">-fmudflap -fmudflapth
-fmudflapir</a> in <cite>Using the GNU Compiler Collection (GCC)</cite>) is
enabled, we generate code to register some variable declarations with
the mudflap runtime.  Specifically, the runtime tracks the lifetimes of
those variable declarations that have their addresses taken, or whose
bounds are unknown at compile time (<code>extern</code>).  This pass generates
new exception handling constructs (<code>try</code>/<code>finally</code>), and so
must run before those are lowered.  In addition, the pass enqueues
declarations of static variables whose lifetimes extend to the entire
program.  The pass is located in <samp>tree-mudflap.c</samp> and is described
by <code>pass_mudflap_1</code>.
</p>
</li><li> OpenMP lowering

<p>If OpenMP generation (<samp>-fopenmp</samp>) is enabled, this pass lowers
OpenMP constructs into GIMPLE.
</p>
<p>Lowering of OpenMP constructs involves creating replacement
expressions for local variables that have been mapped using data
sharing clauses, exposing the control flow of most synchronization
directives and adding region markers to facilitate the creation of the
control flow graph.  The pass is located in <samp>omp-low.c</samp> and is
described by <code>pass_lower_omp</code>.
</p>
</li><li> OpenMP expansion

<p>If OpenMP generation (<samp>-fopenmp</samp>) is enabled, this pass expands
parallel regions into their own functions to be invoked by the thread
library.  The pass is located in <samp>omp-low.c</samp> and is described by
<code>pass_expand_omp</code>.
</p>
</li><li> Lower control flow

<p>This pass flattens <code>if</code> statements (<code>COND_EXPR</code>)
and moves lexical bindings (<code>BIND_EXPR</code>) out of line.  After
this pass, all <code>if</code> statements will have exactly two <code>goto</code>
statements in its <code>then</code> and <code>else</code> arms.  Lexical binding
information for each statement will be found in <code>TREE_BLOCK</code> rather
than being inferred from its position under a <code>BIND_EXPR</code>.  This
pass is found in <samp>gimple-low.c</samp> and is described by
<code>pass_lower_cf</code>.
</p>
</li><li> Lower exception handling control flow

<p>This pass decomposes high-level exception handling constructs
(<code>TRY_FINALLY_EXPR</code> and <code>TRY_CATCH_EXPR</code>) into a form
that explicitly represents the control flow involved.  After this
pass, <code>lookup_stmt_eh_region</code> will return a non-negative
number for any statement that may have EH control flow semantics;
examine <code>tree_can_throw_internal</code> or <code>tree_can_throw_external</code>
for exact semantics.  Exact control flow may be extracted from
<code>foreach_reachable_handler</code>.  The EH region nesting tree is defined
in <samp>except.h</samp> and built in <samp>except.c</samp>.  The lowering pass
itself is in <samp>tree-eh.c</samp> and is described by <code>pass_lower_eh</code>.
</p>
</li><li> Build the control flow graph

<p>This pass decomposes a function into basic blocks and creates all of
the edges that connect them.  It is located in <samp>tree-cfg.c</samp> and
is described by <code>pass_build_cfg</code>.
</p>
</li><li> Find all referenced variables

<p>This pass walks the entire function and collects an array of all
variables referenced in the function, <code>referenced_vars</code>.  The
index at which a variable is found in the array is used as a UID
for the variable within this function.  This data is needed by the
SSA rewriting routines.  The pass is located in <samp>tree-dfa.c</samp>
and is described by <code>pass_referenced_vars</code>.
</p>
</li><li> Enter static single assignment form

<p>This pass rewrites the function such that it is in SSA form.  After
this pass, all <code>is_gimple_reg</code> variables will be referenced by
<code>SSA_NAME</code>, and all occurrences of other variables will be
annotated with <code>VDEFS</code> and <code>VUSES</code>; PHI nodes will have
been inserted as necessary for each basic block.  This pass is
located in <samp>tree-ssa.c</samp> and is described by <code>pass_build_ssa</code>.
</p>
</li><li> Warn for uninitialized variables

<p>This pass scans the function for uses of <code>SSA_NAME</code>s that
are fed by default definition.  For non-parameter variables, such
uses are uninitialized.  The pass is run twice, before and after
optimization (if turned on).  In the first pass we only warn for uses that are
positively uninitialized; in the second pass we warn for uses that
are possibly uninitialized.  The pass is located in <samp>tree-ssa.c</samp>
and is defined by <code>pass_early_warn_uninitialized</code> and
<code>pass_late_warn_uninitialized</code>.
</p>
</li><li> Dead code elimination

<p>This pass scans the function for statements without side effects whose
result is unused.  It does not do memory life analysis, so any value
that is stored in memory is considered used.  The pass is run multiple
times throughout the optimization process.  It is located in
<samp>tree-ssa-dce.c</samp> and is described by <code>pass_dce</code>.
</p>
</li><li> Dominator optimizations

<p>This pass performs trivial dominator-based copy and constant propagation,
expression simplification, and jump threading.  It is run multiple times
throughout the optimization process.  It is located in <samp>tree-ssa-dom.c</samp>
and is described by <code>pass_dominator</code>.
</p>
</li><li> Forward propagation of single-use variables

<p>This pass attempts to remove redundant computation by substituting
variables that are used once into the expression that uses them and
seeing if the result can be simplified.  It is located in
<samp>tree-ssa-forwprop.c</samp> and is described by <code>pass_forwprop</code>.
</p>
</li><li> Copy Renaming

<p>This pass attempts to change the name of compiler temporaries involved in
copy operations such that SSA-&gt;normal can coalesce the copy away.  When compiler
temporaries are copies of user variables, it also renames the compiler
temporary to the user variable resulting in better use of user symbols.  It is
located in <samp>tree-ssa-copyrename.c</samp> and is described by
<code>pass_copyrename</code>.
</p>
</li><li> PHI node optimizations

<p>This pass recognizes forms of PHI inputs that can be represented as
conditional expressions and rewrites them into straight line code.
It is located in <samp>tree-ssa-phiopt.c</samp> and is described by
<code>pass_phiopt</code>.
</p>
</li><li> May-alias optimization

<p>This pass performs a flow sensitive SSA-based points-to analysis.
The resulting may-alias, must-alias, and escape analysis information
is used to promote variables from in-memory addressable objects to
non-aliased variables that can be renamed into SSA form.  We also
update the <code>VDEF</code>/<code>VUSE</code> memory tags for non-renameable
aggregates so that we get fewer false kills.  The pass is located
in <samp>tree-ssa-alias.c</samp> and is described by <code>pass_may_alias</code>.
</p>
<p>Interprocedural points-to information is located in
<samp>tree-ssa-structalias.c</samp> and described by <code>pass_ipa_pta</code>.
</p>
</li><li> Profiling

<p>This pass rewrites the function in order to collect runtime block
and value profiling data.  Such data may be fed back into the compiler
on a subsequent run so as to allow optimization based on expected
execution frequencies.  The pass is located in <samp>predict.c</samp> and
is described by <code>pass_profile</code>.
</p>
</li><li> Lower complex arithmetic

<p>This pass rewrites complex arithmetic operations into their component
scalar arithmetic operations.  The pass is located in <samp>tree-complex.c</samp>
and is described by <code>pass_lower_complex</code>.
</p>
</li><li> Scalar replacement of aggregates

<p>This pass rewrites suitable non-aliased local aggregate variables into
a set of scalar variables.  The resulting scalar variables are
rewritten into SSA form, which allows subsequent optimization passes
to do a significantly better job with them.  The pass is located in
<samp>tree-sra.c</samp> and is described by <code>pass_sra</code>.
</p>
</li><li> Dead store elimination

<p>This pass eliminates stores to memory that are subsequently overwritten
by another store, without any intervening loads.  The pass is located
in <samp>tree-ssa-dse.c</samp> and is described by <code>pass_dse</code>.
</p>
</li><li> Tail recursion elimination

<p>This pass transforms tail recursion into a loop.  It is located in
<samp>tree-tailcall.c</samp> and is described by <code>pass_tail_recursion</code>.
</p>
</li><li> Forward store motion

<p>This pass sinks stores and assignments down the flowgraph closer to their
use point.  The pass is located in <samp>tree-ssa-sink.c</samp> and is
described by <code>pass_sink_code</code>.
</p>
</li><li> Partial redundancy elimination

<p>This pass eliminates partially redundant computations, as well as
performing load motion.  The pass is located in <samp>tree-ssa-pre.c</samp>
and is described by <code>pass_pre</code>.
</p>
<p>Just before partial redundancy elimination, if
<samp>-funsafe-math-optimizations</samp> is on, GCC tries to convert
divisions to multiplications by the reciprocal.  The pass is located
in <samp>tree-ssa-math-opts.c</samp> and is described by
<code>pass_cse_reciprocal</code>.
</p>
</li><li> Full redundancy elimination

<p>This is a simpler form of PRE that only eliminates redundancies that
occur on all paths.  It is located in <samp>tree-ssa-pre.c</samp> and
described by <code>pass_fre</code>.
</p>
</li><li> Loop optimization

<p>The main driver of the pass is placed in <samp>tree-ssa-loop.c</samp>
and described by <code>pass_loop</code>.
</p>
<p>The optimizations performed by this pass are:
</p>
<p>Loop invariant motion.  This pass moves only invariants that
would be hard to handle on RTL level (function calls, operations that expand to
nontrivial sequences of insns).  With <samp>-funswitch-loops</samp> it also moves
operands of conditions that are invariant out of the loop, so that we can use
just trivial invariantness analysis in loop unswitching.  The pass also includes
store motion.  The pass is implemented in <samp>tree-ssa-loop-im.c</samp>.
</p>
<p>Canonical induction variable creation.  This pass creates a simple counter
for number of iterations of the loop and replaces the exit condition of the
loop using it, in case when a complicated analysis is necessary to determine
the number of iterations.  Later optimizations then may determine the number
easily.  The pass is implemented in <samp>tree-ssa-loop-ivcanon.c</samp>.
</p>
<p>Induction variable optimizations.  This pass performs standard induction
variable optimizations, including strength reduction, induction variable
merging and induction variable elimination.  The pass is implemented in
<samp>tree-ssa-loop-ivopts.c</samp>.
</p>
<p>Loop unswitching.  This pass moves the conditional jumps that are invariant
out of the loops.  To achieve this, a duplicate of the loop is created for
each possible outcome of conditional jump(s).  The pass is implemented in
<samp>tree-ssa-loop-unswitch.c</samp>.  This pass should eventually replace the
RTL level loop unswitching in <samp>loop-unswitch.c</samp>, but currently
the RTL level pass is not completely redundant yet due to deficiencies
in tree level alias analysis.
</p>
<p>The optimizations also use various utility functions contained in
<samp>tree-ssa-loop-manip.c</samp>, <samp>cfgloop.c</samp>, <samp>cfgloopanal.c</samp> and
<samp>cfgloopmanip.c</samp>.
</p>
<p>Vectorization.  This pass transforms loops to operate on vector types
instead of scalar types.  Data parallelism across loop iterations is exploited
to group data elements from consecutive iterations into a vector and operate
on them in parallel.  Depending on available target support the loop is
conceptually unrolled by a factor <code>VF</code> (vectorization factor), which is
the number of elements operated upon in parallel in each iteration, and the
<code>VF</code> copies of each scalar operation are fused to form a vector operation.
Additional loop transformations such as peeling and versioning may take place
to align the number of iterations, and to align the memory accesses in the
loop.
The pass is implemented in <samp>tree-vectorizer.c</samp> (the main driver),
<samp>tree-vect-loop.c</samp> and <samp>tree-vect-loop-manip.c</samp> (loop specific parts
and general loop utilities), <samp>tree-vect-slp</samp> (loop-aware SLP
functionality), <samp>tree-vect-stmts.c</samp> and <samp>tree-vect-data-refs.c</samp>.
Analysis of data references is in <samp>tree-data-ref.c</samp>.
</p>
<p>SLP Vectorization.  This pass performs vectorization of straight-line code. The
pass is implemented in <samp>tree-vectorizer.c</samp> (the main driver),
<samp>tree-vect-slp.c</samp>, <samp>tree-vect-stmts.c</samp> and
<samp>tree-vect-data-refs.c</samp>.
</p>
<p>Autoparallelization.  This pass splits the loop iteration space to run
into several threads.  The pass is implemented in <samp>tree-parloops.c</samp>.
</p>
<p>Graphite is a loop transformation framework based on the polyhedral
model.  Graphite stands for Gimple Represented as Polyhedra.  The
internals of this infrastructure are documented in
<a href="http://gcc.gnu.org/wiki/Graphite">http://gcc.gnu.org/wiki/Graphite</a><!-- /@w -->.  The passes working on
this representation are implemented in the various <samp>graphite-*</samp>
files.
</p>
</li><li> Tree level if-conversion for vectorizer

<p>This pass applies if-conversion to simple loops to help vectorizer.
We identify if convertible loops, if-convert statements and merge
basic blocks in one big block.  The idea is to present loop in such
form so that vectorizer can have one to one mapping between statements
and available vector operations.  This pass is located in
<samp>tree-if-conv.c</samp> and is described by <code>pass_if_conversion</code>.
</p>
</li><li> Conditional constant propagation

<p>This pass relaxes a lattice of values in order to identify those
that must be constant even in the presence of conditional branches.
The pass is located in <samp>tree-ssa-ccp.c</samp> and is described
by <code>pass_ccp</code>.
</p>
<p>A related pass that works on memory loads and stores, and not just
register values, is located in <samp>tree-ssa-ccp.c</samp> and described by
<code>pass_store_ccp</code>.
</p>
</li><li> Conditional copy propagation

<p>This is similar to constant propagation but the lattice of values is
the &ldquo;copy-of&rdquo; relation.  It eliminates redundant copies from the
code.  The pass is located in <samp>tree-ssa-copy.c</samp> and described by
<code>pass_copy_prop</code>.
</p>
<p>A related pass that works on memory copies, and not just register
copies, is located in <samp>tree-ssa-copy.c</samp> and described by
<code>pass_store_copy_prop</code>.
</p>
</li><li> Value range propagation

<p>This transformation is similar to constant propagation but
instead of propagating single constant values, it propagates
known value ranges.  The implementation is based on Patterson&rsquo;s
range propagation algorithm (Accurate Static Branch Prediction by
Value Range Propagation, J. R. C. Patterson, PLDI &rsquo;95).  In
contrast to Patterson&rsquo;s algorithm, this implementation does not
propagate branch probabilities nor it uses more than a single
range per SSA name. This means that the current implementation
cannot be used for branch prediction (though adapting it would
not be difficult).  The pass is located in <samp>tree-vrp.c</samp> and is
described by <code>pass_vrp</code>.
</p>
</li><li> Folding built-in functions

<p>This pass simplifies built-in functions, as applicable, with constant
arguments or with inferable string lengths.  It is located in
<samp>tree-ssa-ccp.c</samp> and is described by <code>pass_fold_builtins</code>.
</p>
</li><li> Split critical edges

<p>This pass identifies critical edges and inserts empty basic blocks
such that the edge is no longer critical.  The pass is located in
<samp>tree-cfg.c</samp> and is described by <code>pass_split_crit_edges</code>.
</p>
</li><li> Control dependence dead code elimination

<p>This pass is a stronger form of dead code elimination that can
eliminate unnecessary control flow statements.   It is located
in <samp>tree-ssa-dce.c</samp> and is described by <code>pass_cd_dce</code>.
</p>
</li><li> Tail call elimination

<p>This pass identifies function calls that may be rewritten into
jumps.  No code transformation is actually applied here, but the
data and control flow problem is solved.  The code transformation
requires target support, and so is delayed until RTL.  In the
meantime <code>CALL_EXPR_TAILCALL</code> is set indicating the possibility.
The pass is located in <samp>tree-tailcall.c</samp> and is described by
<code>pass_tail_calls</code>.  The RTL transformation is handled by
<code>fixup_tail_calls</code> in <samp>calls.c</samp>.
</p>
</li><li> Warn for function return without value

<p>For non-void functions, this pass locates return statements that do
not specify a value and issues a warning.  Such a statement may have
been injected by falling off the end of the function.  This pass is
run last so that we have as much time as possible to prove that the
statement is not reachable.  It is located in <samp>tree-cfg.c</samp> and
is described by <code>pass_warn_function_return</code>.
</p>
</li><li> Mudflap statement annotation

<p>If mudflap is enabled, we rewrite some memory accesses with code to
validate that the memory access is correct.  In particular, expressions
involving pointer dereferences (<code>INDIRECT_REF</code>, <code>ARRAY_REF</code>,
etc.) are replaced by code that checks the selected address range
against the mudflap runtime&rsquo;s database of valid regions.  This check
includes an inline lookup into a direct-mapped cache, based on
shift/mask operations of the pointer value, with a fallback function
call into the runtime.  The pass is located in <samp>tree-mudflap.c</samp> and
is described by <code>pass_mudflap_2</code>.
</p>
</li><li> Leave static single assignment form

<p>This pass rewrites the function such that it is in normal form.  At
the same time, we eliminate as many single-use temporaries as possible,
so the intermediate language is no longer GIMPLE, but GENERIC.  The
pass is located in <samp>tree-outof-ssa.c</samp> and is described by
<code>pass_del_ssa</code>.
</p>
</li><li> Merge PHI nodes that feed into one another

<p>This is part of the CFG cleanup passes.  It attempts to join PHI nodes
from a forwarder CFG block into another block with PHI nodes.  The
pass is located in <samp>tree-cfgcleanup.c</samp> and is described by
<code>pass_merge_phi</code>.
</p>
</li><li> Return value optimization

<p>If a function always returns the same local variable, and that local
variable is an aggregate type, then the variable is replaced with the
return value for the function (i.e., the function&rsquo;s DECL_RESULT).  This
is equivalent to the C++ named return value optimization applied to
GIMPLE.  The pass is located in <samp>tree-nrv.c</samp> and is described by
<code>pass_nrv</code>.
</p>
</li><li> Return slot optimization

<p>If a function returns a memory object and is called as <code>var =
foo()</code>, this pass tries to change the call so that the address of
<code>var</code> is sent to the caller to avoid an extra memory copy.  This
pass is located in <code>tree-nrv.c</code> and is described by
<code>pass_return_slot</code>.
</p>
</li><li> Optimize calls to <code>__builtin_object_size</code>

<p>This is a propagation pass similar to CCP that tries to remove calls
to <code>__builtin_object_size</code> when the size of the object can be
computed at compile-time.  This pass is located in
<samp>tree-object-size.c</samp> and is described by
<code>pass_object_sizes</code>.
</p>
</li><li> Loop invariant motion

<p>This pass removes expensive loop-invariant computations out of loops.
The pass is located in <samp>tree-ssa-loop.c</samp> and described by
<code>pass_lim</code>.
</p>
</li><li> Loop nest optimizations

<p>This is a family of loop transformations that works on loop nests.  It
includes loop interchange, scaling, skewing and reversal and they are
all geared to the optimization of data locality in array traversals
and the removal of dependencies that hamper optimizations such as loop
parallelization and vectorization.  The pass is located in
<samp>tree-loop-linear.c</samp> and described by
<code>pass_linear_transform</code>.
</p>
</li><li> Removal of empty loops

<p>This pass removes loops with no code in them.  The pass is located in
<samp>tree-ssa-loop-ivcanon.c</samp> and described by
<code>pass_empty_loop</code>.
</p>
</li><li> Unrolling of small loops

<p>This pass completely unrolls loops with few iterations.  The pass
is located in <samp>tree-ssa-loop-ivcanon.c</samp> and described by
<code>pass_complete_unroll</code>.
</p>
</li><li> Predictive commoning

<p>This pass makes the code reuse the computations from the previous
iterations of the loops, especially loads and stores to memory.
It does so by storing the values of these computations to a bank
of temporary variables that are rotated at the end of loop.  To avoid
the need for this rotation, the loop is then unrolled and the copies
of the loop body are rewritten to use the appropriate version of
the temporary variable.  This pass is located in <samp>tree-predcom.c</samp>
and described by <code>pass_predcom</code>.
</p>
</li><li> Array prefetching

<p>This pass issues prefetch instructions for array references inside
loops.  The pass is located in <samp>tree-ssa-loop-prefetch.c</samp> and
described by <code>pass_loop_prefetch</code>.
</p>
</li><li> Reassociation

<p>This pass rewrites arithmetic expressions to enable optimizations that
operate on them, like redundancy elimination and vectorization.  The
pass is located in <samp>tree-ssa-reassoc.c</samp> and described by
<code>pass_reassoc</code>.
</p>
</li><li> Optimization of <code>stdarg</code> functions

<p>This pass tries to avoid the saving of register arguments into the
stack on entry to <code>stdarg</code> functions.  If the function doesn&rsquo;t
use any <code>va_start</code> macros, no registers need to be saved.  If
<code>va_start</code> macros are used, the <code>va_list</code> variables don&rsquo;t
escape the function, it is only necessary to save registers that will
be used in <code>va_arg</code> macros.  For instance, if <code>va_arg</code> is
only used with integral types in the function, floating point
registers don&rsquo;t need to be saved.  This pass is located in
<code>tree-stdarg.c</code> and described by <code>pass_stdarg</code>.
</p>
</li></ul>

<hr>
<a name="RTL-passes"></a>
<div class="header">
<p>
Previous: <a href="#Tree-SSA-passes" accesskey="p" rel="prev">Tree SSA passes</a>, Up: <a href="#Passes" accesskey="u" rel="up">Passes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-passes-1"></a>
<h3 class="section">9.5 RTL passes</h3>

<p>The following briefly describes the RTL generation and optimization
passes that are run after the Tree optimization passes.
</p>
<ul>
<li> RTL generation

<p>The source files for RTL generation include
<samp>stmt.c</samp>,
<samp>calls.c</samp>,
<samp>expr.c</samp>,
<samp>explow.c</samp>,
<samp>expmed.c</samp>,
<samp>function.c</samp>,
<samp>optabs.c</samp>
and <samp>emit-rtl.c</samp>.
Also, the file
<samp>insn-emit.c</samp>, generated from the machine description by the
program <code>genemit</code>, is used in this pass.  The header file
<samp>expr.h</samp> is used for communication within this pass.
</p>
<a name="index-genflags"></a>
<a name="index-gencodes"></a>
<p>The header files <samp>insn-flags.h</samp> and <samp>insn-codes.h</samp>,
generated from the machine description by the programs <code>genflags</code>
and <code>gencodes</code>, tell this pass which standard names are available
for use and which patterns correspond to them.
</p>
</li><li> Generation of exception landing pads

<p>This pass generates the glue that handles communication between the
exception handling library routines and the exception handlers within
the function.  Entry points in the function that are invoked by the
exception handling library are called <em>landing pads</em>.  The code
for this pass is located in <samp>except.c</samp>.
</p>
</li><li> Control flow graph cleanup

<p>This pass removes unreachable code, simplifies jumps to next, jumps to
jump, jumps across jumps, etc.  The pass is run multiple times.
For historical reasons, it is occasionally referred to as the &ldquo;jump
optimization pass&rdquo;.  The bulk of the code for this pass is in
<samp>cfgcleanup.c</samp>, and there are support routines in <samp>cfgrtl.c</samp>
and <samp>jump.c</samp>.
</p>
</li><li> Forward propagation of single-def values

<p>This pass attempts to remove redundant computation by substituting
variables that come from a single definition, and
seeing if the result can be simplified.  It performs copy propagation
and addressing mode selection.  The pass is run twice, with values
being propagated into loops only on the second run.  The code is
located in <samp>fwprop.c</samp>.
</p>
</li><li> Common subexpression elimination

<p>This pass removes redundant computation within basic blocks, and
optimizes addressing modes based on cost.  The pass is run twice.
The code for this pass is located in <samp>cse.c</samp>.
</p>
</li><li> Global common subexpression elimination

<p>This pass performs two
different types of GCSE  depending on whether you are optimizing for
size or not (LCM based GCSE tends to increase code size for a gain in
speed, while Morel-Renvoise based GCSE does not).
When optimizing for size, GCSE is done using Morel-Renvoise Partial
Redundancy Elimination, with the exception that it does not try to move
invariants out of loops&mdash;that is left to  the loop optimization pass.
If MR PRE GCSE is done, code hoisting (aka unification) is also done, as
well as load motion.
If you are optimizing for speed, LCM (lazy code motion) based GCSE is
done.  LCM is based on the work of Knoop, Ruthing, and Steffen.  LCM
based GCSE also does loop invariant code motion.  We also perform load
and store motion when optimizing for speed.
Regardless of which type of GCSE is used, the GCSE pass also performs
global constant and  copy propagation.
The source file for this pass is <samp>gcse.c</samp>, and the LCM routines
are in <samp>lcm.c</samp>.
</p>
</li><li> Loop optimization

<p>This pass performs several loop related optimizations.
The source files <samp>cfgloopanal.c</samp> and <samp>cfgloopmanip.c</samp> contain
generic loop analysis and manipulation code.  Initialization and finalization
of loop structures is handled by <samp>loop-init.c</samp>.
A loop invariant motion pass is implemented in <samp>loop-invariant.c</samp>.
Basic block level optimizations&mdash;unrolling, peeling and unswitching loops&mdash;
are implemented in <samp>loop-unswitch.c</samp> and <samp>loop-unroll.c</samp>.
Replacing of the exit condition of loops by special machine-dependent
instructions is handled by <samp>loop-doloop.c</samp>.
</p>
</li><li> Jump bypassing

<p>This pass is an aggressive form of GCSE that transforms the control
flow graph of a function by propagating constants into conditional
branch instructions.  The source file for this pass is <samp>gcse.c</samp>.
</p>
</li><li> If conversion

<p>This pass attempts to replace conditional branches and surrounding
assignments with arithmetic, boolean value producing comparison
instructions, and conditional move instructions.  In the very last
invocation after reload/LRA, it will generate predicated instructions
when supported by the target.  The code is located in <samp>ifcvt.c</samp>.
</p>
</li><li> Web construction

<p>This pass splits independent uses of each pseudo-register.  This can
improve effect of the other transformation, such as CSE or register
allocation.  The code for this pass is located in <samp>web.c</samp>.
</p>
</li><li> Instruction combination

<p>This pass attempts to combine groups of two or three instructions that
are related by data flow into single instructions.  It combines the
RTL expressions for the instructions by substitution, simplifies the
result using algebra, and then attempts to match the result against
the machine description.  The code is located in <samp>combine.c</samp>.
</p>
</li><li> Register movement

<p>This pass looks for cases where matching constraints would force an
instruction to need a reload, and this reload would be a
register-to-register move.  It then attempts to change the registers
used by the instruction to avoid the move instruction.  The code is
located in <samp>regmove.c</samp>.
</p>
</li><li> Mode switching optimization

<p>This pass looks for instructions that require the processor to be in a
specific &ldquo;mode&rdquo; and minimizes the number of mode changes required to
satisfy all users.  What these modes are, and what they apply to are
completely target-specific.  The code for this pass is located in
<samp>mode-switching.c</samp>.
</p>
</li><li> <a name="index-modulo-scheduling"></a>
<a name="index-sms_002c-swing_002c-software-pipelining"></a>
Modulo scheduling

<p>This pass looks at innermost loops and reorders their instructions
by overlapping different iterations.  Modulo scheduling is performed
immediately before instruction scheduling.  The code for this pass is
located in <samp>modulo-sched.c</samp>.
</p>
</li><li> Instruction scheduling

<p>This pass looks for instructions whose output will not be available by
the time that it is used in subsequent instructions.  Memory loads and
floating point instructions often have this behavior on RISC machines.
It re-orders instructions within a basic block to try to separate the
definition and use of items that otherwise would cause pipeline
stalls.  This pass is performed twice, before and after register
allocation.  The code for this pass is located in <samp>haifa-sched.c</samp>,
<samp>sched-deps.c</samp>, <samp>sched-ebb.c</samp>, <samp>sched-rgn.c</samp> and
<samp>sched-vis.c</samp>.
</p>
</li><li> Register allocation

<p>These passes make sure that all occurrences of pseudo registers are
eliminated, either by allocating them to a hard register, replacing
them by an equivalent expression (e.g. a constant) or by placing
them on the stack.  This is done in several subpasses:
</p>
<ul>
<li> Register move optimizations.  This pass makes some simple RTL code
transformations which improve the subsequent register allocation.  The
source file is <samp>regmove.c</samp>.

</li><li> The integrated register allocator (<acronym>IRA</acronym>).  It is called
integrated because coalescing, register live range splitting, and hard
register preferencing are done on-the-fly during coloring.  It also
has better integration with the reload/LRA pass.  Pseudo-registers spilled
by the allocator or the reload/LRA have still a chance to get
hard-registers if the reload/LRA evicts some pseudo-registers from
hard-registers.  The allocator helps to choose better pseudos for
spilling based on their live ranges and to coalesce stack slots
allocated for the spilled pseudo-registers.  IRA is a regional
register allocator which is transformed into Chaitin-Briggs allocator
if there is one region.  By default, IRA chooses regions using
register pressure but the user can force it to use one region or
regions corresponding to all loops.

<p>Source files of the allocator are <samp>ira.c</samp>, <samp>ira-build.c</samp>,
<samp>ira-costs.c</samp>, <samp>ira-conflicts.c</samp>, <samp>ira-color.c</samp>,
<samp>ira-emit.c</samp>, <samp>ira-lives</samp>, plus header files <samp>ira.h</samp>
and <samp>ira-int.h</samp> used for the communication between the allocator
and the rest of the compiler and between the IRA files.
</p>
</li><li> <a name="index-reloading"></a>
Reloading.  This pass renumbers pseudo registers with the hardware
registers numbers they were allocated.  Pseudo registers that did not
get hard registers are replaced with stack slots.  Then it finds
instructions that are invalid because a value has failed to end up in
a register, or has ended up in a register of the wrong kind.  It fixes
up these instructions by reloading the problematical values
temporarily into registers.  Additional instructions are generated to
do the copying.

<p>The reload pass also optionally eliminates the frame pointer and inserts
instructions to save and restore call-clobbered registers around calls.
</p>
<p>Source files are <samp>reload.c</samp> and <samp>reload1.c</samp>, plus the header
<samp>reload.h</samp> used for communication between them.
</p>
</li><li> <a name="index-Local-Register-Allocator-_0028LRA_0029"></a>
This pass is a modern replacement of the reload pass.  Source files
are <samp>lra.c</samp>, <samp>lra-assign.c</samp>, <samp>lra-coalesce.c</samp>,
<samp>lra-constraints.c</samp>, <samp>lra-eliminations.c</samp>,
<samp>lra-equivs.c</samp>, <samp>lra-lives.c</samp>, <samp>lra-saves.c</samp>,
<samp>lra-spills.c</samp>, the header <samp>lra-int.h</samp> used for
communication between them, and the header <samp>lra.h</samp> used for
communication between LRA and the rest of compiler.

<p>Unlike the reload pass, intermediate LRA decisions are reflected in
RTL as much as possible.  This reduces the number of target-dependent
macros and hooks, leaving instruction constraints as the primary
source of control.
</p>
<p>LRA is run on targets for which TARGET_LRA_P returns true.
</p></li></ul>

</li><li> Basic block reordering

<p>This pass implements profile guided code positioning.  If profile
information is not available, various types of static analysis are
performed to make the predictions normally coming from the profile
feedback (IE execution frequency, branch probability, etc).  It is
implemented in the file <samp>bb-reorder.c</samp>, and the various
prediction routines are in <samp>predict.c</samp>.
</p>
</li><li> Variable tracking

<p>This pass computes where the variables are stored at each
position in code and generates notes describing the variable locations
to RTL code.  The location lists are then generated according to these
notes to debug information if the debugging information format supports
location lists.  The code is located in <samp>var-tracking.c</samp>.
</p>
</li><li> Delayed branch scheduling

<p>This optional pass attempts to find instructions that can go into the
delay slots of other instructions, usually jumps and calls.  The code
for this pass is located in <samp>reorg.c</samp>.
</p>
</li><li> Branch shortening

<p>On many RISC machines, branch instructions have a limited range.
Thus, longer sequences of instructions must be used for long branches.
In this pass, the compiler figures out what how far each instruction
will be from each other instruction, and therefore whether the usual
instructions, or the longer sequences, must be used for each branch.
The code for this pass is located in <samp>final.c</samp>.
</p>
</li><li> Register-to-stack conversion

<p>Conversion from usage of some hard registers to usage of a register
stack may be done at this point.  Currently, this is supported only
for the floating-point registers of the Intel 80387 coprocessor.  The
code for this pass is located in <samp>reg-stack.c</samp>.
</p>
</li><li> Final

<p>This pass outputs the assembler code for the function.  The source files
are <samp>final.c</samp> plus <samp>insn-output.c</samp>; the latter is generated
automatically from the machine description by the tool <samp>genoutput</samp>.
The header file <samp>conditions.h</samp> is used for communication between
these files.  If mudflap is enabled, the queue of deferred declarations
and any addressed constants (e.g., string literals) is processed by
<code>mudflap_finish_file</code> into a synthetic constructor function
containing calls into the mudflap runtime.
</p>
</li><li> Debugging information output

<p>This is run after final because it must output the stack slot offsets
for pseudo registers that did not get hard registers.  Source files
are <samp>dbxout.c</samp> for DBX symbol table format, <samp>sdbout.c</samp> for
SDB symbol table format, <samp>dwarfout.c</samp> for DWARF symbol table
format, files <samp>dwarf2out.c</samp> and <samp>dwarf2asm.c</samp> for DWARF2
symbol table format, and <samp>vmsdbgout.c</samp> for VMS debug symbol table
format.
</p>
</li></ul>

<hr>
<a name="RTL"></a>
<div class="header">
<p>
Next: <a href="#GENERIC" accesskey="n" rel="next">GENERIC</a>, Previous: <a href="#Passes" accesskey="p" rel="prev">Passes</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Representation"></a>
<h2 class="chapter">10 RTL Representation</h2>
<a name="index-RTL-representation"></a>
<a name="index-representation-of-RTL"></a>
<a name="index-Register-Transfer-Language-_0028RTL_0029"></a>

<p>The last part of the compiler work is done on a low-level intermediate
representation called Register Transfer Language.  In this language, the
instructions to be output are described, pretty much one by one, in an
algebraic form that describes what the instruction does.
</p>
<p>RTL is inspired by Lisp lists.  It has both an internal form, made up of
structures that point at other structures, and a textual form that is used
in the machine description and in printed debugging dumps.  The textual
form uses nested parentheses to indicate the pointers in the internal form.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#RTL-Objects" accesskey="1">RTL Objects</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions vs vectors vs strings vs integers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#RTL-Classes" accesskey="2">RTL Classes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Categories of RTL expression objects, and their structure.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Accessors" accesskey="3">Accessors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros to access expression operands or vector elts.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Special-Accessors" accesskey="4">Special Accessors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros to access specific annotations on RTL.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Flags" accesskey="5">Flags</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Other flags in an RTL expression.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Machine-Modes" accesskey="6">Machine Modes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Describing the size and format of a datum.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Constants" accesskey="7">Constants</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions with constant values.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Regs-and-Memory" accesskey="8">Regs and Memory</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions representing register contents or memory.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Arithmetic" accesskey="9">Arithmetic</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions representing arithmetic on other expressions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Comparisons">Comparisons</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions representing comparison of expressions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Bit_002dFields">Bit-Fields</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions representing bit-fields in memory or reg.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Vector-Operations">Vector Operations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions involving vector datatypes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Conversions">Conversions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Extending, truncating, floating or fixing.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#RTL-Declarations">RTL Declarations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Declaring volatility, constancy, etc.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Side-Effects">Side Effects</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions for storing in registers, etc.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Incdec">Incdec</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Embedded side-effects for autoincrement addressing.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Assembler">Assembler</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Representing <code>asm</code> with operands.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Debug-Information">Debug Information</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expressions representing debugging information.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Insns">Insns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Expression types for entire insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Calls">Calls</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">RTL representation of function call insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Sharing">Sharing</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Some expressions are unique; others *must* be copied.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Reading-RTL">Reading RTL</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Reading textual RTL from a file.
</td></tr>
</table>

<hr>
<a name="RTL-Objects"></a>
<div class="header">
<p>
Next: <a href="#RTL-Classes" accesskey="n" rel="next">RTL Classes</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Object-Types"></a>
<h3 class="section">10.1 RTL Object Types</h3>
<a name="index-RTL-object-types"></a>

<a name="index-RTL-integers"></a>
<a name="index-RTL-strings"></a>
<a name="index-RTL-vectors"></a>
<a name="index-RTL-expression"></a>
<a name="index-RTX-_0028See-RTL_0029"></a>
<p>RTL uses five kinds of objects: expressions, integers, wide integers,
strings and vectors.  Expressions are the most important ones.  An RTL
expression (&ldquo;RTX&rdquo;, for short) is a C structure, but it is usually
referred to with a pointer; a type that is given the typedef name
<code>rtx</code>.
</p>
<p>An integer is simply an <code>int</code>; their written form uses decimal
digits.  A wide integer is an integral object whose type is
<code>HOST_WIDE_INT</code>; their written form uses decimal digits.
</p>
<p>A string is a sequence of characters.  In core it is represented as a
<code>char *</code> in usual C fashion, and it is written in C syntax as well.
However, strings in RTL may never be null.  If you write an empty string in
a machine description, it is represented in core as a null pointer rather
than as a pointer to a null character.  In certain contexts, these null
pointers instead of strings are valid.  Within RTL code, strings are most
commonly found inside <code>symbol_ref</code> expressions, but they appear in
other contexts in the RTL expressions that make up machine descriptions.
</p>
<p>In a machine description, strings are normally written with double
quotes, as you would in C.  However, strings in machine descriptions may
extend over many lines, which is invalid C, and adjacent string
constants are not concatenated as they are in C.  Any string constant
may be surrounded with a single set of parentheses.  Sometimes this
makes the machine description easier to read.
</p>
<p>There is also a special syntax for strings, which can be useful when C
code is embedded in a machine description.  Wherever a string can
appear, it is also valid to write a C-style brace block.  The entire
brace block, including the outermost pair of braces, is considered to be
the string constant.  Double quote characters inside the braces are not
special.  Therefore, if you write string constants in the C code, you
need not escape each quote character with a backslash.
</p>
<p>A vector contains an arbitrary number of pointers to expressions.  The
number of elements in the vector is explicitly present in the vector.
The written form of a vector consists of square brackets
(&lsquo;<samp>[&hellip;]</samp>&rsquo;) surrounding the elements, in sequence and with
whitespace separating them.  Vectors of length zero are not created;
null pointers are used instead.
</p>
<a name="index-expression-codes"></a>
<a name="index-codes_002c-RTL-expression"></a>
<a name="index-GET_005fCODE"></a>
<a name="index-PUT_005fCODE"></a>
<p>Expressions are classified by <em>expression codes</em> (also called RTX
codes).  The expression code is a name defined in <samp>rtl.def</samp>, which is
also (in uppercase) a C enumeration constant.  The possible expression
codes and their meanings are machine-independent.  The code of an RTX can
be extracted with the macro <code>GET_CODE (<var>x</var>)</code> and altered with
<code>PUT_CODE (<var>x</var>, <var>newcode</var>)</code>.
</p>
<p>The expression code determines how many operands the expression contains,
and what kinds of objects they are.  In RTL, unlike Lisp, you cannot tell
by looking at an operand what kind of object it is.  Instead, you must know
from its context&mdash;from the expression code of the containing expression.
For example, in an expression of code <code>subreg</code>, the first operand is
to be regarded as an expression and the second operand as an integer.  In
an expression of code <code>plus</code>, there are two operands, both of which
are to be regarded as expressions.  In a <code>symbol_ref</code> expression,
there is one operand, which is to be regarded as a string.
</p>
<p>Expressions are written as parentheses containing the name of the
expression type, its flags and machine mode if any, and then the operands
of the expression (separated by spaces).
</p>
<p>Expression code names in the &lsquo;<samp>md</samp>&rsquo; file are written in lowercase,
but when they appear in C code they are written in uppercase.  In this
manual, they are shown as follows: <code>const_int</code>.
</p>
<a name="index-_0028nil_0029"></a>
<a name="index-nil"></a>
<p>In a few contexts a null pointer is valid where an expression is normally
wanted.  The written form of this is <code>(nil)</code>.
</p>
<hr>
<a name="RTL-Classes"></a>
<div class="header">
<p>
Next: <a href="#Accessors" accesskey="n" rel="next">Accessors</a>, Previous: <a href="#RTL-Objects" accesskey="p" rel="prev">RTL Objects</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Classes-and-Formats"></a>
<h3 class="section">10.2 RTL Classes and Formats</h3>
<a name="index-RTL-classes"></a>
<a name="index-classes-of-RTX-codes"></a>
<a name="index-RTX-codes_002c-classes-of"></a>
<a name="index-GET_005fRTX_005fCLASS"></a>

<p>The various expression codes are divided into several <em>classes</em>,
which are represented by single characters.  You can determine the class
of an RTX code with the macro <code>GET_RTX_CLASS (<var>code</var>)</code>.
Currently, <samp>rtl.def</samp> defines these classes:
</p>
<dl compact="compact">
<dt><code>RTX_OBJ</code></dt>
<dd><p>An RTX code that represents an actual object, such as a register
(<code>REG</code>) or a memory location (<code>MEM</code>, <code>SYMBOL_REF</code>).
<code>LO_SUM</code>) is also included; instead, <code>SUBREG</code> and
<code>STRICT_LOW_PART</code> are not in this class, but in class <code>x</code>.
</p>
</dd>
<dt><code>RTX_CONST_OBJ</code></dt>
<dd><p>An RTX code that represents a constant object.  <code>HIGH</code> is also
included in this class.
</p>
</dd>
<dt><code>RTX_COMPARE</code></dt>
<dd><p>An RTX code for a non-symmetric comparison, such as <code>GEU</code> or
<code>LT</code>.
</p>
</dd>
<dt><code>RTX_COMM_COMPARE</code></dt>
<dd><p>An RTX code for a symmetric (commutative) comparison, such as <code>EQ</code>
or <code>ORDERED</code>.
</p>
</dd>
<dt><code>RTX_UNARY</code></dt>
<dd><p>An RTX code for a unary arithmetic operation, such as <code>NEG</code>,
<code>NOT</code>, or <code>ABS</code>.  This category also includes value extension
(sign or zero) and conversions between integer and floating point.
</p>
</dd>
<dt><code>RTX_COMM_ARITH</code></dt>
<dd><p>An RTX code for a commutative binary operation, such as <code>PLUS</code> or
<code>AND</code>.  <code>NE</code> and <code>EQ</code> are comparisons, so they have class
<code>&lt;</code>.
</p>
</dd>
<dt><code>RTX_BIN_ARITH</code></dt>
<dd><p>An RTX code for a non-commutative binary operation, such as <code>MINUS</code>,
<code>DIV</code>, or <code>ASHIFTRT</code>.
</p>
</dd>
<dt><code>RTX_BITFIELD_OPS</code></dt>
<dd><p>An RTX code for a bit-field operation.  Currently only
<code>ZERO_EXTRACT</code> and <code>SIGN_EXTRACT</code>.  These have three inputs
and are lvalues (so they can be used for insertion as well).
See <a href="#Bit_002dFields">Bit-Fields</a>.
</p>
</dd>
<dt><code>RTX_TERNARY</code></dt>
<dd><p>An RTX code for other three input operations.  Currently only
<code>IF_THEN_ELSE</code>,  <code>VEC_MERGE</code>, <code>SIGN_EXTRACT</code>,
<code>ZERO_EXTRACT</code>, and <code>FMA</code>.
</p>
</dd>
<dt><code>RTX_INSN</code></dt>
<dd><p>An RTX code for an entire instruction:  <code>INSN</code>, <code>JUMP_INSN</code>, and
<code>CALL_INSN</code>.  See <a href="#Insns">Insns</a>.
</p>
</dd>
<dt><code>RTX_MATCH</code></dt>
<dd><p>An RTX code for something that matches in insns, such as
<code>MATCH_DUP</code>.  These only occur in machine descriptions.
</p>
</dd>
<dt><code>RTX_AUTOINC</code></dt>
<dd><p>An RTX code for an auto-increment addressing mode, such as
<code>POST_INC</code>.
</p>
</dd>
<dt><code>RTX_EXTRA</code></dt>
<dd><p>All other RTX codes.  This category includes the remaining codes used
only in machine descriptions (<code>DEFINE_*</code>, etc.).  It also includes
all the codes describing side effects (<code>SET</code>, <code>USE</code>,
<code>CLOBBER</code>, etc.) and the non-insns that may appear on an insn
chain, such as <code>NOTE</code>, <code>BARRIER</code>, and <code>CODE_LABEL</code>.
<code>SUBREG</code> is also part of this class.
</p></dd>
</dl>

<a name="index-RTL-format"></a>
<p>For each expression code, <samp>rtl.def</samp> specifies the number of
contained objects and their kinds using a sequence of characters
called the <em>format</em> of the expression code.  For example,
the format of <code>subreg</code> is &lsquo;<samp>ei</samp>&rsquo;.
</p>
<a name="index-RTL-format-characters"></a>
<p>These are the most commonly used format characters:
</p>
<dl compact="compact">
<dt><code>e</code></dt>
<dd><p>An expression (actually a pointer to an expression).
</p>
</dd>
<dt><code>i</code></dt>
<dd><p>An integer.
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>A wide integer.
</p>
</dd>
<dt><code>s</code></dt>
<dd><p>A string.
</p>
</dd>
<dt><code>E</code></dt>
<dd><p>A vector of expressions.
</p></dd>
</dl>

<p>A few other format characters are used occasionally:
</p>
<dl compact="compact">
<dt><code>u</code></dt>
<dd><p>&lsquo;<samp>u</samp>&rsquo; is equivalent to &lsquo;<samp>e</samp>&rsquo; except that it is printed differently
in debugging dumps.  It is used for pointers to insns.
</p>
</dd>
<dt><code>n</code></dt>
<dd><p>&lsquo;<samp>n</samp>&rsquo; is equivalent to &lsquo;<samp>i</samp>&rsquo; except that it is printed differently
in debugging dumps.  It is used for the line number or code number of a
<code>note</code> insn.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>&lsquo;<samp>S</samp>&rsquo; indicates a string which is optional.  In the RTL objects in
core, &lsquo;<samp>S</samp>&rsquo; is equivalent to &lsquo;<samp>s</samp>&rsquo;, but when the object is read,
from an &lsquo;<samp>md</samp>&rsquo; file, the string value of this operand may be omitted.
An omitted string is taken to be the null string.
</p>
</dd>
<dt><code>V</code></dt>
<dd><p>&lsquo;<samp>V</samp>&rsquo; indicates a vector which is optional.  In the RTL objects in
core, &lsquo;<samp>V</samp>&rsquo; is equivalent to &lsquo;<samp>E</samp>&rsquo;, but when the object is read
from an &lsquo;<samp>md</samp>&rsquo; file, the vector value of this operand may be omitted.
An omitted vector is effectively the same as a vector of no elements.
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>&lsquo;<samp>B</samp>&rsquo; indicates a pointer to basic block structure.
</p>
</dd>
<dt><code>0</code></dt>
<dd><p>&lsquo;<samp>0</samp>&rsquo; means a slot whose contents do not fit any normal category.
&lsquo;<samp>0</samp>&rsquo; slots are not printed at all in dumps, and are often used in
special ways by small parts of the compiler.
</p></dd>
</dl>

<p>There are macros to get the number of operands and the format
of an expression code:
</p>
<dl compact="compact">
<dd><a name="index-GET_005fRTX_005fLENGTH"></a>
</dd>
<dt><code>GET_RTX_LENGTH (<var>code</var>)</code></dt>
<dd><p>Number of operands of an RTX of code <var>code</var>.
</p>
<a name="index-GET_005fRTX_005fFORMAT"></a>
</dd>
<dt><code>GET_RTX_FORMAT (<var>code</var>)</code></dt>
<dd><p>The format of an RTX of code <var>code</var>, as a C string.
</p></dd>
</dl>

<p>Some classes of RTX codes always have the same format.  For example, it
is safe to assume that all comparison operations have format <code>ee</code>.
</p>
<dl compact="compact">
<dt><code>1</code></dt>
<dd><p>All codes of this class have format <code>e</code>.
</p>
</dd>
<dt><code>&lt;</code></dt>
<dt><code>c</code></dt>
<dt><code>2</code></dt>
<dd><p>All codes of these classes have format <code>ee</code>.
</p>
</dd>
<dt><code>b</code></dt>
<dt><code>3</code></dt>
<dd><p>All codes of these classes have format <code>eee</code>.
</p>
</dd>
<dt><code>i</code></dt>
<dd><p>All codes of this class have formats that begin with <code>iuueiee</code>.
See <a href="#Insns">Insns</a>.  Note that not all RTL objects linked onto an insn chain
are of class <code>i</code>.
</p>
</dd>
<dt><code>o</code></dt>
<dt><code>m</code></dt>
<dt><code>x</code></dt>
<dd><p>You can make no assumptions about the format of these codes.
</p></dd>
</dl>

<hr>
<a name="Accessors"></a>
<div class="header">
<p>
Next: <a href="#Special-Accessors" accesskey="n" rel="next">Special Accessors</a>, Previous: <a href="#RTL-Classes" accesskey="p" rel="prev">RTL Classes</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Access-to-Operands"></a>
<h3 class="section">10.3 Access to Operands</h3>
<a name="index-accessors"></a>
<a name="index-access-to-operands"></a>
<a name="index-operand-access"></a>

<a name="index-XEXP"></a>
<a name="index-XINT"></a>
<a name="index-XWINT"></a>
<a name="index-XSTR"></a>
<p>Operands of expressions are accessed using the macros <code>XEXP</code>,
<code>XINT</code>, <code>XWINT</code> and <code>XSTR</code>.  Each of these macros takes
two arguments: an expression-pointer (RTX) and an operand number
(counting from zero).  Thus,
</p>
<div class="smallexample">
<pre class="smallexample">XEXP (<var>x</var>, 2)
</pre></div>

<p>accesses operand 2 of expression <var>x</var>, as an expression.
</p>
<div class="smallexample">
<pre class="smallexample">XINT (<var>x</var>, 2)
</pre></div>

<p>accesses the same operand as an integer.  <code>XSTR</code>, used in the same
fashion, would access it as a string.
</p>
<p>Any operand can be accessed as an integer, as an expression or as a string.
You must choose the correct method of access for the kind of value actually
stored in the operand.  You would do this based on the expression code of
the containing expression.  That is also how you would know how many
operands there are.
</p>
<p>For example, if <var>x</var> is a <code>subreg</code> expression, you know that it has
two operands which can be correctly accessed as <code>XEXP (<var>x</var>, 0)</code>
and <code>XINT (<var>x</var>, 1)</code>.  If you did <code>XINT (<var>x</var>, 0)</code>, you
would get the address of the expression operand but cast as an integer;
that might occasionally be useful, but it would be cleaner to write
<code>(int) XEXP (<var>x</var>, 0)</code>.  <code>XEXP (<var>x</var>, 1)</code> would also
compile without error, and would return the second, integer operand cast as
an expression pointer, which would probably result in a crash when
accessed.  Nothing stops you from writing <code>XEXP (<var>x</var>, 28)</code> either,
but this will access memory past the end of the expression with
unpredictable results.
</p>
<p>Access to operands which are vectors is more complicated.  You can use the
macro <code>XVEC</code> to get the vector-pointer itself, or the macros
<code>XVECEXP</code> and <code>XVECLEN</code> to access the elements and length of a
vector.
</p>
<dl compact="compact">
<dd><a name="index-XVEC"></a>
</dd>
<dt><code>XVEC (<var>exp</var>, <var>idx</var>)</code></dt>
<dd><p>Access the vector-pointer which is operand number <var>idx</var> in <var>exp</var>.
</p>
<a name="index-XVECLEN"></a>
</dd>
<dt><code>XVECLEN (<var>exp</var>, <var>idx</var>)</code></dt>
<dd><p>Access the length (number of elements) in the vector which is
in operand number <var>idx</var> in <var>exp</var>.  This value is an <code>int</code>.
</p>
<a name="index-XVECEXP"></a>
</dd>
<dt><code>XVECEXP (<var>exp</var>, <var>idx</var>, <var>eltnum</var>)</code></dt>
<dd><p>Access element number <var>eltnum</var> in the vector which is
in operand number <var>idx</var> in <var>exp</var>.  This value is an RTX.
</p>
<p>It is up to you to make sure that <var>eltnum</var> is not negative
and is less than <code>XVECLEN (<var>exp</var>, <var>idx</var>)</code>.
</p></dd>
</dl>

<p>All the macros defined in this section expand into lvalues and therefore
can be used to assign the operands, lengths and vector elements as well as
to access them.
</p>
<hr>
<a name="Special-Accessors"></a>
<div class="header">
<p>
Next: <a href="#Flags" accesskey="n" rel="next">Flags</a>, Previous: <a href="#Accessors" accesskey="p" rel="prev">Accessors</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Access-to-Special-Operands"></a>
<h3 class="section">10.4 Access to Special Operands</h3>
<a name="index-access-to-special-operands"></a>

<p>Some RTL nodes have special annotations associated with them.
</p>
<dl compact="compact">
<dt><code>MEM</code></dt>
<dd><dl compact="compact">
<dd><a name="index-MEM_005fALIAS_005fSET"></a>
</dd>
<dt><code>MEM_ALIAS_SET (<var>x</var>)</code></dt>
<dd><p>If 0, <var>x</var> is not in any alias set, and may alias anything.  Otherwise,
<var>x</var> can only alias <code>MEM</code>s in a conflicting alias set.  This value
is set in a language-dependent manner in the front-end, and should not be
altered in the back-end.  In some front-ends, these numbers may correspond
in some way to types, or other language-level entities, but they need not,
and the back-end makes no such assumptions.
These set numbers are tested with <code>alias_sets_conflict_p</code>.
</p>
<a name="index-MEM_005fEXPR"></a>
</dd>
<dt><code>MEM_EXPR (<var>x</var>)</code></dt>
<dd><p>If this register is known to hold the value of some user-level
declaration, this is that tree node.  It may also be a
<code>COMPONENT_REF</code>, in which case this is some field reference,
and <code>TREE_OPERAND (<var>x</var>, 0)</code> contains the declaration,
or another <code>COMPONENT_REF</code>, or null if there is no compile-time
object associated with the reference.
</p>
<a name="index-MEM_005fOFFSET_005fKNOWN_005fP"></a>
</dd>
<dt><code>MEM_OFFSET_KNOWN_P (<var>x</var>)</code></dt>
<dd><p>True if the offset of the memory reference from <code>MEM_EXPR</code> is known.
&lsquo;<samp>MEM_OFFSET (<var>x</var>)</samp>&rsquo; provides the offset if so.
</p>
<a name="index-MEM_005fOFFSET"></a>
</dd>
<dt><code>MEM_OFFSET (<var>x</var>)</code></dt>
<dd><p>The offset from the start of <code>MEM_EXPR</code>.  The value is only valid if
&lsquo;<samp>MEM_OFFSET_KNOWN_P (<var>x</var>)</samp>&rsquo; is true.
</p>
<a name="index-MEM_005fSIZE_005fKNOWN_005fP"></a>
</dd>
<dt><code>MEM_SIZE_KNOWN_P (<var>x</var>)</code></dt>
<dd><p>True if the size of the memory reference is known.
&lsquo;<samp>MEM_SIZE (<var>x</var>)</samp>&rsquo; provides its size if so.
</p>
<a name="index-MEM_005fSIZE"></a>
</dd>
<dt><code>MEM_SIZE (<var>x</var>)</code></dt>
<dd><p>The size in bytes of the memory reference.
This is mostly relevant for <code>BLKmode</code> references as otherwise
the size is implied by the mode.  The value is only valid if
&lsquo;<samp>MEM_SIZE_KNOWN_P (<var>x</var>)</samp>&rsquo; is true.
</p>
<a name="index-MEM_005fALIGN"></a>
</dd>
<dt><code>MEM_ALIGN (<var>x</var>)</code></dt>
<dd><p>The known alignment in bits of the memory reference.
</p>
<a name="index-MEM_005fADDR_005fSPACE"></a>
</dd>
<dt><code>MEM_ADDR_SPACE (<var>x</var>)</code></dt>
<dd><p>The address space of the memory reference.  This will commonly be zero
for the generic address space.
</p></dd>
</dl>

</dd>
<dt><code>REG</code></dt>
<dd><dl compact="compact">
<dd><a name="index-ORIGINAL_005fREGNO"></a>
</dd>
<dt><code>ORIGINAL_REGNO (<var>x</var>)</code></dt>
<dd><p>This field holds the number the register &ldquo;originally&rdquo; had; for a
pseudo register turned into a hard reg this will hold the old pseudo
register number.
</p>
<a name="index-REG_005fEXPR"></a>
</dd>
<dt><code>REG_EXPR (<var>x</var>)</code></dt>
<dd><p>If this register is known to hold the value of some user-level
declaration, this is that tree node.
</p>
<a name="index-REG_005fOFFSET"></a>
</dd>
<dt><code>REG_OFFSET (<var>x</var>)</code></dt>
<dd><p>If this register is known to hold the value of some user-level
declaration, this is the offset into that logical storage.
</p></dd>
</dl>

</dd>
<dt><code>SYMBOL_REF</code></dt>
<dd><dl compact="compact">
<dd><a name="index-SYMBOL_005fREF_005fDECL"></a>
</dd>
<dt><code>SYMBOL_REF_DECL (<var>x</var>)</code></dt>
<dd><p>If the <code>symbol_ref</code> <var>x</var> was created for a <code>VAR_DECL</code> or
a <code>FUNCTION_DECL</code>, that tree is recorded here.  If this value is
null, then <var>x</var> was created by back end code generation routines,
and there is no associated front end symbol table entry.
</p>
<p><code>SYMBOL_REF_DECL</code> may also point to a tree of class <code>'c'</code>,
that is, some sort of constant.  In this case, the <code>symbol_ref</code>
is an entry in the per-file constant pool; again, there is no associated
front end symbol table entry.
</p>
<a name="index-SYMBOL_005fREF_005fCONSTANT"></a>
</dd>
<dt><code>SYMBOL_REF_CONSTANT (<var>x</var>)</code></dt>
<dd><p>If &lsquo;<samp>CONSTANT_POOL_ADDRESS_P (<var>x</var>)</samp>&rsquo; is true, this is the constant
pool entry for <var>x</var>.  It is null otherwise.
</p>
<a name="index-SYMBOL_005fREF_005fDATA"></a>
</dd>
<dt><code>SYMBOL_REF_DATA (<var>x</var>)</code></dt>
<dd><p>A field of opaque type used to store <code>SYMBOL_REF_DECL</code> or
<code>SYMBOL_REF_CONSTANT</code>.
</p>
<a name="index-SYMBOL_005fREF_005fFLAGS"></a>
</dd>
<dt><code>SYMBOL_REF_FLAGS (<var>x</var>)</code></dt>
<dd><p>In a <code>symbol_ref</code>, this is used to communicate various predicates
about the symbol.  Some of these are common enough to be computed by
common code, some are specific to the target.  The common bits are:
</p>
<dl compact="compact">
<dd><a name="index-SYMBOL_005fREF_005fFUNCTION_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fFUNCTION"></a>
</dd>
<dt><code>SYMBOL_FLAG_FUNCTION</code></dt>
<dd><p>Set if the symbol refers to a function.
</p>
<a name="index-SYMBOL_005fREF_005fLOCAL_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fLOCAL"></a>
</dd>
<dt><code>SYMBOL_FLAG_LOCAL</code></dt>
<dd><p>Set if the symbol is local to this &ldquo;module&rdquo;.
See <code>TARGET_BINDS_LOCAL_P</code>.
</p>
<a name="index-SYMBOL_005fREF_005fEXTERNAL_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fEXTERNAL"></a>
</dd>
<dt><code>SYMBOL_FLAG_EXTERNAL</code></dt>
<dd><p>Set if this symbol is not defined in this translation unit.
Note that this is not the inverse of <code>SYMBOL_FLAG_LOCAL</code>.
</p>
<a name="index-SYMBOL_005fREF_005fSMALL_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fSMALL"></a>
</dd>
<dt><code>SYMBOL_FLAG_SMALL</code></dt>
<dd><p>Set if the symbol is located in the small data section.
See <code>TARGET_IN_SMALL_DATA_P</code>.
</p>
<a name="index-SYMBOL_005fFLAG_005fTLS_005fSHIFT"></a>
<a name="index-SYMBOL_005fREF_005fTLS_005fMODEL"></a>
</dd>
<dt><code>SYMBOL_REF_TLS_MODEL (<var>x</var>)</code></dt>
<dd><p>This is a multi-bit field accessor that returns the <code>tls_model</code>
to be used for a thread-local storage symbol.  It returns zero for
non-thread-local symbols.
</p>
<a name="index-SYMBOL_005fREF_005fHAS_005fBLOCK_005fINFO_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fHAS_005fBLOCK_005fINFO"></a>
</dd>
<dt><code>SYMBOL_FLAG_HAS_BLOCK_INFO</code></dt>
<dd><p>Set if the symbol has <code>SYMBOL_REF_BLOCK</code> and
<code>SYMBOL_REF_BLOCK_OFFSET</code> fields.
</p>
<a name="index-SYMBOL_005fREF_005fANCHOR_005fP"></a>
<a name="index-SYMBOL_005fFLAG_005fANCHOR"></a>
<a name="index-_002dfsection_002danchors"></a>
</dd>
<dt><code>SYMBOL_FLAG_ANCHOR</code></dt>
<dd><p>Set if the symbol is used as a section anchor.  &ldquo;Section anchors&rdquo;
are symbols that have a known position within an <code>object_block</code>
and that can be used to access nearby members of that block.
They are used to implement <samp>-fsection-anchors</samp>.
</p>
<p>If this flag is set, then <code>SYMBOL_FLAG_HAS_BLOCK_INFO</code> will be too.
</p></dd>
</dl>

<p>Bits beginning with <code>SYMBOL_FLAG_MACH_DEP</code> are available for
the target&rsquo;s use.
</p></dd>
</dl>

<a name="index-SYMBOL_005fREF_005fBLOCK"></a>
</dd>
<dt><code>SYMBOL_REF_BLOCK (<var>x</var>)</code></dt>
<dd><p>If &lsquo;<samp>SYMBOL_REF_HAS_BLOCK_INFO_P (<var>x</var>)</samp>&rsquo;, this is the
&lsquo;<samp>object_block</samp>&rsquo; structure to which the symbol belongs,
or <code>NULL</code> if it has not been assigned a block.
</p>
<a name="index-SYMBOL_005fREF_005fBLOCK_005fOFFSET"></a>
</dd>
<dt><code>SYMBOL_REF_BLOCK_OFFSET (<var>x</var>)</code></dt>
<dd><p>If &lsquo;<samp>SYMBOL_REF_HAS_BLOCK_INFO_P (<var>x</var>)</samp>&rsquo;, this is the offset of <var>x</var>
from the first object in &lsquo;<samp>SYMBOL_REF_BLOCK (<var>x</var>)</samp>&rsquo;.  The value is
negative if <var>x</var> has not yet been assigned to a block, or it has not
been given an offset within that block.
</p></dd>
</dl>

<hr>
<a name="Flags"></a>
<div class="header">
<p>
Next: <a href="#Machine-Modes" accesskey="n" rel="next">Machine Modes</a>, Previous: <a href="#Special-Accessors" accesskey="p" rel="prev">Special Accessors</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Flags-in-an-RTL-Expression"></a>
<h3 class="section">10.5 Flags in an RTL Expression</h3>
<a name="index-flags-in-RTL-expression"></a>

<p>RTL expressions contain several flags (one-bit bit-fields)
that are used in certain types of expression.  Most often they
are accessed with the following macros, which expand into lvalues.
</p>
<dl compact="compact">
<dd><a name="index-CONSTANT_005fPOOL_005fADDRESS_005fP"></a>
<a name="index-symbol_005fref-and-_002fu"></a>
<a name="index-unchanging_002c-in-symbol_005fref"></a>
</dd>
<dt><code>CONSTANT_POOL_ADDRESS_P (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>symbol_ref</code> if it refers to part of the current
function&rsquo;s constant pool.  For most targets these addresses are in a
<code>.rodata</code> section entirely separate from the function, but for
some targets the addresses are close to the beginning of the function.
In either case GCC assumes these addresses can be addressed directly,
perhaps with the help of base registers.
Stored in the <code>unchanging</code> field and printed as &lsquo;<samp>/u</samp>&rsquo;.
</p>
<a name="index-RTL_005fCONST_005fCALL_005fP"></a>
<a name="index-call_005finsn-and-_002fu"></a>
<a name="index-unchanging_002c-in-call_005finsn"></a>
</dd>
<dt><code>RTL_CONST_CALL_P (<var>x</var>)</code></dt>
<dd><p>In a <code>call_insn</code> indicates that the insn represents a call to a
const function.  Stored in the <code>unchanging</code> field and printed as
&lsquo;<samp>/u</samp>&rsquo;.
</p>
<a name="index-RTL_005fPURE_005fCALL_005fP"></a>
<a name="index-call_005finsn-and-_002fi"></a>
<a name="index-return_005fval_002c-in-call_005finsn"></a>
</dd>
<dt><code>RTL_PURE_CALL_P (<var>x</var>)</code></dt>
<dd><p>In a <code>call_insn</code> indicates that the insn represents a call to a
pure function.  Stored in the <code>return_val</code> field and printed as
&lsquo;<samp>/i</samp>&rsquo;.
</p>
<a name="index-RTL_005fCONST_005fOR_005fPURE_005fCALL_005fP"></a>
<a name="index-call_005finsn-and-_002fu-or-_002fi"></a>
</dd>
<dt><code>RTL_CONST_OR_PURE_CALL_P (<var>x</var>)</code></dt>
<dd><p>In a <code>call_insn</code>, true if <code>RTL_CONST_CALL_P</code> or
<code>RTL_PURE_CALL_P</code> is true.
</p>
<a name="index-RTL_005fLOOPING_005fCONST_005fOR_005fPURE_005fCALL_005fP"></a>
<a name="index-call_005finsn-and-_002fc"></a>
<a name="index-call_002c-in-call_005finsn"></a>
</dd>
<dt><code>RTL_LOOPING_CONST_OR_PURE_CALL_P (<var>x</var>)</code></dt>
<dd><p>In a <code>call_insn</code> indicates that the insn represents a possibly
infinite looping call to a const or pure function.  Stored in the
<code>call</code> field and printed as &lsquo;<samp>/c</samp>&rsquo;.  Only true if one of
<code>RTL_CONST_CALL_P</code> or <code>RTL_PURE_CALL_P</code> is true.
</p>
<a name="index-INSN_005fANNULLED_005fBRANCH_005fP"></a>
<a name="index-jump_005finsn-and-_002fu"></a>
<a name="index-call_005finsn-and-_002fu-1"></a>
<a name="index-insn-and-_002fu"></a>
<a name="index-unchanging_002c-in-jump_005finsn_002c-call_005finsn-and-insn"></a>
</dd>
<dt><code>INSN_ANNULLED_BRANCH_P (<var>x</var>)</code></dt>
<dd><p>In a <code>jump_insn</code>, <code>call_insn</code>, or <code>insn</code> indicates
that the branch is an annulling one.  See the discussion under
<code>sequence</code> below.  Stored in the <code>unchanging</code> field and
printed as &lsquo;<samp>/u</samp>&rsquo;.
</p>
<a name="index-INSN_005fDELETED_005fP"></a>
<a name="index-insn-and-_002fv"></a>
<a name="index-call_005finsn-and-_002fv"></a>
<a name="index-jump_005finsn-and-_002fv"></a>
<a name="index-code_005flabel-and-_002fv"></a>
<a name="index-barrier-and-_002fv"></a>
<a name="index-note-and-_002fv"></a>
<a name="index-volatil_002c-in-insn_002c-call_005finsn_002c-jump_005finsn_002c-code_005flabel_002c-barrier_002c-and-note"></a>
</dd>
<dt><code>INSN_DELETED_P (<var>x</var>)</code></dt>
<dd><p>In an <code>insn</code>, <code>call_insn</code>, <code>jump_insn</code>, <code>code_label</code>,
<code>barrier</code>, or <code>note</code>,
nonzero if the insn has been deleted.  Stored in the
<code>volatil</code> field and printed as &lsquo;<samp>/v</samp>&rsquo;.
</p>
<a name="index-INSN_005fFROM_005fTARGET_005fP"></a>
<a name="index-insn-and-_002fs"></a>
<a name="index-jump_005finsn-and-_002fs"></a>
<a name="index-call_005finsn-and-_002fs"></a>
<a name="index-in_005fstruct_002c-in-insn-and-jump_005finsn-and-call_005finsn"></a>
</dd>
<dt><code>INSN_FROM_TARGET_P (<var>x</var>)</code></dt>
<dd><p>In an <code>insn</code> or <code>jump_insn</code> or <code>call_insn</code> in a delay
slot of a branch, indicates that the insn
is from the target of the branch.  If the branch insn has
<code>INSN_ANNULLED_BRANCH_P</code> set, this insn will only be executed if
the branch is taken.  For annulled branches with
<code>INSN_FROM_TARGET_P</code> clear, the insn will be executed only if the
branch is not taken.  When <code>INSN_ANNULLED_BRANCH_P</code> is not set,
this insn will always be executed.  Stored in the <code>in_struct</code>
field and printed as &lsquo;<samp>/s</samp>&rsquo;.
</p>
<a name="index-LABEL_005fPRESERVE_005fP"></a>
<a name="index-code_005flabel-and-_002fi"></a>
<a name="index-note-and-_002fi"></a>
<a name="index-in_005fstruct_002c-in-code_005flabel-and-note"></a>
</dd>
<dt><code>LABEL_PRESERVE_P (<var>x</var>)</code></dt>
<dd><p>In a <code>code_label</code> or <code>note</code>, indicates that the label is referenced by
code or data not visible to the RTL of a given function.
Labels referenced by a non-local goto will have this bit set.  Stored
in the <code>in_struct</code> field and printed as &lsquo;<samp>/s</samp>&rsquo;.
</p>
<a name="index-LABEL_005fREF_005fNONLOCAL_005fP"></a>
<a name="index-label_005fref-and-_002fv"></a>
<a name="index-reg_005flabel-and-_002fv"></a>
<a name="index-volatil_002c-in-label_005fref-and-reg_005flabel"></a>
</dd>
<dt><code>LABEL_REF_NONLOCAL_P (<var>x</var>)</code></dt>
<dd><p>In <code>label_ref</code> and <code>reg_label</code> expressions, nonzero if this is
a reference to a non-local label.
Stored in the <code>volatil</code> field and printed as &lsquo;<samp>/v</samp>&rsquo;.
</p>
<a name="index-MEM_005fKEEP_005fALIAS_005fSET_005fP"></a>
<a name="index-mem-and-_002fj"></a>
<a name="index-jump_002c-in-mem"></a>
</dd>
<dt><code>MEM_KEEP_ALIAS_SET_P (<var>x</var>)</code></dt>
<dd><p>In <code>mem</code> expressions, 1 if we should keep the alias set for this
mem unchanged when we access a component.  Set to 1, for example, when we
are already in a non-addressable component of an aggregate.
Stored in the <code>jump</code> field and printed as &lsquo;<samp>/j</samp>&rsquo;.
</p>
<a name="index-MEM_005fVOLATILE_005fP"></a>
<a name="index-mem-and-_002fv"></a>
<a name="index-asm_005finput-and-_002fv"></a>
<a name="index-asm_005foperands-and-_002fv"></a>
<a name="index-volatil_002c-in-mem_002c-asm_005foperands_002c-and-asm_005finput"></a>
</dd>
<dt><code>MEM_VOLATILE_P (<var>x</var>)</code></dt>
<dd><p>In <code>mem</code>, <code>asm_operands</code>, and <code>asm_input</code> expressions,
nonzero for volatile memory references.
Stored in the <code>volatil</code> field and printed as &lsquo;<samp>/v</samp>&rsquo;.
</p>
<a name="index-MEM_005fNOTRAP_005fP"></a>
<a name="index-mem-and-_002fc"></a>
<a name="index-call_002c-in-mem"></a>
</dd>
<dt><code>MEM_NOTRAP_P (<var>x</var>)</code></dt>
<dd><p>In <code>mem</code>, nonzero for memory references that will not trap.
Stored in the <code>call</code> field and printed as &lsquo;<samp>/c</samp>&rsquo;.
</p>
<a name="index-MEM_005fPOINTER"></a>
<a name="index-mem-and-_002ff"></a>
<a name="index-frame_005frelated_002c-in-mem"></a>
</dd>
<dt><code>MEM_POINTER (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>mem</code> if the memory reference holds a pointer.
Stored in the <code>frame_related</code> field and printed as &lsquo;<samp>/f</samp>&rsquo;.
</p>
<a name="index-REG_005fFUNCTION_005fVALUE_005fP"></a>
<a name="index-reg-and-_002fi"></a>
<a name="index-return_005fval_002c-in-reg"></a>
</dd>
<dt><code>REG_FUNCTION_VALUE_P (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>reg</code> if it is the place in which this function&rsquo;s
value is going to be returned.  (This happens only in a hard
register.)  Stored in the <code>return_val</code> field and printed as
&lsquo;<samp>/i</samp>&rsquo;.
</p>
<a name="index-REG_005fPOINTER"></a>
<a name="index-reg-and-_002ff"></a>
<a name="index-frame_005frelated_002c-in-reg"></a>
</dd>
<dt><code>REG_POINTER (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>reg</code> if the register holds a pointer.  Stored in the
<code>frame_related</code> field and printed as &lsquo;<samp>/f</samp>&rsquo;.
</p>
<a name="index-REG_005fUSERVAR_005fP"></a>
<a name="index-reg-and-_002fv"></a>
<a name="index-volatil_002c-in-reg"></a>
</dd>
<dt><code>REG_USERVAR_P (<var>x</var>)</code></dt>
<dd><p>In a <code>reg</code>, nonzero if it corresponds to a variable present in
the user&rsquo;s source code.  Zero for temporaries generated internally by
the compiler.  Stored in the <code>volatil</code> field and printed as
&lsquo;<samp>/v</samp>&rsquo;.
</p>
<p>The same hard register may be used also for collecting the values of
functions called by this one, but <code>REG_FUNCTION_VALUE_P</code> is zero
in this kind of use.
</p>
<a name="index-RTX_005fFRAME_005fRELATED_005fP"></a>
<a name="index-insn-and-_002ff"></a>
<a name="index-call_005finsn-and-_002ff"></a>
<a name="index-jump_005finsn-and-_002ff"></a>
<a name="index-barrier-and-_002ff"></a>
<a name="index-set-and-_002ff"></a>
<a name="index-frame_005frelated_002c-in-insn_002c-call_005finsn_002c-jump_005finsn_002c-barrier_002c-and-set"></a>
</dd>
<dt><code>RTX_FRAME_RELATED_P (<var>x</var>)</code></dt>
<dd><p>Nonzero in an <code>insn</code>, <code>call_insn</code>, <code>jump_insn</code>,
<code>barrier</code>, or <code>set</code> which is part of a function prologue
and sets the stack pointer, sets the frame pointer, or saves a register.
This flag should also be set on an instruction that sets up a temporary
register to use in place of the frame pointer.
Stored in the <code>frame_related</code> field and printed as &lsquo;<samp>/f</samp>&rsquo;.
</p>
<p>In particular, on RISC targets where there are limits on the sizes of
immediate constants, it is sometimes impossible to reach the register
save area directly from the stack pointer.  In that case, a temporary
register is used that is near enough to the register save area, and the
Canonical Frame Address, i.e., DWARF2&rsquo;s logical frame pointer, register
must (temporarily) be changed to be this temporary register.  So, the
instruction that sets this temporary register must be marked as
<code>RTX_FRAME_RELATED_P</code>.
</p>
<p>If the marked instruction is overly complex (defined in terms of what
<code>dwarf2out_frame_debug_expr</code> can handle), you will also have to
create a <code>REG_FRAME_RELATED_EXPR</code> note and attach it to the
instruction.  This note should contain a simple expression of the
computation performed by this instruction, i.e., one that
<code>dwarf2out_frame_debug_expr</code> can handle.
</p>
<p>This flag is required for exception handling support on targets with RTL
prologues.
</p>
<a name="index-MEM_005fREADONLY_005fP"></a>
<a name="index-mem-and-_002fu"></a>
<a name="index-unchanging_002c-in-mem"></a>
</dd>
<dt><code>MEM_READONLY_P (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>mem</code>, if the memory is statically allocated and read-only.
</p>
<p>Read-only in this context means never modified during the lifetime of the
program, not necessarily in ROM or in write-disabled pages.  A common
example of the later is a shared library&rsquo;s global offset table.  This
table is initialized by the runtime loader, so the memory is technically
writable, but after control is transferred from the runtime loader to the
application, this memory will never be subsequently modified.
</p>
<p>Stored in the <code>unchanging</code> field and printed as &lsquo;<samp>/u</samp>&rsquo;.
</p>
<a name="index-SCHED_005fGROUP_005fP"></a>
<a name="index-insn-and-_002fs-1"></a>
<a name="index-call_005finsn-and-_002fs-1"></a>
<a name="index-jump_005finsn-and-_002fs-1"></a>
<a name="index-in_005fstruct_002c-in-insn_002c-jump_005finsn-and-call_005finsn"></a>
</dd>
<dt><code>SCHED_GROUP_P (<var>x</var>)</code></dt>
<dd><p>During instruction scheduling, in an <code>insn</code>, <code>call_insn</code> or
<code>jump_insn</code>, indicates that the
previous insn must be scheduled together with this insn.  This is used to
ensure that certain groups of instructions will not be split up by the
instruction scheduling pass, for example, <code>use</code> insns before
a <code>call_insn</code> may not be separated from the <code>call_insn</code>.
Stored in the <code>in_struct</code> field and printed as &lsquo;<samp>/s</samp>&rsquo;.
</p>
<a name="index-SET_005fIS_005fRETURN_005fP"></a>
<a name="index-insn-and-_002fj"></a>
<a name="index-jump_002c-in-insn"></a>
</dd>
<dt><code>SET_IS_RETURN_P (<var>x</var>)</code></dt>
<dd><p>For a <code>set</code>, nonzero if it is for a return.
Stored in the <code>jump</code> field and printed as &lsquo;<samp>/j</samp>&rsquo;.
</p>
<a name="index-SIBLING_005fCALL_005fP"></a>
<a name="index-call_005finsn-and-_002fj"></a>
<a name="index-jump_002c-in-call_005finsn"></a>
</dd>
<dt><code>SIBLING_CALL_P (<var>x</var>)</code></dt>
<dd><p>For a <code>call_insn</code>, nonzero if the insn is a sibling call.
Stored in the <code>jump</code> field and printed as &lsquo;<samp>/j</samp>&rsquo;.
</p>
<a name="index-STRING_005fPOOL_005fADDRESS_005fP"></a>
<a name="index-symbol_005fref-and-_002ff"></a>
<a name="index-frame_005frelated_002c-in-symbol_005fref"></a>
</dd>
<dt><code>STRING_POOL_ADDRESS_P (<var>x</var>)</code></dt>
<dd><p>For a <code>symbol_ref</code> expression, nonzero if it addresses this function&rsquo;s
string constant pool.
Stored in the <code>frame_related</code> field and printed as &lsquo;<samp>/f</samp>&rsquo;.
</p>
<a name="index-SUBREG_005fPROMOTED_005fUNSIGNED_005fP"></a>
<a name="index-subreg-and-_002fu-and-_002fv"></a>
<a name="index-unchanging_002c-in-subreg"></a>
<a name="index-volatil_002c-in-subreg"></a>
</dd>
<dt><code>SUBREG_PROMOTED_UNSIGNED_P (<var>x</var>)</code></dt>
<dd><p>Returns a value greater then zero for a <code>subreg</code> that has
<code>SUBREG_PROMOTED_VAR_P</code> nonzero if the object being referenced is kept
zero-extended, zero if it is kept sign-extended, and less then zero if it is
extended some other way via the <code>ptr_extend</code> instruction.
Stored in the <code>unchanging</code>
field and <code>volatil</code> field, printed as &lsquo;<samp>/u</samp>&rsquo; and &lsquo;<samp>/v</samp>&rsquo;.
This macro may only be used to get the value it may not be used to change
the value.  Use <code>SUBREG_PROMOTED_UNSIGNED_SET</code> to change the value.
</p>
<a name="index-SUBREG_005fPROMOTED_005fUNSIGNED_005fSET"></a>
<a name="index-subreg-and-_002fu"></a>
<a name="index-unchanging_002c-in-subreg-1"></a>
<a name="index-volatil_002c-in-subreg-1"></a>
</dd>
<dt><code>SUBREG_PROMOTED_UNSIGNED_SET (<var>x</var>)</code></dt>
<dd><p>Set the <code>unchanging</code> and <code>volatil</code> fields in a <code>subreg</code>
to reflect zero, sign, or other extension.  If <code>volatil</code> is
zero, then <code>unchanging</code> as nonzero means zero extension and as
zero means sign extension.  If <code>volatil</code> is nonzero then some
other type of extension was done via the <code>ptr_extend</code> instruction.
</p>
<a name="index-SUBREG_005fPROMOTED_005fVAR_005fP"></a>
<a name="index-subreg-and-_002fs"></a>
<a name="index-in_005fstruct_002c-in-subreg"></a>
</dd>
<dt><code>SUBREG_PROMOTED_VAR_P (<var>x</var>)</code></dt>
<dd><p>Nonzero in a <code>subreg</code> if it was made when accessing an object that
was promoted to a wider mode in accord with the <code>PROMOTED_MODE</code> machine
description macro (see <a href="#Storage-Layout">Storage Layout</a>).  In this case, the mode of
the <code>subreg</code> is the declared mode of the object and the mode of
<code>SUBREG_REG</code> is the mode of the register that holds the object.
Promoted variables are always either sign- or zero-extended to the wider
mode on every assignment.  Stored in the <code>in_struct</code> field and
printed as &lsquo;<samp>/s</samp>&rsquo;.
</p>
<a name="index-SYMBOL_005fREF_005fUSED"></a>
<a name="index-used_002c-in-symbol_005fref"></a>
</dd>
<dt><code>SYMBOL_REF_USED (<var>x</var>)</code></dt>
<dd><p>In a <code>symbol_ref</code>, indicates that <var>x</var> has been used.  This is
normally only used to ensure that <var>x</var> is only declared external
once.  Stored in the <code>used</code> field.
</p>
<a name="index-SYMBOL_005fREF_005fWEAK"></a>
<a name="index-symbol_005fref-and-_002fi"></a>
<a name="index-return_005fval_002c-in-symbol_005fref"></a>
</dd>
<dt><code>SYMBOL_REF_WEAK (<var>x</var>)</code></dt>
<dd><p>In a <code>symbol_ref</code>, indicates that <var>x</var> has been declared weak.
Stored in the <code>return_val</code> field and printed as &lsquo;<samp>/i</samp>&rsquo;.
</p>
<a name="index-SYMBOL_005fREF_005fFLAG"></a>
<a name="index-symbol_005fref-and-_002fv"></a>
<a name="index-volatil_002c-in-symbol_005fref"></a>
</dd>
<dt><code>SYMBOL_REF_FLAG (<var>x</var>)</code></dt>
<dd><p>In a <code>symbol_ref</code>, this is used as a flag for machine-specific purposes.
Stored in the <code>volatil</code> field and printed as &lsquo;<samp>/v</samp>&rsquo;.
</p>
<p>Most uses of <code>SYMBOL_REF_FLAG</code> are historic and may be subsumed
by <code>SYMBOL_REF_FLAGS</code>.  Certainly use of <code>SYMBOL_REF_FLAGS</code>
is mandatory if the target requires more than one bit of storage.
</p>
<a name="index-PREFETCH_005fSCHEDULE_005fBARRIER_005fP"></a>
<a name="index-prefetch-and-_002fv"></a>
<a name="index-volatile_002c-in-prefetch"></a>
</dd>
<dt><code>PREFETCH_SCHEDULE_BARRIER_P (<var>x</var>)</code></dt>
<dd><p>In a <code>prefetch</code>, indicates that the prefetch is a scheduling barrier.
No other INSNs will be moved over it.
Stored in the <code>volatil</code> field and printed as &lsquo;<samp>/v</samp>&rsquo;.
</p></dd>
</dl>

<p>These are the fields to which the above macros refer:
</p>
<dl compact="compact">
<dd><a name="index-call"></a>
<a name="index-_002fc-in-RTL-dump"></a>
</dd>
<dt><code>call</code></dt>
<dd><p>In a <code>mem</code>, 1 means that the memory reference will not trap.
</p>
<p>In a <code>call</code>, 1 means that this pure or const call may possibly
infinite loop.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/c</samp>&rsquo;.
</p>
<a name="index-frame_005frelated"></a>
<a name="index-_002ff-in-RTL-dump"></a>
</dd>
<dt><code>frame_related</code></dt>
<dd><p>In an <code>insn</code> or <code>set</code> expression, 1 means that it is part of
a function prologue and sets the stack pointer, sets the frame pointer,
saves a register, or sets up a temporary register to use in place of the
frame pointer.
</p>
<p>In <code>reg</code> expressions, 1 means that the register holds a pointer.
</p>
<p>In <code>mem</code> expressions, 1 means that the memory reference holds a pointer.
</p>
<p>In <code>symbol_ref</code> expressions, 1 means that the reference addresses
this function&rsquo;s string constant pool.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/f</samp>&rsquo;.
</p>
<a name="index-in_005fstruct"></a>
<a name="index-_002fs-in-RTL-dump"></a>
</dd>
<dt><code>in_struct</code></dt>
<dd><p>In <code>reg</code> expressions, it is 1 if the register has its entire life
contained within the test expression of some loop.
</p>
<p>In <code>subreg</code> expressions, 1 means that the <code>subreg</code> is accessing
an object that has had its mode promoted from a wider mode.
</p>
<p>In <code>label_ref</code> expressions, 1 means that the referenced label is
outside the innermost loop containing the insn in which the <code>label_ref</code>
was found.
</p>
<p>In <code>code_label</code> expressions, it is 1 if the label may never be deleted.
This is used for labels which are the target of non-local gotos.  Such a
label that would have been deleted is replaced with a <code>note</code> of type
<code>NOTE_INSN_DELETED_LABEL</code>.
</p>
<p>In an <code>insn</code> during dead-code elimination, 1 means that the insn is
dead code.
</p>
<p>In an <code>insn</code> or <code>jump_insn</code> during reorg for an insn in the
delay slot of a branch,
1 means that this insn is from the target of the branch.
</p>
<p>In an <code>insn</code> during instruction scheduling, 1 means that this insn
must be scheduled as part of a group together with the previous insn.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/s</samp>&rsquo;.
</p>
<a name="index-return_005fval"></a>
<a name="index-_002fi-in-RTL-dump"></a>
</dd>
<dt><code>return_val</code></dt>
<dd><p>In <code>reg</code> expressions, 1 means the register contains
the value to be returned by the current function.  On
machines that pass parameters in registers, the same register number
may be used for parameters as well, but this flag is not set on such
uses.
</p>
<p>In <code>symbol_ref</code> expressions, 1 means the referenced symbol is weak.
</p>
<p>In <code>call</code> expressions, 1 means the call is pure.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/i</samp>&rsquo;.
</p>
<a name="index-jump"></a>
<a name="index-_002fj-in-RTL-dump"></a>
</dd>
<dt><code>jump</code></dt>
<dd><p>In a <code>mem</code> expression, 1 means we should keep the alias set for this
mem unchanged when we access a component.
</p>
<p>In a <code>set</code>, 1 means it is for a return.
</p>
<p>In a <code>call_insn</code>, 1 means it is a sibling call.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/j</samp>&rsquo;.
</p>
<a name="index-unchanging"></a>
<a name="index-_002fu-in-RTL-dump"></a>
</dd>
<dt><code>unchanging</code></dt>
<dd><p>In <code>reg</code> and <code>mem</code> expressions, 1 means
that the value of the expression never changes.
</p>
<p>In <code>subreg</code> expressions, it is 1 if the <code>subreg</code> references an
unsigned object whose mode has been promoted to a wider mode.
</p>
<p>In an <code>insn</code> or <code>jump_insn</code> in the delay slot of a branch
instruction, 1 means an annulling branch should be used.
</p>
<p>In a <code>symbol_ref</code> expression, 1 means that this symbol addresses
something in the per-function constant pool.
</p>
<p>In a <code>call_insn</code> 1 means that this instruction is a call to a const
function.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/u</samp>&rsquo;.
</p>
<a name="index-used"></a>
</dd>
<dt><code>used</code></dt>
<dd><p>This flag is used directly (without an access macro) at the end of RTL
generation for a function, to count the number of times an expression
appears in insns.  Expressions that appear more than once are copied,
according to the rules for shared structure (see <a href="#Sharing">Sharing</a>).
</p>
<p>For a <code>reg</code>, it is used directly (without an access macro) by the
leaf register renumbering code to ensure that each register is only
renumbered once.
</p>
<p>In a <code>symbol_ref</code>, it indicates that an external declaration for
the symbol has already been written.
</p>
<a name="index-volatil"></a>
<a name="index-_002fv-in-RTL-dump"></a>
</dd>
<dt><code>volatil</code></dt>
<dd><a name="index-volatile-memory-references"></a>
<p>In a <code>mem</code>, <code>asm_operands</code>, or <code>asm_input</code>
expression, it is 1 if the memory
reference is volatile.  Volatile memory references may not be deleted,
reordered or combined.
</p>
<p>In a <code>symbol_ref</code> expression, it is used for machine-specific
purposes.
</p>
<p>In a <code>reg</code> expression, it is 1 if the value is a user-level variable.
0 indicates an internal compiler temporary.
</p>
<p>In an <code>insn</code>, 1 means the insn has been deleted.
</p>
<p>In <code>label_ref</code> and <code>reg_label</code> expressions, 1 means a reference
to a non-local label.
</p>
<p>In <code>prefetch</code> expressions, 1 means that the containing insn is a
scheduling barrier.
</p>
<p>In an RTL dump, this flag is represented as &lsquo;<samp>/v</samp>&rsquo;.
</p></dd>
</dl>

<hr>
<a name="Machine-Modes"></a>
<div class="header">
<p>
Next: <a href="#Constants" accesskey="n" rel="next">Constants</a>, Previous: <a href="#Flags" accesskey="p" rel="prev">Flags</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Machine-Modes-1"></a>
<h3 class="section">10.6 Machine Modes</h3>
<a name="index-machine-modes"></a>

<a name="index-enum-machine_005fmode"></a>
<p>A machine mode describes a size of data object and the representation used
for it.  In the C code, machine modes are represented by an enumeration
type, <code>enum machine_mode</code>, defined in <samp>machmode.def</samp>.  Each RTL
expression has room for a machine mode and so do certain kinds of tree
expressions (declarations and types, to be precise).
</p>
<p>In debugging dumps and machine descriptions, the machine mode of an RTL
expression is written after the expression code with a colon to separate
them.  The letters &lsquo;<samp>mode</samp>&rsquo; which appear at the end of each machine mode
name are omitted.  For example, <code>(reg:SI 38)</code> is a <code>reg</code>
expression with machine mode <code>SImode</code>.  If the mode is
<code>VOIDmode</code>, it is not written at all.
</p>
<p>Here is a table of machine modes.  The term &ldquo;byte&rdquo; below refers to an
object of <code>BITS_PER_UNIT</code> bits (see <a href="#Storage-Layout">Storage Layout</a>).
</p>
<dl compact="compact">
<dd><a name="index-BImode"></a>
</dd>
<dt><code>BImode</code></dt>
<dd><p>&ldquo;Bit&rdquo; mode represents a single bit, for predicate registers.
</p>
<a name="index-QImode"></a>
</dd>
<dt><code>QImode</code></dt>
<dd><p>&ldquo;Quarter-Integer&rdquo; mode represents a single byte treated as an integer.
</p>
<a name="index-HImode"></a>
</dd>
<dt><code>HImode</code></dt>
<dd><p>&ldquo;Half-Integer&rdquo; mode represents a two-byte integer.
</p>
<a name="index-PSImode"></a>
</dd>
<dt><code>PSImode</code></dt>
<dd><p>&ldquo;Partial Single Integer&rdquo; mode represents an integer which occupies
four bytes but which doesn&rsquo;t really use all four.  On some machines,
this is the right mode to use for pointers.
</p>
<a name="index-SImode"></a>
</dd>
<dt><code>SImode</code></dt>
<dd><p>&ldquo;Single Integer&rdquo; mode represents a four-byte integer.
</p>
<a name="index-PDImode"></a>
</dd>
<dt><code>PDImode</code></dt>
<dd><p>&ldquo;Partial Double Integer&rdquo; mode represents an integer which occupies
eight bytes but which doesn&rsquo;t really use all eight.  On some machines,
this is the right mode to use for certain pointers.
</p>
<a name="index-DImode"></a>
</dd>
<dt><code>DImode</code></dt>
<dd><p>&ldquo;Double Integer&rdquo; mode represents an eight-byte integer.
</p>
<a name="index-TImode"></a>
</dd>
<dt><code>TImode</code></dt>
<dd><p>&ldquo;Tetra Integer&rdquo; (?) mode represents a sixteen-byte integer.
</p>
<a name="index-OImode"></a>
</dd>
<dt><code>OImode</code></dt>
<dd><p>&ldquo;Octa Integer&rdquo; (?) mode represents a thirty-two-byte integer.
</p>
<a name="index-QFmode"></a>
</dd>
<dt><code>QFmode</code></dt>
<dd><p>&ldquo;Quarter-Floating&rdquo; mode represents a quarter-precision (single byte)
floating point number.
</p>
<a name="index-HFmode"></a>
</dd>
<dt><code>HFmode</code></dt>
<dd><p>&ldquo;Half-Floating&rdquo; mode represents a half-precision (two byte) floating
point number.
</p>
<a name="index-TQFmode"></a>
</dd>
<dt><code>TQFmode</code></dt>
<dd><p>&ldquo;Three-Quarter-Floating&rdquo; (?) mode represents a three-quarter-precision
(three byte) floating point number.
</p>
<a name="index-SFmode"></a>
</dd>
<dt><code>SFmode</code></dt>
<dd><p>&ldquo;Single Floating&rdquo; mode represents a four byte floating point number.
In the common case, of a processor with IEEE arithmetic and 8-bit bytes,
this is a single-precision IEEE floating point number; it can also be
used for double-precision (on processors with 16-bit bytes) and
single-precision VAX and IBM types.
</p>
<a name="index-DFmode"></a>
</dd>
<dt><code>DFmode</code></dt>
<dd><p>&ldquo;Double Floating&rdquo; mode represents an eight byte floating point number.
In the common case, of a processor with IEEE arithmetic and 8-bit bytes,
this is a double-precision IEEE floating point number.
</p>
<a name="index-XFmode"></a>
</dd>
<dt><code>XFmode</code></dt>
<dd><p>&ldquo;Extended Floating&rdquo; mode represents an IEEE extended floating point
number.  This mode only has 80 meaningful bits (ten bytes).  Some
processors require such numbers to be padded to twelve bytes, others
to sixteen; this mode is used for either.
</p>
<a name="index-SDmode"></a>
</dd>
<dt><code>SDmode</code></dt>
<dd><p>&ldquo;Single Decimal Floating&rdquo; mode represents a four byte decimal
floating point number (as distinct from conventional binary floating
point).
</p>
<a name="index-DDmode"></a>
</dd>
<dt><code>DDmode</code></dt>
<dd><p>&ldquo;Double Decimal Floating&rdquo; mode represents an eight byte decimal
floating point number.
</p>
<a name="index-TDmode"></a>
</dd>
<dt><code>TDmode</code></dt>
<dd><p>&ldquo;Tetra Decimal Floating&rdquo; mode represents a sixteen byte decimal
floating point number all 128 of whose bits are meaningful.
</p>
<a name="index-TFmode"></a>
</dd>
<dt><code>TFmode</code></dt>
<dd><p>&ldquo;Tetra Floating&rdquo; mode represents a sixteen byte floating point number
all 128 of whose bits are meaningful.  One common use is the
IEEE quad-precision format.
</p>
<a name="index-QQmode"></a>
</dd>
<dt><code>QQmode</code></dt>
<dd><p>&ldquo;Quarter-Fractional&rdquo; mode represents a single byte treated as a signed
fractional number.  The default format is &ldquo;s.7&rdquo;.
</p>
<a name="index-HQmode"></a>
</dd>
<dt><code>HQmode</code></dt>
<dd><p>&ldquo;Half-Fractional&rdquo; mode represents a two-byte signed fractional number.
The default format is &ldquo;s.15&rdquo;.
</p>
<a name="index-SQmode"></a>
</dd>
<dt><code>SQmode</code></dt>
<dd><p>&ldquo;Single Fractional&rdquo; mode represents a four-byte signed fractional number.
The default format is &ldquo;s.31&rdquo;.
</p>
<a name="index-DQmode"></a>
</dd>
<dt><code>DQmode</code></dt>
<dd><p>&ldquo;Double Fractional&rdquo; mode represents an eight-byte signed fractional number.
The default format is &ldquo;s.63&rdquo;.
</p>
<a name="index-TQmode"></a>
</dd>
<dt><code>TQmode</code></dt>
<dd><p>&ldquo;Tetra Fractional&rdquo; mode represents a sixteen-byte signed fractional number.
The default format is &ldquo;s.127&rdquo;.
</p>
<a name="index-UQQmode"></a>
</dd>
<dt><code>UQQmode</code></dt>
<dd><p>&ldquo;Unsigned Quarter-Fractional&rdquo; mode represents a single byte treated as an
unsigned fractional number.  The default format is &ldquo;.8&rdquo;.
</p>
<a name="index-UHQmode"></a>
</dd>
<dt><code>UHQmode</code></dt>
<dd><p>&ldquo;Unsigned Half-Fractional&rdquo; mode represents a two-byte unsigned fractional
number.  The default format is &ldquo;.16&rdquo;.
</p>
<a name="index-USQmode"></a>
</dd>
<dt><code>USQmode</code></dt>
<dd><p>&ldquo;Unsigned Single Fractional&rdquo; mode represents a four-byte unsigned fractional
number.  The default format is &ldquo;.32&rdquo;.
</p>
<a name="index-UDQmode"></a>
</dd>
<dt><code>UDQmode</code></dt>
<dd><p>&ldquo;Unsigned Double Fractional&rdquo; mode represents an eight-byte unsigned
fractional number.  The default format is &ldquo;.64&rdquo;.
</p>
<a name="index-UTQmode"></a>
</dd>
<dt><code>UTQmode</code></dt>
<dd><p>&ldquo;Unsigned Tetra Fractional&rdquo; mode represents a sixteen-byte unsigned
fractional number.  The default format is &ldquo;.128&rdquo;.
</p>
<a name="index-HAmode"></a>
</dd>
<dt><code>HAmode</code></dt>
<dd><p>&ldquo;Half-Accumulator&rdquo; mode represents a two-byte signed accumulator.
The default format is &ldquo;s8.7&rdquo;.
</p>
<a name="index-SAmode"></a>
</dd>
<dt><code>SAmode</code></dt>
<dd><p>&ldquo;Single Accumulator&rdquo; mode represents a four-byte signed accumulator.
The default format is &ldquo;s16.15&rdquo;.
</p>
<a name="index-DAmode"></a>
</dd>
<dt><code>DAmode</code></dt>
<dd><p>&ldquo;Double Accumulator&rdquo; mode represents an eight-byte signed accumulator.
The default format is &ldquo;s32.31&rdquo;.
</p>
<a name="index-TAmode"></a>
</dd>
<dt><code>TAmode</code></dt>
<dd><p>&ldquo;Tetra Accumulator&rdquo; mode represents a sixteen-byte signed accumulator.
The default format is &ldquo;s64.63&rdquo;.
</p>
<a name="index-UHAmode"></a>
</dd>
<dt><code>UHAmode</code></dt>
<dd><p>&ldquo;Unsigned Half-Accumulator&rdquo; mode represents a two-byte unsigned accumulator.
The default format is &ldquo;8.8&rdquo;.
</p>
<a name="index-USAmode"></a>
</dd>
<dt><code>USAmode</code></dt>
<dd><p>&ldquo;Unsigned Single Accumulator&rdquo; mode represents a four-byte unsigned
accumulator.  The default format is &ldquo;16.16&rdquo;.
</p>
<a name="index-UDAmode"></a>
</dd>
<dt><code>UDAmode</code></dt>
<dd><p>&ldquo;Unsigned Double Accumulator&rdquo; mode represents an eight-byte unsigned
accumulator.  The default format is &ldquo;32.32&rdquo;.
</p>
<a name="index-UTAmode"></a>
</dd>
<dt><code>UTAmode</code></dt>
<dd><p>&ldquo;Unsigned Tetra Accumulator&rdquo; mode represents a sixteen-byte unsigned
accumulator.  The default format is &ldquo;64.64&rdquo;.
</p>
<a name="index-CCmode"></a>
</dd>
<dt><code>CCmode</code></dt>
<dd><p>&ldquo;Condition Code&rdquo; mode represents the value of a condition code, which
is a machine-specific set of bits used to represent the result of a
comparison operation.  Other machine-specific modes may also be used for
the condition code.  These modes are not used on machines that use
<code>cc0</code> (see <a href="#Condition-Code">Condition Code</a>).
</p>
<a name="index-BLKmode"></a>
</dd>
<dt><code>BLKmode</code></dt>
<dd><p>&ldquo;Block&rdquo; mode represents values that are aggregates to which none of
the other modes apply.  In RTL, only memory references can have this mode,
and only if they appear in string-move or vector instructions.  On machines
which have no such instructions, <code>BLKmode</code> will not appear in RTL.
</p>
<a name="index-VOIDmode"></a>
</dd>
<dt><code>VOIDmode</code></dt>
<dd><p>Void mode means the absence of a mode or an unspecified mode.
For example, RTL expressions of code <code>const_int</code> have mode
<code>VOIDmode</code> because they can be taken to have whatever mode the context
requires.  In debugging dumps of RTL, <code>VOIDmode</code> is expressed by
the absence of any mode.
</p>
<a name="index-QCmode"></a>
<a name="index-HCmode"></a>
<a name="index-SCmode"></a>
<a name="index-DCmode"></a>
<a name="index-XCmode"></a>
<a name="index-TCmode"></a>
</dd>
<dt><code>QCmode, HCmode, SCmode, DCmode, XCmode, TCmode</code></dt>
<dd><p>These modes stand for a complex number represented as a pair of floating
point values.  The floating point values are in <code>QFmode</code>,
<code>HFmode</code>, <code>SFmode</code>, <code>DFmode</code>, <code>XFmode</code>, and
<code>TFmode</code>, respectively.
</p>
<a name="index-CQImode"></a>
<a name="index-CHImode"></a>
<a name="index-CSImode"></a>
<a name="index-CDImode"></a>
<a name="index-CTImode"></a>
<a name="index-COImode"></a>
</dd>
<dt><code>CQImode, CHImode, CSImode, CDImode, CTImode, COImode</code></dt>
<dd><p>These modes stand for a complex number represented as a pair of integer
values.  The integer values are in <code>QImode</code>, <code>HImode</code>,
<code>SImode</code>, <code>DImode</code>, <code>TImode</code>, and <code>OImode</code>,
respectively.
</p></dd>
</dl>

<p>The machine description defines <code>Pmode</code> as a C macro which expands
into the machine mode used for addresses.  Normally this is the mode
whose size is <code>BITS_PER_WORD</code>, <code>SImode</code> on 32-bit machines.
</p>
<p>The only modes which a machine description <i>must</i> support are
<code>QImode</code>, and the modes corresponding to <code>BITS_PER_WORD</code>,
<code>FLOAT_TYPE_SIZE</code> and <code>DOUBLE_TYPE_SIZE</code>.
The compiler will attempt to use <code>DImode</code> for 8-byte structures and
unions, but this can be prevented by overriding the definition of
<code>MAX_FIXED_MODE_SIZE</code>.  Alternatively, you can have the compiler
use <code>TImode</code> for 16-byte structures and unions.  Likewise, you can
arrange for the C type <code>short int</code> to avoid using <code>HImode</code>.
</p>
<a name="index-mode-classes"></a>
<p>Very few explicit references to machine modes remain in the compiler and
these few references will soon be removed.  Instead, the machine modes
are divided into mode classes.  These are represented by the enumeration
type <code>enum mode_class</code> defined in <samp>machmode.h</samp>.  The possible
mode classes are:
</p>
<dl compact="compact">
<dd><a name="index-MODE_005fINT"></a>
</dd>
<dt><code>MODE_INT</code></dt>
<dd><p>Integer modes.  By default these are <code>BImode</code>, <code>QImode</code>,
<code>HImode</code>, <code>SImode</code>, <code>DImode</code>, <code>TImode</code>, and
<code>OImode</code>.
</p>
<a name="index-MODE_005fPARTIAL_005fINT"></a>
</dd>
<dt><code>MODE_PARTIAL_INT</code></dt>
<dd><p>The &ldquo;partial integer&rdquo; modes, <code>PQImode</code>, <code>PHImode</code>,
<code>PSImode</code> and <code>PDImode</code>.
</p>
<a name="index-MODE_005fFLOAT"></a>
</dd>
<dt><code>MODE_FLOAT</code></dt>
<dd><p>Floating point modes.  By default these are <code>QFmode</code>,
<code>HFmode</code>, <code>TQFmode</code>, <code>SFmode</code>, <code>DFmode</code>,
<code>XFmode</code> and <code>TFmode</code>.
</p>
<a name="index-MODE_005fDECIMAL_005fFLOAT"></a>
</dd>
<dt><code>MODE_DECIMAL_FLOAT</code></dt>
<dd><p>Decimal floating point modes.  By default these are <code>SDmode</code>,
<code>DDmode</code> and <code>TDmode</code>.
</p>
<a name="index-MODE_005fFRACT"></a>
</dd>
<dt><code>MODE_FRACT</code></dt>
<dd><p>Signed fractional modes.  By default these are <code>QQmode</code>, <code>HQmode</code>,
<code>SQmode</code>, <code>DQmode</code> and <code>TQmode</code>.
</p>
<a name="index-MODE_005fUFRACT"></a>
</dd>
<dt><code>MODE_UFRACT</code></dt>
<dd><p>Unsigned fractional modes.  By default these are <code>UQQmode</code>, <code>UHQmode</code>,
<code>USQmode</code>, <code>UDQmode</code> and <code>UTQmode</code>.
</p>
<a name="index-MODE_005fACCUM"></a>
</dd>
<dt><code>MODE_ACCUM</code></dt>
<dd><p>Signed accumulator modes.  By default these are <code>HAmode</code>,
<code>SAmode</code>, <code>DAmode</code> and <code>TAmode</code>.
</p>
<a name="index-MODE_005fUACCUM"></a>
</dd>
<dt><code>MODE_UACCUM</code></dt>
<dd><p>Unsigned accumulator modes.  By default these are <code>UHAmode</code>,
<code>USAmode</code>, <code>UDAmode</code> and <code>UTAmode</code>.
</p>
<a name="index-MODE_005fCOMPLEX_005fINT"></a>
</dd>
<dt><code>MODE_COMPLEX_INT</code></dt>
<dd><p>Complex integer modes.  (These are not currently implemented).
</p>
<a name="index-MODE_005fCOMPLEX_005fFLOAT"></a>
</dd>
<dt><code>MODE_COMPLEX_FLOAT</code></dt>
<dd><p>Complex floating point modes.  By default these are <code>QCmode</code>,
<code>HCmode</code>, <code>SCmode</code>, <code>DCmode</code>, <code>XCmode</code>, and
<code>TCmode</code>.
</p>
<a name="index-MODE_005fFUNCTION"></a>
</dd>
<dt><code>MODE_FUNCTION</code></dt>
<dd><p>Algol or Pascal function variables including a static chain.
(These are not currently implemented).
</p>
<a name="index-MODE_005fCC"></a>
</dd>
<dt><code>MODE_CC</code></dt>
<dd><p>Modes representing condition code values.  These are <code>CCmode</code> plus
any <code>CC_MODE</code> modes listed in the <samp><var>machine</var>-modes.def</samp>.
See <a href="#Jump-Patterns">Jump Patterns</a>,
also see <a href="#Condition-Code">Condition Code</a>.
</p>
<a name="index-MODE_005fRANDOM"></a>
</dd>
<dt><code>MODE_RANDOM</code></dt>
<dd><p>This is a catchall mode class for modes which don&rsquo;t fit into the above
classes.  Currently <code>VOIDmode</code> and <code>BLKmode</code> are in
<code>MODE_RANDOM</code>.
</p></dd>
</dl>

<p>Here are some C macros that relate to machine modes:
</p>
<dl compact="compact">
<dd><a name="index-GET_005fMODE"></a>
</dd>
<dt><code>GET_MODE (<var>x</var>)</code></dt>
<dd><p>Returns the machine mode of the RTX <var>x</var>.
</p>
<a name="index-PUT_005fMODE"></a>
</dd>
<dt><code>PUT_MODE (<var>x</var>, <var>newmode</var>)</code></dt>
<dd><p>Alters the machine mode of the RTX <var>x</var> to be <var>newmode</var>.
</p>
<a name="index-NUM_005fMACHINE_005fMODES"></a>
</dd>
<dt><code>NUM_MACHINE_MODES</code></dt>
<dd><p>Stands for the number of machine modes available on the target
machine.  This is one greater than the largest numeric value of any
machine mode.
</p>
<a name="index-GET_005fMODE_005fNAME"></a>
</dd>
<dt><code>GET_MODE_NAME (<var>m</var>)</code></dt>
<dd><p>Returns the name of mode <var>m</var> as a string.
</p>
<a name="index-GET_005fMODE_005fCLASS"></a>
</dd>
<dt><code>GET_MODE_CLASS (<var>m</var>)</code></dt>
<dd><p>Returns the mode class of mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fWIDER_005fMODE"></a>
</dd>
<dt><code>GET_MODE_WIDER_MODE (<var>m</var>)</code></dt>
<dd><p>Returns the next wider natural mode.  For example, the expression
<code>GET_MODE_WIDER_MODE (QImode)</code> returns <code>HImode</code>.
</p>
<a name="index-GET_005fMODE_005fSIZE"></a>
</dd>
<dt><code>GET_MODE_SIZE (<var>m</var>)</code></dt>
<dd><p>Returns the size in bytes of a datum of mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fBITSIZE"></a>
</dd>
<dt><code>GET_MODE_BITSIZE (<var>m</var>)</code></dt>
<dd><p>Returns the size in bits of a datum of mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fIBIT"></a>
</dd>
<dt><code>GET_MODE_IBIT (<var>m</var>)</code></dt>
<dd><p>Returns the number of integral bits of a datum of fixed-point mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fFBIT"></a>
</dd>
<dt><code>GET_MODE_FBIT (<var>m</var>)</code></dt>
<dd><p>Returns the number of fractional bits of a datum of fixed-point mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fMASK"></a>
</dd>
<dt><code>GET_MODE_MASK (<var>m</var>)</code></dt>
<dd><p>Returns a bitmask containing 1 for all bits in a word that fit within
mode <var>m</var>.  This macro can only be used for modes whose bitsize is
less than or equal to <code>HOST_BITS_PER_INT</code>.
</p>
<a name="index-GET_005fMODE_005fALIGNMENT"></a>
</dd>
<dt><code>GET_MODE_ALIGNMENT (<var>m</var>)</code></dt>
<dd><p>Return the required alignment, in bits, for an object of mode <var>m</var>.
</p>
<a name="index-GET_005fMODE_005fUNIT_005fSIZE"></a>
</dd>
<dt><code>GET_MODE_UNIT_SIZE (<var>m</var>)</code></dt>
<dd><p>Returns the size in bytes of the subunits of a datum of mode <var>m</var>.
This is the same as <code>GET_MODE_SIZE</code> except in the case of complex
modes.  For them, the unit size is the size of the real or imaginary
part.
</p>
<a name="index-GET_005fMODE_005fNUNITS"></a>
</dd>
<dt><code>GET_MODE_NUNITS (<var>m</var>)</code></dt>
<dd><p>Returns the number of units contained in a mode, i.e.,
<code>GET_MODE_SIZE</code> divided by <code>GET_MODE_UNIT_SIZE</code>.
</p>
<a name="index-GET_005fCLASS_005fNARROWEST_005fMODE"></a>
</dd>
<dt><code>GET_CLASS_NARROWEST_MODE (<var>c</var>)</code></dt>
<dd><p>Returns the narrowest mode in mode class <var>c</var>.
</p></dd>
</dl>

<a name="index-byte_005fmode"></a>
<a name="index-word_005fmode"></a>
<p>The global variables <code>byte_mode</code> and <code>word_mode</code> contain modes
whose classes are <code>MODE_INT</code> and whose bitsizes are either
<code>BITS_PER_UNIT</code> or <code>BITS_PER_WORD</code>, respectively.  On 32-bit
machines, these are <code>QImode</code> and <code>SImode</code>, respectively.
</p>
<hr>
<a name="Constants"></a>
<div class="header">
<p>
Next: <a href="#Regs-and-Memory" accesskey="n" rel="next">Regs and Memory</a>, Previous: <a href="#Machine-Modes" accesskey="p" rel="prev">Machine Modes</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constant-Expression-Types"></a>
<h3 class="section">10.7 Constant Expression Types</h3>
<a name="index-RTL-constants"></a>
<a name="index-RTL-constant-expression-types"></a>

<p>The simplest RTL expressions are those that represent constant values.
</p>
<dl compact="compact">
<dd><a name="index-const_005fint"></a>
</dd>
<dt><code>(const_int <var>i</var>)</code></dt>
<dd><p>This type of expression represents the integer value <var>i</var>.  <var>i</var>
is customarily accessed with the macro <code>INTVAL</code> as in
<code>INTVAL (<var>exp</var>)</code>, which is equivalent to <code>XWINT (<var>exp</var>, 0)</code>.
</p>
<p>Constants generated for modes with fewer bits than in
<code>HOST_WIDE_INT</code> must be sign extended to full width (e.g., with
<code>gen_int_mode</code>).  For constants for modes with more bits than in
<code>HOST_WIDE_INT</code> the implied high order bits of that constant are
copies of the top bit.  Note however that values are neither
inherently signed nor inherently unsigned; where necessary, signedness
is determined by the rtl operation instead.
</p>
<a name="index-const0_005frtx"></a>
<a name="index-const1_005frtx"></a>
<a name="index-const2_005frtx"></a>
<a name="index-constm1_005frtx"></a>
<p>There is only one expression object for the integer value zero; it is
the value of the variable <code>const0_rtx</code>.  Likewise, the only
expression for integer value one is found in <code>const1_rtx</code>, the only
expression for integer value two is found in <code>const2_rtx</code>, and the
only expression for integer value negative one is found in
<code>constm1_rtx</code>.  Any attempt to create an expression of code
<code>const_int</code> and value zero, one, two or negative one will return
<code>const0_rtx</code>, <code>const1_rtx</code>, <code>const2_rtx</code> or
<code>constm1_rtx</code> as appropriate.
</p>
<a name="index-const_005ftrue_005frtx"></a>
<p>Similarly, there is only one object for the integer whose value is
<code>STORE_FLAG_VALUE</code>.  It is found in <code>const_true_rtx</code>.  If
<code>STORE_FLAG_VALUE</code> is one, <code>const_true_rtx</code> and
<code>const1_rtx</code> will point to the same object.  If
<code>STORE_FLAG_VALUE</code> is -1, <code>const_true_rtx</code> and
<code>constm1_rtx</code> will point to the same object.
</p>
<a name="index-const_005fdouble"></a>
</dd>
<dt><code>(const_double:<var>m</var> <var>i0</var> <var>i1</var> &hellip;)</code></dt>
<dd><p>Represents either a floating-point constant of mode <var>m</var> or an
integer constant too large to fit into <code>HOST_BITS_PER_WIDE_INT</code>
bits but small enough to fit within twice that number of bits (GCC
does not provide a mechanism to represent even larger constants).  In
the latter case, <var>m</var> will be <code>VOIDmode</code>.  For integral values
constants for modes with more bits than twice the number in
<code>HOST_WIDE_INT</code> the implied high order bits of that constant are
copies of the top bit of <code>CONST_DOUBLE_HIGH</code>.  Note however that
integral values are neither inherently signed nor inherently unsigned;
where necessary, signedness is determined by the rtl operation
instead.
</p>
<a name="index-CONST_005fDOUBLE_005fLOW"></a>
<p>If <var>m</var> is <code>VOIDmode</code>, the bits of the value are stored in
<var>i0</var> and <var>i1</var>.  <var>i0</var> is customarily accessed with the macro
<code>CONST_DOUBLE_LOW</code> and <var>i1</var> with <code>CONST_DOUBLE_HIGH</code>.
</p>
<p>If the constant is floating point (regardless of its precision), then
the number of integers used to store the value depends on the size of
<code>REAL_VALUE_TYPE</code> (see <a href="#Floating-Point">Floating Point</a>).  The integers
represent a floating point number, but not precisely in the target
machine&rsquo;s or host machine&rsquo;s floating point format.  To convert them to
the precise bit pattern used by the target machine, use the macro
<code>REAL_VALUE_TO_TARGET_DOUBLE</code> and friends (see <a href="#Data-Output">Data Output</a>).
</p>
<a name="index-const_005ffixed"></a>
</dd>
<dt><code>(const_fixed:<var>m</var> &hellip;)</code></dt>
<dd><p>Represents a fixed-point constant of mode <var>m</var>.
The operand is a data structure of type <code>struct fixed_value</code> and
is accessed with the macro <code>CONST_FIXED_VALUE</code>.  The high part of
data is accessed with <code>CONST_FIXED_VALUE_HIGH</code>; the low part is
accessed with <code>CONST_FIXED_VALUE_LOW</code>.
</p>
<a name="index-const_005fvector"></a>
</dd>
<dt><code>(const_vector:<var>m</var> [<var>x0</var> <var>x1</var> &hellip;])</code></dt>
<dd><p>Represents a vector constant.  The square brackets stand for the vector
containing the constant elements.  <var>x0</var>, <var>x1</var> and so on are
the <code>const_int</code>, <code>const_double</code> or <code>const_fixed</code> elements.
</p>
<p>The number of units in a <code>const_vector</code> is obtained with the macro
<code>CONST_VECTOR_NUNITS</code> as in <code>CONST_VECTOR_NUNITS (<var>v</var>)</code>.
</p>
<p>Individual elements in a vector constant are accessed with the macro
<code>CONST_VECTOR_ELT</code> as in <code>CONST_VECTOR_ELT (<var>v</var>, <var>n</var>)</code>
where <var>v</var> is the vector constant and <var>n</var> is the element
desired.
</p>
<a name="index-const_005fstring"></a>
</dd>
<dt><code>(const_string <var>str</var>)</code></dt>
<dd><p>Represents a constant string with value <var>str</var>.  Currently this is
used only for insn attributes (see <a href="#Insn-Attributes">Insn Attributes</a>) since constant
strings in C are placed in memory.
</p>
<a name="index-symbol_005fref"></a>
</dd>
<dt><code>(symbol_ref:<var>mode</var> <var>symbol</var>)</code></dt>
<dd><p>Represents the value of an assembler label for data.  <var>symbol</var> is
a string that describes the name of the assembler label.  If it starts
with a &lsquo;<samp>*</samp>&rsquo;, the label is the rest of <var>symbol</var> not including
the &lsquo;<samp>*</samp>&rsquo;.  Otherwise, the label is <var>symbol</var>, usually prefixed
with &lsquo;<samp>_</samp>&rsquo;.
</p>
<p>The <code>symbol_ref</code> contains a mode, which is usually <code>Pmode</code>.
Usually that is the only mode for which a symbol is directly valid.
</p>
<a name="index-label_005fref"></a>
</dd>
<dt><code>(label_ref:<var>mode</var> <var>label</var>)</code></dt>
<dd><p>Represents the value of an assembler label for code.  It contains one
operand, an expression, which must be a <code>code_label</code> or a <code>note</code>
of type <code>NOTE_INSN_DELETED_LABEL</code> that appears in the instruction
sequence to identify the place where the label should go.
</p>
<p>The reason for using a distinct expression type for code label
references is so that jump optimization can distinguish them.
</p>
<p>The <code>label_ref</code> contains a mode, which is usually <code>Pmode</code>.
Usually that is the only mode for which a label is directly valid.
</p>
<a name="index-const"></a>
</dd>
<dt><code>(const:<var>m</var> <var>exp</var>)</code></dt>
<dd><p>Represents a constant that is the result of an assembly-time
arithmetic computation.  The operand, <var>exp</var>, is an expression that
contains only constants (<code>const_int</code>, <code>symbol_ref</code> and
<code>label_ref</code> expressions) combined with <code>plus</code> and
<code>minus</code>.  However, not all combinations are valid, since the
assembler cannot do arbitrary arithmetic on relocatable symbols.
</p>
<p><var>m</var> should be <code>Pmode</code>.
</p>
<a name="index-high"></a>
</dd>
<dt><code>(high:<var>m</var> <var>exp</var>)</code></dt>
<dd><p>Represents the high-order bits of <var>exp</var>, usually a
<code>symbol_ref</code>.  The number of bits is machine-dependent and is
normally the number of bits specified in an instruction that initializes
the high order bits of a register.  It is used with <code>lo_sum</code> to
represent the typical two-instruction sequence used in RISC machines to
reference a global memory location.
</p>
<p><var>m</var> should be <code>Pmode</code>.
</p></dd>
</dl>

<a name="index-CONST0_005fRTX"></a>
<a name="index-CONST1_005fRTX"></a>
<a name="index-CONST2_005fRTX"></a>
<p>The macro <code>CONST0_RTX (<var>mode</var>)</code> refers to an expression with
value 0 in mode <var>mode</var>.  If mode <var>mode</var> is of mode class
<code>MODE_INT</code>, it returns <code>const0_rtx</code>.  If mode <var>mode</var> is of
mode class <code>MODE_FLOAT</code>, it returns a <code>CONST_DOUBLE</code>
expression in mode <var>mode</var>.  Otherwise, it returns a
<code>CONST_VECTOR</code> expression in mode <var>mode</var>.  Similarly, the macro
<code>CONST1_RTX (<var>mode</var>)</code> refers to an expression with value 1 in
mode <var>mode</var> and similarly for <code>CONST2_RTX</code>.  The
<code>CONST1_RTX</code> and <code>CONST2_RTX</code> macros are undefined
for vector modes.
</p>
<hr>
<a name="Regs-and-Memory"></a>
<div class="header">
<p>
Next: <a href="#Arithmetic" accesskey="n" rel="next">Arithmetic</a>, Previous: <a href="#Constants" accesskey="p" rel="prev">Constants</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Registers-and-Memory"></a>
<h3 class="section">10.8 Registers and Memory</h3>
<a name="index-RTL-register-expressions"></a>
<a name="index-RTL-memory-expressions"></a>

<p>Here are the RTL expression types for describing access to machine
registers and to main memory.
</p>
<dl compact="compact">
<dd><a name="index-reg"></a>
<a name="index-hard-registers"></a>
<a name="index-pseudo-registers"></a>
</dd>
<dt><code>(reg:<var>m</var> <var>n</var>)</code></dt>
<dd><p>For small values of the integer <var>n</var> (those that are less than
<code>FIRST_PSEUDO_REGISTER</code>), this stands for a reference to machine
register number <var>n</var>: a <em>hard register</em>.  For larger values of
<var>n</var>, it stands for a temporary value or <em>pseudo register</em>.
The compiler&rsquo;s strategy is to generate code assuming an unlimited
number of such pseudo registers, and later convert them into hard
registers or into memory references.
</p>
<p><var>m</var> is the machine mode of the reference.  It is necessary because
machines can generally refer to each register in more than one mode.
For example, a register may contain a full word but there may be
instructions to refer to it as a half word or as a single byte, as
well as instructions to refer to it as a floating point number of
various precisions.
</p>
<p>Even for a register that the machine can access in only one mode,
the mode must always be specified.
</p>
<p>The symbol <code>FIRST_PSEUDO_REGISTER</code> is defined by the machine
description, since the number of hard registers on the machine is an
invariant characteristic of the machine.  Note, however, that not
all of the machine registers must be general registers.  All the
machine registers that can be used for storage of data are given
hard register numbers, even those that can be used only in certain
instructions or can hold only certain types of data.
</p>
<p>A hard register may be accessed in various modes throughout one
function, but each pseudo register is given a natural mode
and is accessed only in that mode.  When it is necessary to describe
an access to a pseudo register using a nonnatural mode, a <code>subreg</code>
expression is used.
</p>
<p>A <code>reg</code> expression with a machine mode that specifies more than
one word of data may actually stand for several consecutive registers.
If in addition the register number specifies a hardware register, then
it actually represents several consecutive hardware registers starting
with the specified one.
</p>
<p>Each pseudo register number used in a function&rsquo;s RTL code is
represented by a unique <code>reg</code> expression.
</p>
<a name="index-FIRST_005fVIRTUAL_005fREGISTER"></a>
<a name="index-LAST_005fVIRTUAL_005fREGISTER"></a>
<p>Some pseudo register numbers, those within the range of
<code>FIRST_VIRTUAL_REGISTER</code> to <code>LAST_VIRTUAL_REGISTER</code> only
appear during the RTL generation phase and are eliminated before the
optimization phases.  These represent locations in the stack frame that
cannot be determined until RTL generation for the function has been
completed.  The following virtual register numbers are defined:
</p>
<dl compact="compact">
<dd><a name="index-VIRTUAL_005fINCOMING_005fARGS_005fREGNUM"></a>
</dd>
<dt><code>VIRTUAL_INCOMING_ARGS_REGNUM</code></dt>
<dd><p>This points to the first word of the incoming arguments passed on the
stack.  Normally these arguments are placed there by the caller, but the
callee may have pushed some arguments that were previously passed in
registers.
</p>
<a name="index-FIRST_005fPARM_005fOFFSET-and-virtual-registers"></a>
<a name="index-ARG_005fPOINTER_005fREGNUM-and-virtual-registers"></a>
<p>When RTL generation is complete, this virtual register is replaced
by the sum of the register given by <code>ARG_POINTER_REGNUM</code> and the
value of <code>FIRST_PARM_OFFSET</code>.
</p>
<a name="index-VIRTUAL_005fSTACK_005fVARS_005fREGNUM"></a>
<a name="index-FRAME_005fGROWS_005fDOWNWARD-and-virtual-registers"></a>
</dd>
<dt><code>VIRTUAL_STACK_VARS_REGNUM</code></dt>
<dd><p>If <code>FRAME_GROWS_DOWNWARD</code> is defined to a nonzero value, this points
to immediately above the first variable on the stack.  Otherwise, it points
to the first variable on the stack.
</p>
<a name="index-STARTING_005fFRAME_005fOFFSET-and-virtual-registers"></a>
<a name="index-FRAME_005fPOINTER_005fREGNUM-and-virtual-registers"></a>
<p><code>VIRTUAL_STACK_VARS_REGNUM</code> is replaced with the sum of the
register given by <code>FRAME_POINTER_REGNUM</code> and the value
<code>STARTING_FRAME_OFFSET</code>.
</p>
<a name="index-VIRTUAL_005fSTACK_005fDYNAMIC_005fREGNUM"></a>
</dd>
<dt><code>VIRTUAL_STACK_DYNAMIC_REGNUM</code></dt>
<dd><p>This points to the location of dynamically allocated memory on the stack
immediately after the stack pointer has been adjusted by the amount of
memory desired.
</p>
<a name="index-STACK_005fDYNAMIC_005fOFFSET-and-virtual-registers"></a>
<a name="index-STACK_005fPOINTER_005fREGNUM-and-virtual-registers"></a>
<p>This virtual register is replaced by the sum of the register given by
<code>STACK_POINTER_REGNUM</code> and the value <code>STACK_DYNAMIC_OFFSET</code>.
</p>
<a name="index-VIRTUAL_005fOUTGOING_005fARGS_005fREGNUM"></a>
</dd>
<dt><code>VIRTUAL_OUTGOING_ARGS_REGNUM</code></dt>
<dd><p>This points to the location in the stack at which outgoing arguments
should be written when the stack is pre-pushed (arguments pushed using
push insns should always use <code>STACK_POINTER_REGNUM</code>).
</p>
<a name="index-STACK_005fPOINTER_005fOFFSET-and-virtual-registers"></a>
<p>This virtual register is replaced by the sum of the register given by
<code>STACK_POINTER_REGNUM</code> and the value <code>STACK_POINTER_OFFSET</code>.
</p></dd>
</dl>

<a name="index-subreg"></a>
</dd>
<dt><code>(subreg:<var>m1</var> <var>reg:m2</var> <var>bytenum</var>)</code></dt>
<dd>
<p><code>subreg</code> expressions are used to refer to a register in a machine
mode other than its natural one, or to refer to one register of
a multi-part <code>reg</code> that actually refers to several registers.
</p>
<p>Each pseudo register has a natural mode.  If it is necessary to
operate on it in a different mode, the register must be
enclosed in a <code>subreg</code>.
</p>
<p>There are currently three supported types for the first operand of a
<code>subreg</code>:
</p><ul>
<li> pseudo registers
This is the most common case.  Most <code>subreg</code>s have pseudo
<code>reg</code>s as their first operand.

</li><li> mem
<code>subreg</code>s of <code>mem</code> were common in earlier versions of GCC and
are still supported.  During the reload pass these are replaced by plain
<code>mem</code>s.  On machines that do not do instruction scheduling, use of
<code>subreg</code>s of <code>mem</code> are still used, but this is no longer
recommended.  Such <code>subreg</code>s are considered to be
<code>register_operand</code>s rather than <code>memory_operand</code>s before and
during reload.  Because of this, the scheduling passes cannot properly
schedule instructions with <code>subreg</code>s of <code>mem</code>, so for machines
that do scheduling, <code>subreg</code>s of <code>mem</code> should never be used.
To support this, the combine and recog passes have explicit code to
inhibit the creation of <code>subreg</code>s of <code>mem</code> when
<code>INSN_SCHEDULING</code> is defined.

<p>The use of <code>subreg</code>s of <code>mem</code> after the reload pass is an area
that is not well understood and should be avoided.  There is still some
code in the compiler to support this, but this code has possibly rotted.
This use of <code>subreg</code>s is discouraged and will most likely not be
supported in the future.
</p>
</li><li> hard registers
It is seldom necessary to wrap hard registers in <code>subreg</code>s; such
registers would normally reduce to a single <code>reg</code> rtx.  This use of
<code>subreg</code>s is discouraged and may not be supported in the future.

</li></ul>

<p><code>subreg</code>s of <code>subreg</code>s are not supported.  Using
<code>simplify_gen_subreg</code> is the recommended way to avoid this problem.
</p>
<p><code>subreg</code>s come in two distinct flavors, each having its own
usage and rules:
</p>
<dl compact="compact">
<dt>Paradoxical subregs</dt>
<dd><p>When <var>m1</var> is strictly wider than <var>m2</var>, the <code>subreg</code>
expression is called <em>paradoxical</em>.  The canonical test for this
class of <code>subreg</code> is:
</p>
<div class="smallexample">
<pre class="smallexample">GET_MODE_SIZE (<var>m1</var>) &gt; GET_MODE_SIZE (<var>m2</var>)
</pre></div>

<p>Paradoxical <code>subreg</code>s can be used as both lvalues and rvalues.
When used as an lvalue, the low-order bits of the source value
are stored in <var>reg</var> and the high-order bits are discarded.
When used as an rvalue, the low-order bits of the <code>subreg</code> are
taken from <var>reg</var> while the high-order bits may or may not be
defined.
</p>
<p>The high-order bits of rvalues are in the following circumstances:
</p>
<ul>
<li> <code>subreg</code>s of <code>mem</code>
When <var>m2</var> is smaller than a word, the macro <code>LOAD_EXTEND_OP</code>,
can control how the high-order bits are defined.

</li><li> <code>subreg</code> of <code>reg</code>s
The upper bits are defined when <code>SUBREG_PROMOTED_VAR_P</code> is true.
<code>SUBREG_PROMOTED_UNSIGNED_P</code> describes what the upper bits hold.
Such subregs usually represent local variables, register variables
and parameter pseudo variables that have been promoted to a wider mode.

</li></ul>

<p><var>bytenum</var> is always zero for a paradoxical <code>subreg</code>, even on
big-endian targets.
</p>
<p>For example, the paradoxical <code>subreg</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(set (subreg:SI (reg:HI <var>x</var>) 0) <var>y</var>)
</pre></div>

<p>stores the lower 2 bytes of <var>y</var> in <var>x</var> and discards the upper
2 bytes.  A subsequent:
</p>
<div class="smallexample">
<pre class="smallexample">(set <var>z</var> (subreg:SI (reg:HI <var>x</var>) 0))
</pre></div>

<p>would set the lower two bytes of <var>z</var> to <var>y</var> and set the upper
two bytes to an unknown value assuming <code>SUBREG_PROMOTED_VAR_P</code> is
false.
</p>
</dd>
<dt>Normal subregs</dt>
<dd><p>When <var>m1</var> is at least as narrow as <var>m2</var> the <code>subreg</code>
expression is called <em>normal</em>.
</p>
<p>Normal <code>subreg</code>s restrict consideration to certain bits of
<var>reg</var>.  There are two cases.  If <var>m1</var> is smaller than a word,
the <code>subreg</code> refers to the least-significant part (or
<em>lowpart</em>) of one word of <var>reg</var>.  If <var>m1</var> is word-sized or
greater, the <code>subreg</code> refers to one or more complete words.
</p>
<p>When used as an lvalue, <code>subreg</code> is a word-based accessor.
Storing to a <code>subreg</code> modifies all the words of <var>reg</var> that
overlap the <code>subreg</code>, but it leaves the other words of <var>reg</var>
alone.
</p>
<p>When storing to a normal <code>subreg</code> that is smaller than a word,
the other bits of the referenced word are usually left in an undefined
state.  This laxity makes it easier to generate efficient code for
such instructions.  To represent an instruction that preserves all the
bits outside of those in the <code>subreg</code>, use <code>strict_low_part</code>
or <code>zero_extract</code> around the <code>subreg</code>.
</p>
<p><var>bytenum</var> must identify the offset of the first byte of the
<code>subreg</code> from the start of <var>reg</var>, assuming that <var>reg</var> is
laid out in memory order.  The memory order of bytes is defined by
two target macros, <code>WORDS_BIG_ENDIAN</code> and <code>BYTES_BIG_ENDIAN</code>:
</p>
<ul>
<li> <a name="index-WORDS_005fBIG_005fENDIAN_002c-effect-on-subreg"></a>
<code>WORDS_BIG_ENDIAN</code>, if set to 1, says that byte number zero is
part of the most significant word; otherwise, it is part of the least
significant word.

</li><li> <a name="index-BYTES_005fBIG_005fENDIAN_002c-effect-on-subreg"></a>
<code>BYTES_BIG_ENDIAN</code>, if set to 1, says that byte number zero is
the most significant byte within a word; otherwise, it is the least
significant byte within a word.
</li></ul>

<a name="index-FLOAT_005fWORDS_005fBIG_005fENDIAN_002c-_0028lack-of_0029-effect-on-subreg"></a>
<p>On a few targets, <code>FLOAT_WORDS_BIG_ENDIAN</code> disagrees with
<code>WORDS_BIG_ENDIAN</code>.  However, most parts of the compiler treat
floating point values as if they had the same endianness as integer
values.  This works because they handle them solely as a collection of
integer values, with no particular numerical value.  Only real.c and
the runtime libraries care about <code>FLOAT_WORDS_BIG_ENDIAN</code>.
</p>
<p>Thus,
</p>
<div class="smallexample">
<pre class="smallexample">(subreg:HI (reg:SI <var>x</var>) 2)
</pre></div>

<p>on a <code>BYTES_BIG_ENDIAN</code>, &lsquo;<samp>UNITS_PER_WORD == 4</samp>&rsquo; target is the same as
</p>
<div class="smallexample">
<pre class="smallexample">(subreg:HI (reg:SI <var>x</var>) 0)
</pre></div>

<p>on a little-endian, &lsquo;<samp>UNITS_PER_WORD == 4</samp>&rsquo; target.  Both
<code>subreg</code>s access the lower two bytes of register <var>x</var>.
</p>
</dd>
</dl>

<p>A <code>MODE_PARTIAL_INT</code> mode behaves as if it were as wide as the
corresponding <code>MODE_INT</code> mode, except that it has an unknown
number of undefined bits.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">(subreg:PSI (reg:SI 0) 0)
</pre></div>

<p>accesses the whole of &lsquo;<samp>(reg:SI 0)</samp>&rsquo;, but the exact relationship
between the <code>PSImode</code> value and the <code>SImode</code> value is not
defined.  If we assume &lsquo;<samp>UNITS_PER_WORD &lt;= 4</samp>&rsquo;, then the following
two <code>subreg</code>s:
</p>
<div class="smallexample">
<pre class="smallexample">(subreg:PSI (reg:DI 0) 0)
(subreg:PSI (reg:DI 0) 4)
</pre></div>

<p>represent independent 4-byte accesses to the two halves of
&lsquo;<samp>(reg:DI 0)</samp>&rsquo;.  Both <code>subreg</code>s have an unknown number
of undefined bits.
</p>
<p>If &lsquo;<samp>UNITS_PER_WORD &lt;= 2</samp>&rsquo; then these two <code>subreg</code>s:
</p>
<div class="smallexample">
<pre class="smallexample">(subreg:HI (reg:PSI 0) 0)
(subreg:HI (reg:PSI 0) 2)
</pre></div>

<p>represent independent 2-byte accesses that together span the whole
of &lsquo;<samp>(reg:PSI 0)</samp>&rsquo;.  Storing to the first <code>subreg</code> does not
affect the value of the second, and vice versa.  &lsquo;<samp>(reg:PSI 0)</samp>&rsquo;
has an unknown number of undefined bits, so the assignment:
</p>
<div class="smallexample">
<pre class="smallexample">(set (subreg:HI (reg:PSI 0) 0) (reg:HI 4))
</pre></div>

<p>does not guarantee that &lsquo;<samp>(subreg:HI (reg:PSI 0) 0)</samp>&rsquo; has the
value &lsquo;<samp>(reg:HI 4)</samp>&rsquo;.
</p>
<a name="index-CANNOT_005fCHANGE_005fMODE_005fCLASS-and-subreg-semantics"></a>
<p>The rules above apply to both pseudo <var>reg</var>s and hard <var>reg</var>s.
If the semantics are not correct for particular combinations of
<var>m1</var>, <var>m2</var> and hard <var>reg</var>, the target-specific code
must ensure that those combinations are never used.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">CANNOT_CHANGE_MODE_CLASS (<var>m2</var>, <var>m1</var>, <var>class</var>)
</pre></div>

<p>must be true for every class <var>class</var> that includes <var>reg</var>.
</p>
<a name="index-SUBREG_005fREG"></a>
<a name="index-SUBREG_005fBYTE"></a>
<p>The first operand of a <code>subreg</code> expression is customarily accessed
with the <code>SUBREG_REG</code> macro and the second operand is customarily
accessed with the <code>SUBREG_BYTE</code> macro.
</p>
<p>It has been several years since a platform in which
<code>BYTES_BIG_ENDIAN</code> not equal to <code>WORDS_BIG_ENDIAN</code> has
been tested.  Anyone wishing to support such a platform in the future
may be confronted with code rot.
</p>
<a name="index-scratch"></a>
<a name="index-scratch-operands"></a>
</dd>
<dt><code>(scratch:<var>m</var>)</code></dt>
<dd><p>This represents a scratch register that will be required for the
execution of a single instruction and not used subsequently.  It is
converted into a <code>reg</code> by either the local register allocator or
the reload pass.
</p>
<p><code>scratch</code> is usually present inside a <code>clobber</code> operation
(see <a href="#Side-Effects">Side Effects</a>).
</p>
<a name="index-cc0"></a>
<a name="index-condition-code-register"></a>
</dd>
<dt><code>(cc0)</code></dt>
<dd><p>This refers to the machine&rsquo;s condition code register.  It has no
operands and may not have a machine mode.  There are two ways to use it:
</p>
<ul>
<li> To stand for a complete set of condition code flags.  This is best on
most machines, where each comparison sets the entire series of flags.

<p>With this technique, <code>(cc0)</code> may be validly used in only two
contexts: as the destination of an assignment (in test and compare
instructions) and in comparison operators comparing against zero
(<code>const_int</code> with value zero; that is to say, <code>const0_rtx</code>).
</p>
</li><li> To stand for a single flag that is the result of a single condition.
This is useful on machines that have only a single flag bit, and in
which comparison instructions must specify the condition to test.

<p>With this technique, <code>(cc0)</code> may be validly used in only two
contexts: as the destination of an assignment (in test and compare
instructions) where the source is a comparison operator, and as the
first operand of <code>if_then_else</code> (in a conditional branch).
</p></li></ul>

<a name="index-cc0_005frtx"></a>
<p>There is only one expression object of code <code>cc0</code>; it is the
value of the variable <code>cc0_rtx</code>.  Any attempt to create an
expression of code <code>cc0</code> will return <code>cc0_rtx</code>.
</p>
<p>Instructions can set the condition code implicitly.  On many machines,
nearly all instructions set the condition code based on the value that
they compute or store.  It is not necessary to record these actions
explicitly in the RTL because the machine description includes a
prescription for recognizing the instructions that do so (by means of
the macro <code>NOTICE_UPDATE_CC</code>).  See <a href="#Condition-Code">Condition Code</a>.  Only
instructions whose sole purpose is to set the condition code, and
instructions that use the condition code, need mention <code>(cc0)</code>.
</p>
<p>On some machines, the condition code register is given a register number
and a <code>reg</code> is used instead of <code>(cc0)</code>.  This is usually the
preferable approach if only a small subset of instructions modify the
condition code.  Other machines store condition codes in general
registers; in such cases a pseudo register should be used.
</p>
<p>Some machines, such as the SPARC and RS/6000, have two sets of
arithmetic instructions, one that sets and one that does not set the
condition code.  This is best handled by normally generating the
instruction that does not set the condition code, and making a pattern
that both performs the arithmetic and sets the condition code register
(which would not be <code>(cc0)</code> in this case).  For examples, search
for &lsquo;<samp>addcc</samp>&rsquo; and &lsquo;<samp>andcc</samp>&rsquo; in <samp>sparc.md</samp>.
</p>
<a name="index-pc"></a>
</dd>
<dt><code>(pc)</code></dt>
<dd><a name="index-program-counter"></a>
<p>This represents the machine&rsquo;s program counter.  It has no operands and
may not have a machine mode.  <code>(pc)</code> may be validly used only in
certain specific contexts in jump instructions.
</p>
<a name="index-pc_005frtx"></a>
<p>There is only one expression object of code <code>pc</code>; it is the value
of the variable <code>pc_rtx</code>.  Any attempt to create an expression of
code <code>pc</code> will return <code>pc_rtx</code>.
</p>
<p>All instructions that do not jump alter the program counter implicitly
by incrementing it, but there is no need to mention this in the RTL.
</p>
<a name="index-mem"></a>
</dd>
<dt><code>(mem:<var>m</var> <var>addr</var> <var>alias</var>)</code></dt>
<dd><p>This RTX represents a reference to main memory at an address
represented by the expression <var>addr</var>.  <var>m</var> specifies how large
a unit of memory is accessed.  <var>alias</var> specifies an alias set for the
reference.  In general two items are in different alias sets if they cannot
reference the same memory address.
</p>
<p>The construct <code>(mem:BLK (scratch))</code> is considered to alias all
other memories.  Thus it may be used as a memory barrier in epilogue
stack deallocation patterns.
</p>
<a name="index-concat"></a>
</dd>
<dt><code>(concat<var>m</var> <var>rtx</var> <var>rtx</var>)</code></dt>
<dd><p>This RTX represents the concatenation of two other RTXs.  This is used
for complex values.  It should only appear in the RTL attached to
declarations and during RTL generation.  It should not appear in the
ordinary insn chain.
</p>
<a name="index-concatn"></a>
</dd>
<dt><code>(concatn<var>m</var> [<var>rtx</var> &hellip;])</code></dt>
<dd><p>This RTX represents the concatenation of all the <var>rtx</var> to make a
single value.  Like <code>concat</code>, this should only appear in
declarations, and not in the insn chain.
</p></dd>
</dl>

<hr>
<a name="Arithmetic"></a>
<div class="header">
<p>
Next: <a href="#Comparisons" accesskey="n" rel="next">Comparisons</a>, Previous: <a href="#Regs-and-Memory" accesskey="p" rel="prev">Regs and Memory</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Expressions-for-Arithmetic"></a>
<h3 class="section">10.9 RTL Expressions for Arithmetic</h3>
<a name="index-arithmetic_002c-in-RTL"></a>
<a name="index-math_002c-in-RTL"></a>
<a name="index-RTL-expressions-for-arithmetic"></a>

<p>Unless otherwise specified, all the operands of arithmetic expressions
must be valid for mode <var>m</var>.  An operand is valid for mode <var>m</var>
if it has mode <var>m</var>, or if it is a <code>const_int</code> or
<code>const_double</code> and <var>m</var> is a mode of class <code>MODE_INT</code>.
</p>
<p>For commutative binary operations, constants should be placed in the
second operand.
</p>
<dl compact="compact">
<dd><a name="index-plus"></a>
<a name="index-ss_005fplus"></a>
<a name="index-us_005fplus"></a>
<a name="index-RTL-sum"></a>
<a name="index-RTL-addition"></a>
<a name="index-RTL-addition-with-signed-saturation"></a>
<a name="index-RTL-addition-with-unsigned-saturation"></a>
</dd>
<dt><code>(plus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(ss_plus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(us_plus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd>
<p>These three expressions all represent the sum of the values
represented by <var>x</var> and <var>y</var> carried out in machine mode
<var>m</var>.  They differ in their behavior on overflow of integer modes.
<code>plus</code> wraps round modulo the width of <var>m</var>; <code>ss_plus</code>
saturates at the maximum signed value representable in <var>m</var>;
<code>us_plus</code> saturates at the maximum unsigned value.
</p>

<a name="index-lo_005fsum"></a>
</dd>
<dt><code>(lo_sum:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd>
<p>This expression represents the sum of <var>x</var> and the low-order bits
of <var>y</var>.  It is used with <code>high</code> (see <a href="#Constants">Constants</a>) to
represent the typical two-instruction sequence used in RISC machines
to reference a global memory location.
</p>
<p>The number of low order bits is machine-dependent but is
normally the number of bits in a <code>Pmode</code> item minus the number of
bits set by <code>high</code>.
</p>
<p><var>m</var> should be <code>Pmode</code>.
</p>
<a name="index-minus"></a>
<a name="index-ss_005fminus"></a>
<a name="index-us_005fminus"></a>
<a name="index-RTL-difference"></a>
<a name="index-RTL-subtraction"></a>
<a name="index-RTL-subtraction-with-signed-saturation"></a>
<a name="index-RTL-subtraction-with-unsigned-saturation"></a>
</dd>
<dt><code>(minus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(ss_minus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(us_minus:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd>
<p>These three expressions represent the result of subtracting <var>y</var>
from <var>x</var>, carried out in mode <var>M</var>.  Behavior on overflow is
the same as for the three variants of <code>plus</code> (see above).
</p>
<a name="index-compare"></a>
<a name="index-RTL-comparison"></a>
</dd>
<dt><code>(compare:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the result of subtracting <var>y</var> from <var>x</var> for purposes
of comparison.  The result is computed without overflow, as if with
infinite precision.
</p>
<p>Of course, machines can&rsquo;t really subtract with infinite precision.
However, they can pretend to do so when only the sign of the result will
be used, which is the case when the result is stored in the condition
code.  And that is the <em>only</em> way this kind of expression may
validly be used: as a value to be stored in the condition codes, either
<code>(cc0)</code> or a register.  See <a href="#Comparisons">Comparisons</a>.
</p>
<p>The mode <var>m</var> is not related to the modes of <var>x</var> and <var>y</var>, but
instead is the mode of the condition code value.  If <code>(cc0)</code> is
used, it is <code>VOIDmode</code>.  Otherwise it is some mode in class
<code>MODE_CC</code>, often <code>CCmode</code>.  See <a href="#Condition-Code">Condition Code</a>.  If <var>m</var>
is <code>VOIDmode</code> or <code>CCmode</code>, the operation returns sufficient
information (in an unspecified format) so that any comparison operator
can be applied to the result of the <code>COMPARE</code> operation.  For other
modes in class <code>MODE_CC</code>, the operation only returns a subset of
this information.
</p>
<p>Normally, <var>x</var> and <var>y</var> must have the same mode.  Otherwise,
<code>compare</code> is valid only if the mode of <var>x</var> is in class
<code>MODE_INT</code> and <var>y</var> is a <code>const_int</code> or
<code>const_double</code> with mode <code>VOIDmode</code>.  The mode of <var>x</var>
determines what mode the comparison is to be done in; thus it must not
be <code>VOIDmode</code>.
</p>
<p>If one of the operands is a constant, it should be placed in the
second operand and the comparison code adjusted as appropriate.
</p>
<p>A <code>compare</code> specifying two <code>VOIDmode</code> constants is not valid
since there is no way to know in what mode the comparison is to be
performed; the comparison must either be folded during the compilation
or the first operand must be loaded into a register while its mode is
still known.
</p>
<a name="index-neg"></a>
<a name="index-ss_005fneg"></a>
<a name="index-us_005fneg"></a>
<a name="index-negation"></a>
<a name="index-negation-with-signed-saturation"></a>
<a name="index-negation-with-unsigned-saturation"></a>
</dd>
<dt><code>(neg:<var>m</var> <var>x</var>)</code></dt>
<dt><code>(ss_neg:<var>m</var> <var>x</var>)</code></dt>
<dt><code>(us_neg:<var>m</var> <var>x</var>)</code></dt>
<dd><p>These two expressions represent the negation (subtraction from zero) of
the value represented by <var>x</var>, carried out in mode <var>m</var>.  They
differ in the behavior on overflow of integer modes.  In the case of
<code>neg</code>, the negation of the operand may be a number not representable
in mode <var>m</var>, in which case it is truncated to <var>m</var>.  <code>ss_neg</code>
and <code>us_neg</code> ensure that an out-of-bounds result saturates to the
maximum or minimum signed or unsigned value.
</p>
<a name="index-mult"></a>
<a name="index-ss_005fmult"></a>
<a name="index-us_005fmult"></a>
<a name="index-multiplication"></a>
<a name="index-product"></a>
<a name="index-multiplication-with-signed-saturation"></a>
<a name="index-multiplication-with-unsigned-saturation"></a>
</dd>
<dt><code>(mult:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(ss_mult:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(us_mult:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the signed product of the values represented by <var>x</var> and
<var>y</var> carried out in machine mode <var>m</var>.
<code>ss_mult</code> and <code>us_mult</code> ensure that an out-of-bounds result
saturates to the maximum or minimum signed or unsigned value.
</p>
<p>Some machines support a multiplication that generates a product wider
than the operands.  Write the pattern for this as
</p>
<div class="smallexample">
<pre class="smallexample">(mult:<var>m</var> (sign_extend:<var>m</var> <var>x</var>) (sign_extend:<var>m</var> <var>y</var>))
</pre></div>

<p>where <var>m</var> is wider than the modes of <var>x</var> and <var>y</var>, which need
not be the same.
</p>
<p>For unsigned widening multiplication, use the same idiom, but with
<code>zero_extend</code> instead of <code>sign_extend</code>.
</p>
<a name="index-fma"></a>
</dd>
<dt><code>(fma:<var>m</var> <var>x</var> <var>y</var> <var>z</var>)</code></dt>
<dd><p>Represents the <code>fma</code>, <code>fmaf</code>, and <code>fmal</code> builtin
functions that do a combined multiply of <var>x</var> and <var>y</var> and then
adding to<var>z</var> without doing an intermediate rounding step.
</p>
<a name="index-div"></a>
<a name="index-ss_005fdiv"></a>
<a name="index-division"></a>
<a name="index-signed-division"></a>
<a name="index-signed-division-with-signed-saturation"></a>
<a name="index-quotient"></a>
</dd>
<dt><code>(div:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(ss_div:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the quotient in signed division of <var>x</var> by <var>y</var>,
carried out in machine mode <var>m</var>.  If <var>m</var> is a floating point
mode, it represents the exact quotient; otherwise, the integerized
quotient.
<code>ss_div</code> ensures that an out-of-bounds result saturates to the maximum
or minimum signed value.
</p>
<p>Some machines have division instructions in which the operands and
quotient widths are not all the same; you should represent
such instructions using <code>truncate</code> and <code>sign_extend</code> as in,
</p>
<div class="smallexample">
<pre class="smallexample">(truncate:<var>m1</var> (div:<var>m2</var> <var>x</var> (sign_extend:<var>m2</var> <var>y</var>)))
</pre></div>

<a name="index-udiv"></a>
<a name="index-unsigned-division"></a>
<a name="index-unsigned-division-with-unsigned-saturation"></a>
<a name="index-division-1"></a>
</dd>
<dt><code>(udiv:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(us_div:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>div</code> but represents unsigned division.
<code>us_div</code> ensures that an out-of-bounds result saturates to the maximum
or minimum unsigned value.
</p>
<a name="index-mod"></a>
<a name="index-umod"></a>
<a name="index-remainder"></a>
<a name="index-division-2"></a>
</dd>
<dt><code>(mod:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(umod:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>div</code> and <code>udiv</code> but represent the remainder instead of
the quotient.
</p>
<a name="index-smin"></a>
<a name="index-smax"></a>
<a name="index-signed-minimum"></a>
<a name="index-signed-maximum"></a>
</dd>
<dt><code>(smin:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(smax:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the smaller (for <code>smin</code>) or larger (for <code>smax</code>) of
<var>x</var> and <var>y</var>, interpreted as signed values in mode <var>m</var>.
When used with floating point, if both operands are zeros, or if either
operand is <code>NaN</code>, then it is unspecified which of the two operands
is returned as the result.
</p>
<a name="index-umin"></a>
<a name="index-umax"></a>
<a name="index-unsigned-minimum-and-maximum"></a>
</dd>
<dt><code>(umin:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(umax:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>smin</code> and <code>smax</code>, but the values are interpreted as unsigned
integers.
</p>
<a name="index-not"></a>
<a name="index-complement_002c-bitwise"></a>
<a name="index-bitwise-complement"></a>
</dd>
<dt><code>(not:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the bitwise complement of the value represented by <var>x</var>,
carried out in mode <var>m</var>, which must be a fixed-point machine mode.
</p>
<a name="index-and"></a>
<a name="index-logical_002dand_002c-bitwise"></a>
<a name="index-bitwise-logical_002dand"></a>
</dd>
<dt><code>(and:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the bitwise logical-and of the values represented by
<var>x</var> and <var>y</var>, carried out in machine mode <var>m</var>, which must be
a fixed-point machine mode.
</p>
<a name="index-ior"></a>
<a name="index-inclusive_002dor_002c-bitwise"></a>
<a name="index-bitwise-inclusive_002dor"></a>
</dd>
<dt><code>(ior:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the bitwise inclusive-or of the values represented by <var>x</var>
and <var>y</var>, carried out in machine mode <var>m</var>, which must be a
fixed-point mode.
</p>
<a name="index-xor"></a>
<a name="index-exclusive_002dor_002c-bitwise"></a>
<a name="index-bitwise-exclusive_002dor"></a>
</dd>
<dt><code>(xor:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Represents the bitwise exclusive-or of the values represented by <var>x</var>
and <var>y</var>, carried out in machine mode <var>m</var>, which must be a
fixed-point mode.
</p>
<a name="index-ashift"></a>
<a name="index-ss_005fashift"></a>
<a name="index-us_005fashift"></a>
<a name="index-left-shift"></a>
<a name="index-shift"></a>
<a name="index-arithmetic-shift"></a>
<a name="index-arithmetic-shift-with-signed-saturation"></a>
<a name="index-arithmetic-shift-with-unsigned-saturation"></a>
</dd>
<dt><code>(ashift:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dt><code>(ss_ashift:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dt><code>(us_ashift:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dd><p>These three expressions represent the result of arithmetically shifting <var>x</var>
left by <var>c</var> places.  They differ in their behavior on overflow of integer
modes.  An <code>ashift</code> operation is a plain shift with no special behavior
in case of a change in the sign bit; <code>ss_ashift</code> and <code>us_ashift</code>
saturates to the minimum or maximum representable value if any of the bits
shifted out differs from the final sign bit.
</p>
<p><var>x</var> have mode <var>m</var>, a fixed-point machine mode.  <var>c</var>
be a fixed-point mode or be a constant with mode <code>VOIDmode</code>; which
mode is determined by the mode called for in the machine description
entry for the left-shift instruction.  For example, on the VAX, the mode
of <var>c</var> is <code>QImode</code> regardless of <var>m</var>.
</p>
<a name="index-lshiftrt"></a>
<a name="index-right-shift"></a>
<a name="index-ashiftrt"></a>
</dd>
<dt><code>(lshiftrt:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dt><code>(ashiftrt:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dd><p>Like <code>ashift</code> but for right shift.  Unlike the case for left shift,
these two operations are distinct.
</p>
<a name="index-rotate-1"></a>
<a name="index-rotate"></a>
<a name="index-left-rotate"></a>
<a name="index-rotatert"></a>
<a name="index-right-rotate"></a>
</dd>
<dt><code>(rotate:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dt><code>(rotatert:<var>m</var> <var>x</var> <var>c</var>)</code></dt>
<dd><p>Similar but represent left and right rotate.  If <var>c</var> is a constant,
use <code>rotate</code>.
</p>
<a name="index-abs"></a>
<a name="index-ss_005fabs"></a>
<a name="index-absolute-value"></a>
</dd>
<dt><code>(abs:<var>m</var> <var>x</var>)</code></dt>
<dt><code>(ss_abs:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the absolute value of <var>x</var>, computed in mode <var>m</var>.
<code>ss_abs</code> ensures that an out-of-bounds result saturates to the
maximum signed value.
</p>

<a name="index-sqrt"></a>
<a name="index-square-root"></a>
</dd>
<dt><code>(sqrt:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the square root of <var>x</var>, computed in mode <var>m</var>.
Most often <var>m</var> will be a floating point mode.
</p>
<a name="index-ffs"></a>
</dd>
<dt><code>(ffs:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents one plus the index of the least significant 1-bit in
<var>x</var>, represented as an integer of mode <var>m</var>.  (The value is
zero if <var>x</var> is zero.)  The mode of <var>x</var> must be <var>m</var>
or <code>VOIDmode</code>.
</p>
<a name="index-clrsb"></a>
</dd>
<dt><code>(clrsb:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the number of redundant leading sign bits in <var>x</var>,
represented as an integer of mode <var>m</var>, starting at the most
significant bit position.  This is one less than the number of leading
sign bits (either 0 or 1), with no special cases.  The mode of <var>x</var>
must be <var>m</var> or <code>VOIDmode</code>.
</p>
<a name="index-clz"></a>
</dd>
<dt><code>(clz:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the number of leading 0-bits in <var>x</var>, represented as an
integer of mode <var>m</var>, starting at the most significant bit position.
If <var>x</var> is zero, the value is determined by
<code>CLZ_DEFINED_VALUE_AT_ZERO</code> (see <a href="#Misc">Misc</a>).  Note that this is one of
the few expressions that is not invariant under widening.  The mode of
<var>x</var> must be <var>m</var> or <code>VOIDmode</code>.
</p>
<a name="index-ctz"></a>
</dd>
<dt><code>(ctz:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the number of trailing 0-bits in <var>x</var>, represented as an
integer of mode <var>m</var>, starting at the least significant bit position.
If <var>x</var> is zero, the value is determined by
<code>CTZ_DEFINED_VALUE_AT_ZERO</code> (see <a href="#Misc">Misc</a>).  Except for this case,
<code>ctz(x)</code> is equivalent to <code>ffs(<var>x</var>) - 1</code>.  The mode of
<var>x</var> must be <var>m</var> or <code>VOIDmode</code>.
</p>
<a name="index-popcount"></a>
</dd>
<dt><code>(popcount:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the number of 1-bits in <var>x</var>, represented as an integer of
mode <var>m</var>.  The mode of <var>x</var> must be <var>m</var> or <code>VOIDmode</code>.
</p>
<a name="index-parity"></a>
</dd>
<dt><code>(parity:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the number of 1-bits modulo 2 in <var>x</var>, represented as an
integer of mode <var>m</var>.  The mode of <var>x</var> must be <var>m</var> or
<code>VOIDmode</code>.
</p>
<a name="index-bswap"></a>
</dd>
<dt><code>(bswap:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the value <var>x</var> with the order of bytes reversed, carried out
in mode <var>m</var>, which must be a fixed-point machine mode.
The mode of <var>x</var> must be <var>m</var> or <code>VOIDmode</code>.
</p></dd>
</dl>

<hr>
<a name="Comparisons"></a>
<div class="header">
<p>
Next: <a href="#Bit_002dFields" accesskey="n" rel="next">Bit-Fields</a>, Previous: <a href="#Arithmetic" accesskey="p" rel="prev">Arithmetic</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Comparison-Operations"></a>
<h3 class="section">10.10 Comparison Operations</h3>
<a name="index-RTL-comparison-operations"></a>

<p>Comparison operators test a relation on two operands and are considered
to represent a machine-dependent nonzero value described by, but not
necessarily equal to, <code>STORE_FLAG_VALUE</code> (see <a href="#Misc">Misc</a>)
if the relation holds, or zero if it does not, for comparison operators
whose results have a &lsquo;MODE_INT&rsquo; mode,
<code>FLOAT_STORE_FLAG_VALUE</code> (see <a href="#Misc">Misc</a>) if the relation holds, or
zero if it does not, for comparison operators that return floating-point
values, and a vector of either <code>VECTOR_STORE_FLAG_VALUE</code> (see <a href="#Misc">Misc</a>)
if the relation holds, or of zeros if it does not, for comparison operators
that return vector results.
The mode of the comparison operation is independent of the mode
of the data being compared.  If the comparison operation is being tested
(e.g., the first operand of an <code>if_then_else</code>), the mode must be
<code>VOIDmode</code>.
</p>
<a name="index-condition-codes"></a>
<p>There are two ways that comparison operations may be used.  The
comparison operators may be used to compare the condition codes
<code>(cc0)</code> against zero, as in <code>(eq (cc0) (const_int 0))</code>.  Such
a construct actually refers to the result of the preceding instruction
in which the condition codes were set.  The instruction setting the
condition code must be adjacent to the instruction using the condition
code; only <code>note</code> insns may separate them.
</p>
<p>Alternatively, a comparison operation may directly compare two data
objects.  The mode of the comparison is determined by the operands; they
must both be valid for a common machine mode.  A comparison with both
operands constant would be invalid as the machine mode could not be
deduced from it, but such a comparison should never exist in RTL due to
constant folding.
</p>
<p>In the example above, if <code>(cc0)</code> were last set to
<code>(compare <var>x</var> <var>y</var>)</code>, the comparison operation is
identical to <code>(eq <var>x</var> <var>y</var>)</code>.  Usually only one style
of comparisons is supported on a particular machine, but the combine
pass will try to merge the operations to produce the <code>eq</code> shown
in case it exists in the context of the particular insn involved.
</p>
<p>Inequality comparisons come in two flavors, signed and unsigned.  Thus,
there are distinct expression codes <code>gt</code> and <code>gtu</code> for signed and
unsigned greater-than.  These can produce different results for the same
pair of integer values: for example, 1 is signed greater-than -1 but not
unsigned greater-than, because -1 when regarded as unsigned is actually
<code>0xffffffff</code> which is greater than 1.
</p>
<p>The signed comparisons are also used for floating point values.  Floating
point comparisons are distinguished by the machine modes of the operands.
</p>
<dl compact="compact">
<dd><a name="index-eq"></a>
<a name="index-equal"></a>
</dd>
<dt><code>(eq:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p><code>STORE_FLAG_VALUE</code> if the values represented by <var>x</var> and <var>y</var>
are equal, otherwise 0.
</p>
<a name="index-ne"></a>
<a name="index-not-equal"></a>
</dd>
<dt><code>(ne:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p><code>STORE_FLAG_VALUE</code> if the values represented by <var>x</var> and <var>y</var>
are not equal, otherwise 0.
</p>
<a name="index-gt"></a>
<a name="index-greater-than"></a>
</dd>
<dt><code>(gt:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p><code>STORE_FLAG_VALUE</code> if the <var>x</var> is greater than <var>y</var>.  If they
are fixed-point, the comparison is done in a signed sense.
</p>
<a name="index-gtu"></a>
<a name="index-greater-than-1"></a>
<a name="index-unsigned-greater-than"></a>
</dd>
<dt><code>(gtu:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>gt</code> but does unsigned comparison, on fixed-point numbers only.
</p>
<a name="index-lt"></a>
<a name="index-less-than"></a>
<a name="index-ltu"></a>
<a name="index-unsigned-less-than"></a>
</dd>
<dt><code>(lt:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(ltu:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>gt</code> and <code>gtu</code> but test for &ldquo;less than&rdquo;.
</p>
<a name="index-ge"></a>
<a name="index-greater-than-2"></a>
<a name="index-geu"></a>
<a name="index-unsigned-greater-than-1"></a>
</dd>
<dt><code>(ge:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(geu:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>gt</code> and <code>gtu</code> but test for &ldquo;greater than or equal&rdquo;.
</p>
<a name="index-le"></a>
<a name="index-less-than-or-equal"></a>
<a name="index-leu"></a>
<a name="index-unsigned-less-than-1"></a>
</dd>
<dt><code>(le:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dt><code>(leu:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd><p>Like <code>gt</code> and <code>gtu</code> but test for &ldquo;less than or equal&rdquo;.
</p>
<a name="index-if_005fthen_005felse"></a>
</dd>
<dt><code>(if_then_else <var>cond</var> <var>then</var> <var>else</var>)</code></dt>
<dd><p>This is not a comparison operation but is listed here because it is
always used in conjunction with a comparison operation.  To be
precise, <var>cond</var> is a comparison expression.  This expression
represents a choice, according to <var>cond</var>, between the value
represented by <var>then</var> and the one represented by <var>else</var>.
</p>
<p>On most machines, <code>if_then_else</code> expressions are valid only
to express conditional jumps.
</p>
<a name="index-cond"></a>
</dd>
<dt><code>(cond [<var>test1</var> <var>value1</var> <var>test2</var> <var>value2</var> &hellip;] <var>default</var>)</code></dt>
<dd><p>Similar to <code>if_then_else</code>, but more general.  Each of <var>test1</var>,
<var>test2</var>, &hellip; is performed in turn.  The result of this expression is
the <var>value</var> corresponding to the first nonzero test, or <var>default</var> if
none of the tests are nonzero expressions.
</p>
<p>This is currently not valid for instruction patterns and is supported only
for insn attributes.  See <a href="#Insn-Attributes">Insn Attributes</a>.
</p></dd>
</dl>

<hr>
<a name="Bit_002dFields"></a>
<div class="header">
<p>
Next: <a href="#Vector-Operations" accesskey="n" rel="next">Vector Operations</a>, Previous: <a href="#Comparisons" accesskey="p" rel="prev">Comparisons</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Bit_002dFields-1"></a>
<h3 class="section">10.11 Bit-Fields</h3>
<a name="index-bit_002dfields"></a>

<p>Special expression codes exist to represent bit-field instructions.
</p>
<dl compact="compact">
<dd><a name="index-sign_005fextract"></a>
<a name="index-BITS_005fBIG_005fENDIAN_002c-effect-on-sign_005fextract"></a>
</dd>
<dt><code>(sign_extract:<var>m</var> <var>loc</var> <var>size</var> <var>pos</var>)</code></dt>
<dd><p>This represents a reference to a sign-extended bit-field contained or
starting in <var>loc</var> (a memory or register reference).  The bit-field
is <var>size</var> bits wide and starts at bit <var>pos</var>.  The compilation
option <code>BITS_BIG_ENDIAN</code> says which end of the memory unit
<var>pos</var> counts from.
</p>
<p>If <var>loc</var> is in memory, its mode must be a single-byte integer mode.
If <var>loc</var> is in a register, the mode to use is specified by the
operand of the <code>insv</code> or <code>extv</code> pattern
(see <a href="#Standard-Names">Standard Names</a>) and is usually a full-word integer mode,
which is the default if none is specified.
</p>
<p>The mode of <var>pos</var> is machine-specific and is also specified
in the <code>insv</code> or <code>extv</code> pattern.
</p>
<p>The mode <var>m</var> is the same as the mode that would be used for
<var>loc</var> if it were a register.
</p>
<p>A <code>sign_extract</code> can not appear as an lvalue, or part thereof,
in RTL.
</p>
<a name="index-zero_005fextract"></a>
</dd>
<dt><code>(zero_extract:<var>m</var> <var>loc</var> <var>size</var> <var>pos</var>)</code></dt>
<dd><p>Like <code>sign_extract</code> but refers to an unsigned or zero-extended
bit-field.  The same sequence of bits are extracted, but they
are filled to an entire word with zeros instead of by sign-extension.
</p>
<p>Unlike <code>sign_extract</code>, this type of expressions can be lvalues
in RTL; they may appear on the left side of an assignment, indicating
insertion of a value into the specified bit-field.
</p></dd>
</dl>

<hr>
<a name="Vector-Operations"></a>
<div class="header">
<p>
Next: <a href="#Conversions" accesskey="n" rel="next">Conversions</a>, Previous: <a href="#Bit_002dFields" accesskey="p" rel="prev">Bit-Fields</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Vector-Operations-1"></a>
<h3 class="section">10.12 Vector Operations</h3>
<a name="index-vector-operations"></a>

<p>All normal RTL expressions can be used with vector modes; they are
interpreted as operating on each part of the vector independently.
Additionally, there are a few new expressions to describe specific vector
operations.
</p>
<dl compact="compact">
<dd><a name="index-vec_005fmerge"></a>
</dd>
<dt><code>(vec_merge:<var>m</var> <var>vec1</var> <var>vec2</var> <var>items</var>)</code></dt>
<dd><p>This describes a merge operation between two vectors.  The result is a vector
of mode <var>m</var>; its elements are selected from either <var>vec1</var> or
<var>vec2</var>.  Which elements are selected is described by <var>items</var>, which
is a bit mask represented by a <code>const_int</code>; a zero bit indicates the
corresponding element in the result vector is taken from <var>vec2</var> while
a set bit indicates it is taken from <var>vec1</var>.
</p>
<a name="index-vec_005fselect"></a>
</dd>
<dt><code>(vec_select:<var>m</var> <var>vec1</var> <var>selection</var>)</code></dt>
<dd><p>This describes an operation that selects parts of a vector.  <var>vec1</var> is
the source vector, and <var>selection</var> is a <code>parallel</code> that contains a
<code>const_int</code> for each of the subparts of the result vector, giving the
number of the source subpart that should be stored into it.
The result mode <var>m</var> is either the submode for a single element of
<var>vec1</var> (if only one subpart is selected), or another vector mode
with that element submode (if multiple subparts are selected).
</p>
<a name="index-vec_005fconcat"></a>
</dd>
<dt><code>(vec_concat:<var>m</var> <var>x1</var> <var>x2</var>)</code></dt>
<dd><p>Describes a vector concat operation.  The result is a concatenation of the
vectors or scalars <var>x1</var> and <var>x2</var>; its length is the sum of the
lengths of the two inputs.
</p>
<a name="index-vec_005fduplicate"></a>
</dd>
<dt><code>(vec_duplicate:<var>m</var> <var>x</var>)</code></dt>
<dd><p>This operation converts a scalar into a vector or a small vector into a
larger one by duplicating the input values.  The output vector mode must have
the same submodes as the input vector mode or the scalar modes, and the
number of output parts must be an integer multiple of the number of input
parts.
</p>
</dd>
</dl>

<hr>
<a name="Conversions"></a>
<div class="header">
<p>
Next: <a href="#RTL-Declarations" accesskey="n" rel="next">RTL Declarations</a>, Previous: <a href="#Vector-Operations" accesskey="p" rel="prev">Vector Operations</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Conversions-1"></a>
<h3 class="section">10.13 Conversions</h3>
<a name="index-conversions"></a>
<a name="index-machine-mode-conversions"></a>

<p>All conversions between machine modes must be represented by
explicit conversion operations.  For example, an expression
which is the sum of a byte and a full word cannot be written as
<code>(plus:SI (reg:QI 34) (reg:SI 80))</code> because the <code>plus</code>
operation requires two operands of the same machine mode.
Therefore, the byte-sized operand is enclosed in a conversion
operation, as in
</p>
<div class="smallexample">
<pre class="smallexample">(plus:SI (sign_extend:SI (reg:QI 34)) (reg:SI 80))
</pre></div>

<p>The conversion operation is not a mere placeholder, because there
may be more than one way of converting from a given starting mode
to the desired final mode.  The conversion operation code says how
to do it.
</p>
<p>For all conversion operations, <var>x</var> must not be <code>VOIDmode</code>
because the mode in which to do the conversion would not be known.
The conversion must either be done at compile-time or <var>x</var>
must be placed into a register.
</p>
<dl compact="compact">
<dd><a name="index-sign_005fextend"></a>
</dd>
<dt><code>(sign_extend:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of sign-extending the value <var>x</var>
to machine mode <var>m</var>.  <var>m</var> must be a fixed-point mode
and <var>x</var> a fixed-point value of a mode narrower than <var>m</var>.
</p>
<a name="index-zero_005fextend"></a>
</dd>
<dt><code>(zero_extend:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of zero-extending the value <var>x</var>
to machine mode <var>m</var>.  <var>m</var> must be a fixed-point mode
and <var>x</var> a fixed-point value of a mode narrower than <var>m</var>.
</p>
<a name="index-float_005fextend"></a>
</dd>
<dt><code>(float_extend:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of extending the value <var>x</var>
to machine mode <var>m</var>.  <var>m</var> must be a floating point mode
and <var>x</var> a floating point value of a mode narrower than <var>m</var>.
</p>
<a name="index-truncate"></a>
</dd>
<dt><code>(truncate:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of truncating the value <var>x</var>
to machine mode <var>m</var>.  <var>m</var> must be a fixed-point mode
and <var>x</var> a fixed-point value of a mode wider than <var>m</var>.
</p>
<a name="index-ss_005ftruncate"></a>
</dd>
<dt><code>(ss_truncate:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of truncating the value <var>x</var>
to machine mode <var>m</var>, using signed saturation in the case of
overflow.  Both <var>m</var> and the mode of <var>x</var> must be fixed-point
modes.
</p>
<a name="index-us_005ftruncate"></a>
</dd>
<dt><code>(us_truncate:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of truncating the value <var>x</var>
to machine mode <var>m</var>, using unsigned saturation in the case of
overflow.  Both <var>m</var> and the mode of <var>x</var> must be fixed-point
modes.
</p>
<a name="index-float_005ftruncate"></a>
</dd>
<dt><code>(float_truncate:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of truncating the value <var>x</var>
to machine mode <var>m</var>.  <var>m</var> must be a floating point mode
and <var>x</var> a floating point value of a mode wider than <var>m</var>.
</p>
<a name="index-float"></a>
</dd>
<dt><code>(float:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting fixed point value <var>x</var>,
regarded as signed, to floating point mode <var>m</var>.
</p>
<a name="index-unsigned_005ffloat"></a>
</dd>
<dt><code>(unsigned_float:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting fixed point value <var>x</var>,
regarded as unsigned, to floating point mode <var>m</var>.
</p>
<a name="index-fix"></a>
</dd>
<dt><code>(fix:<var>m</var> <var>x</var>)</code></dt>
<dd><p>When <var>m</var> is a floating-point mode, represents the result of
converting floating point value <var>x</var> (valid for mode <var>m</var>) to an
integer, still represented in floating point mode <var>m</var>, by rounding
towards zero.
</p>
<p>When <var>m</var> is a fixed-point mode, represents the result of
converting floating point value <var>x</var> to mode <var>m</var>, regarded as
signed.  How rounding is done is not specified, so this operation may
be used validly in compiling C code only for integer-valued operands.
</p>
<a name="index-unsigned_005ffix"></a>
</dd>
<dt><code>(unsigned_fix:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting floating point value <var>x</var> to
fixed point mode <var>m</var>, regarded as unsigned.  How rounding is done
is not specified.
</p>
<a name="index-fract_005fconvert"></a>
</dd>
<dt><code>(fract_convert:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting fixed-point value <var>x</var> to
fixed-point mode <var>m</var>, signed integer value <var>x</var> to
fixed-point mode <var>m</var>, floating-point value <var>x</var> to
fixed-point mode <var>m</var>, fixed-point value <var>x</var> to integer mode <var>m</var>
regarded as signed, or fixed-point value <var>x</var> to floating-point mode <var>m</var>.
When overflows or underflows happen, the results are undefined.
</p>
<a name="index-sat_005ffract"></a>
</dd>
<dt><code>(sat_fract:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting fixed-point value <var>x</var> to
fixed-point mode <var>m</var>, signed integer value <var>x</var> to
fixed-point mode <var>m</var>, or floating-point value <var>x</var> to
fixed-point mode <var>m</var>.
When overflows or underflows happen, the results are saturated to the
maximum or the minimum.
</p>
<a name="index-unsigned_005ffract_005fconvert"></a>
</dd>
<dt><code>(unsigned_fract_convert:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting fixed-point value <var>x</var> to
integer mode <var>m</var> regarded as unsigned, or unsigned integer value <var>x</var> to
fixed-point mode <var>m</var>.
When overflows or underflows happen, the results are undefined.
</p>
<a name="index-unsigned_005fsat_005ffract"></a>
</dd>
<dt><code>(unsigned_sat_fract:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the result of converting unsigned integer value <var>x</var> to
fixed-point mode <var>m</var>.
When overflows or underflows happen, the results are saturated to the
maximum or the minimum.
</p></dd>
</dl>

<hr>
<a name="RTL-Declarations"></a>
<div class="header">
<p>
Next: <a href="#Side-Effects" accesskey="n" rel="next">Side Effects</a>, Previous: <a href="#Conversions" accesskey="p" rel="prev">Conversions</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Declarations-1"></a>
<h3 class="section">10.14 Declarations</h3>
<a name="index-RTL-declarations"></a>
<a name="index-declarations_002c-RTL"></a>

<p>Declaration expression codes do not represent arithmetic operations
but rather state assertions about their operands.
</p>
<dl compact="compact">
<dd><a name="index-strict_005flow_005fpart"></a>
<a name="index-subreg_002c-in-strict_005flow_005fpart"></a>
</dd>
<dt><code>(strict_low_part (subreg:<var>m</var> (reg:<var>n</var> <var>r</var>) 0))</code></dt>
<dd><p>This expression code is used in only one context: as the destination operand of a
<code>set</code> expression.  In addition, the operand of this expression
must be a non-paradoxical <code>subreg</code> expression.
</p>
<p>The presence of <code>strict_low_part</code> says that the part of the
register which is meaningful in mode <var>n</var>, but is not part of
mode <var>m</var>, is not to be altered.  Normally, an assignment to such
a subreg is allowed to have undefined effects on the rest of the
register when <var>m</var> is less than a word.
</p></dd>
</dl>

<hr>
<a name="Side-Effects"></a>
<div class="header">
<p>
Next: <a href="#Incdec" accesskey="n" rel="next">Incdec</a>, Previous: <a href="#RTL-Declarations" accesskey="p" rel="prev">RTL Declarations</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Side-Effect-Expressions"></a>
<h3 class="section">10.15 Side Effect Expressions</h3>
<a name="index-RTL-side-effect-expressions"></a>

<p>The expression codes described so far represent values, not actions.
But machine instructions never produce values; they are meaningful
only for their side effects on the state of the machine.  Special
expression codes are used to represent side effects.
</p>
<p>The body of an instruction is always one of these side effect codes;
the codes described above, which represent values, appear only as
the operands of these.
</p>
<dl compact="compact">
<dd><a name="index-set"></a>
</dd>
<dt><code>(set <var>lval</var> <var>x</var>)</code></dt>
<dd><p>Represents the action of storing the value of <var>x</var> into the place
represented by <var>lval</var>.  <var>lval</var> must be an expression
representing a place that can be stored in: <code>reg</code> (or <code>subreg</code>,
<code>strict_low_part</code> or <code>zero_extract</code>), <code>mem</code>, <code>pc</code>,
<code>parallel</code>, or <code>cc0</code>.
</p>
<p>If <var>lval</var> is a <code>reg</code>, <code>subreg</code> or <code>mem</code>, it has a
machine mode; then <var>x</var> must be valid for that mode.
</p>
<p>If <var>lval</var> is a <code>reg</code> whose machine mode is less than the full
width of the register, then it means that the part of the register
specified by the machine mode is given the specified value and the
rest of the register receives an undefined value.  Likewise, if
<var>lval</var> is a <code>subreg</code> whose machine mode is narrower than
the mode of the register, the rest of the register can be changed in
an undefined way.
</p>
<p>If <var>lval</var> is a <code>strict_low_part</code> of a subreg, then the part
of the register specified by the machine mode of the <code>subreg</code> is
given the value <var>x</var> and the rest of the register is not changed.
</p>
<p>If <var>lval</var> is a <code>zero_extract</code>, then the referenced part of
the bit-field (a memory or register reference) specified by the
<code>zero_extract</code> is given the value <var>x</var> and the rest of the
bit-field is not changed.  Note that <code>sign_extract</code> can not
appear in <var>lval</var>.
</p>
<p>If <var>lval</var> is <code>(cc0)</code>, it has no machine mode, and <var>x</var> may
be either a <code>compare</code> expression or a value that may have any mode.
The latter case represents a &ldquo;test&rdquo; instruction.  The expression
<code>(set (cc0) (reg:<var>m</var> <var>n</var>))</code> is equivalent to
<code>(set (cc0) (compare (reg:<var>m</var> <var>n</var>) (const_int 0)))</code>.
Use the former expression to save space during the compilation.
</p>
<p>If <var>lval</var> is a <code>parallel</code>, it is used to represent the case of
a function returning a structure in multiple registers.  Each element
of the <code>parallel</code> is an <code>expr_list</code> whose first operand is a
<code>reg</code> and whose second operand is a <code>const_int</code> representing the
offset (in bytes) into the structure at which the data in that register
corresponds.  The first element may be null to indicate that the structure
is also passed partly in memory.
</p>
<a name="index-jump-instructions-and-set"></a>
<a name="index-if_005fthen_005felse-usage"></a>
<p>If <var>lval</var> is <code>(pc)</code>, we have a jump instruction, and the
possibilities for <var>x</var> are very limited.  It may be a
<code>label_ref</code> expression (unconditional jump).  It may be an
<code>if_then_else</code> (conditional jump), in which case either the
second or the third operand must be <code>(pc)</code> (for the case which
does not jump) and the other of the two must be a <code>label_ref</code>
(for the case which does jump).  <var>x</var> may also be a <code>mem</code> or
<code>(plus:SI (pc) <var>y</var>)</code>, where <var>y</var> may be a <code>reg</code> or a
<code>mem</code>; these unusual patterns are used to represent jumps through
branch tables.
</p>
<p>If <var>lval</var> is neither <code>(cc0)</code> nor <code>(pc)</code>, the mode of
<var>lval</var> must not be <code>VOIDmode</code> and the mode of <var>x</var> must be
valid for the mode of <var>lval</var>.
</p>
<a name="index-SET_005fDEST"></a>
<a name="index-SET_005fSRC"></a>
<p><var>lval</var> is customarily accessed with the <code>SET_DEST</code> macro and
<var>x</var> with the <code>SET_SRC</code> macro.
</p>
<a name="index-return"></a>
</dd>
<dt><code>(return)</code></dt>
<dd><p>As the sole expression in a pattern, represents a return from the
current function, on machines where this can be done with one
instruction, such as VAXen.  On machines where a multi-instruction
&ldquo;epilogue&rdquo; must be executed in order to return from the function,
returning is done by jumping to a label which precedes the epilogue, and
the <code>return</code> expression code is never used.
</p>
<p>Inside an <code>if_then_else</code> expression, represents the value to be
placed in <code>pc</code> to return to the caller.
</p>
<p>Note that an insn pattern of <code>(return)</code> is logically equivalent to
<code>(set (pc) (return))</code>, but the latter form is never used.
</p>
<a name="index-simple_005freturn"></a>
</dd>
<dt><code>(simple_return)</code></dt>
<dd><p>Like <code>(return)</code>, but truly represents only a function return, while
<code>(return)</code> may represent an insn that also performs other functions
of the function epilogue.  Like <code>(return)</code>, this may also occur in
conditional jumps.
</p>
<a name="index-call-1"></a>
</dd>
<dt><code>(call <var>function</var> <var>nargs</var>)</code></dt>
<dd><p>Represents a function call.  <var>function</var> is a <code>mem</code> expression
whose address is the address of the function to be called.
<var>nargs</var> is an expression which can be used for two purposes: on
some machines it represents the number of bytes of stack argument; on
others, it represents the number of argument registers.
</p>
<p>Each machine has a standard machine mode which <var>function</var> must
have.  The machine description defines macro <code>FUNCTION_MODE</code> to
expand into the requisite mode name.  The purpose of this mode is to
specify what kind of addressing is allowed, on machines where the
allowed kinds of addressing depend on the machine mode being
addressed.
</p>
<a name="index-clobber"></a>
</dd>
<dt><code>(clobber <var>x</var>)</code></dt>
<dd><p>Represents the storing or possible storing of an unpredictable,
undescribed value into <var>x</var>, which must be a <code>reg</code>,
<code>scratch</code>, <code>parallel</code> or <code>mem</code> expression.
</p>
<p>One place this is used is in string instructions that store standard
values into particular hard registers.  It may not be worth the
trouble to describe the values that are stored, but it is essential to
inform the compiler that the registers will be altered, lest it
attempt to keep data in them across the string instruction.
</p>
<p>If <var>x</var> is <code>(mem:BLK (const_int 0))</code> or
<code>(mem:BLK (scratch))</code>, it means that all memory
locations must be presumed clobbered.  If <var>x</var> is a <code>parallel</code>,
it has the same meaning as a <code>parallel</code> in a <code>set</code> expression.
</p>
<p>Note that the machine description classifies certain hard registers as
&ldquo;call-clobbered&rdquo;.  All function call instructions are assumed by
default to clobber these registers, so there is no need to use
<code>clobber</code> expressions to indicate this fact.  Also, each function
call is assumed to have the potential to alter any memory location,
unless the function is declared <code>const</code>.
</p>
<p>If the last group of expressions in a <code>parallel</code> are each a
<code>clobber</code> expression whose arguments are <code>reg</code> or
<code>match_scratch</code> (see <a href="#RTL-Template">RTL Template</a>) expressions, the combiner
phase can add the appropriate <code>clobber</code> expressions to an insn it
has constructed when doing so will cause a pattern to be matched.
</p>
<p>This feature can be used, for example, on a machine that whose multiply
and add instructions don&rsquo;t use an MQ register but which has an
add-accumulate instruction that does clobber the MQ register.  Similarly,
a combined instruction might require a temporary register while the
constituent instructions might not.
</p>
<p>When a <code>clobber</code> expression for a register appears inside a
<code>parallel</code> with other side effects, the register allocator
guarantees that the register is unoccupied both before and after that
insn if it is a hard register clobber.  For pseudo-register clobber,
the register allocator and the reload pass do not assign the same hard
register to the clobber and the input operands if there is an insn
alternative containing the &lsquo;<samp>&amp;</samp>&rsquo; constraint (see <a href="#Modifiers">Modifiers</a>) for
the clobber and the hard register is in register classes of the
clobber in the alternative.  You can clobber either a specific hard
register, a pseudo register, or a <code>scratch</code> expression; in the
latter two cases, GCC will allocate a hard register that is available
there for use as a temporary.
</p>
<p>For instructions that require a temporary register, you should use
<code>scratch</code> instead of a pseudo-register because this will allow the
combiner phase to add the <code>clobber</code> when required.  You do this by
coding (<code>clobber</code> (<code>match_scratch</code> &hellip;)).  If you do
clobber a pseudo register, use one which appears nowhere else&mdash;generate
a new one each time.  Otherwise, you may confuse CSE.
</p>
<p>There is one other known use for clobbering a pseudo register in a
<code>parallel</code>: when one of the input operands of the insn is also
clobbered by the insn.  In this case, using the same pseudo register in
the clobber and elsewhere in the insn produces the expected results.
</p>
<a name="index-use"></a>
</dd>
<dt><code>(use <var>x</var>)</code></dt>
<dd><p>Represents the use of the value of <var>x</var>.  It indicates that the
value in <var>x</var> at this point in the program is needed, even though
it may not be apparent why this is so.  Therefore, the compiler will
not attempt to delete previous instructions whose only effect is to
store a value in <var>x</var>.  <var>x</var> must be a <code>reg</code> expression.
</p>
<p>In some situations, it may be tempting to add a <code>use</code> of a
register in a <code>parallel</code> to describe a situation where the value
of a special register will modify the behavior of the instruction.
A hypothetical example might be a pattern for an addition that can
either wrap around or use saturating addition depending on the value
of a special control register:
</p>
<div class="smallexample">
<pre class="smallexample">(parallel [(set (reg:SI 2) (unspec:SI [(reg:SI 3)
                                       (reg:SI 4)] 0))
           (use (reg:SI 1))])
</pre></div>


<p>This will not work, several of the optimizers only look at expressions
locally; it is very likely that if you have multiple insns with
identical inputs to the <code>unspec</code>, they will be optimized away even
if register 1 changes in between.
</p>
<p>This means that <code>use</code> can <em>only</em> be used to describe
that the register is live.  You should think twice before adding
<code>use</code> statements, more often you will want to use <code>unspec</code>
instead.  The <code>use</code> RTX is most commonly useful to describe that
a fixed register is implicitly used in an insn.  It is also safe to use
in patterns where the compiler knows for other reasons that the result
of the whole pattern is variable, such as &lsquo;<samp>movmem<var>m</var></samp>&rsquo; or
&lsquo;<samp>call</samp>&rsquo; patterns.
</p>
<p>During the reload phase, an insn that has a <code>use</code> as pattern
can carry a reg_equal note.  These <code>use</code> insns will be deleted
before the reload phase exits.
</p>
<p>During the delayed branch scheduling phase, <var>x</var> may be an insn.
This indicates that <var>x</var> previously was located at this place in the
code and its data dependencies need to be taken into account.  These
<code>use</code> insns will be deleted before the delayed branch scheduling
phase exits.
</p>
<a name="index-parallel"></a>
</dd>
<dt><code>(parallel [<var>x0</var> <var>x1</var> &hellip;])</code></dt>
<dd><p>Represents several side effects performed in parallel.  The square
brackets stand for a vector; the operand of <code>parallel</code> is a
vector of expressions.  <var>x0</var>, <var>x1</var> and so on are individual
side effect expressions&mdash;expressions of code <code>set</code>, <code>call</code>,
<code>return</code>, <code>simple_return</code>, <code>clobber</code> or <code>use</code>.
</p>
<p>&ldquo;In parallel&rdquo; means that first all the values used in the individual
side-effects are computed, and second all the actual side-effects are
performed.  For example,
</p>
<div class="smallexample">
<pre class="smallexample">(parallel [(set (reg:SI 1) (mem:SI (reg:SI 1)))
           (set (mem:SI (reg:SI 1)) (reg:SI 1))])
</pre></div>

<p>says unambiguously that the values of hard register 1 and the memory
location addressed by it are interchanged.  In both places where
<code>(reg:SI 1)</code> appears as a memory address it refers to the value
in register 1 <em>before</em> the execution of the insn.
</p>
<p>It follows that it is <em>incorrect</em> to use <code>parallel</code> and
expect the result of one <code>set</code> to be available for the next one.
For example, people sometimes attempt to represent a jump-if-zero
instruction this way:
</p>
<div class="smallexample">
<pre class="smallexample">(parallel [(set (cc0) (reg:SI 34))
           (set (pc) (if_then_else
                        (eq (cc0) (const_int 0))
                        (label_ref &hellip;)
                        (pc)))])
</pre></div>

<p>But this is incorrect, because it says that the jump condition depends
on the condition code value <em>before</em> this instruction, not on the
new value that is set by this instruction.
</p>
<a name="index-peephole-optimization_002c-RTL-representation"></a>
<p>Peephole optimization, which takes place together with final assembly
code output, can produce insns whose patterns consist of a <code>parallel</code>
whose elements are the operands needed to output the resulting
assembler code&mdash;often <code>reg</code>, <code>mem</code> or constant expressions.
This would not be well-formed RTL at any other stage in compilation,
but it is ok then because no further optimization remains to be done.
However, the definition of the macro <code>NOTICE_UPDATE_CC</code>, if
any, must deal with such insns if you define any peephole optimizations.
</p>
<a name="index-cond_005fexec"></a>
</dd>
<dt><code>(cond_exec [<var>cond</var> <var>expr</var>])</code></dt>
<dd><p>Represents a conditionally executed expression.  The <var>expr</var> is
executed only if the <var>cond</var> is nonzero.  The <var>cond</var> expression
must not have side-effects, but the <var>expr</var> may very well have
side-effects.
</p>
<a name="index-sequence"></a>
</dd>
<dt><code>(sequence [<var>insns</var> &hellip;])</code></dt>
<dd><p>Represents a sequence of insns.  Each of the <var>insns</var> that appears
in the vector is suitable for appearing in the chain of insns, so it
must be an <code>insn</code>, <code>jump_insn</code>, <code>call_insn</code>,
<code>code_label</code>, <code>barrier</code> or <code>note</code>.
</p>
<p>A <code>sequence</code> RTX is never placed in an actual insn during RTL
generation.  It represents the sequence of insns that result from a
<code>define_expand</code> <em>before</em> those insns are passed to
<code>emit_insn</code> to insert them in the chain of insns.  When actually
inserted, the individual sub-insns are separated out and the
<code>sequence</code> is forgotten.
</p>
<p>After delay-slot scheduling is completed, an insn and all the insns that
reside in its delay slots are grouped together into a <code>sequence</code>.
The insn requiring the delay slot is the first insn in the vector;
subsequent insns are to be placed in the delay slot.
</p>
<p><code>INSN_ANNULLED_BRANCH_P</code> is set on an insn in a delay slot to
indicate that a branch insn should be used that will conditionally annul
the effect of the insns in the delay slots.  In such a case,
<code>INSN_FROM_TARGET_P</code> indicates that the insn is from the target of
the branch and should be executed only if the branch is taken; otherwise
the insn should be executed only if the branch is not taken.
See <a href="#Delay-Slots">Delay Slots</a>.
</p></dd>
</dl>

<p>These expression codes appear in place of a side effect, as the body of
an insn, though strictly speaking they do not always describe side
effects as such:
</p>
<dl compact="compact">
<dd><a name="index-asm_005finput"></a>
</dd>
<dt><code>(asm_input <var>s</var>)</code></dt>
<dd><p>Represents literal assembler code as described by the string <var>s</var>.
</p>
<a name="index-unspec"></a>
<a name="index-unspec_005fvolatile"></a>
</dd>
<dt><code>(unspec [<var>operands</var> &hellip;] <var>index</var>)</code></dt>
<dt><code>(unspec_volatile [<var>operands</var> &hellip;] <var>index</var>)</code></dt>
<dd><p>Represents a machine-specific operation on <var>operands</var>.  <var>index</var>
selects between multiple machine-specific operations.
<code>unspec_volatile</code> is used for volatile operations and operations
that may trap; <code>unspec</code> is used for other operations.
</p>
<p>These codes may appear inside a <code>pattern</code> of an
insn, inside a <code>parallel</code>, or inside an expression.
</p>
<a name="index-addr_005fvec"></a>
</dd>
<dt><code>(addr_vec:<var>m</var> [<var>lr0</var> <var>lr1</var> &hellip;])</code></dt>
<dd><p>Represents a table of jump addresses.  The vector elements <var>lr0</var>,
etc., are <code>label_ref</code> expressions.  The mode <var>m</var> specifies
how much space is given to each address; normally <var>m</var> would be
<code>Pmode</code>.
</p>
<a name="index-addr_005fdiff_005fvec"></a>
</dd>
<dt><code>(addr_diff_vec:<var>m</var> <var>base</var> [<var>lr0</var> <var>lr1</var> &hellip;] <var>min</var> <var>max</var> <var>flags</var>)</code></dt>
<dd><p>Represents a table of jump addresses expressed as offsets from
<var>base</var>.  The vector elements <var>lr0</var>, etc., are <code>label_ref</code>
expressions and so is <var>base</var>.  The mode <var>m</var> specifies how much
space is given to each address-difference.  <var>min</var> and <var>max</var>
are set up by branch shortening and hold a label with a minimum and a
maximum address, respectively.  <var>flags</var> indicates the relative
position of <var>base</var>, <var>min</var> and <var>max</var> to the containing insn
and of <var>min</var> and <var>max</var> to <var>base</var>.  See rtl.def for details.
</p>
<a name="index-prefetch"></a>
</dd>
<dt><code>(prefetch:<var>m</var> <var>addr</var> <var>rw</var> <var>locality</var>)</code></dt>
<dd><p>Represents prefetch of memory at address <var>addr</var>.
Operand <var>rw</var> is 1 if the prefetch is for data to be written, 0 otherwise;
targets that do not support write prefetches should treat this as a normal
prefetch.
Operand <var>locality</var> specifies the amount of temporal locality; 0 if there
is none or 1, 2, or 3 for increasing levels of temporal locality;
targets that do not support locality hints should ignore this.
</p>
<p>This insn is used to minimize cache-miss latency by moving data into a
cache before it is accessed.  It should use only non-faulting data prefetch
instructions.
</p></dd>
</dl>

<hr>
<a name="Incdec"></a>
<div class="header">
<p>
Next: <a href="#Assembler" accesskey="n" rel="next">Assembler</a>, Previous: <a href="#Side-Effects" accesskey="p" rel="prev">Side Effects</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Embedded-Side_002dEffects-on-Addresses"></a>
<h3 class="section">10.16 Embedded Side-Effects on Addresses</h3>
<a name="index-RTL-preincrement"></a>
<a name="index-RTL-postincrement"></a>
<a name="index-RTL-predecrement"></a>
<a name="index-RTL-postdecrement"></a>

<p>Six special side-effect expression codes appear as memory addresses.
</p>
<dl compact="compact">
<dd><a name="index-pre_005fdec"></a>
</dd>
<dt><code>(pre_dec:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the side effect of decrementing <var>x</var> by a standard
amount and represents also the value that <var>x</var> has after being
decremented.  <var>x</var> must be a <code>reg</code> or <code>mem</code>, but most
machines allow only a <code>reg</code>.  <var>m</var> must be the machine mode
for pointers on the machine in use.  The amount <var>x</var> is decremented
by is the length in bytes of the machine mode of the containing memory
reference of which this expression serves as the address.  Here is an
example of its use:
</p>
<div class="smallexample">
<pre class="smallexample">(mem:DF (pre_dec:SI (reg:SI 39)))
</pre></div>

<p>This says to decrement pseudo register 39 by the length of a <code>DFmode</code>
value and use the result to address a <code>DFmode</code> value.
</p>
<a name="index-pre_005finc"></a>
</dd>
<dt><code>(pre_inc:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Similar, but specifies incrementing <var>x</var> instead of decrementing it.
</p>
<a name="index-post_005fdec"></a>
</dd>
<dt><code>(post_dec:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Represents the same side effect as <code>pre_dec</code> but a different
value.  The value represented here is the value <var>x</var> has <i>before</i>
being decremented.
</p>
<a name="index-post_005finc"></a>
</dd>
<dt><code>(post_inc:<var>m</var> <var>x</var>)</code></dt>
<dd><p>Similar, but specifies incrementing <var>x</var> instead of decrementing it.
</p>
<a name="index-post_005fmodify"></a>
</dd>
<dt><code>(post_modify:<var>m</var> <var>x</var> <var>y</var>)</code></dt>
<dd>
<p>Represents the side effect of setting <var>x</var> to <var>y</var> and
represents <var>x</var> before <var>x</var> is modified.  <var>x</var> must be a
<code>reg</code> or <code>mem</code>, but most machines allow only a <code>reg</code>.
<var>m</var> must be the machine mode for pointers on the machine in use.
</p>
<p>The expression <var>y</var> must be one of three forms:
<code>(plus:<var>m</var> <var>x</var> <var>z</var>)</code>,
<code>(minus:<var>m</var> <var>x</var> <var>z</var>)</code>, or
<code>(plus:<var>m</var> <var>x</var> <var>i</var>)</code>,
where <var>z</var> is an index register and <var>i</var> is a constant.
</p>
<p>Here is an example of its use:
</p>
<div class="smallexample">
<pre class="smallexample">(mem:SF (post_modify:SI (reg:SI 42) (plus (reg:SI 42)
                                          (reg:SI 48))))
</pre></div>

<p>This says to modify pseudo register 42 by adding the contents of pseudo
register 48 to it, after the use of what ever 42 points to.
</p>
<a name="index-pre_005fmodify"></a>
</dd>
<dt><code>(pre_modify:<var>m</var> <var>x</var> <var>expr</var>)</code></dt>
<dd><p>Similar except side effects happen before the use.
</p></dd>
</dl>

<p>These embedded side effect expressions must be used with care.  Instruction
patterns may not use them.  Until the &lsquo;<samp>flow</samp>&rsquo; pass of the compiler,
they may occur only to represent pushes onto the stack.  The &lsquo;<samp>flow</samp>&rsquo;
pass finds cases where registers are incremented or decremented in one
instruction and used as an address shortly before or after; these cases are
then transformed to use pre- or post-increment or -decrement.
</p>
<p>If a register used as the operand of these expressions is used in
another address in an insn, the original value of the register is used.
Uses of the register outside of an address are not permitted within the
same insn as a use in an embedded side effect expression because such
insns behave differently on different machines and hence must be treated
as ambiguous and disallowed.
</p>
<p>An instruction that can be represented with an embedded side effect
could also be represented using <code>parallel</code> containing an additional
<code>set</code> to describe how the address register is altered.  This is not
done because machines that allow these operations at all typically
allow them wherever a memory address is called for.  Describing them as
additional parallel stores would require doubling the number of entries
in the machine description.
</p>
<hr>
<a name="Assembler"></a>
<div class="header">
<p>
Next: <a href="#Debug-Information" accesskey="n" rel="next">Debug Information</a>, Previous: <a href="#Incdec" accesskey="p" rel="prev">Incdec</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Assembler-Instructions-as-Expressions"></a>
<h3 class="section">10.17 Assembler Instructions as Expressions</h3>
<a name="index-assembler-instructions-in-RTL"></a>

<a name="index-asm_005foperands_002c-usage"></a>
<p>The RTX code <code>asm_operands</code> represents a value produced by a
user-specified assembler instruction.  It is used to represent
an <code>asm</code> statement with arguments.  An <code>asm</code> statement with
a single output operand, like this:
</p>
<div class="smallexample">
<pre class="smallexample">asm (&quot;foo %1,%2,%0&quot; : &quot;=a&quot; (outputvar) : &quot;g&quot; (x + y), &quot;di&quot; (*z));
</pre></div>

<p>is represented using a single <code>asm_operands</code> RTX which represents
the value that is stored in <code>outputvar</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(set <var>rtx-for-outputvar</var>
     (asm_operands &quot;foo %1,%2,%0&quot; &quot;a&quot; 0
                   [<var>rtx-for-addition-result</var> <var>rtx-for-*z</var>]
                   [(asm_input:<var>m1</var> &quot;g&quot;)
                    (asm_input:<var>m2</var> &quot;di&quot;)]))
</pre></div>

<p>Here the operands of the <code>asm_operands</code> RTX are the assembler
template string, the output-operand&rsquo;s constraint, the index-number of the
output operand among the output operands specified, a vector of input
operand RTX&rsquo;s, and a vector of input-operand modes and constraints.  The
mode <var>m1</var> is the mode of the sum <code>x+y</code>; <var>m2</var> is that of
<code>*z</code>.
</p>
<p>When an <code>asm</code> statement has multiple output values, its insn has
several such <code>set</code> RTX&rsquo;s inside of a <code>parallel</code>.  Each <code>set</code>
contains an <code>asm_operands</code>; all of these share the same assembler
template and vectors, but each contains the constraint for the respective
output operand.  They are also distinguished by the output-operand index
number, which is 0, 1, &hellip; for successive output operands.
</p>
<hr>
<a name="Debug-Information"></a>
<div class="header">
<p>
Next: <a href="#Insns" accesskey="n" rel="next">Insns</a>, Previous: <a href="#Assembler" accesskey="p" rel="prev">Assembler</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Variable-Location-Debug-Information-in-RTL"></a>
<h3 class="section">10.18 Variable Location Debug Information in RTL</h3>
<a name="index-Variable-Location-Debug-Information-in-RTL"></a>

<p>Variable tracking relies on <code>MEM_EXPR</code> and <code>REG_EXPR</code>
annotations to determine what user variables memory and register
references refer to.
</p>
<p>Variable tracking at assignments uses these notes only when they refer
to variables that live at fixed locations (e.g., addressable
variables, global non-automatic variables).  For variables whose
location may vary, it relies on the following types of notes.
</p>
<dl compact="compact">
<dd><a name="index-var_005flocation"></a>
</dd>
<dt><code>(var_location:<var>mode</var> <var>var</var> <var>exp</var> <var>stat</var>)</code></dt>
<dd><p>Binds variable <code>var</code>, a tree, to value <var>exp</var>, an RTL
expression.  It appears only in <code>NOTE_INSN_VAR_LOCATION</code> and
<code>DEBUG_INSN</code>s, with slightly different meanings.  <var>mode</var>, if
present, represents the mode of <var>exp</var>, which is useful if it is a
modeless expression.  <var>stat</var> is only meaningful in notes,
indicating whether the variable is known to be initialized or
uninitialized.
</p>
<a name="index-debug_005fexpr"></a>
</dd>
<dt><code>(debug_expr:<var>mode</var> <var>decl</var>)</code></dt>
<dd><p>Stands for the value bound to the <code>DEBUG_EXPR_DECL</code> <var>decl</var>,
that points back to it, within value expressions in
<code>VAR_LOCATION</code> nodes.
</p>
</dd>
</dl>

<hr>
<a name="Insns"></a>
<div class="header">
<p>
Next: <a href="#Calls" accesskey="n" rel="next">Calls</a>, Previous: <a href="#Debug-Information" accesskey="p" rel="prev">Debug Information</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Insns-1"></a>
<h3 class="section">10.19 Insns</h3>
<a name="index-insns"></a>

<p>The RTL representation of the code for a function is a doubly-linked
chain of objects called <em>insns</em>.  Insns are expressions with
special codes that are used for no other purpose.  Some insns are
actual instructions; others represent dispatch tables for <code>switch</code>
statements; others represent labels to jump to or various sorts of
declarative information.
</p>
<p>In addition to its own specific data, each insn must have a unique
id-number that distinguishes it from all other insns in the current
function (after delayed branch scheduling, copies of an insn with the
same id-number may be present in multiple places in a function, but
these copies will always be identical and will only appear inside a
<code>sequence</code>), and chain pointers to the preceding and following
insns.  These three fields occupy the same position in every insn,
independent of the expression code of the insn.  They could be accessed
with <code>XEXP</code> and <code>XINT</code>, but instead three special macros are
always used:
</p>
<dl compact="compact">
<dd><a name="index-INSN_005fUID"></a>
</dd>
<dt><code>INSN_UID (<var>i</var>)</code></dt>
<dd><p>Accesses the unique id of insn <var>i</var>.
</p>
<a name="index-PREV_005fINSN"></a>
</dd>
<dt><code>PREV_INSN (<var>i</var>)</code></dt>
<dd><p>Accesses the chain pointer to the insn preceding <var>i</var>.
If <var>i</var> is the first insn, this is a null pointer.
</p>
<a name="index-NEXT_005fINSN"></a>
</dd>
<dt><code>NEXT_INSN (<var>i</var>)</code></dt>
<dd><p>Accesses the chain pointer to the insn following <var>i</var>.
If <var>i</var> is the last insn, this is a null pointer.
</p></dd>
</dl>

<a name="index-get_005finsns"></a>
<a name="index-get_005flast_005finsn"></a>
<p>The first insn in the chain is obtained by calling <code>get_insns</code>; the
last insn is the result of calling <code>get_last_insn</code>.  Within the
chain delimited by these insns, the <code>NEXT_INSN</code> and
<code>PREV_INSN</code> pointers must always correspond: if <var>insn</var> is not
the first insn,
</p>
<div class="smallexample">
<pre class="smallexample">NEXT_INSN (PREV_INSN (<var>insn</var>)) == <var>insn</var>
</pre></div>

<p>is always true and if <var>insn</var> is not the last insn,
</p>
<div class="smallexample">
<pre class="smallexample">PREV_INSN (NEXT_INSN (<var>insn</var>)) == <var>insn</var>
</pre></div>

<p>is always true.
</p>
<p>After delay slot scheduling, some of the insns in the chain might be
<code>sequence</code> expressions, which contain a vector of insns.  The value
of <code>NEXT_INSN</code> in all but the last of these insns is the next insn
in the vector; the value of <code>NEXT_INSN</code> of the last insn in the vector
is the same as the value of <code>NEXT_INSN</code> for the <code>sequence</code> in
which it is contained.  Similar rules apply for <code>PREV_INSN</code>.
</p>
<p>This means that the above invariants are not necessarily true for insns
inside <code>sequence</code> expressions.  Specifically, if <var>insn</var> is the
first insn in a <code>sequence</code>, <code>NEXT_INSN (PREV_INSN (<var>insn</var>))</code>
is the insn containing the <code>sequence</code> expression, as is the value
of <code>PREV_INSN (NEXT_INSN (<var>insn</var>))</code> if <var>insn</var> is the last
insn in the <code>sequence</code> expression.  You can use these expressions
to find the containing <code>sequence</code> expression.
</p>
<p>Every insn has one of the following expression codes:
</p>
<dl compact="compact">
<dd><a name="index-insn"></a>
</dd>
<dt><code>insn</code></dt>
<dd><p>The expression code <code>insn</code> is used for instructions that do not jump
and do not do function calls.  <code>sequence</code> expressions are always
contained in insns with code <code>insn</code> even if one of those insns
should jump or do function calls.
</p>
<p>Insns with code <code>insn</code> have four additional fields beyond the three
mandatory ones listed above.  These four are described in a table below.
</p>
<a name="index-jump_005finsn"></a>
</dd>
<dt><code>jump_insn</code></dt>
<dd><p>The expression code <code>jump_insn</code> is used for instructions that may
jump (or, more generally, may contain <code>label_ref</code> expressions to
which <code>pc</code> can be set in that instruction).  If there is an
instruction to return from the current function, it is recorded as a
<code>jump_insn</code>.
</p>
<a name="index-JUMP_005fLABEL"></a>
<p><code>jump_insn</code> insns have the same extra fields as <code>insn</code> insns,
accessed in the same way and in addition contain a field
<code>JUMP_LABEL</code> which is defined once jump optimization has completed.
</p>
<p>For simple conditional and unconditional jumps, this field contains
the <code>code_label</code> to which this insn will (possibly conditionally)
branch.  In a more complex jump, <code>JUMP_LABEL</code> records one of the
labels that the insn refers to; other jump target labels are recorded
as <code>REG_LABEL_TARGET</code> notes.  The exception is <code>addr_vec</code>
and <code>addr_diff_vec</code>, where <code>JUMP_LABEL</code> is <code>NULL_RTX</code>
and the only way to find the labels is to scan the entire body of the
insn.
</p>
<p>Return insns count as jumps, but since they do not refer to any
labels, their <code>JUMP_LABEL</code> is <code>NULL_RTX</code>.
</p>
<a name="index-call_005finsn"></a>
</dd>
<dt><code>call_insn</code></dt>
<dd><p>The expression code <code>call_insn</code> is used for instructions that may do
function calls.  It is important to distinguish these instructions because
they imply that certain registers and memory locations may be altered
unpredictably.
</p>
<a name="index-CALL_005fINSN_005fFUNCTION_005fUSAGE"></a>
<p><code>call_insn</code> insns have the same extra fields as <code>insn</code> insns,
accessed in the same way and in addition contain a field
<code>CALL_INSN_FUNCTION_USAGE</code>, which contains a list (chain of
<code>expr_list</code> expressions) containing <code>use</code>, <code>clobber</code> and
sometimes <code>set</code> expressions that denote hard registers and
<code>mem</code>s used or clobbered by the called function.
</p>
<p>A <code>mem</code> generally points to a stack slot in which arguments passed
to the libcall by reference (see <a href="#Register-Arguments">TARGET_PASS_BY_REFERENCE</a>) are stored.  If the argument is
caller-copied (see <a href="#Register-Arguments">TARGET_CALLEE_COPIES</a>),
the stack slot will be mentioned in <code>clobber</code> and <code>use</code>
entries; if it&rsquo;s callee-copied, only a <code>use</code> will appear, and the
<code>mem</code> may point to addresses that are not stack slots.
</p>
<p>Registers occurring inside a <code>clobber</code> in this list augment
registers specified in <code>CALL_USED_REGISTERS</code> (see <a href="#Register-Basics">Register Basics</a>).
</p>
<p>If the list contains a <code>set</code> involving two registers, it indicates
that the function returns one of its arguments.  Such a <code>set</code> may
look like a no-op if the same register holds the argument and the return
value.
</p>
<a name="index-code_005flabel"></a>
<a name="index-CODE_005fLABEL_005fNUMBER"></a>
</dd>
<dt><code>code_label</code></dt>
<dd><p>A <code>code_label</code> insn represents a label that a jump insn can jump
to.  It contains two special fields of data in addition to the three
standard ones.  <code>CODE_LABEL_NUMBER</code> is used to hold the <em>label
number</em>, a number that identifies this label uniquely among all the
labels in the compilation (not just in the current function).
Ultimately, the label is represented in the assembler output as an
assembler label, usually of the form &lsquo;<samp>L<var>n</var></samp>&rsquo; where <var>n</var> is
the label number.
</p>
<p>When a <code>code_label</code> appears in an RTL expression, it normally
appears within a <code>label_ref</code> which represents the address of
the label, as a number.
</p>
<p>Besides as a <code>code_label</code>, a label can also be represented as a
<code>note</code> of type <code>NOTE_INSN_DELETED_LABEL</code>.
</p>
<a name="index-LABEL_005fNUSES"></a>
<p>The field <code>LABEL_NUSES</code> is only defined once the jump optimization
phase is completed.  It contains the number of times this label is
referenced in the current function.
</p>
<a name="index-LABEL_005fKIND"></a>
<a name="index-SET_005fLABEL_005fKIND"></a>
<a name="index-LABEL_005fALT_005fENTRY_005fP"></a>
<a name="index-alternate-entry-points"></a>
<p>The field <code>LABEL_KIND</code> differentiates four different types of
labels: <code>LABEL_NORMAL</code>, <code>LABEL_STATIC_ENTRY</code>,
<code>LABEL_GLOBAL_ENTRY</code>, and <code>LABEL_WEAK_ENTRY</code>.  The only labels
that do not have type <code>LABEL_NORMAL</code> are <em>alternate entry
points</em> to the current function.  These may be static (visible only in
the containing translation unit), global (exposed to all translation
units), or weak (global, but can be overridden by another symbol with the
same name).
</p>
<p>Much of the compiler treats all four kinds of label identically.  Some
of it needs to know whether or not a label is an alternate entry point;
for this purpose, the macro <code>LABEL_ALT_ENTRY_P</code> is provided.  It is
equivalent to testing whether &lsquo;<samp>LABEL_KIND (label) == LABEL_NORMAL</samp>&rsquo;.
The only place that cares about the distinction between static, global,
and weak alternate entry points, besides the front-end code that creates
them, is the function <code>output_alternate_entry_point</code>, in
<samp>final.c</samp>.
</p>
<p>To set the kind of a label, use the <code>SET_LABEL_KIND</code> macro.
</p>
<a name="index-barrier"></a>
</dd>
<dt><code>barrier</code></dt>
<dd><p>Barriers are placed in the instruction stream when control cannot flow
past them.  They are placed after unconditional jump instructions to
indicate that the jumps are unconditional and after calls to
<code>volatile</code> functions, which do not return (e.g., <code>exit</code>).
They contain no information beyond the three standard fields.
</p>
<a name="index-note"></a>
<a name="index-NOTE_005fLINE_005fNUMBER"></a>
<a name="index-NOTE_005fSOURCE_005fFILE"></a>
</dd>
<dt><code>note</code></dt>
<dd><p><code>note</code> insns are used to represent additional debugging and
declarative information.  They contain two nonstandard fields, an
integer which is accessed with the macro <code>NOTE_LINE_NUMBER</code> and a
string accessed with <code>NOTE_SOURCE_FILE</code>.
</p>
<p>If <code>NOTE_LINE_NUMBER</code> is positive, the note represents the
position of a source line and <code>NOTE_SOURCE_FILE</code> is the source file name
that the line came from.  These notes control generation of line
number data in the assembler output.
</p>
<p>Otherwise, <code>NOTE_LINE_NUMBER</code> is not really a line number but a
code with one of the following values (and <code>NOTE_SOURCE_FILE</code>
must contain a null pointer):
</p>
<dl compact="compact">
<dd><a name="index-NOTE_005fINSN_005fDELETED"></a>
</dd>
<dt><code>NOTE_INSN_DELETED</code></dt>
<dd><p>Such a note is completely ignorable.  Some passes of the compiler
delete insns by altering them into notes of this kind.
</p>
<a name="index-NOTE_005fINSN_005fDELETED_005fLABEL"></a>
</dd>
<dt><code>NOTE_INSN_DELETED_LABEL</code></dt>
<dd><p>This marks what used to be a <code>code_label</code>, but was not used for other
purposes than taking its address and was transformed to mark that no
code jumps to it.
</p>
<a name="index-NOTE_005fINSN_005fBLOCK_005fBEG"></a>
<a name="index-NOTE_005fINSN_005fBLOCK_005fEND"></a>
</dd>
<dt><code>NOTE_INSN_BLOCK_BEG</code></dt>
<dt><code>NOTE_INSN_BLOCK_END</code></dt>
<dd><p>These types of notes indicate the position of the beginning and end
of a level of scoping of variable names.  They control the output
of debugging information.
</p>
<a name="index-NOTE_005fINSN_005fEH_005fREGION_005fBEG"></a>
<a name="index-NOTE_005fINSN_005fEH_005fREGION_005fEND"></a>
</dd>
<dt><code>NOTE_INSN_EH_REGION_BEG</code></dt>
<dt><code>NOTE_INSN_EH_REGION_END</code></dt>
<dd><p>These types of notes indicate the position of the beginning and end of a
level of scoping for exception handling.  <code>NOTE_BLOCK_NUMBER</code>
identifies which <code>CODE_LABEL</code> or <code>note</code> of type
<code>NOTE_INSN_DELETED_LABEL</code> is associated with the given region.
</p>
<a name="index-NOTE_005fINSN_005fLOOP_005fBEG"></a>
<a name="index-NOTE_005fINSN_005fLOOP_005fEND"></a>
</dd>
<dt><code>NOTE_INSN_LOOP_BEG</code></dt>
<dt><code>NOTE_INSN_LOOP_END</code></dt>
<dd><p>These types of notes indicate the position of the beginning and end
of a <code>while</code> or <code>for</code> loop.  They enable the loop optimizer
to find loops quickly.
</p>
<a name="index-NOTE_005fINSN_005fLOOP_005fCONT"></a>
</dd>
<dt><code>NOTE_INSN_LOOP_CONT</code></dt>
<dd><p>Appears at the place in a loop that <code>continue</code> statements jump to.
</p>
<a name="index-NOTE_005fINSN_005fLOOP_005fVTOP"></a>
</dd>
<dt><code>NOTE_INSN_LOOP_VTOP</code></dt>
<dd><p>This note indicates the place in a loop where the exit test begins for
those loops in which the exit test has been duplicated.  This position
becomes another virtual start of the loop when considering loop
invariants.
</p>
<a name="index-NOTE_005fINSN_005fFUNCTION_005fBEG"></a>
</dd>
<dt><code>NOTE_INSN_FUNCTION_BEG</code></dt>
<dd><p>Appears at the start of the function body, after the function
prologue.
</p>
<a name="index-NOTE_005fINSN_005fVAR_005fLOCATION"></a>
<a name="index-NOTE_005fVAR_005fLOCATION"></a>
</dd>
<dt><code>NOTE_INSN_VAR_LOCATION</code></dt>
<dd><p>This note is used to generate variable location debugging information.
It indicates that the user variable in its <code>VAR_LOCATION</code> operand
is at the location given in the RTL expression, or holds a value that
can be computed by evaluating the RTL expression from that static
point in the program up to the next such note for the same user
variable.
</p>
</dd>
</dl>

<p>These codes are printed symbolically when they appear in debugging dumps.
</p>
<a name="index-debug_005finsn"></a>
<a name="index-INSN_005fVAR_005fLOCATION"></a>
</dd>
<dt><code>debug_insn</code></dt>
<dd><p>The expression code <code>debug_insn</code> is used for pseudo-instructions
that hold debugging information for variable tracking at assignments
(see <samp>-fvar-tracking-assignments</samp> option).  They are the RTL
representation of <code>GIMPLE_DEBUG</code> statements
(<a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a>), with a <code>VAR_LOCATION</code> operand that
binds a user variable tree to an RTL representation of the
<code>value</code> in the corresponding statement.  A <code>DEBUG_EXPR</code> in
it stands for the value bound to the corresponding
<code>DEBUG_EXPR_DECL</code>.
</p>
<p>Throughout optimization passes, binding information is kept in
pseudo-instruction form, so that, unlike notes, it gets the same
treatment and adjustments that regular instructions would.  It is the
variable tracking pass that turns these pseudo-instructions into var
location notes, analyzing control flow, value equivalences and changes
to registers and memory referenced in value expressions, propagating
the values of debug temporaries and determining expressions that can
be used to compute the value of each user variable at as many points
(ranges, actually) in the program as possible.
</p>
<p>Unlike <code>NOTE_INSN_VAR_LOCATION</code>, the value expression in an
<code>INSN_VAR_LOCATION</code> denotes a value at that specific point in the
program, rather than an expression that can be evaluated at any later
point before an overriding <code>VAR_LOCATION</code> is encountered.  E.g.,
if a user variable is bound to a <code>REG</code> and then a subsequent insn
modifies the <code>REG</code>, the note location would keep mapping the user
variable to the register across the insn, whereas the insn location
would keep the variable bound to the value, so that the variable
tracking pass would emit another location note for the variable at the
point in which the register is modified.
</p>
</dd>
</dl>

<a name="index-TImode_002c-in-insn"></a>
<a name="index-HImode_002c-in-insn"></a>
<a name="index-QImode_002c-in-insn"></a>
<p>The machine mode of an insn is normally <code>VOIDmode</code>, but some
phases use the mode for various purposes.
</p>
<p>The common subexpression elimination pass sets the mode of an insn to
<code>QImode</code> when it is the first insn in a block that has already
been processed.
</p>
<p>The second Haifa scheduling pass, for targets that can multiple issue,
sets the mode of an insn to <code>TImode</code> when it is believed that the
instruction begins an issue group.  That is, when the instruction
cannot issue simultaneously with the previous.  This may be relied on
by later passes, in particular machine-dependent reorg.
</p>
<p>Here is a table of the extra fields of <code>insn</code>, <code>jump_insn</code>
and <code>call_insn</code> insns:
</p>
<dl compact="compact">
<dd><a name="index-PATTERN"></a>
</dd>
<dt><code>PATTERN (<var>i</var>)</code></dt>
<dd><p>An expression for the side effect performed by this insn.  This must
be one of the following codes: <code>set</code>, <code>call</code>, <code>use</code>,
<code>clobber</code>, <code>return</code>, <code>simple_return</code>, <code>asm_input</code>,
<code>asm_output</code>, <code>addr_vec</code>, <code>addr_diff_vec</code>,
<code>trap_if</code>, <code>unspec</code>, <code>unspec_volatile</code>,
<code>parallel</code>, <code>cond_exec</code>, or <code>sequence</code>.  If it is a
<code>parallel</code>, each element of the <code>parallel</code> must be one these
codes, except that <code>parallel</code> expressions cannot be nested and
<code>addr_vec</code> and <code>addr_diff_vec</code> are not permitted inside a
<code>parallel</code> expression.
</p>
<a name="index-INSN_005fCODE"></a>
</dd>
<dt><code>INSN_CODE (<var>i</var>)</code></dt>
<dd><p>An integer that says which pattern in the machine description matches
this insn, or -1 if the matching has not yet been attempted.
</p>
<p>Such matching is never attempted and this field remains -1 on an insn
whose pattern consists of a single <code>use</code>, <code>clobber</code>,
<code>asm_input</code>, <code>addr_vec</code> or <code>addr_diff_vec</code> expression.
</p>
<a name="index-asm_005fnoperands"></a>
<p>Matching is also never attempted on insns that result from an <code>asm</code>
statement.  These contain at least one <code>asm_operands</code> expression.
The function <code>asm_noperands</code> returns a non-negative value for
such insns.
</p>
<p>In the debugging output, this field is printed as a number followed by
a symbolic representation that locates the pattern in the <samp>md</samp>
file as some small positive or negative offset from a named pattern.
</p>
<a name="index-LOG_005fLINKS"></a>
</dd>
<dt><code>LOG_LINKS (<var>i</var>)</code></dt>
<dd><p>A list (chain of <code>insn_list</code> expressions) giving information about
dependencies between instructions within a basic block.  Neither a jump
nor a label may come between the related insns.  These are only used by
the schedulers and by combine.  This is a deprecated data structure.
Def-use and use-def chains are now preferred.
</p>
<a name="index-REG_005fNOTES"></a>
</dd>
<dt><code>REG_NOTES (<var>i</var>)</code></dt>
<dd><p>A list (chain of <code>expr_list</code> and <code>insn_list</code> expressions)
giving miscellaneous information about the insn.  It is often
information pertaining to the registers used in this insn.
</p></dd>
</dl>

<p>The <code>LOG_LINKS</code> field of an insn is a chain of <code>insn_list</code>
expressions.  Each of these has two operands: the first is an insn,
and the second is another <code>insn_list</code> expression (the next one in
the chain).  The last <code>insn_list</code> in the chain has a null pointer
as second operand.  The significant thing about the chain is which
insns appear in it (as first operands of <code>insn_list</code>
expressions).  Their order is not significant.
</p>
<p>This list is originally set up by the flow analysis pass; it is a null
pointer until then.  Flow only adds links for those data dependencies
which can be used for instruction combination.  For each insn, the flow
analysis pass adds a link to insns which store into registers values
that are used for the first time in this insn.
</p>
<p>The <code>REG_NOTES</code> field of an insn is a chain similar to the
<code>LOG_LINKS</code> field but it includes <code>expr_list</code> expressions in
addition to <code>insn_list</code> expressions.  There are several kinds of
register notes, which are distinguished by the machine mode, which in a
register note is really understood as being an <code>enum reg_note</code>.
The first operand <var>op</var> of the note is data whose meaning depends on
the kind of note.
</p>
<a name="index-REG_005fNOTE_005fKIND"></a>
<a name="index-PUT_005fREG_005fNOTE_005fKIND"></a>
<p>The macro <code>REG_NOTE_KIND (<var>x</var>)</code> returns the kind of
register note.  Its counterpart, the macro <code>PUT_REG_NOTE_KIND
(<var>x</var>, <var>newkind</var>)</code> sets the register note type of <var>x</var> to be
<var>newkind</var>.
</p>
<p>Register notes are of three classes: They may say something about an
input to an insn, they may say something about an output of an insn, or
they may create a linkage between two insns.  There are also a set
of values that are only used in <code>LOG_LINKS</code>.
</p>
<p>These register notes annotate inputs to an insn:
</p>
<dl compact="compact">
<dd><a name="index-REG_005fDEAD"></a>
</dd>
<dt><code>REG_DEAD</code></dt>
<dd><p>The value in <var>op</var> dies in this insn; that is to say, altering the
value immediately after this insn would not affect the future behavior
of the program.
</p>
<p>It does not follow that the register <var>op</var> has no useful value after
this insn since <var>op</var> is not necessarily modified by this insn.
Rather, no subsequent instruction uses the contents of <var>op</var>.
</p>
<a name="index-REG_005fUNUSED"></a>
</dd>
<dt><code>REG_UNUSED</code></dt>
<dd><p>The register <var>op</var> being set by this insn will not be used in a
subsequent insn.  This differs from a <code>REG_DEAD</code> note, which
indicates that the value in an input will not be used subsequently.
These two notes are independent; both may be present for the same
register.
</p>
<a name="index-REG_005fINC"></a>
</dd>
<dt><code>REG_INC</code></dt>
<dd><p>The register <var>op</var> is incremented (or decremented; at this level
there is no distinction) by an embedded side effect inside this insn.
This means it appears in a <code>post_inc</code>, <code>pre_inc</code>,
<code>post_dec</code> or <code>pre_dec</code> expression.
</p>
<a name="index-REG_005fNONNEG"></a>
</dd>
<dt><code>REG_NONNEG</code></dt>
<dd><p>The register <var>op</var> is known to have a nonnegative value when this
insn is reached.  This is used so that decrement and branch until zero
instructions, such as the m68k dbra, can be matched.
</p>
<p>The <code>REG_NONNEG</code> note is added to insns only if the machine
description has a &lsquo;<samp>decrement_and_branch_until_zero</samp>&rsquo; pattern.
</p>
<a name="index-REG_005fLABEL_005fOPERAND"></a>
</dd>
<dt><code>REG_LABEL_OPERAND</code></dt>
<dd><p>This insn uses <var>op</var>, a <code>code_label</code> or a <code>note</code> of type
<code>NOTE_INSN_DELETED_LABEL</code>, but is not a <code>jump_insn</code>, or it
is a <code>jump_insn</code> that refers to the operand as an ordinary
operand.  The label may still eventually be a jump target, but if so
in an indirect jump in a subsequent insn.  The presence of this note
allows jump optimization to be aware that <var>op</var> is, in fact, being
used, and flow optimization to build an accurate flow graph.
</p>
<a name="index-REG_005fLABEL_005fTARGET"></a>
</dd>
<dt><code>REG_LABEL_TARGET</code></dt>
<dd><p>This insn is a <code>jump_insn</code> but not an <code>addr_vec</code> or
<code>addr_diff_vec</code>.  It uses <var>op</var>, a <code>code_label</code> as a
direct or indirect jump target.  Its purpose is similar to that of
<code>REG_LABEL_OPERAND</code>.  This note is only present if the insn has
multiple targets; the last label in the insn (in the highest numbered
insn-field) goes into the <code>JUMP_LABEL</code> field and does not have a
<code>REG_LABEL_TARGET</code> note.  See <a href="#Insns">JUMP_LABEL</a>.
</p>
<a name="index-REG_005fCROSSING_005fJUMP"></a>
</dd>
<dt><code>REG_CROSSING_JUMP</code></dt>
<dd><p>This insn is a branching instruction (either an unconditional jump or
an indirect jump) which crosses between hot and cold sections, which
could potentially be very far apart in the executable.  The presence
of this note indicates to other optimizations that this branching
instruction should not be &ldquo;collapsed&rdquo; into a simpler branching
construct.  It is used when the optimization to partition basic blocks
into hot and cold sections is turned on.
</p>
<a name="index-REG_005fSETJMP"></a>
</dd>
<dt><code>REG_SETJMP</code></dt>
<dd><p>Appears attached to each <code>CALL_INSN</code> to <code>setjmp</code> or a
related function.
</p></dd>
</dl>

<p>The following notes describe attributes of outputs of an insn:
</p>
<dl compact="compact">
<dd><a name="index-REG_005fEQUIV"></a>
<a name="index-REG_005fEQUAL"></a>
</dd>
<dt><code>REG_EQUIV</code></dt>
<dt><code>REG_EQUAL</code></dt>
<dd><p>This note is only valid on an insn that sets only one register and
indicates that that register will be equal to <var>op</var> at run time; the
scope of this equivalence differs between the two types of notes.  The
value which the insn explicitly copies into the register may look
different from <var>op</var>, but they will be equal at run time.  If the
output of the single <code>set</code> is a <code>strict_low_part</code> expression,
the note refers to the register that is contained in <code>SUBREG_REG</code>
of the <code>subreg</code> expression.
</p>
<p>For <code>REG_EQUIV</code>, the register is equivalent to <var>op</var> throughout
the entire function, and could validly be replaced in all its
occurrences by <var>op</var>.  (&ldquo;Validly&rdquo; here refers to the data flow of
the program; simple replacement may make some insns invalid.)  For
example, when a constant is loaded into a register that is never
assigned any other value, this kind of note is used.
</p>
<p>When a parameter is copied into a pseudo-register at entry to a function,
a note of this kind records that the register is equivalent to the stack
slot where the parameter was passed.  Although in this case the register
may be set by other insns, it is still valid to replace the register
by the stack slot throughout the function.
</p>
<p>A <code>REG_EQUIV</code> note is also used on an instruction which copies a
register parameter into a pseudo-register at entry to a function, if
there is a stack slot where that parameter could be stored.  Although
other insns may set the pseudo-register, it is valid for the compiler to
replace the pseudo-register by stack slot throughout the function,
provided the compiler ensures that the stack slot is properly
initialized by making the replacement in the initial copy instruction as
well.  This is used on machines for which the calling convention
allocates stack space for register parameters.  See
<code>REG_PARM_STACK_SPACE</code> in <a href="#Stack-Arguments">Stack Arguments</a>.
</p>
<p>In the case of <code>REG_EQUAL</code>, the register that is set by this insn
will be equal to <var>op</var> at run time at the end of this insn but not
necessarily elsewhere in the function.  In this case, <var>op</var>
is typically an arithmetic expression.  For example, when a sequence of
insns such as a library call is used to perform an arithmetic operation,
this kind of note is attached to the insn that produces or copies the
final value.
</p>
<p>These two notes are used in different ways by the compiler passes.
<code>REG_EQUAL</code> is used by passes prior to register allocation (such as
common subexpression elimination and loop optimization) to tell them how
to think of that value.  <code>REG_EQUIV</code> notes are used by register
allocation to indicate that there is an available substitute expression
(either a constant or a <code>mem</code> expression for the location of a
parameter on the stack) that may be used in place of a register if
insufficient registers are available.
</p>
<p>Except for stack homes for parameters, which are indicated by a
<code>REG_EQUIV</code> note and are not useful to the early optimization
passes and pseudo registers that are equivalent to a memory location
throughout their entire life, which is not detected until later in
the compilation, all equivalences are initially indicated by an attached
<code>REG_EQUAL</code> note.  In the early stages of register allocation, a
<code>REG_EQUAL</code> note is changed into a <code>REG_EQUIV</code> note if
<var>op</var> is a constant and the insn represents the only set of its
destination register.
</p>
<p>Thus, compiler passes prior to register allocation need only check for
<code>REG_EQUAL</code> notes and passes subsequent to register allocation
need only check for <code>REG_EQUIV</code> notes.
</p></dd>
</dl>

<p>These notes describe linkages between insns.  They occur in pairs: one
insn has one of a pair of notes that points to a second insn, which has
the inverse note pointing back to the first insn.
</p>
<dl compact="compact">
<dd><a name="index-REG_005fCC_005fSETTER"></a>
<a name="index-REG_005fCC_005fUSER"></a>
</dd>
<dt><code>REG_CC_SETTER</code></dt>
<dt><code>REG_CC_USER</code></dt>
<dd><p>On machines that use <code>cc0</code>, the insns which set and use <code>cc0</code>
set and use <code>cc0</code> are adjacent.  However, when branch delay slot
filling is done, this may no longer be true.  In this case a
<code>REG_CC_USER</code> note will be placed on the insn setting <code>cc0</code> to
point to the insn using <code>cc0</code> and a <code>REG_CC_SETTER</code> note will
be placed on the insn using <code>cc0</code> to point to the insn setting
<code>cc0</code>.
</p></dd>
</dl>

<p>These values are only used in the <code>LOG_LINKS</code> field, and indicate
the type of dependency that each link represents.  Links which indicate
a data dependence (a read after write dependence) do not use any code,
they simply have mode <code>VOIDmode</code>, and are printed without any
descriptive text.
</p>
<dl compact="compact">
<dd><a name="index-REG_005fDEP_005fTRUE"></a>
</dd>
<dt><code>REG_DEP_TRUE</code></dt>
<dd><p>This indicates a true dependence (a read after write dependence).
</p>
<a name="index-REG_005fDEP_005fOUTPUT"></a>
</dd>
<dt><code>REG_DEP_OUTPUT</code></dt>
<dd><p>This indicates an output dependence (a write after write dependence).
</p>
<a name="index-REG_005fDEP_005fANTI"></a>
</dd>
<dt><code>REG_DEP_ANTI</code></dt>
<dd><p>This indicates an anti dependence (a write after read dependence).
</p>
</dd>
</dl>

<p>These notes describe information gathered from gcov profile data.  They
are stored in the <code>REG_NOTES</code> field of an insn as an
<code>expr_list</code>.
</p>
<dl compact="compact">
<dd><a name="index-REG_005fBR_005fPROB"></a>
</dd>
<dt><code>REG_BR_PROB</code></dt>
<dd><p>This is used to specify the ratio of branches to non-branches of a
branch insn according to the profile data.  The value is stored as a
value between 0 and REG_BR_PROB_BASE; larger values indicate a higher
probability that the branch will be taken.
</p>
<a name="index-REG_005fBR_005fPRED"></a>
</dd>
<dt><code>REG_BR_PRED</code></dt>
<dd><p>These notes are found in JUMP insns after delayed branch scheduling
has taken place.  They indicate both the direction and the likelihood
of the JUMP.  The format is a bitmask of ATTR_FLAG_* values.
</p>
<a name="index-REG_005fFRAME_005fRELATED_005fEXPR"></a>
</dd>
<dt><code>REG_FRAME_RELATED_EXPR</code></dt>
<dd><p>This is used on an RTX_FRAME_RELATED_P insn wherein the attached expression
is used in place of the actual insn pattern.  This is done in cases where
the pattern is either complex or misleading.
</p></dd>
</dl>

<p>For convenience, the machine mode in an <code>insn_list</code> or
<code>expr_list</code> is printed using these symbolic codes in debugging dumps.
</p>
<a name="index-insn_005flist"></a>
<a name="index-expr_005flist"></a>
<p>The only difference between the expression codes <code>insn_list</code> and
<code>expr_list</code> is that the first operand of an <code>insn_list</code> is
assumed to be an insn and is printed in debugging dumps as the insn&rsquo;s
unique id; the first operand of an <code>expr_list</code> is printed in the
ordinary way as an expression.
</p>
<hr>
<a name="Calls"></a>
<div class="header">
<p>
Next: <a href="#Sharing" accesskey="n" rel="next">Sharing</a>, Previous: <a href="#Insns" accesskey="p" rel="prev">Insns</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Representation-of-Function_002dCall-Insns"></a>
<h3 class="section">10.20 RTL Representation of Function-Call Insns</h3>
<a name="index-calling-functions-in-RTL"></a>
<a name="index-RTL-function_002dcall-insns"></a>
<a name="index-function_002dcall-insns"></a>

<p>Insns that call subroutines have the RTL expression code <code>call_insn</code>.
These insns must satisfy special rules, and their bodies must use a special
RTL expression code, <code>call</code>.
</p>
<a name="index-call-usage"></a>
<p>A <code>call</code> expression has two operands, as follows:
</p>
<div class="smallexample">
<pre class="smallexample">(call (mem:<var>fm</var> <var>addr</var>) <var>nbytes</var>)
</pre></div>

<p>Here <var>nbytes</var> is an operand that represents the number of bytes of
argument data being passed to the subroutine, <var>fm</var> is a machine mode
(which must equal as the definition of the <code>FUNCTION_MODE</code> macro in
the machine description) and <var>addr</var> represents the address of the
subroutine.
</p>
<p>For a subroutine that returns no value, the <code>call</code> expression as
shown above is the entire body of the insn, except that the insn might
also contain <code>use</code> or <code>clobber</code> expressions.
</p>
<a name="index-BLKmode_002c-and-function-return-values"></a>
<p>For a subroutine that returns a value whose mode is not <code>BLKmode</code>,
the value is returned in a hard register.  If this register&rsquo;s number is
<var>r</var>, then the body of the call insn looks like this:
</p>
<div class="smallexample">
<pre class="smallexample">(set (reg:<var>m</var> <var>r</var>)
     (call (mem:<var>fm</var> <var>addr</var>) <var>nbytes</var>))
</pre></div>

<p>This RTL expression makes it clear (to the optimizer passes) that the
appropriate register receives a useful value in this insn.
</p>
<p>When a subroutine returns a <code>BLKmode</code> value, it is handled by
passing to the subroutine the address of a place to store the value.
So the call insn itself does not &ldquo;return&rdquo; any value, and it has the
same RTL form as a call that returns nothing.
</p>
<p>On some machines, the call instruction itself clobbers some register,
for example to contain the return address.  <code>call_insn</code> insns
on these machines should have a body which is a <code>parallel</code>
that contains both the <code>call</code> expression and <code>clobber</code>
expressions that indicate which registers are destroyed.  Similarly,
if the call instruction requires some register other than the stack
pointer that is not explicitly mentioned in its RTL, a <code>use</code>
subexpression should mention that register.
</p>
<p>Functions that are called are assumed to modify all registers listed in
the configuration macro <code>CALL_USED_REGISTERS</code> (see <a href="#Register-Basics">Register Basics</a>) and, with the exception of <code>const</code> functions and library
calls, to modify all of memory.
</p>
<p>Insns containing just <code>use</code> expressions directly precede the
<code>call_insn</code> insn to indicate which registers contain inputs to the
function.  Similarly, if registers other than those in
<code>CALL_USED_REGISTERS</code> are clobbered by the called function, insns
containing a single <code>clobber</code> follow immediately after the call to
indicate which registers.
</p>
<hr>
<a name="Sharing"></a>
<div class="header">
<p>
Next: <a href="#Reading-RTL" accesskey="n" rel="next">Reading RTL</a>, Previous: <a href="#Calls" accesskey="p" rel="prev">Calls</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Structure-Sharing-Assumptions"></a>
<h3 class="section">10.21 Structure Sharing Assumptions</h3>
<a name="index-sharing-of-RTL-components"></a>
<a name="index-RTL-structure-sharing-assumptions"></a>

<p>The compiler assumes that certain kinds of RTL expressions are unique;
there do not exist two distinct objects representing the same value.
In other cases, it makes an opposite assumption: that no RTL expression
object of a certain kind appears in more than one place in the
containing structure.
</p>
<p>These assumptions refer to a single function; except for the RTL
objects that describe global variables and external functions,
and a few standard objects such as small integer constants,
no RTL objects are common to two functions.
</p>
<ul>
<li> <a name="index-reg_002c-RTL-sharing"></a>
Each pseudo-register has only a single <code>reg</code> object to represent it,
and therefore only a single machine mode.

</li><li> <a name="index-symbolic-label"></a>
<a name="index-symbol_005fref_002c-RTL-sharing"></a>
For any symbolic label, there is only one <code>symbol_ref</code> object
referring to it.

</li><li> <a name="index-const_005fint_002c-RTL-sharing"></a>
All <code>const_int</code> expressions with equal values are shared.

</li><li> <a name="index-pc_002c-RTL-sharing"></a>
There is only one <code>pc</code> expression.

</li><li> <a name="index-cc0_002c-RTL-sharing"></a>
There is only one <code>cc0</code> expression.

</li><li> <a name="index-const_005fdouble_002c-RTL-sharing"></a>
There is only one <code>const_double</code> expression with value 0 for
each floating point mode.  Likewise for values 1 and 2.

</li><li> <a name="index-const_005fvector_002c-RTL-sharing"></a>
There is only one <code>const_vector</code> expression with value 0 for
each vector mode, be it an integer or a double constant vector.

</li><li> <a name="index-label_005fref_002c-RTL-sharing"></a>
<a name="index-scratch_002c-RTL-sharing"></a>
No <code>label_ref</code> or <code>scratch</code> appears in more than one place in
the RTL structure; in other words, it is safe to do a tree-walk of all
the insns in the function and assume that each time a <code>label_ref</code>
or <code>scratch</code> is seen it is distinct from all others that are seen.

</li><li> <a name="index-mem_002c-RTL-sharing"></a>
Only one <code>mem</code> object is normally created for each static
variable or stack slot, so these objects are frequently shared in all
the places they appear.  However, separate but equal objects for these
variables are occasionally made.

</li><li> <a name="index-asm_005foperands_002c-RTL-sharing"></a>
When a single <code>asm</code> statement has multiple output operands, a
distinct <code>asm_operands</code> expression is made for each output operand.
However, these all share the vector which contains the sequence of input
operands.  This sharing is used later on to test whether two
<code>asm_operands</code> expressions come from the same statement, so all
optimizations must carefully preserve the sharing if they copy the
vector at all.

</li><li> No RTL object appears in more than one place in the RTL structure
except as described above.  Many passes of the compiler rely on this
by assuming that they can modify RTL objects in place without unwanted
side-effects on other insns.

</li><li> <a name="index-unshare_005fall_005frtl"></a>
During initial RTL generation, shared structure is freely introduced.
After all the RTL for a function has been generated, all shared
structure is copied by <code>unshare_all_rtl</code> in <samp>emit-rtl.c</samp>,
after which the above rules are guaranteed to be followed.

</li><li> <a name="index-copy_005frtx_005fif_005fshared"></a>
During the combiner pass, shared structure within an insn can exist
temporarily.  However, the shared structure is copied before the
combiner is finished with the insn.  This is done by calling
<code>copy_rtx_if_shared</code>, which is a subroutine of
<code>unshare_all_rtl</code>.
</li></ul>

<hr>
<a name="Reading-RTL"></a>
<div class="header">
<p>
Previous: <a href="#Sharing" accesskey="p" rel="prev">Sharing</a>, Up: <a href="#RTL" accesskey="u" rel="up">RTL</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Reading-RTL-1"></a>
<h3 class="section">10.22 Reading RTL</h3>

<p>To read an RTL object from a file, call <code>read_rtx</code>.  It takes one
argument, a stdio stream, and returns a single RTL object.  This routine
is defined in <samp>read-rtl.c</samp>.  It is not available in the compiler
itself, only the various programs that generate the compiler back end
from the machine description.
</p>
<p>People frequently have the idea of using RTL stored as text in a file as
an interface between a language front end and the bulk of GCC.  This
idea is not feasible.
</p>
<p>GCC was designed to use RTL internally only.  Correct RTL for a given
program is very dependent on the particular target machine.  And the RTL
does not contain all the information about the program.
</p>
<p>The proper way to interface GCC to a new language front end is with
the &ldquo;tree&rdquo; data structure, described in the files <samp>tree.h</samp> and
<samp>tree.def</samp>.  The documentation for this structure (see <a href="#GENERIC">GENERIC</a>)
is incomplete.
</p>

<hr>
<a name="GENERIC"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE" accesskey="n" rel="next">GIMPLE</a>, Previous: <a href="#RTL" accesskey="p" rel="prev">RTL</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GENERIC-1"></a>
<h2 class="chapter">11 GENERIC</h2>
<a name="index-GENERIC-1"></a>

<p>The purpose of GENERIC is simply to provide a
language-independent way of representing an entire function in
trees.  To this end, it was necessary to add a few new tree codes
to the back end, but most everything was already there.  If you
can express it with the codes in <code>gcc/tree.def</code>, it&rsquo;s
GENERIC.
</p>
<p>Early on, there was a great deal of debate about how to think
about statements in a tree IL.  In GENERIC, a statement is
defined as any expression whose value, if any, is ignored.  A
statement will always have <code>TREE_SIDE_EFFECTS</code> set (or it
will be discarded), but a non-statement expression may also have
side effects.  A <code>CALL_EXPR</code>, for instance.
</p>
<p>It would be possible for some local optimizations to work on the
GENERIC form of a function; indeed, the adapted tree inliner
works fine on GENERIC, but the current compiler performs inlining
after lowering to GIMPLE (a restricted form described in the next
section). Indeed, currently the frontends perform this lowering
before handing off to <code>tree_rest_of_compilation</code>, but this
seems inelegant.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Deficiencies" accesskey="1">Deficiencies</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Topics net yet covered in this document.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tree-overview" accesskey="2">Tree overview</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">All about <code>tree</code>s.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Types" accesskey="3">Types</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Fundamental and aggregate types.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Declarations" accesskey="4">Declarations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Type declarations and variables.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Attributes" accesskey="5">Attributes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Declaration and type attributes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Expression-trees" accesskey="6">Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Operating on data.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Statements" accesskey="7">Statements</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Control flow and related trees.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Functions" accesskey="8">Functions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Function bodies, linkage, and other aspects.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Language_002ddependent-trees" accesskey="9">Language-dependent trees</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Topics and trees specific to language front ends.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#C-and-C_002b_002b-Trees">C and C++ Trees</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Trees specific to C and C++.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Java-Trees">Java Trees</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Trees specific to Java.
</td></tr>
</table>


<hr>
<a name="Deficiencies"></a>
<div class="header">
<p>
Next: <a href="#Tree-overview" accesskey="n" rel="next">Tree overview</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Deficiencies-1"></a>
<h3 class="section">11.1 Deficiencies</h3>

<p>There are many places in which this document is incomplet and incorrekt.
It is, as of yet, only <em>preliminary</em> documentation.
</p>

<hr>
<a name="Tree-overview"></a>
<div class="header">
<p>
Next: <a href="#Types" accesskey="n" rel="next">Types</a>, Previous: <a href="#Deficiencies" accesskey="p" rel="prev">Deficiencies</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Overview-1"></a>
<h3 class="section">11.2 Overview</h3>
<a name="index-tree"></a>
<a name="index-TREE_005fCODE"></a>

<p>The central data structure used by the internal representation is the
<code>tree</code>.  These nodes, while all of the C type <code>tree</code>, are of
many varieties.  A <code>tree</code> is a pointer type, but the object to
which it points may be of a variety of types.  From this point forward,
we will refer to trees in ordinary type, rather than in <code>this
font</code>, except when talking about the actual C type <code>tree</code>.
</p>
<p>You can tell what kind of node a particular tree is by using the
<code>TREE_CODE</code> macro.  Many, many macros take trees as input and
return trees as output.  However, most macros require a certain kind of
tree node as input.  In other words, there is a type-system for trees,
but it is not reflected in the C type-system.
</p>
<p>For safety, it is useful to configure GCC with <samp>--enable-checking</samp>.
Although this results in a significant performance penalty (since all
tree types are checked at run-time), and is therefore inappropriate in a
release version, it is extremely helpful during the development process.
</p>
<p>Many macros behave as predicates.  Many, although not all, of these
predicates end in &lsquo;<samp>_P</samp>&rsquo;.  Do not rely on the result type of these
macros being of any particular type.  You may, however, rely on the fact
that the type can be compared to <code>0</code>, so that statements like
</p><div class="smallexample">
<pre class="smallexample">if (TEST_P (t) &amp;&amp; !TEST_P (y))
  x = 1;
</pre></div>
<p>and
</p><div class="smallexample">
<pre class="smallexample">int i = (TEST_P (t) != 0);
</pre></div>
<p>are legal.  Macros that return <code>int</code> values now may be changed to
return <code>tree</code> values, or other pointers in the future.  Even those
that continue to return <code>int</code> may return multiple nonzero codes
where previously they returned only zero and one.  Therefore, you should
not write code like
</p><div class="smallexample">
<pre class="smallexample">if (TEST_P (t) == 1)
</pre></div>
<p>as this code is not guaranteed to work correctly in the future.
</p>
<p>You should not take the address of values returned by the macros or
functions described here.  In particular, no guarantee is given that the
values are lvalues.
</p>
<p>In general, the names of macros are all in uppercase, while the names of
functions are entirely in lowercase.  There are rare exceptions to this
rule.  You should assume that any macro or function whose name is made
up entirely of uppercase letters may evaluate its arguments more than
once.  You may assume that a macro or function whose name is made up
entirely of lowercase letters will evaluate its arguments only once.
</p>
<p>The <code>error_mark_node</code> is a special tree.  Its tree code is
<code>ERROR_MARK</code>, but since there is only ever one node with that code,
the usual practice is to compare the tree against
<code>error_mark_node</code>.  (This test is just a test for pointer
equality.)  If an error has occurred during front-end processing the
flag <code>errorcount</code> will be set.  If the front end has encountered
code it cannot handle, it will issue a message to the user and set
<code>sorrycount</code>.  When these flags are set, any macro or function
which normally returns a tree of a particular kind may instead return
the <code>error_mark_node</code>.  Thus, if you intend to do any processing of
erroneous code, you must be prepared to deal with the
<code>error_mark_node</code>.
</p>
<p>Occasionally, a particular tree slot (like an operand to an expression,
or a particular field in a declaration) will be referred to as
&ldquo;reserved for the back end&rdquo;.  These slots are used to store RTL when
the tree is converted to RTL for use by the GCC back end.  However, if
that process is not taking place (e.g., if the front end is being hooked
up to an intelligent editor), then those slots may be used by the
back end presently in use.
</p>
<p>If you encounter situations that do not match this documentation, such
as tree nodes of types not mentioned here, or macros documented to
return entities of a particular kind that instead return entities of
some different kind, you have found a bug, either in the front end or in
the documentation.  Please report these bugs as you would any other
bug.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Macros-and-Functions" accesskey="1">Macros and Functions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros and functions that can be used with all trees.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Identifiers" accesskey="2">Identifiers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The names of things.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Containers" accesskey="3">Containers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Lists and vectors.
</td></tr>
</table>


<hr>
<a name="Macros-and-Functions"></a>
<div class="header">
<p>
Next: <a href="#Identifiers" accesskey="n" rel="next">Identifiers</a>, Up: <a href="#Tree-overview" accesskey="u" rel="up">Tree overview</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Trees"></a>
<h4 class="subsection">11.2.1 Trees</h4>
<a name="index-tree-1"></a>
<a name="index-TREE_005fCHAIN"></a>
<a name="index-TREE_005fTYPE"></a>

<p>All GENERIC trees have two fields in common.  First, <code>TREE_CHAIN</code>
is a pointer that can be used as a singly-linked list to other trees.
The other is <code>TREE_TYPE</code>.  Many trees store the type of an
expression or declaration in this field.
</p>
<p>These are some other functions for handling trees:
</p>
<dl compact="compact">
<dt><code>tree_size</code>
<a name="index-tree_005fsize"></a>
</dt>
<dd><p>Return the number of bytes a tree takes.
</p>
</dd>
<dt><code>build0</code>
<a name="index-build0"></a>
</dt>
<dt><code>build1</code>
<a name="index-build1"></a>
</dt>
<dt><code>build2</code>
<a name="index-build2"></a>
</dt>
<dt><code>build3</code>
<a name="index-build3"></a>
</dt>
<dt><code>build4</code>
<a name="index-build4"></a>
</dt>
<dt><code>build5</code>
<a name="index-build5"></a>
</dt>
<dt><code>build6</code>
<a name="index-build6"></a>
</dt>
<dd>
<p>These functions build a tree and supply values to put in each
parameter.  The basic signature is &lsquo;<samp>code,&nbsp;type,&nbsp;[operands]<!-- /@w --></samp>&rsquo;.
<code>code</code> is the <code>TREE_CODE</code>, and <code>type</code> is a tree
representing the <code>TREE_TYPE</code>.  These are followed by the
operands, each of which is also a tree.
</p>
</dd>
</dl>



<hr>
<a name="Identifiers"></a>
<div class="header">
<p>
Next: <a href="#Containers" accesskey="n" rel="next">Containers</a>, Previous: <a href="#Macros-and-Functions" accesskey="p" rel="prev">Macros and Functions</a>, Up: <a href="#Tree-overview" accesskey="u" rel="up">Tree overview</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Identifiers-1"></a>
<h4 class="subsection">11.2.2 Identifiers</h4>
<a name="index-identifier"></a>
<a name="index-name"></a>
<a name="index-IDENTIFIER_005fNODE"></a>

<p>An <code>IDENTIFIER_NODE</code> represents a slightly more general concept
that the standard C or C++ concept of identifier.  In particular, an
<code>IDENTIFIER_NODE</code> may contain a &lsquo;<samp>$</samp>&rsquo;, or other extraordinary
characters.
</p>
<p>There are never two distinct <code>IDENTIFIER_NODE</code>s representing the
same identifier.  Therefore, you may use pointer equality to compare
<code>IDENTIFIER_NODE</code>s, rather than using a routine like
<code>strcmp</code>.  Use <code>get_identifier</code> to obtain the unique
<code>IDENTIFIER_NODE</code> for a supplied string.
</p>
<p>You can use the following macros to access identifiers:
</p><dl compact="compact">
<dt><code>IDENTIFIER_POINTER</code>
<a name="index-IDENTIFIER_005fPOINTER"></a>
</dt>
<dd><p>The string represented by the identifier, represented as a
<code>char*</code>.  This string is always <code>NUL</code>-terminated, and contains
no embedded <code>NUL</code> characters.
</p>
</dd>
<dt><code>IDENTIFIER_LENGTH</code>
<a name="index-IDENTIFIER_005fLENGTH"></a>
</dt>
<dd><p>The length of the string returned by <code>IDENTIFIER_POINTER</code>, not
including the trailing <code>NUL</code>.  This value of
<code>IDENTIFIER_LENGTH (x)</code> is always the same as <code>strlen
(IDENTIFIER_POINTER (x))</code>.
</p>
</dd>
<dt><code>IDENTIFIER_OPNAME_P</code>
<a name="index-IDENTIFIER_005fOPNAME_005fP"></a>
</dt>
<dd><p>This predicate holds if the identifier represents the name of an
overloaded operator.  In this case, you should not depend on the
contents of either the <code>IDENTIFIER_POINTER</code> or the
<code>IDENTIFIER_LENGTH</code>.
</p>
</dd>
<dt><code>IDENTIFIER_TYPENAME_P</code>
<a name="index-IDENTIFIER_005fTYPENAME_005fP"></a>
</dt>
<dd><p>This predicate holds if the identifier represents the name of a
user-defined conversion operator.  In this case, the <code>TREE_TYPE</code> of
the <code>IDENTIFIER_NODE</code> holds the type to which the conversion
operator converts.
</p>
</dd>
</dl>


<hr>
<a name="Containers"></a>
<div class="header">
<p>
Previous: <a href="#Identifiers" accesskey="p" rel="prev">Identifiers</a>, Up: <a href="#Tree-overview" accesskey="u" rel="up">Tree overview</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Containers-1"></a>
<h4 class="subsection">11.2.3 Containers</h4>
<a name="index-container"></a>
<a name="index-list"></a>
<a name="index-vector"></a>
<a name="index-TREE_005fLIST"></a>
<a name="index-TREE_005fVEC"></a>
<a name="index-TREE_005fPURPOSE"></a>
<a name="index-TREE_005fVALUE"></a>
<a name="index-TREE_005fVEC_005fLENGTH"></a>
<a name="index-TREE_005fVEC_005fELT"></a>

<p>Two common container data structures can be represented directly with
tree nodes.  A <code>TREE_LIST</code> is a singly linked list containing two
trees per node.  These are the <code>TREE_PURPOSE</code> and <code>TREE_VALUE</code>
of each node.  (Often, the <code>TREE_PURPOSE</code> contains some kind of
tag, or additional information, while the <code>TREE_VALUE</code> contains the
majority of the payload.  In other cases, the <code>TREE_PURPOSE</code> is
simply <code>NULL_TREE</code>, while in still others both the
<code>TREE_PURPOSE</code> and <code>TREE_VALUE</code> are of equal stature.)  Given
one <code>TREE_LIST</code> node, the next node is found by following the
<code>TREE_CHAIN</code>.  If the <code>TREE_CHAIN</code> is <code>NULL_TREE</code>, then
you have reached the end of the list.
</p>
<p>A <code>TREE_VEC</code> is a simple vector.  The <code>TREE_VEC_LENGTH</code> is an
integer (not a tree) giving the number of nodes in the vector.  The
nodes themselves are accessed using the <code>TREE_VEC_ELT</code> macro, which
takes two arguments.  The first is the <code>TREE_VEC</code> in question; the
second is an integer indicating which element in the vector is desired.
The elements are indexed from zero.
</p>

<hr>
<a name="Types"></a>
<div class="header">
<p>
Next: <a href="#Declarations" accesskey="n" rel="next">Declarations</a>, Previous: <a href="#Tree-overview" accesskey="p" rel="prev">Tree overview</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Types-1"></a>
<h3 class="section">11.3 Types</h3>
<a name="index-type"></a>
<a name="index-pointer"></a>
<a name="index-reference"></a>
<a name="index-fundamental-type"></a>
<a name="index-array"></a>
<a name="index-VOID_005fTYPE"></a>
<a name="index-INTEGER_005fTYPE"></a>
<a name="index-TYPE_005fMIN_005fVALUE"></a>
<a name="index-TYPE_005fMAX_005fVALUE"></a>
<a name="index-REAL_005fTYPE"></a>
<a name="index-FIXED_005fPOINT_005fTYPE"></a>
<a name="index-COMPLEX_005fTYPE"></a>
<a name="index-ENUMERAL_005fTYPE"></a>
<a name="index-BOOLEAN_005fTYPE"></a>
<a name="index-POINTER_005fTYPE"></a>
<a name="index-REFERENCE_005fTYPE"></a>
<a name="index-FUNCTION_005fTYPE"></a>
<a name="index-METHOD_005fTYPE"></a>
<a name="index-ARRAY_005fTYPE"></a>
<a name="index-RECORD_005fTYPE"></a>
<a name="index-UNION_005fTYPE"></a>
<a name="index-UNKNOWN_005fTYPE"></a>
<a name="index-OFFSET_005fTYPE"></a>
<a name="index-TYPE_005fUNQUALIFIED"></a>
<a name="index-TYPE_005fQUAL_005fCONST"></a>
<a name="index-TYPE_005fQUAL_005fVOLATILE"></a>
<a name="index-TYPE_005fQUAL_005fRESTRICT"></a>
<a name="index-TYPE_005fMAIN_005fVARIANT"></a>
<a name="index-qualified-type"></a>
<a name="index-TYPE_005fSIZE"></a>
<a name="index-TYPE_005fALIGN"></a>
<a name="index-TYPE_005fPRECISION"></a>
<a name="index-TYPE_005fARG_005fTYPES"></a>
<a name="index-TYPE_005fMETHOD_005fBASETYPE"></a>
<a name="index-TYPE_005fOFFSET_005fBASETYPE"></a>
<a name="index-TREE_005fTYPE-1"></a>
<a name="index-TYPE_005fCONTEXT"></a>
<a name="index-TYPE_005fNAME"></a>
<a name="index-TYPENAME_005fTYPE_005fFULLNAME"></a>
<a name="index-TYPE_005fFIELDS"></a>
<a name="index-TYPE_005fCANONICAL"></a>
<a name="index-TYPE_005fSTRUCTURAL_005fEQUALITY_005fP"></a>
<a name="index-SET_005fTYPE_005fSTRUCTURAL_005fEQUALITY"></a>

<p>All types have corresponding tree nodes.  However, you should not assume
that there is exactly one tree node corresponding to each type.  There
are often multiple nodes corresponding to the same type.
</p>
<p>For the most part, different kinds of types have different tree codes.
(For example, pointer types use a <code>POINTER_TYPE</code> code while arrays
use an <code>ARRAY_TYPE</code> code.)  However, pointers to member functions
use the <code>RECORD_TYPE</code> code.  Therefore, when writing a
<code>switch</code> statement that depends on the code associated with a
particular type, you should take care to handle pointers to member
functions under the <code>RECORD_TYPE</code> case label.
</p>
<p>The following functions and macros deal with cv-qualification of types:
</p><dl compact="compact">
<dt><code>TYPE_MAIN_VARIANT</code>
<a name="index-TYPE_005fMAIN_005fVARIANT-1"></a>
</dt>
<dd><p>This macro returns the unqualified version of a type.  It may be applied
to an unqualified type, but it is not always the identity function in
that case.
</p></dd>
</dl>

<p>A few other macros and functions are usable with all types:
</p><dl compact="compact">
<dt><code>TYPE_SIZE</code>
<a name="index-TYPE_005fSIZE-1"></a>
</dt>
<dd><p>The number of bits required to represent the type, represented as an
<code>INTEGER_CST</code>.  For an incomplete type, <code>TYPE_SIZE</code> will be
<code>NULL_TREE</code>.
</p>
</dd>
<dt><code>TYPE_ALIGN</code>
<a name="index-TYPE_005fALIGN-1"></a>
</dt>
<dd><p>The alignment of the type, in bits, represented as an <code>int</code>.
</p>
</dd>
<dt><code>TYPE_NAME</code>
<a name="index-TYPE_005fNAME-1"></a>
</dt>
<dd><p>This macro returns a declaration (in the form of a <code>TYPE_DECL</code>) for
the type.  (Note this macro does <em>not</em> return an
<code>IDENTIFIER_NODE</code>, as you might expect, given its name!)  You can
look at the <code>DECL_NAME</code> of the <code>TYPE_DECL</code> to obtain the
actual name of the type.  The <code>TYPE_NAME</code> will be <code>NULL_TREE</code>
for a type that is not a built-in type, the result of a typedef, or a
named class type.
</p>
</dd>
<dt><code>TYPE_CANONICAL</code>
<a name="index-TYPE_005fCANONICAL-1"></a>
</dt>
<dd><p>This macro returns the &ldquo;canonical&rdquo; type for the given type
node. Canonical types are used to improve performance in the C++ and
Objective-C++ front ends by allowing efficient comparison between two
type nodes in <code>same_type_p</code>: if the <code>TYPE_CANONICAL</code> values
of the types are equal, the types are equivalent; otherwise, the types
are not equivalent. The notion of equivalence for canonical types is
the same as the notion of type equivalence in the language itself. For
instance,
</p>
<p>When <code>TYPE_CANONICAL</code> is <code>NULL_TREE</code>, there is no canonical
type for the given type node. In this case, comparison between this
type and any other type requires the compiler to perform a deep,
&ldquo;structural&rdquo; comparison to see if the two type nodes have the same
form and properties.
</p>
<p>The canonical type for a node is always the most fundamental type in
the equivalence class of types. For instance, <code>int</code> is its own
canonical type. A typedef <code>I</code> of <code>int</code> will have <code>int</code>
as its canonical type. Similarly, <code>I*</code>&nbsp;and a typedef <code>IP</code>&nbsp;(defined to <code>I*</code>) will has <code>int*</code> as their canonical
type. When building a new type node, be sure to set
<code>TYPE_CANONICAL</code> to the appropriate canonical type. If the new
type is a compound type (built from other types), and any of those
other types require structural equality, use
<code>SET_TYPE_STRUCTURAL_EQUALITY</code> to ensure that the new type also
requires structural equality. Finally, if for some reason you cannot
guarantee that <code>TYPE_CANONICAL</code> will point to the canonical type,
use <code>SET_TYPE_STRUCTURAL_EQUALITY</code> to make sure that the new
type&ndash;and any type constructed based on it&ndash;requires structural
equality. If you suspect that the canonical type system is
miscomparing types, pass <code>--param verify-canonical-types=1</code> to
the compiler or configure with <code>--enable-checking</code> to force the
compiler to verify its canonical-type comparisons against the
structural comparisons; the compiler will then print any warnings if
the canonical types miscompare.
</p>
</dd>
<dt><code>TYPE_STRUCTURAL_EQUALITY_P</code>
<a name="index-TYPE_005fSTRUCTURAL_005fEQUALITY_005fP-1"></a>
</dt>
<dd><p>This predicate holds when the node requires structural equality
checks, e.g., when <code>TYPE_CANONICAL</code> is <code>NULL_TREE</code>.
</p>
</dd>
<dt><code>SET_TYPE_STRUCTURAL_EQUALITY</code>
<a name="index-SET_005fTYPE_005fSTRUCTURAL_005fEQUALITY-1"></a>
</dt>
<dd><p>This macro states that the type node it is given requires structural
equality checks, e.g., it sets <code>TYPE_CANONICAL</code> to
<code>NULL_TREE</code>.
</p>
</dd>
<dt><code>same_type_p</code>
<a name="index-same_005ftype_005fp"></a>
</dt>
<dd><p>This predicate takes two types as input, and holds if they are the same
type.  For example, if one type is a <code>typedef</code> for the other, or
both are <code>typedef</code>s for the same type.  This predicate also holds if
the two trees given as input are simply copies of one another; i.e.,
there is no difference between them at the source level, but, for
whatever reason, a duplicate has been made in the representation.  You
should never use <code>==</code> (pointer equality) to compare types; always
use <code>same_type_p</code> instead.
</p></dd>
</dl>

<p>Detailed below are the various kinds of types, and the macros that can
be used to access them.  Although other kinds of types are used
elsewhere in G++, the types described here are the only ones that you
will encounter while examining the intermediate representation.
</p>
<dl compact="compact">
<dt><code>VOID_TYPE</code></dt>
<dd><p>Used to represent the <code>void</code> type.
</p>
</dd>
<dt><code>INTEGER_TYPE</code></dt>
<dd><p>Used to represent the various integral types, including <code>char</code>,
<code>short</code>, <code>int</code>, <code>long</code>, and <code>long long</code>.  This code
is not used for enumeration types, nor for the <code>bool</code> type.
The <code>TYPE_PRECISION</code> is the number of bits used in
the representation, represented as an <code>unsigned int</code>.  (Note that
in the general case this is not the same value as <code>TYPE_SIZE</code>;
suppose that there were a 24-bit integer type, but that alignment
requirements for the ABI required 32-bit alignment.  Then,
<code>TYPE_SIZE</code> would be an <code>INTEGER_CST</code> for 32, while
<code>TYPE_PRECISION</code> would be 24.)  The integer type is unsigned if
<code>TYPE_UNSIGNED</code> holds; otherwise, it is signed.
</p>
<p>The <code>TYPE_MIN_VALUE</code> is an <code>INTEGER_CST</code> for the smallest
integer that may be represented by this type.  Similarly, the
<code>TYPE_MAX_VALUE</code> is an <code>INTEGER_CST</code> for the largest integer
that may be represented by this type.
</p>
</dd>
<dt><code>REAL_TYPE</code></dt>
<dd><p>Used to represent the <code>float</code>, <code>double</code>, and <code>long
double</code> types.  The number of bits in the floating-point representation
is given by <code>TYPE_PRECISION</code>, as in the <code>INTEGER_TYPE</code> case.
</p>
</dd>
<dt><code>FIXED_POINT_TYPE</code></dt>
<dd><p>Used to represent the <code>short _Fract</code>, <code>_Fract</code>, <code>long
_Fract</code>, <code>long long _Fract</code>, <code>short _Accum</code>, <code>_Accum</code>,
<code>long _Accum</code>, and <code>long long _Accum</code> types.  The number of bits
in the fixed-point representation is given by <code>TYPE_PRECISION</code>,
as in the <code>INTEGER_TYPE</code> case.  There may be padding bits, fractional
bits and integral bits.  The number of fractional bits is given by
<code>TYPE_FBIT</code>, and the number of integral bits is given by <code>TYPE_IBIT</code>.
The fixed-point type is unsigned if <code>TYPE_UNSIGNED</code> holds; otherwise,
it is signed.
The fixed-point type is saturating if <code>TYPE_SATURATING</code> holds; otherwise,
it is not saturating.
</p>
</dd>
<dt><code>COMPLEX_TYPE</code></dt>
<dd><p>Used to represent GCC built-in <code>__complex__</code> data types.  The
<code>TREE_TYPE</code> is the type of the real and imaginary parts.
</p>
</dd>
<dt><code>ENUMERAL_TYPE</code></dt>
<dd><p>Used to represent an enumeration type.  The <code>TYPE_PRECISION</code> gives
(as an <code>int</code>), the number of bits used to represent the type.  If
there are no negative enumeration constants, <code>TYPE_UNSIGNED</code> will
hold.  The minimum and maximum enumeration constants may be obtained
with <code>TYPE_MIN_VALUE</code> and <code>TYPE_MAX_VALUE</code>, respectively; each
of these macros returns an <code>INTEGER_CST</code>.
</p>
<p>The actual enumeration constants themselves may be obtained by looking
at the <code>TYPE_VALUES</code>.  This macro will return a <code>TREE_LIST</code>,
containing the constants.  The <code>TREE_PURPOSE</code> of each node will be
an <code>IDENTIFIER_NODE</code> giving the name of the constant; the
<code>TREE_VALUE</code> will be an <code>INTEGER_CST</code> giving the value
assigned to that constant.  These constants will appear in the order in
which they were declared.  The <code>TREE_TYPE</code> of each of these
constants will be the type of enumeration type itself.
</p>
</dd>
<dt><code>BOOLEAN_TYPE</code></dt>
<dd><p>Used to represent the <code>bool</code> type.
</p>
</dd>
<dt><code>POINTER_TYPE</code></dt>
<dd><p>Used to represent pointer types, and pointer to data member types.  The
<code>TREE_TYPE</code> gives the type to which this type points.
</p>
</dd>
<dt><code>REFERENCE_TYPE</code></dt>
<dd><p>Used to represent reference types.  The <code>TREE_TYPE</code> gives the type
to which this type refers.
</p>
</dd>
<dt><code>FUNCTION_TYPE</code></dt>
<dd><p>Used to represent the type of non-member functions and of static member
functions.  The <code>TREE_TYPE</code> gives the return type of the function.
The <code>TYPE_ARG_TYPES</code> are a <code>TREE_LIST</code> of the argument types.
The <code>TREE_VALUE</code> of each node in this list is the type of the
corresponding argument; the <code>TREE_PURPOSE</code> is an expression for the
default argument value, if any.  If the last node in the list is
<code>void_list_node</code> (a <code>TREE_LIST</code> node whose <code>TREE_VALUE</code>
is the <code>void_type_node</code>), then functions of this type do not take
variable arguments.  Otherwise, they do take a variable number of
arguments.
</p>
<p>Note that in C (but not in C++) a function declared like <code>void f()</code>
is an unprototyped function taking a variable number of arguments; the
<code>TYPE_ARG_TYPES</code> of such a function will be <code>NULL</code>.
</p>
</dd>
<dt><code>METHOD_TYPE</code></dt>
<dd><p>Used to represent the type of a non-static member function.  Like a
<code>FUNCTION_TYPE</code>, the return type is given by the <code>TREE_TYPE</code>.
The type of <code>*this</code>, i.e., the class of which functions of this
type are a member, is given by the <code>TYPE_METHOD_BASETYPE</code>.  The
<code>TYPE_ARG_TYPES</code> is the parameter list, as for a
<code>FUNCTION_TYPE</code>, and includes the <code>this</code> argument.
</p>
</dd>
<dt><code>ARRAY_TYPE</code></dt>
<dd><p>Used to represent array types.  The <code>TREE_TYPE</code> gives the type of
the elements in the array.  If the array-bound is present in the type,
the <code>TYPE_DOMAIN</code> is an <code>INTEGER_TYPE</code> whose
<code>TYPE_MIN_VALUE</code> and <code>TYPE_MAX_VALUE</code> will be the lower and
upper bounds of the array, respectively.  The <code>TYPE_MIN_VALUE</code> will
always be an <code>INTEGER_CST</code> for zero, while the
<code>TYPE_MAX_VALUE</code> will be one less than the number of elements in
the array, i.e., the highest value which may be used to index an element
in the array.
</p>
</dd>
<dt><code>RECORD_TYPE</code></dt>
<dd><p>Used to represent <code>struct</code> and <code>class</code> types, as well as
pointers to member functions and similar constructs in other languages.
<code>TYPE_FIELDS</code> contains the items contained in this type, each of
which can be a <code>FIELD_DECL</code>, <code>VAR_DECL</code>, <code>CONST_DECL</code>, or
<code>TYPE_DECL</code>.  You may not make any assumptions about the ordering
of the fields in the type or whether one or more of them overlap.
</p>
</dd>
<dt><code>UNION_TYPE</code></dt>
<dd><p>Used to represent <code>union</code> types.  Similar to <code>RECORD_TYPE</code>
except that all <code>FIELD_DECL</code> nodes in <code>TYPE_FIELD</code> start at
bit position zero.
</p>
</dd>
<dt><code>QUAL_UNION_TYPE</code></dt>
<dd><p>Used to represent part of a variant record in Ada.  Similar to
<code>UNION_TYPE</code> except that each <code>FIELD_DECL</code> has a
<code>DECL_QUALIFIER</code> field, which contains a boolean expression that
indicates whether the field is present in the object.  The type will only
have one field, so each field&rsquo;s <code>DECL_QUALIFIER</code> is only evaluated
if none of the expressions in the previous fields in <code>TYPE_FIELDS</code>
are nonzero.  Normally these expressions will reference a field in the
outer object using a <code>PLACEHOLDER_EXPR</code>.
</p>
</dd>
<dt><code>LANG_TYPE</code></dt>
<dd><p>This node is used to represent a language-specific type.  The front
end must handle it.
</p>
</dd>
<dt><code>OFFSET_TYPE</code></dt>
<dd><p>This node is used to represent a pointer-to-data member.  For a data
member <code>X::m</code> the <code>TYPE_OFFSET_BASETYPE</code> is <code>X</code> and the
<code>TREE_TYPE</code> is the type of <code>m</code>.
</p>
</dd>
</dl>

<p>There are variables whose values represent some of the basic types.
These include:
</p><dl compact="compact">
<dt><code>void_type_node</code></dt>
<dd><p>A node for <code>void</code>.
</p>
</dd>
<dt><code>integer_type_node</code></dt>
<dd><p>A node for <code>int</code>.
</p>
</dd>
<dt><code>unsigned_type_node.</code></dt>
<dd><p>A node for <code>unsigned int</code>.
</p>
</dd>
<dt><code>char_type_node.</code></dt>
<dd><p>A node for <code>char</code>.
</p></dd>
</dl>
<p>It may sometimes be useful to compare one of these variables with a type
in hand, using <code>same_type_p</code>.
</p>

<hr>
<a name="Declarations"></a>
<div class="header">
<p>
Next: <a href="#Attributes" accesskey="n" rel="next">Attributes</a>, Previous: <a href="#Types" accesskey="p" rel="prev">Types</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Declarations-2"></a>
<h3 class="section">11.4 Declarations</h3>
<a name="index-declaration"></a>
<a name="index-variable"></a>
<a name="index-type-declaration"></a>
<a name="index-LABEL_005fDECL"></a>
<a name="index-CONST_005fDECL"></a>
<a name="index-TYPE_005fDECL"></a>
<a name="index-VAR_005fDECL"></a>
<a name="index-PARM_005fDECL"></a>
<a name="index-DEBUG_005fEXPR_005fDECL"></a>
<a name="index-FIELD_005fDECL"></a>
<a name="index-NAMESPACE_005fDECL"></a>
<a name="index-RESULT_005fDECL"></a>
<a name="index-TEMPLATE_005fDECL"></a>
<a name="index-THUNK_005fDECL"></a>
<a name="index-THUNK_005fDELTA"></a>
<a name="index-DECL_005fINITIAL"></a>
<a name="index-DECL_005fSIZE"></a>
<a name="index-DECL_005fALIGN"></a>
<a name="index-DECL_005fEXTERNAL"></a>

<p>This section covers the various kinds of declarations that appear in the
internal representation, except for declarations of functions
(represented by <code>FUNCTION_DECL</code> nodes), which are described in
<a href="#Functions">Functions</a>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Working-with-declarations" accesskey="1">Working with declarations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros and functions that work on
declarations.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Internal-structure" accesskey="2">Internal structure</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How declaration nodes are represented.
</td></tr>
</table>

<hr>
<a name="Working-with-declarations"></a>
<div class="header">
<p>
Next: <a href="#Internal-structure" accesskey="n" rel="next">Internal structure</a>, Up: <a href="#Declarations" accesskey="u" rel="up">Declarations</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Working-with-declarations-1"></a>
<h4 class="subsection">11.4.1 Working with declarations</h4>

<p>Some macros can be used with any kind of declaration.  These include:
</p><dl compact="compact">
<dt><code>DECL_NAME</code>
<a name="index-DECL_005fNAME"></a>
</dt>
<dd><p>This macro returns an <code>IDENTIFIER_NODE</code> giving the name of the
entity.
</p>
</dd>
<dt><code>TREE_TYPE</code>
<a name="index-TREE_005fTYPE-2"></a>
</dt>
<dd><p>This macro returns the type of the entity declared.
</p>
</dd>
<dt><code>EXPR_FILENAME</code>
<a name="index-EXPR_005fFILENAME"></a>
</dt>
<dd><p>This macro returns the name of the file in which the entity was
declared, as a <code>char*</code>.  For an entity declared implicitly by the
compiler (like <code>__builtin_memcpy</code>), this will be the string
<code>&quot;&lt;internal&gt;&quot;</code>.
</p>
</dd>
<dt><code>EXPR_LINENO</code>
<a name="index-EXPR_005fLINENO"></a>
</dt>
<dd><p>This macro returns the line number at which the entity was declared, as
an <code>int</code>.
</p>
</dd>
<dt><code>DECL_ARTIFICIAL</code>
<a name="index-DECL_005fARTIFICIAL"></a>
</dt>
<dd><p>This predicate holds if the declaration was implicitly generated by the
compiler.  For example, this predicate will hold of an implicitly
declared member function, or of the <code>TYPE_DECL</code> implicitly
generated for a class type.  Recall that in C++ code like:
</p><div class="smallexample">
<pre class="smallexample">struct S {};
</pre></div>
<p>is roughly equivalent to C code like:
</p><div class="smallexample">
<pre class="smallexample">struct S {};
typedef struct S S;
</pre></div>
<p>The implicitly generated <code>typedef</code> declaration is represented by a
<code>TYPE_DECL</code> for which <code>DECL_ARTIFICIAL</code> holds.
</p>
</dd>
</dl>

<p>The various kinds of declarations include:
</p><dl compact="compact">
<dt><code>LABEL_DECL</code></dt>
<dd><p>These nodes are used to represent labels in function bodies.  For more
information, see <a href="#Functions">Functions</a>.  These nodes only appear in block
scopes.
</p>
</dd>
<dt><code>CONST_DECL</code></dt>
<dd><p>These nodes are used to represent enumeration constants.  The value of
the constant is given by <code>DECL_INITIAL</code> which will be an
<code>INTEGER_CST</code> with the same type as the <code>TREE_TYPE</code> of the
<code>CONST_DECL</code>, i.e., an <code>ENUMERAL_TYPE</code>.
</p>
</dd>
<dt><code>RESULT_DECL</code></dt>
<dd><p>These nodes represent the value returned by a function.  When a value is
assigned to a <code>RESULT_DECL</code>, that indicates that the value should
be returned, via bitwise copy, by the function.  You can use
<code>DECL_SIZE</code> and <code>DECL_ALIGN</code> on a <code>RESULT_DECL</code>, just as
with a <code>VAR_DECL</code>.
</p>
</dd>
<dt><code>TYPE_DECL</code></dt>
<dd><p>These nodes represent <code>typedef</code> declarations.  The <code>TREE_TYPE</code>
is the type declared to have the name given by <code>DECL_NAME</code>.  In
some cases, there is no associated name.
</p>
</dd>
<dt><code>VAR_DECL</code></dt>
<dd><p>These nodes represent variables with namespace or block scope, as well
as static data members.  The <code>DECL_SIZE</code> and <code>DECL_ALIGN</code> are
analogous to <code>TYPE_SIZE</code> and <code>TYPE_ALIGN</code>.  For a declaration,
you should always use the <code>DECL_SIZE</code> and <code>DECL_ALIGN</code> rather
than the <code>TYPE_SIZE</code> and <code>TYPE_ALIGN</code> given by the
<code>TREE_TYPE</code>, since special attributes may have been applied to the
variable to give it a particular size and alignment.  You may use the
predicates <code>DECL_THIS_STATIC</code> or <code>DECL_THIS_EXTERN</code> to test
whether the storage class specifiers <code>static</code> or <code>extern</code> were
used to declare a variable.
</p>
<p>If this variable is initialized (but does not require a constructor),
the <code>DECL_INITIAL</code> will be an expression for the initializer.  The
initializer should be evaluated, and a bitwise copy into the variable
performed.  If the <code>DECL_INITIAL</code> is the <code>error_mark_node</code>,
there is an initializer, but it is given by an explicit statement later
in the code; no bitwise copy is required.
</p>
<p>GCC provides an extension that allows either automatic variables, or
global variables, to be placed in particular registers.  This extension
is being used for a particular <code>VAR_DECL</code> if <code>DECL_REGISTER</code>
holds for the <code>VAR_DECL</code>, and if <code>DECL_ASSEMBLER_NAME</code> is not
equal to <code>DECL_NAME</code>.  In that case, <code>DECL_ASSEMBLER_NAME</code> is
the name of the register into which the variable will be placed.
</p>
</dd>
<dt><code>PARM_DECL</code></dt>
<dd><p>Used to represent a parameter to a function.  Treat these nodes
similarly to <code>VAR_DECL</code> nodes.  These nodes only appear in the
<code>DECL_ARGUMENTS</code> for a <code>FUNCTION_DECL</code>.
</p>
<p>The <code>DECL_ARG_TYPE</code> for a <code>PARM_DECL</code> is the type that will
actually be used when a value is passed to this function.  It may be a
wider type than the <code>TREE_TYPE</code> of the parameter; for example, the
ordinary type might be <code>short</code> while the <code>DECL_ARG_TYPE</code> is
<code>int</code>.
</p>
</dd>
<dt><code>DEBUG_EXPR_DECL</code></dt>
<dd><p>Used to represent an anonymous debug-information temporary created to
hold an expression as it is optimized away, so that its value can be
referenced in debug bind statements.
</p>
</dd>
<dt><code>FIELD_DECL</code></dt>
<dd><p>These nodes represent non-static data members.  The <code>DECL_SIZE</code> and
<code>DECL_ALIGN</code> behave as for <code>VAR_DECL</code> nodes.
The position of the field within the parent record is specified by a
combination of three attributes.  <code>DECL_FIELD_OFFSET</code> is the position,
counting in bytes, of the <code>DECL_OFFSET_ALIGN</code>-bit sized word containing
the bit of the field closest to the beginning of the structure.
<code>DECL_FIELD_BIT_OFFSET</code> is the bit offset of the first bit of the field
within this word; this may be nonzero even for fields that are not bit-fields,
since <code>DECL_OFFSET_ALIGN</code> may be greater than the natural alignment
of the field&rsquo;s type.
</p>
<p>If <code>DECL_C_BIT_FIELD</code> holds, this field is a bit-field.  In a bit-field,
<code>DECL_BIT_FIELD_TYPE</code> also contains the type that was originally
specified for it, while DECL_TYPE may be a modified type with lesser precision,
according to the size of the bit field.
</p>
</dd>
<dt><code>NAMESPACE_DECL</code></dt>
<dd><p>Namespaces provide a name hierarchy for other declarations.  They
appear in the <code>DECL_CONTEXT</code> of other <code>_DECL</code> nodes.
</p>
</dd>
</dl>

<hr>
<a name="Internal-structure"></a>
<div class="header">
<p>
Previous: <a href="#Working-with-declarations" accesskey="p" rel="prev">Working with declarations</a>, Up: <a href="#Declarations" accesskey="u" rel="up">Declarations</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Internal-structure-1"></a>
<h4 class="subsection">11.4.2 Internal structure</h4>

<p><code>DECL</code> nodes are represented internally as a hierarchy of
structures.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Current-structure-hierarchy" accesskey="1">Current structure hierarchy</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The current DECL node structure
hierarchy.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Adding-new-DECL-node-types" accesskey="2">Adding new DECL node types</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to add a new DECL node to a
frontend.
</td></tr>
</table>

<hr>
<a name="Current-structure-hierarchy"></a>
<div class="header">
<p>
Next: <a href="#Adding-new-DECL-node-types" accesskey="n" rel="next">Adding new DECL node types</a>, Up: <a href="#Internal-structure" accesskey="u" rel="up">Internal structure</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Current-structure-hierarchy-1"></a>
<h4 class="subsubsection">11.4.2.1 Current structure hierarchy</h4>

<dl compact="compact">
<dt><code>struct tree_decl_minimal</code></dt>
<dd><p>This is the minimal structure to inherit from in order for common
<code>DECL</code> macros to work.  The fields it contains are a unique ID,
source location, context, and name.
</p>
</dd>
<dt><code>struct tree_decl_common</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_minimal</code>.  It
contains fields that most <code>DECL</code> nodes need, such as a field to
store alignment, machine mode, size, and attributes.
</p>
</dd>
<dt><code>struct tree_field_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_common</code>.  It is
used to represent <code>FIELD_DECL</code>.
</p>
</dd>
<dt><code>struct tree_label_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_common</code>.  It is
used to represent <code>LABEL_DECL</code>.
</p>
</dd>
<dt><code>struct tree_translation_unit_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_common</code>.  It is
used to represent <code>TRANSLATION_UNIT_DECL</code>.
</p>
</dd>
<dt><code>struct tree_decl_with_rtl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_common</code>.  It
contains a field to store the low-level RTL associated with a
<code>DECL</code> node.
</p>
</dd>
<dt><code>struct tree_result_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_rtl</code>.  It is
used to represent <code>RESULT_DECL</code>.
</p>
</dd>
<dt><code>struct tree_const_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_rtl</code>.  It is
used to represent <code>CONST_DECL</code>.
</p>
</dd>
<dt><code>struct tree_parm_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_rtl</code>.  It is
used to represent <code>PARM_DECL</code>.
</p>
</dd>
<dt><code>struct tree_decl_with_vis</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_rtl</code>.  It
contains fields necessary to store visibility information, as well as
a section name and assembler name.
</p>
</dd>
<dt><code>struct tree_var_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_vis</code>.  It is
used to represent <code>VAR_DECL</code>.
</p>
</dd>
<dt><code>struct tree_function_decl</code></dt>
<dd><p>This structure inherits from <code>struct tree_decl_with_vis</code>.  It is
used to represent <code>FUNCTION_DECL</code>.
</p>
</dd>
</dl>
<hr>
<a name="Adding-new-DECL-node-types"></a>
<div class="header">
<p>
Previous: <a href="#Current-structure-hierarchy" accesskey="p" rel="prev">Current structure hierarchy</a>, Up: <a href="#Internal-structure" accesskey="u" rel="up">Internal structure</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Adding-new-DECL-node-types-1"></a>
<h4 class="subsubsection">11.4.2.2 Adding new DECL node types</h4>

<p>Adding a new <code>DECL</code> tree consists of the following steps
</p>
<dl compact="compact">
<dt>Add a new tree code for the <code>DECL</code> node</dt>
<dd><p>For language specific <code>DECL</code> nodes, there is a <samp>.def</samp> file
in each frontend directory where the tree code should be added.
For <code>DECL</code> nodes that are part of the middle-end, the code should
be added to <samp>tree.def</samp>.
</p>
</dd>
<dt>Create a new structure type for the <code>DECL</code> node</dt>
<dd><p>These structures should inherit from one of the existing structures in
the language hierarchy by using that structure as the first member.
</p>
<div class="smallexample">
<pre class="smallexample">struct tree_foo_decl
{
   struct tree_decl_with_vis common;
}
</pre></div>

<p>Would create a structure name <code>tree_foo_decl</code> that inherits from
<code>struct tree_decl_with_vis</code>.
</p>
<p>For language specific <code>DECL</code> nodes, this new structure type
should go in the appropriate <samp>.h</samp> file.
For <code>DECL</code> nodes that are part of the middle-end, the structure
type should go in <samp>tree.h</samp>.
</p>
</dd>
<dt>Add a member to the tree structure enumerator for the node</dt>
<dd><p>For garbage collection and dynamic checking purposes, each <code>DECL</code>
node structure type is required to have a unique enumerator value
specified with it.
For language specific <code>DECL</code> nodes, this new enumerator value
should go in the appropriate <samp>.def</samp> file.
For <code>DECL</code> nodes that are part of the middle-end, the enumerator
values are specified in <samp>treestruct.def</samp>.
</p>
</dd>
<dt>Update <code>union tree_node</code></dt>
<dd><p>In order to make your new structure type usable, it must be added to
<code>union tree_node</code>.
For language specific <code>DECL</code> nodes, a new entry should be added
to the appropriate <samp>.h</samp> file of the form
</p><div class="smallexample">
<pre class="smallexample">  struct tree_foo_decl GTY ((tag (&quot;TS_VAR_DECL&quot;))) foo_decl;
</pre></div>
<p>For <code>DECL</code> nodes that are part of the middle-end, the additional
member goes directly into <code>union tree_node</code> in <samp>tree.h</samp>.
</p>
</dd>
<dt>Update dynamic checking info</dt>
<dd><p>In order to be able to check whether accessing a named portion of
<code>union tree_node</code> is legal, and whether a certain <code>DECL</code> node
contains one of the enumerated <code>DECL</code> node structures in the
hierarchy, a simple lookup table is used.
This lookup table needs to be kept up to date with the tree structure
hierarchy, or else checking and containment macros will fail
inappropriately.
</p>
<p>For language specific <code>DECL</code> nodes, their is an <code>init_ts</code>
function in an appropriate <samp>.c</samp> file, which initializes the lookup
table.
Code setting up the table for new <code>DECL</code> nodes should be added
there.
For each <code>DECL</code> tree code and enumerator value representing a
member of the inheritance  hierarchy, the table should contain 1 if
that tree code inherits (directly or indirectly) from that member.
Thus, a <code>FOO_DECL</code> node derived from <code>struct decl_with_rtl</code>,
and enumerator value <code>TS_FOO_DECL</code>, would be set up as follows
</p><div class="smallexample">
<pre class="smallexample">tree_contains_struct[FOO_DECL][TS_FOO_DECL] = 1;
tree_contains_struct[FOO_DECL][TS_DECL_WRTL] = 1;
tree_contains_struct[FOO_DECL][TS_DECL_COMMON] = 1;
tree_contains_struct[FOO_DECL][TS_DECL_MINIMAL] = 1;
</pre></div>

<p>For <code>DECL</code> nodes that are part of the middle-end, the setup code
goes into <samp>tree.c</samp>.
</p>
</dd>
<dt>Add macros to access any new fields and flags</dt>
<dd>
<p>Each added field or flag should have a macro that is used to access
it, that performs appropriate checking to ensure only the right type of
<code>DECL</code> nodes access the field.
</p>
<p>These macros generally take the following form
</p><div class="smallexample">
<pre class="smallexample">#define FOO_DECL_FIELDNAME(NODE) FOO_DECL_CHECK(NODE)-&gt;foo_decl.fieldname
</pre></div>
<p>However, if the structure is simply a base class for further
structures, something like the following should be used
</p><div class="smallexample">
<pre class="smallexample">#define BASE_STRUCT_CHECK(T) CONTAINS_STRUCT_CHECK(T, TS_BASE_STRUCT)
#define BASE_STRUCT_FIELDNAME(NODE) \
   (BASE_STRUCT_CHECK(NODE)-&gt;base_struct.fieldname
</pre></div>

</dd>
</dl>


<hr>
<a name="Attributes"></a>
<div class="header">
<p>
Next: <a href="#Expression-trees" accesskey="n" rel="next">Expression trees</a>, Previous: <a href="#Declarations" accesskey="p" rel="prev">Declarations</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Attributes-in-trees"></a>
<h3 class="section">11.5 Attributes in trees</h3>
<a name="index-attributes"></a>

<p>Attributes, as specified using the <code>__attribute__</code> keyword, are
represented internally as a <code>TREE_LIST</code>.  The <code>TREE_PURPOSE</code>
is the name of the attribute, as an <code>IDENTIFIER_NODE</code>.  The
<code>TREE_VALUE</code> is a <code>TREE_LIST</code> of the arguments of the
attribute, if any, or <code>NULL_TREE</code> if there are no arguments; the
arguments are stored as the <code>TREE_VALUE</code> of successive entries in
the list, and may be identifiers or expressions.  The <code>TREE_CHAIN</code>
of the attribute is the next attribute in a list of attributes applying
to the same declaration or type, or <code>NULL_TREE</code> if there are no
further attributes in the list.
</p>
<p>Attributes may be attached to declarations and to types; these
attributes may be accessed with the following macros.  All attributes
are stored in this way, and many also cause other changes to the
declaration or type or to other internal compiler data structures.
</p>
<dl>
<dt><a name="index-DECL_005fATTRIBUTES"></a>Tree Macro: <em>tree</em> <strong>DECL_ATTRIBUTES</strong> <em>(tree <var>decl</var>)</em></dt>
<dd><p>This macro returns the attributes on the declaration <var>decl</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TYPE_005fATTRIBUTES"></a>Tree Macro: <em>tree</em> <strong>TYPE_ATTRIBUTES</strong> <em>(tree <var>type</var>)</em></dt>
<dd><p>This macro returns the attributes on the type <var>type</var>.
</p></dd></dl>



<hr>
<a name="Expression-trees"></a>
<div class="header">
<p>
Next: <a href="#Statements" accesskey="n" rel="next">Statements</a>, Previous: <a href="#Attributes" accesskey="p" rel="prev">Attributes</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Expressions-1"></a>
<h3 class="section">11.6 Expressions</h3>
<a name="index-expression"></a>
<a name="index-TREE_005fTYPE-3"></a>
<a name="index-TREE_005fOPERAND"></a>

<p>The internal representation for expressions is for the most part quite
straightforward.  However, there are a few facts that one must bear in
mind.  In particular, the expression &ldquo;tree&rdquo; is actually a directed
acyclic graph.  (For example there may be many references to the integer
constant zero throughout the source program; many of these will be
represented by the same expression node.)  You should not rely on
certain kinds of node being shared, nor should you rely on certain kinds of
nodes being unshared.
</p>
<p>The following macros can be used with all expression nodes:
</p>
<dl compact="compact">
<dt><code>TREE_TYPE</code>
<a name="index-TREE_005fTYPE-4"></a>
</dt>
<dd><p>Returns the type of the expression.  This value may not be precisely the
same type that would be given the expression in the original program.
</p></dd>
</dl>

<p>In what follows, some nodes that one might expect to always have type
<code>bool</code> are documented to have either integral or boolean type.  At
some point in the future, the C front end may also make use of this same
intermediate representation, and at this point these nodes will
certainly have integral type.  The previous sentence is not meant to
imply that the C++ front end does not or will not give these nodes
integral type.
</p>
<p>Below, we list the various kinds of expression nodes.  Except where
noted otherwise, the operands to an expression are accessed using the
<code>TREE_OPERAND</code> macro.  For example, to access the first operand to
a binary plus expression <code>expr</code>, use:
</p>
<div class="smallexample">
<pre class="smallexample">TREE_OPERAND (expr, 0)
</pre></div>

<p>As this example indicates, the operands are zero-indexed.
</p>

<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Constant-expressions" accesskey="1">Constants</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Storage-References" accesskey="2">Storage References</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Unary-and-Binary-Expressions" accesskey="3">Unary and Binary Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Vectors" accesskey="4">Vectors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Constant-expressions"></a>
<div class="header">
<p>
Next: <a href="#Storage-References" accesskey="n" rel="next">Storage References</a>, Up: <a href="#Expression-trees" accesskey="u" rel="up">Expression trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constant-expressions-1"></a>
<h4 class="subsection">11.6.1 Constant expressions</h4>
<a name="index-INTEGER_005fCST"></a>
<a name="index-TREE_005fINT_005fCST_005fHIGH"></a>
<a name="index-TREE_005fINT_005fCST_005fLOW"></a>
<a name="index-tree_005fint_005fcst_005flt"></a>
<a name="index-tree_005fint_005fcst_005fequal"></a>
<a name="index-REAL_005fCST"></a>
<a name="index-FIXED_005fCST"></a>
<a name="index-COMPLEX_005fCST"></a>
<a name="index-VECTOR_005fCST"></a>
<a name="index-STRING_005fCST"></a>
<a name="index-TREE_005fSTRING_005fLENGTH"></a>
<a name="index-TREE_005fSTRING_005fPOINTER"></a>

<p>The table below begins with constants, moves on to unary expressions,
then proceeds to binary expressions, and concludes with various other
kinds of expressions:
</p>
<dl compact="compact">
<dt><code>INTEGER_CST</code></dt>
<dd><p>These nodes represent integer constants.  Note that the type of these
constants is obtained with <code>TREE_TYPE</code>; they are not always of type
<code>int</code>.  In particular, <code>char</code> constants are represented with
<code>INTEGER_CST</code> nodes.  The value of the integer constant <code>e</code> is
given by
</p><div class="smallexample">
<pre class="smallexample">((TREE_INT_CST_HIGH (e) &lt;&lt; HOST_BITS_PER_WIDE_INT)
+ TREE_INST_CST_LOW (e))
</pre></div>
<p>HOST_BITS_PER_WIDE_INT is at least thirty-two on all platforms.  Both
<code>TREE_INT_CST_HIGH</code> and <code>TREE_INT_CST_LOW</code> return a
<code>HOST_WIDE_INT</code>.  The value of an <code>INTEGER_CST</code> is interpreted
as a signed or unsigned quantity depending on the type of the constant.
In general, the expression given above will overflow, so it should not
be used to calculate the value of the constant.
</p>
<p>The variable <code>integer_zero_node</code> is an integer constant with value
zero.  Similarly, <code>integer_one_node</code> is an integer constant with
value one.  The <code>size_zero_node</code> and <code>size_one_node</code> variables
are analogous, but have type <code>size_t</code> rather than <code>int</code>.
</p>
<p>The function <code>tree_int_cst_lt</code> is a predicate which holds if its
first argument is less than its second.  Both constants are assumed to
have the same signedness (i.e., either both should be signed or both
should be unsigned.)  The full width of the constant is used when doing
the comparison; the usual rules about promotions and conversions are
ignored.  Similarly, <code>tree_int_cst_equal</code> holds if the two
constants are equal.  The <code>tree_int_cst_sgn</code> function returns the
sign of a constant.  The value is <code>1</code>, <code>0</code>, or <code>-1</code>
according on whether the constant is greater than, equal to, or less
than zero.  Again, the signedness of the constant&rsquo;s type is taken into
account; an unsigned constant is never less than zero, no matter what
its bit-pattern.
</p>
</dd>
<dt><code>REAL_CST</code></dt>
<dd>
<p>FIXME: Talk about how to obtain representations of this constant, do
comparisons, and so forth.
</p>
</dd>
<dt><code>FIXED_CST</code></dt>
<dd>
<p>These nodes represent fixed-point constants.  The type of these constants
is obtained with <code>TREE_TYPE</code>.  <code>TREE_FIXED_CST_PTR</code> points to
a <code>struct fixed_value</code>;  <code>TREE_FIXED_CST</code> returns the structure
itself.  <code>struct fixed_value</code> contains <code>data</code> with the size of two
<code>HOST_BITS_PER_WIDE_INT</code> and <code>mode</code> as the associated fixed-point
machine mode for <code>data</code>.
</p>
</dd>
<dt><code>COMPLEX_CST</code></dt>
<dd><p>These nodes are used to represent complex number constants, that is a
<code>__complex__</code> whose parts are constant nodes.  The
<code>TREE_REALPART</code> and <code>TREE_IMAGPART</code> return the real and the
imaginary parts respectively.
</p>
</dd>
<dt><code>VECTOR_CST</code></dt>
<dd><p>These nodes are used to represent vector constants, whose parts are
constant nodes.  Each individual constant node is either an integer or a
double constant node.  The first operand is a <code>TREE_LIST</code> of the
constant nodes and is accessed through <code>TREE_VECTOR_CST_ELTS</code>.
</p>
</dd>
<dt><code>STRING_CST</code></dt>
<dd><p>These nodes represent string-constants.  The <code>TREE_STRING_LENGTH</code>
returns the length of the string, as an <code>int</code>.  The
<code>TREE_STRING_POINTER</code> is a <code>char*</code> containing the string
itself.  The string may not be <code>NUL</code>-terminated, and it may contain
embedded <code>NUL</code> characters.  Therefore, the
<code>TREE_STRING_LENGTH</code> includes the trailing <code>NUL</code> if it is
present.
</p>
<p>For wide string constants, the <code>TREE_STRING_LENGTH</code> is the number
of bytes in the string, and the <code>TREE_STRING_POINTER</code>
points to an array of the bytes of the string, as represented on the
target system (that is, as integers in the target endianness).  Wide and
non-wide string constants are distinguished only by the <code>TREE_TYPE</code>
of the <code>STRING_CST</code>.
</p>
<p>FIXME: The formats of string constants are not well-defined when the
target system bytes are not the same width as host system bytes.
</p>
</dd>
</dl>

<hr>
<a name="Storage-References"></a>
<div class="header">
<p>
Next: <a href="#Unary-and-Binary-Expressions" accesskey="n" rel="next">Unary and Binary Expressions</a>, Previous: <a href="#Constant-expressions" accesskey="p" rel="prev">Constant expressions</a>, Up: <a href="#Expression-trees" accesskey="u" rel="up">Expression trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="References-to-storage"></a>
<h4 class="subsection">11.6.2 References to storage</h4>
<a name="index-ADDR_005fEXPR"></a>
<a name="index-INDIRECT_005fREF"></a>
<a name="index-MEM_005fREF"></a>
<a name="index-ARRAY_005fREF"></a>
<a name="index-ARRAY_005fRANGE_005fREF"></a>
<a name="index-TARGET_005fMEM_005fREF"></a>
<a name="index-COMPONENT_005fREF"></a>

<dl compact="compact">
<dt><code>ARRAY_REF</code></dt>
<dd><p>These nodes represent array accesses.  The first operand is the array;
the second is the index.  To calculate the address of the memory
accessed, you must scale the index by the size of the type of the array
elements.  The type of these expressions must be the type of a component of
the array.  The third and fourth operands are used after gimplification
to represent the lower bound and component size but should not be used
directly; call <code>array_ref_low_bound</code> and <code>array_ref_element_size</code>
instead.
</p>
</dd>
<dt><code>ARRAY_RANGE_REF</code></dt>
<dd><p>These nodes represent access to a range (or &ldquo;slice&rdquo;) of an array.  The
operands are the same as that for <code>ARRAY_REF</code> and have the same
meanings.  The type of these expressions must be an array whose component
type is the same as that of the first operand.  The range of that array
type determines the amount of data these expressions access.
</p>
</dd>
<dt><code>TARGET_MEM_REF</code></dt>
<dd><p>These nodes represent memory accesses whose address directly map to
an addressing mode of the target architecture.  The first argument
is <code>TMR_SYMBOL</code> and must be a <code>VAR_DECL</code> of an object with
a fixed address.  The second argument is <code>TMR_BASE</code> and the
third one is <code>TMR_INDEX</code>.  The fourth argument is
<code>TMR_STEP</code> and must be an <code>INTEGER_CST</code>.  The fifth
argument is <code>TMR_OFFSET</code> and must be an <code>INTEGER_CST</code>.
Any of the arguments may be NULL if the appropriate component
does not appear in the address.  Address of the <code>TARGET_MEM_REF</code>
is determined in the following way.
</p>
<div class="smallexample">
<pre class="smallexample">&amp;TMR_SYMBOL + TMR_BASE + TMR_INDEX * TMR_STEP + TMR_OFFSET
</pre></div>

<p>The sixth argument is the reference to the original memory access, which
is preserved for the purposes of the RTL alias analysis.  The seventh
argument is a tag representing the results of tree level alias analysis.
</p>
</dd>
<dt><code>ADDR_EXPR</code></dt>
<dd><p>These nodes are used to represent the address of an object.  (These
expressions will always have pointer or reference type.)  The operand may
be another expression, or it may be a declaration.
</p>
<p>As an extension, GCC allows users to take the address of a label.  In
this case, the operand of the <code>ADDR_EXPR</code> will be a
<code>LABEL_DECL</code>.  The type of such an expression is <code>void*</code>.
</p>
<p>If the object addressed is not an lvalue, a temporary is created, and
the address of the temporary is used.
</p>
</dd>
<dt><code>INDIRECT_REF</code></dt>
<dd><p>These nodes are used to represent the object pointed to by a pointer.
The operand is the pointer being dereferenced; it will always have
pointer or reference type.
</p>
</dd>
<dt><code>MEM_REF</code></dt>
<dd><p>These nodes are used to represent the object pointed to by a pointer
offset by a constant.
The first operand is the pointer being dereferenced; it will always have
pointer or reference type.  The second operand is a pointer constant.
Its type is specifying the type to be used for type-based alias analysis.
</p>
</dd>
<dt><code>COMPONENT_REF</code></dt>
<dd><p>These nodes represent non-static data member accesses.  The first
operand is the object (rather than a pointer to it); the second operand
is the <code>FIELD_DECL</code> for the data member.  The third operand represents
the byte offset of the field, but should not be used directly; call
<code>component_ref_field_offset</code> instead.
</p>

</dd>
</dl>

<hr>
<a name="Unary-and-Binary-Expressions"></a>
<div class="header">
<p>
Next: <a href="#Vectors" accesskey="n" rel="next">Vectors</a>, Previous: <a href="#Storage-References" accesskey="p" rel="prev">Storage References</a>, Up: <a href="#Expression-trees" accesskey="u" rel="up">Expression trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Unary-and-Binary-Expressions-1"></a>
<h4 class="subsection">11.6.3 Unary and Binary Expressions</h4>
<a name="index-NEGATE_005fEXPR"></a>
<a name="index-ABS_005fEXPR"></a>
<a name="index-BIT_005fNOT_005fEXPR"></a>
<a name="index-TRUTH_005fNOT_005fEXPR"></a>
<a name="index-PREDECREMENT_005fEXPR"></a>
<a name="index-PREINCREMENT_005fEXPR"></a>
<a name="index-POSTDECREMENT_005fEXPR"></a>
<a name="index-POSTINCREMENT_005fEXPR"></a>
<a name="index-FIX_005fTRUNC_005fEXPR"></a>
<a name="index-FLOAT_005fEXPR"></a>
<a name="index-COMPLEX_005fEXPR"></a>
<a name="index-CONJ_005fEXPR"></a>
<a name="index-REALPART_005fEXPR"></a>
<a name="index-IMAGPART_005fEXPR"></a>
<a name="index-NON_005fLVALUE_005fEXPR"></a>
<a name="index-NOP_005fEXPR"></a>
<a name="index-CONVERT_005fEXPR"></a>
<a name="index-FIXED_005fCONVERT_005fEXPR"></a>
<a name="index-THROW_005fEXPR"></a>
<a name="index-LSHIFT_005fEXPR"></a>
<a name="index-RSHIFT_005fEXPR"></a>
<a name="index-BIT_005fIOR_005fEXPR"></a>
<a name="index-BIT_005fXOR_005fEXPR"></a>
<a name="index-BIT_005fAND_005fEXPR"></a>
<a name="index-TRUTH_005fANDIF_005fEXPR"></a>
<a name="index-TRUTH_005fORIF_005fEXPR"></a>
<a name="index-TRUTH_005fAND_005fEXPR"></a>
<a name="index-TRUTH_005fOR_005fEXPR"></a>
<a name="index-TRUTH_005fXOR_005fEXPR"></a>
<a name="index-POINTER_005fPLUS_005fEXPR"></a>
<a name="index-PLUS_005fEXPR"></a>
<a name="index-MINUS_005fEXPR"></a>
<a name="index-MULT_005fEXPR"></a>
<a name="index-MULT_005fHIGHPART_005fEXPR"></a>
<a name="index-RDIV_005fEXPR"></a>
<a name="index-TRUNC_005fDIV_005fEXPR"></a>
<a name="index-FLOOR_005fDIV_005fEXPR"></a>
<a name="index-CEIL_005fDIV_005fEXPR"></a>
<a name="index-ROUND_005fDIV_005fEXPR"></a>
<a name="index-TRUNC_005fMOD_005fEXPR"></a>
<a name="index-FLOOR_005fMOD_005fEXPR"></a>
<a name="index-CEIL_005fMOD_005fEXPR"></a>
<a name="index-ROUND_005fMOD_005fEXPR"></a>
<a name="index-EXACT_005fDIV_005fEXPR"></a>
<a name="index-LT_005fEXPR"></a>
<a name="index-LE_005fEXPR"></a>
<a name="index-GT_005fEXPR"></a>
<a name="index-GE_005fEXPR"></a>
<a name="index-EQ_005fEXPR"></a>
<a name="index-NE_005fEXPR"></a>
<a name="index-ORDERED_005fEXPR"></a>
<a name="index-UNORDERED_005fEXPR"></a>
<a name="index-UNLT_005fEXPR"></a>
<a name="index-UNLE_005fEXPR"></a>
<a name="index-UNGT_005fEXPR"></a>
<a name="index-UNGE_005fEXPR"></a>
<a name="index-UNEQ_005fEXPR"></a>
<a name="index-LTGT_005fEXPR"></a>
<a name="index-MODIFY_005fEXPR"></a>
<a name="index-INIT_005fEXPR"></a>
<a name="index-COMPOUND_005fEXPR"></a>
<a name="index-COND_005fEXPR"></a>
<a name="index-CALL_005fEXPR"></a>
<a name="index-STMT_005fEXPR"></a>
<a name="index-BIND_005fEXPR"></a>
<a name="index-LOOP_005fEXPR"></a>
<a name="index-EXIT_005fEXPR"></a>
<a name="index-CLEANUP_005fPOINT_005fEXPR"></a>
<a name="index-CONSTRUCTOR"></a>
<a name="index-COMPOUND_005fLITERAL_005fEXPR"></a>
<a name="index-SAVE_005fEXPR"></a>
<a name="index-TARGET_005fEXPR"></a>
<a name="index-VA_005fARG_005fEXPR"></a>

<dl compact="compact">
<dt><code>NEGATE_EXPR</code></dt>
<dd><p>These nodes represent unary negation of the single operand, for both
integer and floating-point types.  The type of negation can be
determined by looking at the type of the expression.
</p>
<p>The behavior of this operation on signed arithmetic overflow is
controlled by the <code>flag_wrapv</code> and <code>flag_trapv</code> variables.
</p>
</dd>
<dt><code>ABS_EXPR</code></dt>
<dd><p>These nodes represent the absolute value of the single operand, for
both integer and floating-point types.  This is typically used to
implement the <code>abs</code>, <code>labs</code> and <code>llabs</code> builtins for
integer types, and the <code>fabs</code>, <code>fabsf</code> and <code>fabsl</code>
builtins for floating point types.  The type of abs operation can
be determined by looking at the type of the expression.
</p>
<p>This node is not used for complex types.  To represent the modulus
or complex abs of a complex value, use the <code>BUILT_IN_CABS</code>,
<code>BUILT_IN_CABSF</code> or <code>BUILT_IN_CABSL</code> builtins, as used
to implement the C99 <code>cabs</code>, <code>cabsf</code> and <code>cabsl</code>
built-in functions.
</p>
</dd>
<dt><code>BIT_NOT_EXPR</code></dt>
<dd><p>These nodes represent bitwise complement, and will always have integral
type.  The only operand is the value to be complemented.
</p>
</dd>
<dt><code>TRUTH_NOT_EXPR</code></dt>
<dd><p>These nodes represent logical negation, and will always have integral
(or boolean) type.  The operand is the value being negated.  The type
of the operand and that of the result are always of <code>BOOLEAN_TYPE</code>
or <code>INTEGER_TYPE</code>.
</p>
</dd>
<dt><code>PREDECREMENT_EXPR</code></dt>
<dt><code>PREINCREMENT_EXPR</code></dt>
<dt><code>POSTDECREMENT_EXPR</code></dt>
<dt><code>POSTINCREMENT_EXPR</code></dt>
<dd><p>These nodes represent increment and decrement expressions.  The value of
the single operand is computed, and the operand incremented or
decremented.  In the case of <code>PREDECREMENT_EXPR</code> and
<code>PREINCREMENT_EXPR</code>, the value of the expression is the value
resulting after the increment or decrement; in the case of
<code>POSTDECREMENT_EXPR</code> and <code>POSTINCREMENT_EXPR</code> is the value
before the increment or decrement occurs.  The type of the operand, like
that of the result, will be either integral, boolean, or floating-point.
</p>
</dd>
<dt><code>FIX_TRUNC_EXPR</code></dt>
<dd><p>These nodes represent conversion of a floating-point value to an
integer.  The single operand will have a floating-point type, while
the complete expression will have an integral (or boolean) type.  The
operand is rounded towards zero.
</p>
</dd>
<dt><code>FLOAT_EXPR</code></dt>
<dd><p>These nodes represent conversion of an integral (or boolean) value to a
floating-point value.  The single operand will have integral type, while
the complete expression will have a floating-point type.
</p>
<p>FIXME: How is the operand supposed to be rounded?  Is this dependent on
<samp>-mieee</samp>?
</p>
</dd>
<dt><code>COMPLEX_EXPR</code></dt>
<dd><p>These nodes are used to represent complex numbers constructed from two
expressions of the same (integer or real) type.  The first operand is the
real part and the second operand is the imaginary part.
</p>
</dd>
<dt><code>CONJ_EXPR</code></dt>
<dd><p>These nodes represent the conjugate of their operand.
</p>
</dd>
<dt><code>REALPART_EXPR</code></dt>
<dt><code>IMAGPART_EXPR</code></dt>
<dd><p>These nodes represent respectively the real and the imaginary parts
of complex numbers (their sole argument).
</p>
</dd>
<dt><code>NON_LVALUE_EXPR</code></dt>
<dd><p>These nodes indicate that their one and only operand is not an lvalue.
A back end can treat these identically to the single operand.
</p>
</dd>
<dt><code>NOP_EXPR</code></dt>
<dd><p>These nodes are used to represent conversions that do not require any
code-generation.  For example, conversion of a <code>char*</code> to an
<code>int*</code> does not require any code be generated; such a conversion is
represented by a <code>NOP_EXPR</code>.  The single operand is the expression
to be converted.  The conversion from a pointer to a reference is also
represented with a <code>NOP_EXPR</code>.
</p>
</dd>
<dt><code>CONVERT_EXPR</code></dt>
<dd><p>These nodes are similar to <code>NOP_EXPR</code>s, but are used in those
situations where code may need to be generated.  For example, if an
<code>int*</code> is converted to an <code>int</code> code may need to be generated
on some platforms.  These nodes are never used for C++-specific
conversions, like conversions between pointers to different classes in
an inheritance hierarchy.  Any adjustments that need to be made in such
cases are always indicated explicitly.  Similarly, a user-defined
conversion is never represented by a <code>CONVERT_EXPR</code>; instead, the
function calls are made explicit.
</p>
</dd>
<dt><code>FIXED_CONVERT_EXPR</code></dt>
<dd><p>These nodes are used to represent conversions that involve fixed-point
values.  For example, from a fixed-point value to another fixed-point value,
from an integer to a fixed-point value, from a fixed-point value to an
integer, from a floating-point value to a fixed-point value, or from
a fixed-point value to a floating-point value.
</p>
</dd>
<dt><code>LSHIFT_EXPR</code></dt>
<dt><code>RSHIFT_EXPR</code></dt>
<dd><p>These nodes represent left and right shifts, respectively.  The first
operand is the value to shift; it will always be of integral type.  The
second operand is an expression for the number of bits by which to
shift.  Right shift should be treated as arithmetic, i.e., the
high-order bits should be zero-filled when the expression has unsigned
type and filled with the sign bit when the expression has signed type.
Note that the result is undefined if the second operand is larger
than or equal to the first operand&rsquo;s type size. Unlike most nodes, these
can have a vector as first operand and a scalar as second operand.
</p>

</dd>
<dt><code>BIT_IOR_EXPR</code></dt>
<dt><code>BIT_XOR_EXPR</code></dt>
<dt><code>BIT_AND_EXPR</code></dt>
<dd><p>These nodes represent bitwise inclusive or, bitwise exclusive or, and
bitwise and, respectively.  Both operands will always have integral
type.
</p>
</dd>
<dt><code>TRUTH_ANDIF_EXPR</code></dt>
<dt><code>TRUTH_ORIF_EXPR</code></dt>
<dd><p>These nodes represent logical &ldquo;and&rdquo; and logical &ldquo;or&rdquo;, respectively.
These operators are not strict; i.e., the second operand is evaluated
only if the value of the expression is not determined by evaluation of
the first operand.  The type of the operands and that of the result are
always of <code>BOOLEAN_TYPE</code> or <code>INTEGER_TYPE</code>.
</p>
</dd>
<dt><code>TRUTH_AND_EXPR</code></dt>
<dt><code>TRUTH_OR_EXPR</code></dt>
<dt><code>TRUTH_XOR_EXPR</code></dt>
<dd><p>These nodes represent logical and, logical or, and logical exclusive or.
They are strict; both arguments are always evaluated.  There are no
corresponding operators in C or C++, but the front end will sometimes
generate these expressions anyhow, if it can tell that strictness does
not matter.  The type of the operands and that of the result are
always of <code>BOOLEAN_TYPE</code> or <code>INTEGER_TYPE</code>.
</p>
</dd>
<dt><code>POINTER_PLUS_EXPR</code></dt>
<dd><p>This node represents pointer arithmetic.  The first operand is always
a pointer/reference type.  The second operand is always an unsigned
integer type compatible with sizetype.  This is the only binary
arithmetic operand that can operate on pointer types.
</p>
</dd>
<dt><code>PLUS_EXPR</code></dt>
<dt><code>MINUS_EXPR</code></dt>
<dt><code>MULT_EXPR</code></dt>
<dd><p>These nodes represent various binary arithmetic operations.
Respectively, these operations are addition, subtraction (of the second
operand from the first) and multiplication.  Their operands may have
either integral or floating type, but there will never be case in which
one operand is of floating type and the other is of integral type.
</p>
<p>The behavior of these operations on signed arithmetic overflow is
controlled by the <code>flag_wrapv</code> and <code>flag_trapv</code> variables.
</p>
</dd>
<dt><code>MULT_HIGHPART_EXPR</code></dt>
<dd><p>This node represents the &ldquo;high-part&rdquo; of a widening multiplication.
For an integral type with <var>b</var> bits of precision, the result is
the most significant <var>b</var> bits of the full <em>2<var>b</var></em> product.
</p>
</dd>
<dt><code>RDIV_EXPR</code></dt>
<dd><p>This node represents a floating point division operation.
</p>
</dd>
<dt><code>TRUNC_DIV_EXPR</code></dt>
<dt><code>FLOOR_DIV_EXPR</code></dt>
<dt><code>CEIL_DIV_EXPR</code></dt>
<dt><code>ROUND_DIV_EXPR</code></dt>
<dd><p>These nodes represent integer division operations that return an integer
result.  <code>TRUNC_DIV_EXPR</code> rounds towards zero, <code>FLOOR_DIV_EXPR</code>
rounds towards negative infinity, <code>CEIL_DIV_EXPR</code> rounds towards
positive infinity and <code>ROUND_DIV_EXPR</code> rounds to the closest integer.
Integer division in C and C++ is truncating, i.e. <code>TRUNC_DIV_EXPR</code>.
</p>
<p>The behavior of these operations on signed arithmetic overflow, when
dividing the minimum signed integer by minus one, is controlled by the
<code>flag_wrapv</code> and <code>flag_trapv</code> variables.
</p>
</dd>
<dt><code>TRUNC_MOD_EXPR</code></dt>
<dt><code>FLOOR_MOD_EXPR</code></dt>
<dt><code>CEIL_MOD_EXPR</code></dt>
<dt><code>ROUND_MOD_EXPR</code></dt>
<dd><p>These nodes represent the integer remainder or modulus operation.
The integer modulus of two operands <code>a</code> and <code>b</code> is
defined as <code>a - (a/b)*b</code> where the division calculated using
the corresponding division operator.  Hence for <code>TRUNC_MOD_EXPR</code>
this definition assumes division using truncation towards zero, i.e.
<code>TRUNC_DIV_EXPR</code>.  Integer remainder in C and C++ uses truncating
division, i.e. <code>TRUNC_MOD_EXPR</code>.
</p>
</dd>
<dt><code>EXACT_DIV_EXPR</code></dt>
<dd><p>The <code>EXACT_DIV_EXPR</code> code is used to represent integer divisions where
the numerator is known to be an exact multiple of the denominator.  This
allows the backend to choose between the faster of <code>TRUNC_DIV_EXPR</code>,
<code>CEIL_DIV_EXPR</code> and <code>FLOOR_DIV_EXPR</code> for the current target.
</p>
</dd>
<dt><code>LT_EXPR</code></dt>
<dt><code>LE_EXPR</code></dt>
<dt><code>GT_EXPR</code></dt>
<dt><code>GE_EXPR</code></dt>
<dt><code>EQ_EXPR</code></dt>
<dt><code>NE_EXPR</code></dt>
<dd><p>These nodes represent the less than, less than or equal to, greater
than, greater than or equal to, equal, and not equal comparison
operators.  The first and second operands will either be both of integral
type, both of floating type or both of vector type.  The result type of
these expressions will always be of integral, boolean or signed integral
vector type.  These operations return the result type&rsquo;s zero value for
false, the result type&rsquo;s one value for true, and a vector whose elements
are zero (false) or minus one (true) for vectors.
</p>
<p>For floating point comparisons, if we honor IEEE NaNs and either operand
is NaN, then <code>NE_EXPR</code> always returns true and the remaining operators
always return false.  On some targets, comparisons against an IEEE NaN,
other than equality and inequality, may generate a floating point exception.
</p>
</dd>
<dt><code>ORDERED_EXPR</code></dt>
<dt><code>UNORDERED_EXPR</code></dt>
<dd><p>These nodes represent non-trapping ordered and unordered comparison
operators.  These operations take two floating point operands and
determine whether they are ordered or unordered relative to each other.
If either operand is an IEEE NaN, their comparison is defined to be
unordered, otherwise the comparison is defined to be ordered.  The
result type of these expressions will always be of integral or boolean
type.  These operations return the result type&rsquo;s zero value for false,
and the result type&rsquo;s one value for true.
</p>
</dd>
<dt><code>UNLT_EXPR</code></dt>
<dt><code>UNLE_EXPR</code></dt>
<dt><code>UNGT_EXPR</code></dt>
<dt><code>UNGE_EXPR</code></dt>
<dt><code>UNEQ_EXPR</code></dt>
<dt><code>LTGT_EXPR</code></dt>
<dd><p>These nodes represent the unordered comparison operators.
These operations take two floating point operands and determine whether
the operands are unordered or are less than, less than or equal to,
greater than, greater than or equal to, or equal respectively.  For
example, <code>UNLT_EXPR</code> returns true if either operand is an IEEE
NaN or the first operand is less than the second.  With the possible
exception of <code>LTGT_EXPR</code>, all of these operations are guaranteed
not to generate a floating point exception.  The result
type of these expressions will always be of integral or boolean type.
These operations return the result type&rsquo;s zero value for false,
and the result type&rsquo;s one value for true.
</p>
</dd>
<dt><code>MODIFY_EXPR</code></dt>
<dd><p>These nodes represent assignment.  The left-hand side is the first
operand; the right-hand side is the second operand.  The left-hand side
will be a <code>VAR_DECL</code>, <code>INDIRECT_REF</code>, <code>COMPONENT_REF</code>, or
other lvalue.
</p>
<p>These nodes are used to represent not only assignment with &lsquo;<samp>=</samp>&rsquo; but
also compound assignments (like &lsquo;<samp>+=</samp>&rsquo;), by reduction to &lsquo;<samp>=</samp>&rsquo;
assignment.  In other words, the representation for &lsquo;<samp>i += 3</samp>&rsquo; looks
just like that for &lsquo;<samp>i = i + 3</samp>&rsquo;.
</p>
</dd>
<dt><code>INIT_EXPR</code></dt>
<dd><p>These nodes are just like <code>MODIFY_EXPR</code>, but are used only when a
variable is initialized, rather than assigned to subsequently.  This
means that we can assume that the target of the initialization is not
used in computing its own value; any reference to the lhs in computing
the rhs is undefined.
</p>
</dd>
<dt><code>COMPOUND_EXPR</code></dt>
<dd><p>These nodes represent comma-expressions.  The first operand is an
expression whose value is computed and thrown away prior to the
evaluation of the second operand.  The value of the entire expression is
the value of the second operand.
</p>
</dd>
<dt><code>COND_EXPR</code></dt>
<dd><p>These nodes represent <code>?:</code> expressions.  The first operand
is of boolean or integral type.  If it evaluates to a nonzero value,
the second operand should be evaluated, and returned as the value of the
expression.  Otherwise, the third operand is evaluated, and returned as
the value of the expression.
</p>
<p>The second operand must have the same type as the entire expression,
unless it unconditionally throws an exception or calls a noreturn
function, in which case it should have void type.  The same constraints
apply to the third operand.  This allows array bounds checks to be
represented conveniently as <code>(i &gt;= 0 &amp;&amp; i &lt; 10) ? i : abort()</code>.
</p>
<p>As a GNU extension, the C language front-ends allow the second
operand of the <code>?:</code> operator may be omitted in the source.
For example, <code>x ? : 3</code> is equivalent to <code>x ? x : 3</code>,
assuming that <code>x</code> is an expression without side-effects.
In the tree representation, however, the second operand is always
present, possibly protected by <code>SAVE_EXPR</code> if the first
argument does cause side-effects.
</p>
</dd>
<dt><code>CALL_EXPR</code></dt>
<dd><p>These nodes are used to represent calls to functions, including
non-static member functions.  <code>CALL_EXPR</code>s are implemented as
expression nodes with a variable number of operands.  Rather than using
<code>TREE_OPERAND</code> to extract them, it is preferable to use the
specialized accessor macros and functions that operate specifically on
<code>CALL_EXPR</code> nodes.
</p>
<p><code>CALL_EXPR_FN</code> returns a pointer to the
function to call; it is always an expression whose type is a
<code>POINTER_TYPE</code>.
</p>
<p>The number of arguments to the call is returned by <code>call_expr_nargs</code>,
while the arguments themselves can be accessed with the <code>CALL_EXPR_ARG</code>
macro.  The arguments are zero-indexed and numbered left-to-right.
You can iterate over the arguments using <code>FOR_EACH_CALL_EXPR_ARG</code>, as in:
</p>
<div class="smallexample">
<pre class="smallexample">tree call, arg;
call_expr_arg_iterator iter;
FOR_EACH_CALL_EXPR_ARG (arg, iter, call)
  /* arg is bound to successive arguments of call.  */
  &hellip;;
</pre></div>

<p>For non-static
member functions, there will be an operand corresponding to the
<code>this</code> pointer.  There will always be expressions corresponding to
all of the arguments, even if the function is declared with default
arguments and some arguments are not explicitly provided at the call
sites.
</p>
<p><code>CALL_EXPR</code>s also have a <code>CALL_EXPR_STATIC_CHAIN</code> operand that
is used to implement nested functions.  This operand is otherwise null.
</p>
</dd>
<dt><code>CLEANUP_POINT_EXPR</code></dt>
<dd><p>These nodes represent full-expressions.  The single operand is an
expression to evaluate.  Any destructor calls engendered by the creation
of temporaries during the evaluation of that expression should be
performed immediately after the expression is evaluated.
</p>
</dd>
<dt><code>CONSTRUCTOR</code></dt>
<dd><p>These nodes represent the brace-enclosed initializers for a structure or
array.  The first operand is reserved for use by the back end.  The
second operand is a <code>TREE_LIST</code>.  If the <code>TREE_TYPE</code> of the
<code>CONSTRUCTOR</code> is a <code>RECORD_TYPE</code> or <code>UNION_TYPE</code>, then
the <code>TREE_PURPOSE</code> of each node in the <code>TREE_LIST</code> will be a
<code>FIELD_DECL</code> and the <code>TREE_VALUE</code> of each node will be the
expression used to initialize that field.
</p>
<p>If the <code>TREE_TYPE</code> of the <code>CONSTRUCTOR</code> is an
<code>ARRAY_TYPE</code>, then the <code>TREE_PURPOSE</code> of each element in the
<code>TREE_LIST</code> will be an <code>INTEGER_CST</code> or a <code>RANGE_EXPR</code> of
two <code>INTEGER_CST</code>s.  A single <code>INTEGER_CST</code> indicates which
element of the array (indexed from zero) is being assigned to.  A
<code>RANGE_EXPR</code> indicates an inclusive range of elements to
initialize.  In both cases the <code>TREE_VALUE</code> is the corresponding
initializer.  It is re-evaluated for each element of a
<code>RANGE_EXPR</code>.  If the <code>TREE_PURPOSE</code> is <code>NULL_TREE</code>, then
the initializer is for the next available array element.
</p>
<p>In the front end, you should not depend on the fields appearing in any
particular order.  However, in the middle end, fields must appear in
declaration order.  You should not assume that all fields will be
represented.  Unrepresented fields will be set to zero.
</p>
</dd>
<dt><code>COMPOUND_LITERAL_EXPR</code></dt>
<dd><a name="index-COMPOUND_005fLITERAL_005fEXPR_005fDECL_005fEXPR"></a>
<a name="index-COMPOUND_005fLITERAL_005fEXPR_005fDECL"></a>
<p>These nodes represent ISO C99 compound literals.  The
<code>COMPOUND_LITERAL_EXPR_DECL_EXPR</code> is a <code>DECL_EXPR</code>
containing an anonymous <code>VAR_DECL</code> for
the unnamed object represented by the compound literal; the
<code>DECL_INITIAL</code> of that <code>VAR_DECL</code> is a <code>CONSTRUCTOR</code>
representing the brace-enclosed list of initializers in the compound
literal.  That anonymous <code>VAR_DECL</code> can also be accessed directly
by the <code>COMPOUND_LITERAL_EXPR_DECL</code> macro.
</p>
</dd>
<dt><code>SAVE_EXPR</code></dt>
<dd>
<p>A <code>SAVE_EXPR</code> represents an expression (possibly involving
side-effects) that is used more than once.  The side-effects should
occur only the first time the expression is evaluated.  Subsequent uses
should just reuse the computed value.  The first operand to the
<code>SAVE_EXPR</code> is the expression to evaluate.  The side-effects should
be executed where the <code>SAVE_EXPR</code> is first encountered in a
depth-first preorder traversal of the expression tree.
</p>
</dd>
<dt><code>TARGET_EXPR</code></dt>
<dd><p>A <code>TARGET_EXPR</code> represents a temporary object.  The first operand
is a <code>VAR_DECL</code> for the temporary variable.  The second operand is
the initializer for the temporary.  The initializer is evaluated and,
if non-void, copied (bitwise) into the temporary.  If the initializer
is void, that means that it will perform the initialization itself.
</p>
<p>Often, a <code>TARGET_EXPR</code> occurs on the right-hand side of an
assignment, or as the second operand to a comma-expression which is
itself the right-hand side of an assignment, etc.  In this case, we say
that the <code>TARGET_EXPR</code> is &ldquo;normal&rdquo;; otherwise, we say it is
&ldquo;orphaned&rdquo;.  For a normal <code>TARGET_EXPR</code> the temporary variable
should be treated as an alias for the left-hand side of the assignment,
rather than as a new temporary variable.
</p>
<p>The third operand to the <code>TARGET_EXPR</code>, if present, is a
cleanup-expression (i.e., destructor call) for the temporary.  If this
expression is orphaned, then this expression must be executed when the
statement containing this expression is complete.  These cleanups must
always be executed in the order opposite to that in which they were
encountered.  Note that if a temporary is created on one branch of a
conditional operator (i.e., in the second or third operand to a
<code>COND_EXPR</code>), the cleanup must be run only if that branch is
actually executed.
</p>
</dd>
<dt><code>VA_ARG_EXPR</code></dt>
<dd><p>This node is used to implement support for the C/C++ variable argument-list
mechanism.  It represents expressions like <code>va_arg (ap, type)</code>.
Its <code>TREE_TYPE</code> yields the tree representation for <code>type</code> and
its sole argument yields the representation for <code>ap</code>.
</p>
</dd>
</dl>

<hr>
<a name="Vectors"></a>
<div class="header">
<p>
Previous: <a href="#Unary-and-Binary-Expressions" accesskey="p" rel="prev">Unary and Binary Expressions</a>, Up: <a href="#Expression-trees" accesskey="u" rel="up">Expression trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Vectors-1"></a>
<h4 class="subsection">11.6.4 Vectors</h4>
<a name="index-VEC_005fLSHIFT_005fEXPR"></a>
<a name="index-VEC_005fRSHIFT_005fEXPR"></a>
<a name="index-VEC_005fWIDEN_005fMULT_005fHI_005fEXPR"></a>
<a name="index-VEC_005fWIDEN_005fMULT_005fLO_005fEXPR"></a>
<a name="index-VEC_005fUNPACK_005fHI_005fEXPR"></a>
<a name="index-VEC_005fUNPACK_005fLO_005fEXPR"></a>
<a name="index-VEC_005fUNPACK_005fFLOAT_005fHI_005fEXPR"></a>
<a name="index-VEC_005fUNPACK_005fFLOAT_005fLO_005fEXPR"></a>
<a name="index-VEC_005fPACK_005fTRUNC_005fEXPR"></a>
<a name="index-VEC_005fPACK_005fSAT_005fEXPR"></a>
<a name="index-VEC_005fPACK_005fFIX_005fTRUNC_005fEXPR"></a>

<dl compact="compact">
<dt><code>VEC_LSHIFT_EXPR</code></dt>
<dt><code>VEC_RSHIFT_EXPR</code></dt>
<dd><p>These nodes represent whole vector left and right shifts, respectively.
The first operand is the vector to shift; it will always be of vector type.
The second operand is an expression for the number of bits by which to
shift.  Note that the result is undefined if the second operand is larger
than or equal to the first operand&rsquo;s type size.
</p>
</dd>
<dt><code>VEC_WIDEN_MULT_HI_EXPR</code></dt>
<dt><code>VEC_WIDEN_MULT_LO_EXPR</code></dt>
<dd><p>These nodes represent widening vector multiplication of the high and low
parts of the two input vectors, respectively.  Their operands are vectors
that contain the same number of elements (<code>N</code>) of the same integral type.
The result is a vector that contains half as many elements, of an integral type
whose size is twice as wide.  In the case of <code>VEC_WIDEN_MULT_HI_EXPR</code> the
high <code>N/2</code> elements of the two vector are multiplied to produce the
vector of <code>N/2</code> products. In the case of <code>VEC_WIDEN_MULT_LO_EXPR</code> the
low <code>N/2</code> elements of the two vector are multiplied to produce the
vector of <code>N/2</code> products.
</p>
</dd>
<dt><code>VEC_UNPACK_HI_EXPR</code></dt>
<dt><code>VEC_UNPACK_LO_EXPR</code></dt>
<dd><p>These nodes represent unpacking of the high and low parts of the input vector,
respectively.  The single operand is a vector that contains <code>N</code> elements
of the same integral or floating point type.  The result is a vector
that contains half as many elements, of an integral or floating point type
whose size is twice as wide.  In the case of <code>VEC_UNPACK_HI_EXPR</code> the
high <code>N/2</code> elements of the vector are extracted and widened (promoted).
In the case of <code>VEC_UNPACK_LO_EXPR</code> the low <code>N/2</code> elements of the
vector are extracted and widened (promoted).
</p>
</dd>
<dt><code>VEC_UNPACK_FLOAT_HI_EXPR</code></dt>
<dt><code>VEC_UNPACK_FLOAT_LO_EXPR</code></dt>
<dd><p>These nodes represent unpacking of the high and low parts of the input vector,
where the values are converted from fixed point to floating point.  The
single operand is a vector that contains <code>N</code> elements of the same
integral type.  The result is a vector that contains half as many elements
of a floating point type whose size is twice as wide.  In the case of
<code>VEC_UNPACK_HI_EXPR</code> the high <code>N/2</code> elements of the vector are
extracted, converted and widened.  In the case of <code>VEC_UNPACK_LO_EXPR</code>
the low <code>N/2</code> elements of the vector are extracted, converted and widened.
</p>
</dd>
<dt><code>VEC_PACK_TRUNC_EXPR</code></dt>
<dd><p>This node represents packing of truncated elements of the two input vectors
into the output vector.  Input operands are vectors that contain the same
number of elements of the same integral or floating point type.  The result
is a vector that contains twice as many elements of an integral or floating
point type whose size is half as wide. The elements of the two vectors are
demoted and merged (concatenated) to form the output vector.
</p>
</dd>
<dt><code>VEC_PACK_SAT_EXPR</code></dt>
<dd><p>This node represents packing of elements of the two input vectors into the
output vector using saturation.  Input operands are vectors that contain
the same number of elements of the same integral type.  The result is a
vector that contains twice as many elements of an integral type whose size
is half as wide.  The elements of the two vectors are demoted and merged
(concatenated) to form the output vector.
</p>
</dd>
<dt><code>VEC_PACK_FIX_TRUNC_EXPR</code></dt>
<dd><p>This node represents packing of elements of the two input vectors into the
output vector, where the values are converted from floating point
to fixed point.  Input operands are vectors that contain the same number
of elements of a floating point type.  The result is a vector that contains
twice as many elements of an integral type whose size is half as wide.  The
elements of the two vectors are merged (concatenated) to form the output
vector.
</p>
</dd>
<dt><code>VEC_COND_EXPR</code></dt>
<dd><p>These nodes represent <code>?:</code> expressions.  The three operands must be
vectors of the same size and number of elements.  The second and third
operands must have the same type as the entire expression.  The first
operand is of signed integral vector type.  If an element of the first
operand evaluates to a zero value, the corresponding element of the
result is taken from the third operand. If it evaluates to a minus one
value, it is taken from the second operand. It should never evaluate to
any other value currently, but optimizations should not rely on that
property. In contrast with a <code>COND_EXPR</code>, all operands are always
evaluated.
</p></dd>
</dl>



<hr>
<a name="Statements"></a>
<div class="header">
<p>
Next: <a href="#Functions" accesskey="n" rel="next">Functions</a>, Previous: <a href="#Expression-trees" accesskey="p" rel="prev">Expression trees</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Statements-1"></a>
<h3 class="section">11.7 Statements</h3>
<a name="index-Statements"></a>

<p>Most statements in GIMPLE are assignment statements, represented by
<code>GIMPLE_ASSIGN</code>.  No other C expressions can appear at statement level;
a reference to a volatile object is converted into a
<code>GIMPLE_ASSIGN</code>.
</p>
<p>There are also several varieties of complex statements.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Basic-Statements" accesskey="1">Basic Statements</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Blocks" accesskey="2">Blocks</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Statement-Sequences" accesskey="3">Statement Sequences</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Empty-Statements" accesskey="4">Empty Statements</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Jumps" accesskey="5">Jumps</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Cleanups" accesskey="6">Cleanups</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#OpenMP" accesskey="7">OpenMP</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Basic-Statements"></a>
<div class="header">
<p>
Next: <a href="#Blocks" accesskey="n" rel="next">Blocks</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Basic-Statements-1"></a>
<h4 class="subsection">11.7.1 Basic Statements</h4>
<a name="index-Basic-Statements"></a>

<dl compact="compact">
<dt><code>ASM_EXPR</code></dt>
<dd>
<p>Used to represent an inline assembly statement.  For an inline assembly
statement like:
</p><div class="smallexample">
<pre class="smallexample">asm (&quot;mov x, y&quot;);
</pre></div>
<p>The <code>ASM_STRING</code> macro will return a <code>STRING_CST</code> node for
<code>&quot;mov x, y&quot;</code>.  If the original statement made use of the
extended-assembly syntax, then <code>ASM_OUTPUTS</code>,
<code>ASM_INPUTS</code>, and <code>ASM_CLOBBERS</code> will be the outputs, inputs,
and clobbers for the statement, represented as <code>STRING_CST</code> nodes.
The extended-assembly syntax looks like:
</p><div class="smallexample">
<pre class="smallexample">asm (&quot;fsinx %1,%0&quot; : &quot;=f&quot; (result) : &quot;f&quot; (angle));
</pre></div>
<p>The first string is the <code>ASM_STRING</code>, containing the instruction
template.  The next two strings are the output and inputs, respectively;
this statement has no clobbers.  As this example indicates, &ldquo;plain&rdquo;
assembly statements are merely a special case of extended assembly
statements; they have no cv-qualifiers, outputs, inputs, or clobbers.
All of the strings will be <code>NUL</code>-terminated, and will contain no
embedded <code>NUL</code>-characters.
</p>
<p>If the assembly statement is declared <code>volatile</code>, or if the
statement was not an extended assembly statement, and is therefore
implicitly volatile, then the predicate <code>ASM_VOLATILE_P</code> will hold
of the <code>ASM_EXPR</code>.
</p>
</dd>
<dt><code>DECL_EXPR</code></dt>
<dd>
<p>Used to represent a local declaration.  The <code>DECL_EXPR_DECL</code> macro
can be used to obtain the entity declared.  This declaration may be a
<code>LABEL_DECL</code>, indicating that the label declared is a local label.
(As an extension, GCC allows the declaration of labels with scope.)  In
C, this declaration may be a <code>FUNCTION_DECL</code>, indicating the
use of the GCC nested function extension.  For more information,
see <a href="#Functions">Functions</a>.
</p>
</dd>
<dt><code>LABEL_EXPR</code></dt>
<dd>
<p>Used to represent a label.  The <code>LABEL_DECL</code> declared by this
statement can be obtained with the <code>LABEL_EXPR_LABEL</code> macro.  The
<code>IDENTIFIER_NODE</code> giving the name of the label can be obtained from
the <code>LABEL_DECL</code> with <code>DECL_NAME</code>.
</p>
</dd>
<dt><code>GOTO_EXPR</code></dt>
<dd>
<p>Used to represent a <code>goto</code> statement.  The <code>GOTO_DESTINATION</code> will
usually be a <code>LABEL_DECL</code>.  However, if the &ldquo;computed goto&rdquo; extension
has been used, the <code>GOTO_DESTINATION</code> will be an arbitrary expression
indicating the destination.  This expression will always have pointer type.
</p>
</dd>
<dt><code>RETURN_EXPR</code></dt>
<dd>
<p>Used to represent a <code>return</code> statement.  Operand 0 represents the
value to return.  It should either be the <code>RESULT_DECL</code> for the
containing function, or a <code>MODIFY_EXPR</code> or <code>INIT_EXPR</code>
setting the function&rsquo;s <code>RESULT_DECL</code>.  It will be
<code>NULL_TREE</code> if the statement was just
</p><div class="smallexample">
<pre class="smallexample">return;
</pre></div>

</dd>
<dt><code>LOOP_EXPR</code></dt>
<dd><p>These nodes represent &ldquo;infinite&rdquo; loops.  The <code>LOOP_EXPR_BODY</code>
represents the body of the loop.  It should be executed forever, unless
an <code>EXIT_EXPR</code> is encountered.
</p>
</dd>
<dt><code>EXIT_EXPR</code></dt>
<dd><p>These nodes represent conditional exits from the nearest enclosing
<code>LOOP_EXPR</code>.  The single operand is the condition; if it is
nonzero, then the loop should be exited.  An <code>EXIT_EXPR</code> will only
appear within a <code>LOOP_EXPR</code>.
</p>
</dd>
<dt><code>SWITCH_STMT</code></dt>
<dd>
<p>Used to represent a <code>switch</code> statement.  The <code>SWITCH_STMT_COND</code>
is the expression on which the switch is occurring.  See the documentation
for an <code>IF_STMT</code> for more information on the representation used
for the condition.  The <code>SWITCH_STMT_BODY</code> is the body of the switch
statement.   The <code>SWITCH_STMT_TYPE</code> is the original type of switch
expression as given in the source, before any compiler conversions.
</p>
</dd>
<dt><code>CASE_LABEL_EXPR</code></dt>
<dd>
<p>Use to represent a <code>case</code> label, range of <code>case</code> labels, or a
<code>default</code> label.  If <code>CASE_LOW</code> is <code>NULL_TREE</code>, then this is a
<code>default</code> label.  Otherwise, if <code>CASE_HIGH</code> is <code>NULL_TREE</code>, then
this is an ordinary <code>case</code> label.  In this case, <code>CASE_LOW</code> is
an expression giving the value of the label.  Both <code>CASE_LOW</code> and
<code>CASE_HIGH</code> are <code>INTEGER_CST</code> nodes.  These values will have
the same type as the condition expression in the switch statement.
</p>
<p>Otherwise, if both <code>CASE_LOW</code> and <code>CASE_HIGH</code> are defined, the
statement is a range of case labels.  Such statements originate with the
extension that allows users to write things of the form:
</p><div class="smallexample">
<pre class="smallexample">case 2 ... 5:
</pre></div>
<p>The first value will be <code>CASE_LOW</code>, while the second will be
<code>CASE_HIGH</code>.
</p>
</dd>
</dl>


<hr>
<a name="Blocks"></a>
<div class="header">
<p>
Next: <a href="#Statement-Sequences" accesskey="n" rel="next">Statement Sequences</a>, Previous: <a href="#Basic-Statements" accesskey="p" rel="prev">Basic Statements</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Blocks-1"></a>
<h4 class="subsection">11.7.2 Blocks</h4>
<a name="index-Blocks"></a>

<p>Block scopes and the variables they declare in GENERIC are
expressed using the <code>BIND_EXPR</code> code, which in previous
versions of GCC was primarily used for the C statement-expression
extension.
</p>
<p>Variables in a block are collected into <code>BIND_EXPR_VARS</code> in
declaration order through their <code>TREE_CHAIN</code> field.  Any runtime
initialization is moved out of <code>DECL_INITIAL</code> and into a
statement in the controlled block.  When gimplifying from C or C++,
this initialization replaces the <code>DECL_STMT</code>.  These variables
will never require cleanups.  The scope of these variables is just the
body
</p>
<p>Variable-length arrays (VLAs) complicate this process, as their
size often refers to variables initialized earlier in the block.
To handle this, we currently split the block at that point, and
move the VLA into a new, inner <code>BIND_EXPR</code>.  This strategy
may change in the future.
</p>
<p>A C++ program will usually contain more <code>BIND_EXPR</code>s than
there are syntactic blocks in the source code, since several C++
constructs have implicit scopes associated with them.  On the
other hand, although the C++ front end uses pseudo-scopes to
handle cleanups for objects with destructors, these don&rsquo;t
translate into the GIMPLE form; multiple declarations at the same
level use the same <code>BIND_EXPR</code>.
</p>
<hr>
<a name="Statement-Sequences"></a>
<div class="header">
<p>
Next: <a href="#Empty-Statements" accesskey="n" rel="next">Empty Statements</a>, Previous: <a href="#Blocks" accesskey="p" rel="prev">Blocks</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Statement-Sequences-1"></a>
<h4 class="subsection">11.7.3 Statement Sequences</h4>
<a name="index-Statement-Sequences"></a>

<p>Multiple statements at the same nesting level are collected into
a <code>STATEMENT_LIST</code>.  Statement lists are modified and
traversed using the interface in &lsquo;<samp>tree-iterator.h</samp>&rsquo;.
</p>
<hr>
<a name="Empty-Statements"></a>
<div class="header">
<p>
Next: <a href="#Jumps" accesskey="n" rel="next">Jumps</a>, Previous: <a href="#Statement-Sequences" accesskey="p" rel="prev">Statement Sequences</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Empty-Statements-1"></a>
<h4 class="subsection">11.7.4 Empty Statements</h4>
<a name="index-Empty-Statements"></a>

<p>Whenever possible, statements with no effect are discarded.  But
if they are nested within another construct which cannot be
discarded for some reason, they are instead replaced with an
empty statement, generated by <code>build_empty_stmt</code>.
Initially, all empty statements were shared, after the pattern of
the Java front end, but this caused a lot of trouble in practice.
</p>
<p>An empty statement is represented as <code>(void)0</code>.
</p>
<hr>
<a name="Jumps"></a>
<div class="header">
<p>
Next: <a href="#Cleanups" accesskey="n" rel="next">Cleanups</a>, Previous: <a href="#Empty-Statements" accesskey="p" rel="prev">Empty Statements</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Jumps-1"></a>
<h4 class="subsection">11.7.5 Jumps</h4>
<a name="index-Jumps"></a>

<p>Other jumps are expressed by either <code>GOTO_EXPR</code> or
<code>RETURN_EXPR</code>.
</p>
<p>The operand of a <code>GOTO_EXPR</code> must be either a label or a
variable containing the address to jump to.
</p>
<p>The operand of a <code>RETURN_EXPR</code> is either <code>NULL_TREE</code>,
<code>RESULT_DECL</code>, or a <code>MODIFY_EXPR</code> which sets the return
value.  It would be nice to move the <code>MODIFY_EXPR</code> into a
separate statement, but the special return semantics in
<code>expand_return</code> make that difficult.  It may still happen in
the future, perhaps by moving most of that logic into
<code>expand_assignment</code>.
</p>
<hr>
<a name="Cleanups"></a>
<div class="header">
<p>
Next: <a href="#OpenMP" accesskey="n" rel="next">OpenMP</a>, Previous: <a href="#Jumps" accesskey="p" rel="prev">Jumps</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Cleanups-1"></a>
<h4 class="subsection">11.7.6 Cleanups</h4>
<a name="index-Cleanups"></a>

<p>Destructors for local C++ objects and similar dynamic cleanups are
represented in GIMPLE by a <code>TRY_FINALLY_EXPR</code>.
<code>TRY_FINALLY_EXPR</code> has two operands, both of which are a sequence
of statements to execute.  The first sequence is executed.  When it
completes the second sequence is executed.
</p>
<p>The first sequence may complete in the following ways:
</p>
<ol>
<li> Execute the last statement in the sequence and fall off the
end.

</li><li> Execute a goto statement (<code>GOTO_EXPR</code>) to an ordinary
label outside the sequence.

</li><li> Execute a return statement (<code>RETURN_EXPR</code>).

</li><li> Throw an exception.  This is currently not explicitly represented in
GIMPLE.

</li></ol>

<p>The second sequence is not executed if the first sequence completes by
calling <code>setjmp</code> or <code>exit</code> or any other function that does
not return.  The second sequence is also not executed if the first
sequence completes via a non-local goto or a computed goto (in general
the compiler does not know whether such a goto statement exits the
first sequence or not, so we assume that it doesn&rsquo;t).
</p>
<p>After the second sequence is executed, if it completes normally by
falling off the end, execution continues wherever the first sequence
would have continued, by falling off the end, or doing a goto, etc.
</p>
<p><code>TRY_FINALLY_EXPR</code> complicates the flow graph, since the cleanup
needs to appear on every edge out of the controlled block; this
reduces the freedom to move code across these edges.  Therefore, the
EH lowering pass which runs before most of the optimization passes
eliminates these expressions by explicitly adding the cleanup to each
edge.  Rethrowing the exception is represented using <code>RESX_EXPR</code>.
</p>
<hr>
<a name="OpenMP"></a>
<div class="header">
<p>
Previous: <a href="#Cleanups" accesskey="p" rel="prev">Cleanups</a>, Up: <a href="#Statements" accesskey="u" rel="up">Statements</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="OpenMP-1"></a>
<h4 class="subsection">11.7.7 OpenMP</h4>
<a name="index-OMP_005fPARALLEL"></a>
<a name="index-OMP_005fFOR"></a>
<a name="index-OMP_005fSECTIONS"></a>
<a name="index-OMP_005fSINGLE"></a>
<a name="index-OMP_005fSECTION"></a>
<a name="index-OMP_005fMASTER"></a>
<a name="index-OMP_005fORDERED"></a>
<a name="index-OMP_005fCRITICAL"></a>
<a name="index-OMP_005fRETURN"></a>
<a name="index-OMP_005fCONTINUE"></a>
<a name="index-OMP_005fATOMIC"></a>
<a name="index-OMP_005fCLAUSE"></a>

<p>All the statements starting with <code>OMP_</code> represent directives and
clauses used by the OpenMP API <a href="http://www.openmp.org/">http://www.openmp.org/</a><!-- /@w -->.
</p>
<dl compact="compact">
<dt><code>OMP_PARALLEL</code></dt>
<dd>
<p>Represents <code>#pragma omp parallel [clause1 &hellip; clauseN]</code>. It
has four operands:
</p>
<p>Operand <code>OMP_PARALLEL_BODY</code> is valid while in GENERIC and
High GIMPLE forms.  It contains the body of code to be executed
by all the threads.  During GIMPLE lowering, this operand becomes
<code>NULL</code> and the body is emitted linearly after
<code>OMP_PARALLEL</code>.
</p>
<p>Operand <code>OMP_PARALLEL_CLAUSES</code> is the list of clauses
associated with the directive.
</p>
<p>Operand <code>OMP_PARALLEL_FN</code> is created by
<code>pass_lower_omp</code>, it contains the <code>FUNCTION_DECL</code>
for the function that will contain the body of the parallel
region.
</p>
<p>Operand <code>OMP_PARALLEL_DATA_ARG</code> is also created by
<code>pass_lower_omp</code>. If there are shared variables to be
communicated to the children threads, this operand will contain
the <code>VAR_DECL</code> that contains all the shared values and
variables.
</p>
</dd>
<dt><code>OMP_FOR</code></dt>
<dd>
<p>Represents <code>#pragma omp for [clause1 &hellip; clauseN]</code>.  It
has 5 operands:
</p>
<p>Operand <code>OMP_FOR_BODY</code> contains the loop body.
</p>
<p>Operand <code>OMP_FOR_CLAUSES</code> is the list of clauses
associated with the directive.
</p>
<p>Operand <code>OMP_FOR_INIT</code> is the loop initialization code of
the form <code>VAR = N1</code>.
</p>
<p>Operand <code>OMP_FOR_COND</code> is the loop conditional expression
of the form <code>VAR {&lt;,&gt;,&lt;=,&gt;=} N2</code>.
</p>
<p>Operand <code>OMP_FOR_INCR</code> is the loop index increment of the
form <code>VAR {+=,-=} INCR</code>.
</p>
<p>Operand <code>OMP_FOR_PRE_BODY</code> contains side-effect code from
operands <code>OMP_FOR_INIT</code>, <code>OMP_FOR_COND</code> and
<code>OMP_FOR_INC</code>.  These side-effects are part of the
<code>OMP_FOR</code> block but must be evaluated before the start of
loop body.
</p>
<p>The loop index variable <code>VAR</code> must be a signed integer variable,
which is implicitly private to each thread.  Bounds
<code>N1</code> and <code>N2</code> and the increment expression
<code>INCR</code> are required to be loop invariant integer
expressions that are evaluated without any synchronization. The
evaluation order, frequency of evaluation and side-effects are
unspecified by the standard.
</p>
</dd>
<dt><code>OMP_SECTIONS</code></dt>
<dd>
<p>Represents <code>#pragma omp sections [clause1 &hellip; clauseN]</code>.
</p>
<p>Operand <code>OMP_SECTIONS_BODY</code> contains the sections body,
which in turn contains a set of <code>OMP_SECTION</code> nodes for
each of the concurrent sections delimited by <code>#pragma omp
section</code>.
</p>
<p>Operand <code>OMP_SECTIONS_CLAUSES</code> is the list of clauses
associated with the directive.
</p>
</dd>
<dt><code>OMP_SECTION</code></dt>
<dd>
<p>Section delimiter for <code>OMP_SECTIONS</code>.
</p>
</dd>
<dt><code>OMP_SINGLE</code></dt>
<dd>
<p>Represents <code>#pragma omp single</code>.
</p>
<p>Operand <code>OMP_SINGLE_BODY</code> contains the body of code to be
executed by a single thread.
</p>
<p>Operand <code>OMP_SINGLE_CLAUSES</code> is the list of clauses
associated with the directive.
</p>
</dd>
<dt><code>OMP_MASTER</code></dt>
<dd>
<p>Represents <code>#pragma omp master</code>.
</p>
<p>Operand <code>OMP_MASTER_BODY</code> contains the body of code to be
executed by the master thread.
</p>
</dd>
<dt><code>OMP_ORDERED</code></dt>
<dd>
<p>Represents <code>#pragma omp ordered</code>.
</p>
<p>Operand <code>OMP_ORDERED_BODY</code> contains the body of code to be
executed in the sequential order dictated by the loop index
variable.
</p>
</dd>
<dt><code>OMP_CRITICAL</code></dt>
<dd>
<p>Represents <code>#pragma omp critical [name]</code>.
</p>
<p>Operand <code>OMP_CRITICAL_BODY</code> is the critical section.
</p>
<p>Operand <code>OMP_CRITICAL_NAME</code> is an optional identifier to
label the critical section.
</p>
</dd>
<dt><code>OMP_RETURN</code></dt>
<dd>
<p>This does not represent any OpenMP directive, it is an artificial
marker to indicate the end of the body of an OpenMP. It is used
by the flow graph (<code>tree-cfg.c</code>) and OpenMP region
building code (<code>omp-low.c</code>).
</p>
</dd>
<dt><code>OMP_CONTINUE</code></dt>
<dd>
<p>Similarly, this instruction does not represent an OpenMP
directive, it is used by <code>OMP_FOR</code> and
<code>OMP_SECTIONS</code> to mark the place where the code needs to
loop to the next iteration (in the case of <code>OMP_FOR</code>) or
the next section (in the case of <code>OMP_SECTIONS</code>).
</p>
<p>In some cases, <code>OMP_CONTINUE</code> is placed right before
<code>OMP_RETURN</code>.  But if there are cleanups that need to
occur right after the looping body, it will be emitted between
<code>OMP_CONTINUE</code> and <code>OMP_RETURN</code>.
</p>
</dd>
<dt><code>OMP_ATOMIC</code></dt>
<dd>
<p>Represents <code>#pragma omp atomic</code>.
</p>
<p>Operand 0 is the address at which the atomic operation is to be
performed.
</p>
<p>Operand 1 is the expression to evaluate.  The gimplifier tries
three alternative code generation strategies.  Whenever possible,
an atomic update built-in is used.  If that fails, a
compare-and-swap loop is attempted.  If that also fails, a
regular critical section around the expression is used.
</p>
</dd>
<dt><code>OMP_CLAUSE</code></dt>
<dd>
<p>Represents clauses associated with one of the <code>OMP_</code> directives.
Clauses are represented by separate sub-codes defined in
<samp>tree.h</samp>.  Clauses codes can be one of:
<code>OMP_CLAUSE_PRIVATE</code>, <code>OMP_CLAUSE_SHARED</code>,
<code>OMP_CLAUSE_FIRSTPRIVATE</code>,
<code>OMP_CLAUSE_LASTPRIVATE</code>, <code>OMP_CLAUSE_COPYIN</code>,
<code>OMP_CLAUSE_COPYPRIVATE</code>, <code>OMP_CLAUSE_IF</code>,
<code>OMP_CLAUSE_NUM_THREADS</code>, <code>OMP_CLAUSE_SCHEDULE</code>,
<code>OMP_CLAUSE_NOWAIT</code>, <code>OMP_CLAUSE_ORDERED</code>,
<code>OMP_CLAUSE_DEFAULT</code>, <code>OMP_CLAUSE_REDUCTION</code>,
<code>OMP_CLAUSE_COLLAPSE</code>, <code>OMP_CLAUSE_UNTIED</code>,
<code>OMP_CLAUSE_FINAL</code>, and <code>OMP_CLAUSE_MERGEABLE</code>.  Each code
represents the corresponding OpenMP clause.
</p>
<p>Clauses associated with the same directive are chained together
via <code>OMP_CLAUSE_CHAIN</code>. Those clauses that accept a list
of variables are restricted to exactly one, accessed with
<code>OMP_CLAUSE_VAR</code>.  Therefore, multiple variables under the
same clause <code>C</code> need to be represented as multiple <code>C</code> clauses
chained together.  This facilitates adding new clauses during
compilation.
</p>
</dd>
</dl>


<hr>
<a name="Functions"></a>
<div class="header">
<p>
Next: <a href="#Language_002ddependent-trees" accesskey="n" rel="next">Language-dependent trees</a>, Previous: <a href="#Statements" accesskey="p" rel="prev">Statements</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Functions-1"></a>
<h3 class="section">11.8 Functions</h3>
<a name="index-function"></a>
<a name="index-FUNCTION_005fDECL"></a>

<p>A function is represented by a <code>FUNCTION_DECL</code> node.  It stores
the basic pieces of the function such as body, parameters, and return
type as well as information on the surrounding context, visibility,
and linkage.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Function-Basics" accesskey="1">Function Basics</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Function names, body, and parameters.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Function-Properties" accesskey="2">Function Properties</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Context, linkage, etc.
</td></tr>
</table>


<hr>
<a name="Function-Basics"></a>
<div class="header">
<p>
Next: <a href="#Function-Properties" accesskey="n" rel="next">Function Properties</a>, Up: <a href="#Functions" accesskey="u" rel="up">Functions</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Function-Basics-1"></a>
<h4 class="subsection">11.8.1 Function Basics</h4>
<a name="index-DECL_005fNAME-1"></a>
<a name="index-DECL_005fASSEMBLER_005fNAME"></a>
<a name="index-TREE_005fPUBLIC"></a>
<a name="index-DECL_005fARTIFICIAL-1"></a>
<a name="index-DECL_005fFUNCTION_005fSPECIFIC_005fTARGET"></a>
<a name="index-DECL_005fFUNCTION_005fSPECIFIC_005fOPTIMIZATION"></a>

<p>A function has four core parts: the name, the parameters, the result,
and the body.  The following macros and functions access these parts
of a <code>FUNCTION_DECL</code> as well as other basic features:
</p><dl compact="compact">
<dt><code>DECL_NAME</code>
<a name="index-DECL_005fNAME-2"></a>
</dt>
<dd><p>This macro returns the unqualified name of the function, as an
<code>IDENTIFIER_NODE</code>.  For an instantiation of a function template,
the <code>DECL_NAME</code> is the unqualified name of the template, not
something like <code>f&lt;int&gt;</code>.  The value of <code>DECL_NAME</code> is
undefined when used on a constructor, destructor, overloaded operator,
or type-conversion operator, or any function that is implicitly
generated by the compiler.  See below for macros that can be used to
distinguish these cases.
</p>
</dd>
<dt><code>DECL_ASSEMBLER_NAME</code>
<a name="index-DECL_005fASSEMBLER_005fNAME-1"></a>
</dt>
<dd><p>This macro returns the mangled name of the function, also an
<code>IDENTIFIER_NODE</code>.  This name does not contain leading underscores
on systems that prefix all identifiers with underscores.  The mangled
name is computed in the same way on all platforms; if special processing
is required to deal with the object file format used on a particular
platform, it is the responsibility of the back end to perform those
modifications.  (Of course, the back end should not modify
<code>DECL_ASSEMBLER_NAME</code> itself.)
</p>
<p>Using <code>DECL_ASSEMBLER_NAME</code> will cause additional memory to be
allocated (for the mangled name of the entity) so it should be used
only when emitting assembly code.  It should not be used within the
optimizers to determine whether or not two declarations are the same,
even though some of the existing optimizers do use it in that way.
These uses will be removed over time.
</p>
</dd>
<dt><code>DECL_ARGUMENTS</code>
<a name="index-DECL_005fARGUMENTS"></a>
</dt>
<dd><p>This macro returns the <code>PARM_DECL</code> for the first argument to the
function.  Subsequent <code>PARM_DECL</code> nodes can be obtained by
following the <code>TREE_CHAIN</code> links.
</p>
</dd>
<dt><code>DECL_RESULT</code>
<a name="index-DECL_005fRESULT"></a>
</dt>
<dd><p>This macro returns the <code>RESULT_DECL</code> for the function.
</p>
</dd>
<dt><code>DECL_SAVED_TREE</code>
<a name="index-DECL_005fSAVED_005fTREE"></a>
</dt>
<dd><p>This macro returns the complete body of the function.
</p>
</dd>
<dt><code>TREE_TYPE</code>
<a name="index-TREE_005fTYPE-5"></a>
</dt>
<dd><p>This macro returns the <code>FUNCTION_TYPE</code> or <code>METHOD_TYPE</code> for
the function.
</p>
</dd>
<dt><code>DECL_INITIAL</code>
<a name="index-DECL_005fINITIAL-1"></a>
</dt>
<dd><p>A function that has a definition in the current translation unit will
have a non-<code>NULL</code> <code>DECL_INITIAL</code>.  However, back ends should not make
use of the particular value given by <code>DECL_INITIAL</code>.
</p>
<p>It should contain a tree of <code>BLOCK</code> nodes that mirrors the scopes
that variables are bound in the function.  Each block contains a list
of decls declared in a basic block, a pointer to a chain of blocks at
the next lower scope level, then a pointer to the next block at the
same level and a backpointer to the parent <code>BLOCK</code> or
<code>FUNCTION_DECL</code>.  So given a function as follows:
</p>
<div class="smallexample">
<pre class="smallexample">void foo()
{
  int a;
  {
    int b;
  }
  int c;
}
</pre></div>

<p>you would get the following:
</p>
<div class="smallexample">
<pre class="smallexample">tree foo = FUNCTION_DECL;
tree decl_a = VAR_DECL;
tree decl_b = VAR_DECL;
tree decl_c = VAR_DECL;
tree block_a = BLOCK;
tree block_b = BLOCK;
tree block_c = BLOCK;
BLOCK_VARS(block_a) = decl_a;
BLOCK_SUBBLOCKS(block_a) = block_b;
BLOCK_CHAIN(block_a) = block_c;
BLOCK_SUPERCONTEXT(block_a) = foo;
BLOCK_VARS(block_b) = decl_b;
BLOCK_SUPERCONTEXT(block_b) = block_a;
BLOCK_VARS(block_c) = decl_c;
BLOCK_SUPERCONTEXT(block_c) = foo;
DECL_INITIAL(foo) = block_a;
</pre></div>

</dd>
</dl>


<hr>
<a name="Function-Properties"></a>
<div class="header">
<p>
Previous: <a href="#Function-Basics" accesskey="p" rel="prev">Function Basics</a>, Up: <a href="#Functions" accesskey="u" rel="up">Functions</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Function-Properties-1"></a>
<h4 class="subsection">11.8.2 Function Properties</h4>
<a name="index-function-properties"></a>
<a name="index-statements"></a>

<p>To determine the scope of a function, you can use the
<code>DECL_CONTEXT</code> macro.  This macro will return the class
(either a <code>RECORD_TYPE</code> or a <code>UNION_TYPE</code>) or namespace (a
<code>NAMESPACE_DECL</code>) of which the function is a member.  For a virtual
function, this macro returns the class in which the function was
actually defined, not the base class in which the virtual declaration
occurred.
</p>
<p>In C, the <code>DECL_CONTEXT</code> for a function maybe another function.
This representation indicates that the GNU nested function extension
is in use.  For details on the semantics of nested functions, see the
GCC Manual.  The nested function can refer to local variables in its
containing function.  Such references are not explicitly marked in the
tree structure; back ends must look at the <code>DECL_CONTEXT</code> for the
referenced <code>VAR_DECL</code>.  If the <code>DECL_CONTEXT</code> for the
referenced <code>VAR_DECL</code> is not the same as the function currently
being processed, and neither <code>DECL_EXTERNAL</code> nor
<code>TREE_STATIC</code> hold, then the reference is to a local variable in
a containing function, and the back end must take appropriate action.
</p>
<dl compact="compact">
<dt><code>DECL_EXTERNAL</code>
<a name="index-DECL_005fEXTERNAL-1"></a>
</dt>
<dd><p>This predicate holds if the function is undefined.
</p>
</dd>
<dt><code>TREE_PUBLIC</code>
<a name="index-TREE_005fPUBLIC-1"></a>
</dt>
<dd><p>This predicate holds if the function has external linkage.
</p>
</dd>
<dt><code>TREE_STATIC</code>
<a name="index-TREE_005fSTATIC"></a>
</dt>
<dd><p>This predicate holds if the function has been defined.
</p>
</dd>
<dt><code>TREE_THIS_VOLATILE</code>
<a name="index-TREE_005fTHIS_005fVOLATILE"></a>
</dt>
<dd><p>This predicate holds if the function does not return normally.
</p>
</dd>
<dt><code>TREE_READONLY</code>
<a name="index-TREE_005fREADONLY"></a>
</dt>
<dd><p>This predicate holds if the function can only read its arguments.
</p>
</dd>
<dt><code>DECL_PURE_P</code>
<a name="index-DECL_005fPURE_005fP"></a>
</dt>
<dd><p>This predicate holds if the function can only read its arguments, but
may also read global memory.
</p>
</dd>
<dt><code>DECL_VIRTUAL_P</code>
<a name="index-DECL_005fVIRTUAL_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is virtual.
</p>
</dd>
<dt><code>DECL_ARTIFICIAL</code>
<a name="index-DECL_005fARTIFICIAL-2"></a>
</dt>
<dd><p>This macro holds if the function was implicitly generated by the
compiler, rather than explicitly declared.  In addition to implicitly
generated class member functions, this macro holds for the special
functions created to implement static initialization and destruction, to
compute run-time type information, and so forth.
</p>
</dd>
<dt><code>DECL_FUNCTION_SPECIFIC_TARGET</code>
<a name="index-DECL_005fFUNCTION_005fSPECIFIC_005fTARGET-1"></a>
</dt>
<dd><p>This macro returns a tree node that holds the target options that are
to be used to compile this particular function or <code>NULL_TREE</code> if
the function is to be compiled with the target options specified on
the command line.
</p>
</dd>
<dt><code>DECL_FUNCTION_SPECIFIC_OPTIMIZATION</code>
<a name="index-DECL_005fFUNCTION_005fSPECIFIC_005fOPTIMIZATION-1"></a>
</dt>
<dd><p>This macro returns a tree node that holds the optimization options
that are to be used to compile this particular function or
<code>NULL_TREE</code> if the function is to be compiled with the
optimization options specified on the command line.
</p>
</dd>
</dl>


<hr>
<a name="Language_002ddependent-trees"></a>
<div class="header">
<p>
Next: <a href="#C-and-C_002b_002b-Trees" accesskey="n" rel="next">C and C++ Trees</a>, Previous: <a href="#Functions" accesskey="p" rel="prev">Functions</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Language_002ddependent-trees-1"></a>
<h3 class="section">11.9 Language-dependent trees</h3>
<a name="index-language_002ddependent-trees"></a>

<p>Front ends may wish to keep some state associated with various GENERIC
trees while parsing.  To support this, trees provide a set of flags
that may be used by the front end.  They are accessed using
<code>TREE_LANG_FLAG_n</code> where &lsquo;<samp>n</samp>&rsquo; is currently 0 through 6.
</p>
<p>If necessary, a front end can use some language-dependent tree
codes in its GENERIC representation, so long as it provides a
hook for converting them to GIMPLE and doesn&rsquo;t expect them to
work with any (hypothetical) optimizers that run before the
conversion to GIMPLE. The intermediate representation used while
parsing C and C++ looks very little like GENERIC, but the C and
C++ gimplifier hooks are perfectly happy to take it as input and
spit out GIMPLE.
</p>


<hr>
<a name="C-and-C_002b_002b-Trees"></a>
<div class="header">
<p>
Next: <a href="#Java-Trees" accesskey="n" rel="next">Java Trees</a>, Previous: <a href="#Language_002ddependent-trees" accesskey="p" rel="prev">Language-dependent trees</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="C-and-C_002b_002b-Trees-1"></a>
<h3 class="section">11.10 C and C++ Trees</h3>

<p>This section documents the internal representation used by GCC to
represent C and C++ source programs.  When presented with a C or C++
source program, GCC parses the program, performs semantic analysis
(including the generation of error messages), and then produces the
internal representation described here.  This representation contains a
complete representation for the entire translation unit provided as
input to the front end.  This representation is then typically processed
by a code-generator in order to produce machine code, but could also be
used in the creation of source browsers, intelligent editors, automatic
documentation generators, interpreters, and any other programs needing
the ability to process C or C++ code.
</p>
<p>This section explains the internal representation.  In particular, it
documents the internal representation for C and C++ source
constructs, and the macros, functions, and variables that can be used to
access these constructs.  The C++ representation is largely a superset
of the representation used in the C front end.  There is only one
construct used in C that does not appear in the C++ front end and that
is the GNU &ldquo;nested function&rdquo; extension.  Many of the macros documented
here do not apply in C because the corresponding language constructs do
not appear in C.
</p>
<p>The C and C++ front ends generate a mix of GENERIC trees and ones
specific to C and C++.  These language-specific trees are higher-level
constructs than the ones in GENERIC to make the parser&rsquo;s job easier.
This section describes those trees that aren&rsquo;t part of GENERIC as well
as aspects of GENERIC trees that are treated in a language-specific
manner.
</p>
<p>If you are developing a &ldquo;back end&rdquo;, be it is a code-generator or some
other tool, that uses this representation, you may occasionally find
that you need to ask questions not easily answered by the functions and
macros available here.  If that situation occurs, it is quite likely
that GCC already supports the functionality you desire, but that the
interface is simply not documented here.  In that case, you should ask
the GCC maintainers (via mail to <a href="mailto:gcc@gcc.gnu.org">gcc@gcc.gnu.org</a>) about
documenting the functionality you require.  Similarly, if you find
yourself writing functions that do not deal directly with your back end,
but instead might be useful to other people using the GCC front end, you
should submit your patches for inclusion in GCC.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Types-for-C_002b_002b" accesskey="1">Types for C++</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Fundamental and aggregate types.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Namespaces" accesskey="2">Namespaces</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Namespaces.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Classes" accesskey="3">Classes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Classes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Functions-for-C_002b_002b" accesskey="4">Functions for C++</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Overloading and accessors for C++.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Statements-for-C_002b_002b" accesskey="5">Statements for C++</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Statements specific to C and C++.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#C_002b_002b-Expressions" accesskey="6">C++ Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">From <code>typeid</code> to <code>throw</code>.
</td></tr>
</table>

<hr>
<a name="Types-for-C_002b_002b"></a>
<div class="header">
<p>
Next: <a href="#Namespaces" accesskey="n" rel="next">Namespaces</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Types-for-C_002b_002b-1"></a>
<h4 class="subsection">11.10.1 Types for C++</h4>
<a name="index-UNKNOWN_005fTYPE-1"></a>
<a name="index-TYPENAME_005fTYPE"></a>
<a name="index-TYPEOF_005fTYPE"></a>
<a name="index-cp_005ftype_005fquals"></a>
<a name="index-TYPE_005fUNQUALIFIED-1"></a>
<a name="index-TYPE_005fQUAL_005fCONST-1"></a>
<a name="index-TYPE_005fQUAL_005fVOLATILE-1"></a>
<a name="index-TYPE_005fQUAL_005fRESTRICT-1"></a>
<a name="index-TYPE_005fMAIN_005fVARIANT-2"></a>
<a name="index-qualified-type-1"></a>
<a name="index-TYPE_005fSIZE-2"></a>
<a name="index-TYPE_005fALIGN-2"></a>
<a name="index-TYPE_005fPRECISION-1"></a>
<a name="index-TYPE_005fARG_005fTYPES-1"></a>
<a name="index-TYPE_005fMETHOD_005fBASETYPE-1"></a>
<a name="index-TYPE_005fPTRDATAMEM_005fP"></a>
<a name="index-TYPE_005fOFFSET_005fBASETYPE-1"></a>
<a name="index-TREE_005fTYPE-6"></a>
<a name="index-TYPE_005fCONTEXT-1"></a>
<a name="index-TYPE_005fNAME-2"></a>
<a name="index-TYPENAME_005fTYPE_005fFULLNAME-1"></a>
<a name="index-TYPE_005fFIELDS-1"></a>
<a name="index-TYPE_005fPTROBV_005fP"></a>

<p>In C++, an array type is not qualified; rather the type of the array
elements is qualified.  This situation is reflected in the intermediate
representation.  The macros described here will always examine the
qualification of the underlying element type when applied to an array
type.  (If the element type is itself an array, then the recursion
continues until a non-array type is found, and the qualification of this
type is examined.)  So, for example, <code>CP_TYPE_CONST_P</code> will hold of
the type <code>const int ()[7]</code>, denoting an array of seven <code>int</code>s.
</p>
<p>The following functions and macros deal with cv-qualification of types:
</p><dl compact="compact">
<dt><code>cp_type_quals</code>
<a name="index-cp_005ftype_005fquals-1"></a>
</dt>
<dd><p>This function returns the set of type qualifiers applied to this type.
This value is <code>TYPE_UNQUALIFIED</code> if no qualifiers have been
applied.  The <code>TYPE_QUAL_CONST</code> bit is set if the type is
<code>const</code>-qualified.  The <code>TYPE_QUAL_VOLATILE</code> bit is set if the
type is <code>volatile</code>-qualified.  The <code>TYPE_QUAL_RESTRICT</code> bit is
set if the type is <code>restrict</code>-qualified.
</p>
</dd>
<dt><code>CP_TYPE_CONST_P</code>
<a name="index-CP_005fTYPE_005fCONST_005fP"></a>
</dt>
<dd><p>This macro holds if the type is <code>const</code>-qualified.
</p>
</dd>
<dt><code>CP_TYPE_VOLATILE_P</code>
<a name="index-CP_005fTYPE_005fVOLATILE_005fP"></a>
</dt>
<dd><p>This macro holds if the type is <code>volatile</code>-qualified.
</p>
</dd>
<dt><code>CP_TYPE_RESTRICT_P</code>
<a name="index-CP_005fTYPE_005fRESTRICT_005fP"></a>
</dt>
<dd><p>This macro holds if the type is <code>restrict</code>-qualified.
</p>
</dd>
<dt><code>CP_TYPE_CONST_NON_VOLATILE_P</code>
<a name="index-CP_005fTYPE_005fCONST_005fNON_005fVOLATILE_005fP"></a>
</dt>
<dd><p>This predicate holds for a type that is <code>const</code>-qualified, but
<em>not</em> <code>volatile</code>-qualified; other cv-qualifiers are ignored as
well: only the <code>const</code>-ness is tested.
</p>
</dd>
</dl>

<p>A few other macros and functions are usable with all types:
</p><dl compact="compact">
<dt><code>TYPE_SIZE</code>
<a name="index-TYPE_005fSIZE-3"></a>
</dt>
<dd><p>The number of bits required to represent the type, represented as an
<code>INTEGER_CST</code>.  For an incomplete type, <code>TYPE_SIZE</code> will be
<code>NULL_TREE</code>.
</p>
</dd>
<dt><code>TYPE_ALIGN</code>
<a name="index-TYPE_005fALIGN-3"></a>
</dt>
<dd><p>The alignment of the type, in bits, represented as an <code>int</code>.
</p>
</dd>
<dt><code>TYPE_NAME</code>
<a name="index-TYPE_005fNAME-3"></a>
</dt>
<dd><p>This macro returns a declaration (in the form of a <code>TYPE_DECL</code>) for
the type.  (Note this macro does <em>not</em> return an
<code>IDENTIFIER_NODE</code>, as you might expect, given its name!)  You can
look at the <code>DECL_NAME</code> of the <code>TYPE_DECL</code> to obtain the
actual name of the type.  The <code>TYPE_NAME</code> will be <code>NULL_TREE</code>
for a type that is not a built-in type, the result of a typedef, or a
named class type.
</p>
</dd>
<dt><code>CP_INTEGRAL_TYPE</code>
<a name="index-CP_005fINTEGRAL_005fTYPE"></a>
</dt>
<dd><p>This predicate holds if the type is an integral type.  Notice that in
C++, enumerations are <em>not</em> integral types.
</p>
</dd>
<dt><code>ARITHMETIC_TYPE_P</code>
<a name="index-ARITHMETIC_005fTYPE_005fP"></a>
</dt>
<dd><p>This predicate holds if the type is an integral type (in the C++ sense)
or a floating point type.
</p>
</dd>
<dt><code>CLASS_TYPE_P</code>
<a name="index-CLASS_005fTYPE_005fP"></a>
</dt>
<dd><p>This predicate holds for a class-type.
</p>
</dd>
<dt><code>TYPE_BUILT_IN</code>
<a name="index-TYPE_005fBUILT_005fIN"></a>
</dt>
<dd><p>This predicate holds for a built-in type.
</p>
</dd>
<dt><code>TYPE_PTRDATAMEM_P</code>
<a name="index-TYPE_005fPTRDATAMEM_005fP-1"></a>
</dt>
<dd><p>This predicate holds if the type is a pointer to data member.
</p>
</dd>
<dt><code>TYPE_PTR_P</code>
<a name="index-TYPE_005fPTR_005fP"></a>
</dt>
<dd><p>This predicate holds if the type is a pointer type, and the pointee is
not a data member.
</p>
</dd>
<dt><code>TYPE_PTRFN_P</code>
<a name="index-TYPE_005fPTRFN_005fP"></a>
</dt>
<dd><p>This predicate holds for a pointer to function type.
</p>
</dd>
<dt><code>TYPE_PTROB_P</code>
<a name="index-TYPE_005fPTROB_005fP"></a>
</dt>
<dd><p>This predicate holds for a pointer to object type.  Note however that it
does not hold for the generic pointer to object type <code>void *</code>.  You
may use <code>TYPE_PTROBV_P</code> to test for a pointer to object type as
well as <code>void *</code>.
</p>
</dd>
</dl>

<p>The table below describes types specific to C and C++ as well as
language-dependent info about GENERIC types.
</p>
<dl compact="compact">
<dt><code>POINTER_TYPE</code></dt>
<dd><p>Used to represent pointer types, and pointer to data member types.  If
<code>TREE_TYPE</code>
is a pointer to data member type, then <code>TYPE_PTRDATAMEM_P</code> will hold.
For a pointer to data member type of the form &lsquo;<samp>T X::*</samp>&rsquo;,
<code>TYPE_PTRMEM_CLASS_TYPE</code> will be the type <code>X</code>, while
<code>TYPE_PTRMEM_POINTED_TO_TYPE</code> will be the type <code>T</code>.
</p>
</dd>
<dt><code>RECORD_TYPE</code></dt>
<dd><p>Used to represent <code>struct</code> and <code>class</code> types in C and C++.  If
<code>TYPE_PTRMEMFUNC_P</code> holds, then this type is a pointer-to-member
type.  In that case, the <code>TYPE_PTRMEMFUNC_FN_TYPE</code> is a
<code>POINTER_TYPE</code> pointing to a <code>METHOD_TYPE</code>.  The
<code>METHOD_TYPE</code> is the type of a function pointed to by the
pointer-to-member function.  If <code>TYPE_PTRMEMFUNC_P</code> does not hold,
this type is a class type.  For more information, see <a href="#Classes">Classes</a>.
</p>
</dd>
<dt><code>UNKNOWN_TYPE</code></dt>
<dd><p>This node is used to represent a type the knowledge of which is
insufficient for a sound processing.
</p>
</dd>
<dt><code>TYPENAME_TYPE</code></dt>
<dd><p>Used to represent a construct of the form <code>typename T::A</code>.  The
<code>TYPE_CONTEXT</code> is <code>T</code>; the <code>TYPE_NAME</code> is an
<code>IDENTIFIER_NODE</code> for <code>A</code>.  If the type is specified via a
template-id, then <code>TYPENAME_TYPE_FULLNAME</code> yields a
<code>TEMPLATE_ID_EXPR</code>.  The <code>TREE_TYPE</code> is non-<code>NULL</code> if the
node is implicitly generated in support for the implicit typename
extension; in which case the <code>TREE_TYPE</code> is a type node for the
base-class.
</p>
</dd>
<dt><code>TYPEOF_TYPE</code></dt>
<dd><p>Used to represent the <code>__typeof__</code> extension.  The
<code>TYPE_FIELDS</code> is the expression the type of which is being
represented.
</p>
</dd>
</dl>



<hr>
<a name="Namespaces"></a>
<div class="header">
<p>
Next: <a href="#Classes" accesskey="n" rel="next">Classes</a>, Previous: <a href="#Types-for-C_002b_002b" accesskey="p" rel="prev">Types for C++</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Namespaces-1"></a>
<h4 class="subsection">11.10.2 Namespaces</h4>
<a name="index-namespace_002c-scope"></a>
<a name="index-NAMESPACE_005fDECL-1"></a>

<p>The root of the entire intermediate representation is the variable
<code>global_namespace</code>.  This is the namespace specified with <code>::</code>
in C++ source code.  All other namespaces, types, variables, functions,
and so forth can be found starting with this namespace.
</p>
<p>However, except for the fact that it is distinguished as the root of the
representation, the global namespace is no different from any other
namespace.  Thus, in what follows, we describe namespaces generally,
rather than the global namespace in particular.
</p>
<p>A namespace is represented by a <code>NAMESPACE_DECL</code> node.
</p>
<p>The following macros and functions can be used on a <code>NAMESPACE_DECL</code>:
</p>
<dl compact="compact">
<dt><code>DECL_NAME</code>
<a name="index-DECL_005fNAME-3"></a>
</dt>
<dd><p>This macro is used to obtain the <code>IDENTIFIER_NODE</code> corresponding to
the unqualified name of the name of the namespace (see <a href="#Identifiers">Identifiers</a>).
The name of the global namespace is &lsquo;<samp>::</samp>&rsquo;, even though in C++ the
global namespace is unnamed.  However, you should use comparison with
<code>global_namespace</code>, rather than <code>DECL_NAME</code> to determine
whether or not a namespace is the global one.  An unnamed namespace
will have a <code>DECL_NAME</code> equal to <code>anonymous_namespace_name</code>.
Within a single translation unit, all unnamed namespaces will have the
same name.
</p>
</dd>
<dt><code>DECL_CONTEXT</code>
<a name="index-DECL_005fCONTEXT"></a>
</dt>
<dd><p>This macro returns the enclosing namespace.  The <code>DECL_CONTEXT</code> for
the <code>global_namespace</code> is <code>NULL_TREE</code>.
</p>
</dd>
<dt><code>DECL_NAMESPACE_ALIAS</code>
<a name="index-DECL_005fNAMESPACE_005fALIAS"></a>
</dt>
<dd><p>If this declaration is for a namespace alias, then
<code>DECL_NAMESPACE_ALIAS</code> is the namespace for which this one is an
alias.
</p>
<p>Do not attempt to use <code>cp_namespace_decls</code> for a namespace which is
an alias.  Instead, follow <code>DECL_NAMESPACE_ALIAS</code> links until you
reach an ordinary, non-alias, namespace, and call
<code>cp_namespace_decls</code> there.
</p>
</dd>
<dt><code>DECL_NAMESPACE_STD_P</code>
<a name="index-DECL_005fNAMESPACE_005fSTD_005fP"></a>
</dt>
<dd><p>This predicate holds if the namespace is the special <code>::std</code>
namespace.
</p>
</dd>
<dt><code>cp_namespace_decls</code>
<a name="index-cp_005fnamespace_005fdecls"></a>
</dt>
<dd><p>This function will return the declarations contained in the namespace,
including types, overloaded functions, other namespaces, and so forth.
If there are no declarations, this function will return
<code>NULL_TREE</code>.  The declarations are connected through their
<code>TREE_CHAIN</code> fields.
</p>
<p>Although most entries on this list will be declarations,
<code>TREE_LIST</code> nodes may also appear.  In this case, the
<code>TREE_VALUE</code> will be an <code>OVERLOAD</code>.  The value of the
<code>TREE_PURPOSE</code> is unspecified; back ends should ignore this value.
As with the other kinds of declarations returned by
<code>cp_namespace_decls</code>, the <code>TREE_CHAIN</code> will point to the next
declaration in this list.
</p>
<p>For more information on the kinds of declarations that can occur on this
list, See <a href="#Declarations">Declarations</a>.  Some declarations will not appear on this
list.  In particular, no <code>FIELD_DECL</code>, <code>LABEL_DECL</code>, or
<code>PARM_DECL</code> nodes will appear here.
</p>
<p>This function cannot be used with namespaces that have
<code>DECL_NAMESPACE_ALIAS</code> set.
</p>
</dd>
</dl>


<hr>
<a name="Classes"></a>
<div class="header">
<p>
Next: <a href="#Functions-for-C_002b_002b" accesskey="n" rel="next">Functions for C++</a>, Previous: <a href="#Namespaces" accesskey="p" rel="prev">Namespaces</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Classes-1"></a>
<h4 class="subsection">11.10.3 Classes</h4>
<a name="index-class_002c-scope"></a>
<a name="index-RECORD_005fTYPE-1"></a>
<a name="index-UNION_005fTYPE-1"></a>
<a name="index-CLASSTYPE_005fDECLARED_005fCLASS"></a>
<a name="index-TYPE_005fBINFO"></a>
<a name="index-BINFO_005fTYPE"></a>
<a name="index-TYPE_005fFIELDS-2"></a>
<a name="index-TYPE_005fVFIELD"></a>
<a name="index-TYPE_005fMETHODS"></a>

<p>Besides namespaces, the other high-level scoping construct in C++ is the
class.  (Throughout this manual the term <em>class</em> is used to mean the
types referred to in the ANSI/ISO C++ Standard as classes; these include
types defined with the <code>class</code>, <code>struct</code>, and <code>union</code>
keywords.)
</p>
<p>A class type is represented by either a <code>RECORD_TYPE</code> or a
<code>UNION_TYPE</code>.  A class declared with the <code>union</code> tag is
represented by a <code>UNION_TYPE</code>, while classes declared with either
the <code>struct</code> or the <code>class</code> tag are represented by
<code>RECORD_TYPE</code>s.  You can use the <code>CLASSTYPE_DECLARED_CLASS</code>
macro to discern whether or not a particular type is a <code>class</code> as
opposed to a <code>struct</code>.  This macro will be true only for classes
declared with the <code>class</code> tag.
</p>
<p>Almost all non-function members are available on the <code>TYPE_FIELDS</code>
list.  Given one member, the next can be found by following the
<code>TREE_CHAIN</code>.  You should not depend in any way on the order in
which fields appear on this list.  All nodes on this list will be
&lsquo;<samp>DECL</samp>&rsquo; nodes.  A <code>FIELD_DECL</code> is used to represent a non-static
data member, a <code>VAR_DECL</code> is used to represent a static data
member, and a <code>TYPE_DECL</code> is used to represent a type.  Note that
the <code>CONST_DECL</code> for an enumeration constant will appear on this
list, if the enumeration type was declared in the class.  (Of course,
the <code>TYPE_DECL</code> for the enumeration type will appear here as well.)
There are no entries for base classes on this list.  In particular,
there is no <code>FIELD_DECL</code> for the &ldquo;base-class portion&rdquo; of an
object.
</p>
<p>The <code>TYPE_VFIELD</code> is a compiler-generated field used to point to
virtual function tables.  It may or may not appear on the
<code>TYPE_FIELDS</code> list.  However, back ends should handle the
<code>TYPE_VFIELD</code> just like all the entries on the <code>TYPE_FIELDS</code>
list.
</p>
<p>The function members are available on the <code>TYPE_METHODS</code> list.
Again, subsequent members are found by following the <code>TREE_CHAIN</code>
field.  If a function is overloaded, each of the overloaded functions
appears; no <code>OVERLOAD</code> nodes appear on the <code>TYPE_METHODS</code>
list.  Implicitly declared functions (including default constructors,
copy constructors, assignment operators, and destructors) will appear on
this list as well.
</p>
<p>Every class has an associated <em>binfo</em>, which can be obtained with
<code>TYPE_BINFO</code>.  Binfos are used to represent base-classes.  The
binfo given by <code>TYPE_BINFO</code> is the degenerate case, whereby every
class is considered to be its own base-class.  The base binfos for a
particular binfo are held in a vector, whose length is obtained with
<code>BINFO_N_BASE_BINFOS</code>.  The base binfos themselves are obtained
with <code>BINFO_BASE_BINFO</code> and <code>BINFO_BASE_ITERATE</code>.  To add a
new binfo, use <code>BINFO_BASE_APPEND</code>.  The vector of base binfos can
be obtained with <code>BINFO_BASE_BINFOS</code>, but normally you do not need
to use that.  The class type associated with a binfo is given by
<code>BINFO_TYPE</code>.  It is not always the case that <code>BINFO_TYPE
(TYPE_BINFO (x))</code>, because of typedefs and qualified types.  Neither is
it the case that <code>TYPE_BINFO (BINFO_TYPE (y))</code> is the same binfo as
<code>y</code>.  The reason is that if <code>y</code> is a binfo representing a
base-class <code>B</code> of a derived class <code>D</code>, then <code>BINFO_TYPE
(y)</code> will be <code>B</code>, and <code>TYPE_BINFO (BINFO_TYPE (y))</code> will be
<code>B</code> as its own base-class, rather than as a base-class of <code>D</code>.
</p>
<p>The access to a base type can be found with <code>BINFO_BASE_ACCESS</code>.
This will produce <code>access_public_node</code>, <code>access_private_node</code>
or <code>access_protected_node</code>.  If bases are always public,
<code>BINFO_BASE_ACCESSES</code> may be <code>NULL</code>.
</p>
<p><code>BINFO_VIRTUAL_P</code> is used to specify whether the binfo is inherited
virtually or not.  The other flags, <code>BINFO_MARKED_P</code> and
<code>BINFO_FLAG_1</code> to <code>BINFO_FLAG_6</code> can be used for language
specific use.
</p>
<p>The following macros can be used on a tree node representing a class-type.
</p>
<dl compact="compact">
<dt><code>LOCAL_CLASS_P</code>
<a name="index-LOCAL_005fCLASS_005fP"></a>
</dt>
<dd><p>This predicate holds if the class is local class <em>i.e.</em> declared
inside a function body.
</p>
</dd>
<dt><code>TYPE_POLYMORPHIC_P</code>
<a name="index-TYPE_005fPOLYMORPHIC_005fP"></a>
</dt>
<dd><p>This predicate holds if the class has at least one virtual function
(declared or inherited).
</p>
</dd>
<dt><code>TYPE_HAS_DEFAULT_CONSTRUCTOR</code>
<a name="index-TYPE_005fHAS_005fDEFAULT_005fCONSTRUCTOR"></a>
</dt>
<dd><p>This predicate holds whenever its argument represents a class-type with
default constructor.
</p>
</dd>
<dt><code>CLASSTYPE_HAS_MUTABLE</code>
<a name="index-CLASSTYPE_005fHAS_005fMUTABLE"></a>
</dt>
<dt><code>TYPE_HAS_MUTABLE_P</code>
<a name="index-TYPE_005fHAS_005fMUTABLE_005fP"></a>
</dt>
<dd><p>These predicates hold for a class-type having a mutable data member.
</p>
</dd>
<dt><code>CLASSTYPE_NON_POD_P</code>
<a name="index-CLASSTYPE_005fNON_005fPOD_005fP"></a>
</dt>
<dd><p>This predicate holds only for class-types that are not PODs.
</p>
</dd>
<dt><code>TYPE_HAS_NEW_OPERATOR</code>
<a name="index-TYPE_005fHAS_005fNEW_005fOPERATOR"></a>
</dt>
<dd><p>This predicate holds for a class-type that defines
<code>operator new</code>.
</p>
</dd>
<dt><code>TYPE_HAS_ARRAY_NEW_OPERATOR</code>
<a name="index-TYPE_005fHAS_005fARRAY_005fNEW_005fOPERATOR"></a>
</dt>
<dd><p>This predicate holds for a class-type for which
<code>operator new[]</code> is defined.
</p>
</dd>
<dt><code>TYPE_OVERLOADS_CALL_EXPR</code>
<a name="index-TYPE_005fOVERLOADS_005fCALL_005fEXPR"></a>
</dt>
<dd><p>This predicate holds for class-type for which the function call
<code>operator()</code> is overloaded.
</p>
</dd>
<dt><code>TYPE_OVERLOADS_ARRAY_REF</code>
<a name="index-TYPE_005fOVERLOADS_005fARRAY_005fREF"></a>
</dt>
<dd><p>This predicate holds for a class-type that overloads
<code>operator[]</code>
</p>
</dd>
<dt><code>TYPE_OVERLOADS_ARROW</code>
<a name="index-TYPE_005fOVERLOADS_005fARROW"></a>
</dt>
<dd><p>This predicate holds for a class-type for which <code>operator-&gt;</code> is
overloaded.
</p>
</dd>
</dl>

<hr>
<a name="Functions-for-C_002b_002b"></a>
<div class="header">
<p>
Next: <a href="#Statements-for-C_002b_002b" accesskey="n" rel="next">Statements for C++</a>, Previous: <a href="#Classes" accesskey="p" rel="prev">Classes</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Functions-for-C_002b_002b-1"></a>
<h4 class="subsection">11.10.4 Functions for C++</h4>
<a name="index-function-1"></a>
<a name="index-FUNCTION_005fDECL-1"></a>
<a name="index-OVERLOAD"></a>
<a name="index-OVL_005fCURRENT"></a>
<a name="index-OVL_005fNEXT"></a>

<p>A function is represented by a <code>FUNCTION_DECL</code> node.  A set of
overloaded functions is sometimes represented by an <code>OVERLOAD</code> node.
</p>
<p>An <code>OVERLOAD</code> node is not a declaration, so none of the
&lsquo;<samp>DECL_</samp>&rsquo; macros should be used on an <code>OVERLOAD</code>.  An
<code>OVERLOAD</code> node is similar to a <code>TREE_LIST</code>.  Use
<code>OVL_CURRENT</code> to get the function associated with an
<code>OVERLOAD</code> node; use <code>OVL_NEXT</code> to get the next
<code>OVERLOAD</code> node in the list of overloaded functions.  The macros
<code>OVL_CURRENT</code> and <code>OVL_NEXT</code> are actually polymorphic; you can
use them to work with <code>FUNCTION_DECL</code> nodes as well as with
overloads.  In the case of a <code>FUNCTION_DECL</code>, <code>OVL_CURRENT</code>
will always return the function itself, and <code>OVL_NEXT</code> will always
be <code>NULL_TREE</code>.
</p>
<p>To determine the scope of a function, you can use the
<code>DECL_CONTEXT</code> macro.  This macro will return the class
(either a <code>RECORD_TYPE</code> or a <code>UNION_TYPE</code>) or namespace (a
<code>NAMESPACE_DECL</code>) of which the function is a member.  For a virtual
function, this macro returns the class in which the function was
actually defined, not the base class in which the virtual declaration
occurred.
</p>
<p>If a friend function is defined in a class scope, the
<code>DECL_FRIEND_CONTEXT</code> macro can be used to determine the class in
which it was defined.  For example, in
</p><div class="smallexample">
<pre class="smallexample">class C { friend void f() {} };
</pre></div>
<p>the <code>DECL_CONTEXT</code> for <code>f</code> will be the
<code>global_namespace</code>, but the <code>DECL_FRIEND_CONTEXT</code> will be the
<code>RECORD_TYPE</code> for <code>C</code>.
</p>

<p>The following macros and functions can be used on a <code>FUNCTION_DECL</code>:
</p><dl compact="compact">
<dt><code>DECL_MAIN_P</code>
<a name="index-DECL_005fMAIN_005fP"></a>
</dt>
<dd><p>This predicate holds for a function that is the program entry point
<code>::code</code>.
</p>
</dd>
<dt><code>DECL_LOCAL_FUNCTION_P</code>
<a name="index-DECL_005fLOCAL_005fFUNCTION_005fP"></a>
</dt>
<dd><p>This predicate holds if the function was declared at block scope, even
though it has a global scope.
</p>
</dd>
<dt><code>DECL_ANTICIPATED</code>
<a name="index-DECL_005fANTICIPATED"></a>
</dt>
<dd><p>This predicate holds if the function is a built-in function but its
prototype is not yet explicitly declared.
</p>
</dd>
<dt><code>DECL_EXTERN_C_FUNCTION_P</code>
<a name="index-DECL_005fEXTERN_005fC_005fFUNCTION_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is declared as an
&lsquo;<code>extern &quot;C&quot;</code>&rsquo; function.
</p>
</dd>
<dt><code>DECL_LINKONCE_P</code>
<a name="index-DECL_005fLINKONCE_005fP"></a>
</dt>
<dd><p>This macro holds if multiple copies of this function may be emitted in
various translation units.  It is the responsibility of the linker to
merge the various copies.  Template instantiations are the most common
example of functions for which <code>DECL_LINKONCE_P</code> holds; G++
instantiates needed templates in all translation units which require them,
and then relies on the linker to remove duplicate instantiations.
</p>
<p>FIXME: This macro is not yet implemented.
</p>
</dd>
<dt><code>DECL_FUNCTION_MEMBER_P</code>
<a name="index-DECL_005fFUNCTION_005fMEMBER_005fP"></a>
</dt>
<dd><p>This macro holds if the function is a member of a class, rather than a
member of a namespace.
</p>
</dd>
<dt><code>DECL_STATIC_FUNCTION_P</code>
<a name="index-DECL_005fSTATIC_005fFUNCTION_005fP"></a>
</dt>
<dd><p>This predicate holds if the function a static member function.
</p>
</dd>
<dt><code>DECL_NONSTATIC_MEMBER_FUNCTION_P</code>
<a name="index-DECL_005fNONSTATIC_005fMEMBER_005fFUNCTION_005fP"></a>
</dt>
<dd><p>This macro holds for a non-static member function.
</p>
</dd>
<dt><code>DECL_CONST_MEMFUNC_P</code>
<a name="index-DECL_005fCONST_005fMEMFUNC_005fP"></a>
</dt>
<dd><p>This predicate holds for a <code>const</code>-member function.
</p>
</dd>
<dt><code>DECL_VOLATILE_MEMFUNC_P</code>
<a name="index-DECL_005fVOLATILE_005fMEMFUNC_005fP"></a>
</dt>
<dd><p>This predicate holds for a <code>volatile</code>-member function.
</p>
</dd>
<dt><code>DECL_CONSTRUCTOR_P</code>
<a name="index-DECL_005fCONSTRUCTOR_005fP"></a>
</dt>
<dd><p>This macro holds if the function is a constructor.
</p>
</dd>
<dt><code>DECL_NONCONVERTING_P</code>
<a name="index-DECL_005fNONCONVERTING_005fP"></a>
</dt>
<dd><p>This predicate holds if the constructor is a non-converting constructor.
</p>
</dd>
<dt><code>DECL_COMPLETE_CONSTRUCTOR_P</code>
<a name="index-DECL_005fCOMPLETE_005fCONSTRUCTOR_005fP"></a>
</dt>
<dd><p>This predicate holds for a function which is a constructor for an object
of a complete type.
</p>
</dd>
<dt><code>DECL_BASE_CONSTRUCTOR_P</code>
<a name="index-DECL_005fBASE_005fCONSTRUCTOR_005fP"></a>
</dt>
<dd><p>This predicate holds for a function which is a constructor for a base
class sub-object.
</p>
</dd>
<dt><code>DECL_COPY_CONSTRUCTOR_P</code>
<a name="index-DECL_005fCOPY_005fCONSTRUCTOR_005fP"></a>
</dt>
<dd><p>This predicate holds for a function which is a copy-constructor.
</p>
</dd>
<dt><code>DECL_DESTRUCTOR_P</code>
<a name="index-DECL_005fDESTRUCTOR_005fP"></a>
</dt>
<dd><p>This macro holds if the function is a destructor.
</p>
</dd>
<dt><code>DECL_COMPLETE_DESTRUCTOR_P</code>
<a name="index-DECL_005fCOMPLETE_005fDESTRUCTOR_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is the destructor for an object a
complete type.
</p>
</dd>
<dt><code>DECL_OVERLOADED_OPERATOR_P</code>
<a name="index-DECL_005fOVERLOADED_005fOPERATOR_005fP"></a>
</dt>
<dd><p>This macro holds if the function is an overloaded operator.
</p>
</dd>
<dt><code>DECL_CONV_FN_P</code>
<a name="index-DECL_005fCONV_005fFN_005fP"></a>
</dt>
<dd><p>This macro holds if the function is a type-conversion operator.
</p>
</dd>
<dt><code>DECL_GLOBAL_CTOR_P</code>
<a name="index-DECL_005fGLOBAL_005fCTOR_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is a file-scope initialization
function.
</p>
</dd>
<dt><code>DECL_GLOBAL_DTOR_P</code>
<a name="index-DECL_005fGLOBAL_005fDTOR_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is a file-scope finalization
function.
</p>
</dd>
<dt><code>DECL_THUNK_P</code>
<a name="index-DECL_005fTHUNK_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is a thunk.
</p>
<p>These functions represent stub code that adjusts the <code>this</code> pointer
and then jumps to another function.  When the jumped-to function
returns, control is transferred directly to the caller, without
returning to the thunk.  The first parameter to the thunk is always the
<code>this</code> pointer; the thunk should add <code>THUNK_DELTA</code> to this
value.  (The <code>THUNK_DELTA</code> is an <code>int</code>, not an
<code>INTEGER_CST</code>.)
</p>
<p>Then, if <code>THUNK_VCALL_OFFSET</code> (an <code>INTEGER_CST</code>) is nonzero
the adjusted <code>this</code> pointer must be adjusted again.  The complete
calculation is given by the following pseudo-code:
</p>
<div class="smallexample">
<pre class="smallexample">this += THUNK_DELTA
if (THUNK_VCALL_OFFSET)
  this += (*((ptrdiff_t **) this))[THUNK_VCALL_OFFSET]
</pre></div>

<p>Finally, the thunk should jump to the location given
by <code>DECL_INITIAL</code>; this will always be an expression for the
address of a function.
</p>
</dd>
<dt><code>DECL_NON_THUNK_FUNCTION_P</code>
<a name="index-DECL_005fNON_005fTHUNK_005fFUNCTION_005fP"></a>
</dt>
<dd><p>This predicate holds if the function is <em>not</em> a thunk function.
</p>
</dd>
<dt><code>GLOBAL_INIT_PRIORITY</code>
<a name="index-GLOBAL_005fINIT_005fPRIORITY"></a>
</dt>
<dd><p>If either <code>DECL_GLOBAL_CTOR_P</code> or <code>DECL_GLOBAL_DTOR_P</code> holds,
then this gives the initialization priority for the function.  The
linker will arrange that all functions for which
<code>DECL_GLOBAL_CTOR_P</code> holds are run in increasing order of priority
before <code>main</code> is called.  When the program exits, all functions for
which <code>DECL_GLOBAL_DTOR_P</code> holds are run in the reverse order.
</p>
</dd>
<dt><code>TYPE_RAISES_EXCEPTIONS</code>
<a name="index-TYPE_005fRAISES_005fEXCEPTIONS"></a>
</dt>
<dd><p>This macro returns the list of exceptions that a (member-)function can
raise.  The returned list, if non <code>NULL</code>, is comprised of nodes
whose <code>TREE_VALUE</code> represents a type.
</p>
</dd>
<dt><code>TYPE_NOTHROW_P</code>
<a name="index-TYPE_005fNOTHROW_005fP"></a>
</dt>
<dd><p>This predicate holds when the exception-specification of its arguments
is of the form &lsquo;<code>()</code>&rsquo;.
</p>
</dd>
<dt><code>DECL_ARRAY_DELETE_OPERATOR_P</code>
<a name="index-DECL_005fARRAY_005fDELETE_005fOPERATOR_005fP"></a>
</dt>
<dd><p>This predicate holds if the function an overloaded
<code>operator delete[]</code>.
</p>
</dd>
</dl>


<hr>
<a name="Statements-for-C_002b_002b"></a>
<div class="header">
<p>
Next: <a href="#C_002b_002b-Expressions" accesskey="n" rel="next">C++ Expressions</a>, Previous: <a href="#Functions-for-C_002b_002b" accesskey="p" rel="prev">Functions for C++</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Statements-for-C_002b_002b-1"></a>
<h4 class="subsection">11.10.5 Statements for C++</h4>
<a name="index-statements-1"></a>
<a name="index-BREAK_005fSTMT"></a>
<a name="index-CLEANUP_005fSTMT"></a>
<a name="index-CLEANUP_005fDECL"></a>
<a name="index-CLEANUP_005fEXPR"></a>
<a name="index-CONTINUE_005fSTMT"></a>
<a name="index-DECL_005fSTMT"></a>
<a name="index-DECL_005fSTMT_005fDECL"></a>
<a name="index-DO_005fSTMT"></a>
<a name="index-DO_005fBODY"></a>
<a name="index-DO_005fCOND"></a>
<a name="index-EMPTY_005fCLASS_005fEXPR"></a>
<a name="index-EXPR_005fSTMT"></a>
<a name="index-EXPR_005fSTMT_005fEXPR"></a>
<a name="index-FOR_005fSTMT"></a>
<a name="index-FOR_005fINIT_005fSTMT"></a>
<a name="index-FOR_005fCOND"></a>
<a name="index-FOR_005fEXPR"></a>
<a name="index-FOR_005fBODY"></a>
<a name="index-HANDLER"></a>
<a name="index-IF_005fSTMT"></a>
<a name="index-IF_005fCOND"></a>
<a name="index-THEN_005fCLAUSE"></a>
<a name="index-ELSE_005fCLAUSE"></a>
<a name="index-RETURN_005fSTMT"></a>
<a name="index-RETURN_005fEXPR"></a>
<a name="index-SUBOBJECT"></a>
<a name="index-SUBOBJECT_005fCLEANUP"></a>
<a name="index-SWITCH_005fSTMT"></a>
<a name="index-SWITCH_005fCOND"></a>
<a name="index-SWITCH_005fBODY"></a>
<a name="index-TRY_005fBLOCK"></a>
<a name="index-TRY_005fSTMTS"></a>
<a name="index-TRY_005fHANDLERS"></a>
<a name="index-HANDLER_005fPARMS"></a>
<a name="index-HANDLER_005fBODY"></a>
<a name="index-USING_005fSTMT"></a>
<a name="index-WHILE_005fSTMT"></a>
<a name="index-WHILE_005fBODY"></a>
<a name="index-WHILE_005fCOND"></a>

<p>A function that has a definition in the current translation unit will
have a non-<code>NULL</code> <code>DECL_INITIAL</code>.  However, back ends should not make
use of the particular value given by <code>DECL_INITIAL</code>.
</p>
<p>The <code>DECL_SAVED_TREE</code> macro will give the complete body of the
function.
</p>
<a name="Statements-2"></a>
<h4 class="subsubsection">11.10.5.1 Statements</h4>

<p>There are tree nodes corresponding to all of the source-level
statement constructs, used within the C and C++ frontends.  These are
enumerated here, together with a list of the various macros that can
be used to obtain information about them.  There are a few macros that
can be used with all statements:
</p>
<dl compact="compact">
<dt><code>STMT_IS_FULL_EXPR_P</code>
<a name="index-STMT_005fIS_005fFULL_005fEXPR_005fP"></a>
</dt>
<dd><p>In C++, statements normally constitute &ldquo;full expressions&rdquo;; temporaries
created during a statement are destroyed when the statement is complete.
However, G++ sometimes represents expressions by statements; these
statements will not have <code>STMT_IS_FULL_EXPR_P</code> set.  Temporaries
created during such statements should be destroyed when the innermost
enclosing statement with <code>STMT_IS_FULL_EXPR_P</code> set is exited.
</p>
</dd>
</dl>

<p>Here is the list of the various statement nodes, and the macros used to
access them.  This documentation describes the use of these nodes in
non-template functions (including instantiations of template functions).
In template functions, the same nodes are used, but sometimes in
slightly different ways.
</p>
<p>Many of the statements have substatements.  For example, a <code>while</code>
loop will have a body, which is itself a statement.  If the substatement
is <code>NULL_TREE</code>, it is considered equivalent to a statement
consisting of a single <code>;</code>, i.e., an expression statement in which
the expression has been omitted.  A substatement may in fact be a list
of statements, connected via their <code>TREE_CHAIN</code>s.  So, you should
always process the statement tree by looping over substatements, like
this:
</p><div class="smallexample">
<pre class="smallexample">void process_stmt (stmt)
     tree stmt;
{
  while (stmt)
    {
      switch (TREE_CODE (stmt))
        {
        case IF_STMT:
          process_stmt (THEN_CLAUSE (stmt));
          /* <span class="roman">More processing here.</span>  */
          break;

        &hellip;
        }

      stmt = TREE_CHAIN (stmt);
    }
}
</pre></div>
<p>In other words, while the <code>then</code> clause of an <code>if</code> statement
in C++ can be only one statement (although that one statement may be a
compound statement), the intermediate representation will sometimes use
several statements chained together.
</p>
<dl compact="compact">
<dt><code>BREAK_STMT</code></dt>
<dd>
<p>Used to represent a <code>break</code> statement.  There are no additional
fields.
</p>
</dd>
<dt><code>CLEANUP_STMT</code></dt>
<dd>
<p>Used to represent an action that should take place upon exit from the
enclosing scope.  Typically, these actions are calls to destructors for
local objects, but back ends cannot rely on this fact.  If these nodes
are in fact representing such destructors, <code>CLEANUP_DECL</code> will be
the <code>VAR_DECL</code> destroyed.  Otherwise, <code>CLEANUP_DECL</code> will be
<code>NULL_TREE</code>.  In any case, the <code>CLEANUP_EXPR</code> is the
expression to execute.  The cleanups executed on exit from a scope
should be run in the reverse order of the order in which the associated
<code>CLEANUP_STMT</code>s were encountered.
</p>
</dd>
<dt><code>CONTINUE_STMT</code></dt>
<dd>
<p>Used to represent a <code>continue</code> statement.  There are no additional
fields.
</p>
</dd>
<dt><code>CTOR_STMT</code></dt>
<dd>
<p>Used to mark the beginning (if <code>CTOR_BEGIN_P</code> holds) or end (if
<code>CTOR_END_P</code> holds of the main body of a constructor.  See also
<code>SUBOBJECT</code> for more information on how to use these nodes.
</p>
</dd>
<dt><code>DO_STMT</code></dt>
<dd>
<p>Used to represent a <code>do</code> loop.  The body of the loop is given by
<code>DO_BODY</code> while the termination condition for the loop is given by
<code>DO_COND</code>.  The condition for a <code>do</code>-statement is always an
expression.
</p>
</dd>
<dt><code>EMPTY_CLASS_EXPR</code></dt>
<dd>
<p>Used to represent a temporary object of a class with no data whose
address is never taken.  (All such objects are interchangeable.)  The
<code>TREE_TYPE</code> represents the type of the object.
</p>
</dd>
<dt><code>EXPR_STMT</code></dt>
<dd>
<p>Used to represent an expression statement.  Use <code>EXPR_STMT_EXPR</code> to
obtain the expression.
</p>
</dd>
<dt><code>FOR_STMT</code></dt>
<dd>
<p>Used to represent a <code>for</code> statement.  The <code>FOR_INIT_STMT</code> is
the initialization statement for the loop.  The <code>FOR_COND</code> is the
termination condition.  The <code>FOR_EXPR</code> is the expression executed
right before the <code>FOR_COND</code> on each loop iteration; often, this
expression increments a counter.  The body of the loop is given by
<code>FOR_BODY</code>.  Note that <code>FOR_INIT_STMT</code> and <code>FOR_BODY</code>
return statements, while <code>FOR_COND</code> and <code>FOR_EXPR</code> return
expressions.
</p>
</dd>
<dt><code>HANDLER</code></dt>
<dd>
<p>Used to represent a C++ <code>catch</code> block.  The <code>HANDLER_TYPE</code>
is the type of exception that will be caught by this handler; it is
equal (by pointer equality) to <code>NULL</code> if this handler is for all
types.  <code>HANDLER_PARMS</code> is the <code>DECL_STMT</code> for the catch
parameter, and <code>HANDLER_BODY</code> is the code for the block itself.
</p>
</dd>
<dt><code>IF_STMT</code></dt>
<dd>
<p>Used to represent an <code>if</code> statement.  The <code>IF_COND</code> is the
expression.
</p>
<p>If the condition is a <code>TREE_LIST</code>, then the <code>TREE_PURPOSE</code> is
a statement (usually a <code>DECL_STMT</code>).  Each time the condition is
evaluated, the statement should be executed.  Then, the
<code>TREE_VALUE</code> should be used as the conditional expression itself.
This representation is used to handle C++ code like this:
</p>
<p>C++ distinguishes between this and <code>COND_EXPR</code> for handling templates.
</p>
<div class="smallexample">
<pre class="smallexample">if (int i = 7) &hellip;
</pre></div>

<p>where there is a new local variable (or variables) declared within the
condition.
</p>
<p>The <code>THEN_CLAUSE</code> represents the statement given by the <code>then</code>
condition, while the <code>ELSE_CLAUSE</code> represents the statement given
by the <code>else</code> condition.
</p>
</dd>
<dt><code>SUBOBJECT</code></dt>
<dd>
<p>In a constructor, these nodes are used to mark the point at which a
subobject of <code>this</code> is fully constructed.  If, after this point, an
exception is thrown before a <code>CTOR_STMT</code> with <code>CTOR_END_P</code> set
is encountered, the <code>SUBOBJECT_CLEANUP</code> must be executed.  The
cleanups must be executed in the reverse order in which they appear.
</p>
</dd>
<dt><code>SWITCH_STMT</code></dt>
<dd>
<p>Used to represent a <code>switch</code> statement.  The <code>SWITCH_STMT_COND</code>
is the expression on which the switch is occurring.  See the documentation
for an <code>IF_STMT</code> for more information on the representation used
for the condition.  The <code>SWITCH_STMT_BODY</code> is the body of the switch
statement.   The <code>SWITCH_STMT_TYPE</code> is the original type of switch
expression as given in the source, before any compiler conversions.
</p>
</dd>
<dt><code>TRY_BLOCK</code></dt>
<dd><p>Used to represent a <code>try</code> block.  The body of the try block is
given by <code>TRY_STMTS</code>.  Each of the catch blocks is a <code>HANDLER</code>
node.  The first handler is given by <code>TRY_HANDLERS</code>.  Subsequent
handlers are obtained by following the <code>TREE_CHAIN</code> link from one
handler to the next.  The body of the handler is given by
<code>HANDLER_BODY</code>.
</p>
<p>If <code>CLEANUP_P</code> holds of the <code>TRY_BLOCK</code>, then the
<code>TRY_HANDLERS</code> will not be a <code>HANDLER</code> node.  Instead, it will
be an expression that should be executed if an exception is thrown in
the try block.  It must rethrow the exception after executing that code.
And, if an exception is thrown while the expression is executing,
<code>terminate</code> must be called.
</p>
</dd>
<dt><code>USING_STMT</code></dt>
<dd><p>Used to represent a <code>using</code> directive.  The namespace is given by
<code>USING_STMT_NAMESPACE</code>, which will be a NAMESPACE_DECL.  This node
is needed inside template functions, to implement using directives
during instantiation.
</p>
</dd>
<dt><code>WHILE_STMT</code></dt>
<dd>
<p>Used to represent a <code>while</code> loop.  The <code>WHILE_COND</code> is the
termination condition for the loop.  See the documentation for an
<code>IF_STMT</code> for more information on the representation used for the
condition.
</p>
<p>The <code>WHILE_BODY</code> is the body of the loop.
</p>
</dd>
</dl>

<hr>
<a name="C_002b_002b-Expressions"></a>
<div class="header">
<p>
Previous: <a href="#Statements-for-C_002b_002b" accesskey="p" rel="prev">Statements for C++</a>, Up: <a href="#C-and-C_002b_002b-Trees" accesskey="u" rel="up">C and C++ Trees</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="C_002b_002b-Expressions-1"></a>
<h4 class="subsection">11.10.6 C++ Expressions</h4>

<p>This section describes expressions specific to the C and C++ front
ends.
</p>
<dl compact="compact">
<dt><code>TYPEID_EXPR</code></dt>
<dd>
<p>Used to represent a <code>typeid</code> expression.
</p>
</dd>
<dt><code>NEW_EXPR</code></dt>
<dt><code>VEC_NEW_EXPR</code></dt>
<dd>
<p>Used to represent a call to <code>new</code> and <code>new[]</code> respectively.
</p>
</dd>
<dt><code>DELETE_EXPR</code></dt>
<dt><code>VEC_DELETE_EXPR</code></dt>
<dd>
<p>Used to represent a call to <code>delete</code> and <code>delete[]</code> respectively.
</p>
</dd>
<dt><code>MEMBER_REF</code></dt>
<dd>
<p>Represents a reference to a member of a class.
</p>
</dd>
<dt><code>THROW_EXPR</code></dt>
<dd>
<p>Represents an instance of <code>throw</code> in the program.  Operand 0,
which is the expression to throw, may be <code>NULL_TREE</code>.
</p>

</dd>
<dt><code>AGGR_INIT_EXPR</code></dt>
<dd><p>An <code>AGGR_INIT_EXPR</code> represents the initialization as the return
value of a function call, or as the result of a constructor.  An
<code>AGGR_INIT_EXPR</code> will only appear as a full-expression, or as the
second operand of a <code>TARGET_EXPR</code>.  <code>AGGR_INIT_EXPR</code>s have
a representation similar to that of <code>CALL_EXPR</code>s.  You can use
the <code>AGGR_INIT_EXPR_FN</code> and <code>AGGR_INIT_EXPR_ARG</code> macros to access
the function to call and the arguments to pass.
</p>
<p>If <code>AGGR_INIT_VIA_CTOR_P</code> holds of the <code>AGGR_INIT_EXPR</code>, then
the initialization is via a constructor call.  The address of the
<code>AGGR_INIT_EXPR_SLOT</code> operand, which is always a <code>VAR_DECL</code>,
is taken, and this value replaces the first argument in the argument
list.
</p>
<p>In either case, the expression is void.
</p>

</dd>
</dl>


<hr>
<a name="Java-Trees"></a>
<div class="header">
<p>
Previous: <a href="#C-and-C_002b_002b-Trees" accesskey="p" rel="prev">C and C++ Trees</a>, Up: <a href="#GENERIC" accesskey="u" rel="up">GENERIC</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Java-Trees-1"></a>
<h3 class="section">11.11 Java Trees</h3>

<hr>
<a name="GIMPLE"></a>
<div class="header">
<p>
Next: <a href="#Tree-SSA" accesskey="n" rel="next">Tree SSA</a>, Previous: <a href="#GENERIC" accesskey="p" rel="prev">GENERIC</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE-1"></a>
<h2 class="chapter">12 GIMPLE</h2>
<a name="index-GIMPLE-2"></a>

<p>GIMPLE is a three-address representation derived from GENERIC by
breaking down GENERIC expressions into tuples of no more than 3
operands (with some exceptions like function calls).  GIMPLE was
heavily influenced by the SIMPLE IL used by the McCAT compiler
project at McGill University, though we have made some different
choices.  For one thing, SIMPLE doesn&rsquo;t support <code>goto</code>.
</p>
<p>Temporaries are introduced to hold intermediate values needed to
compute complex expressions. Additionally, all the control
structures used in GENERIC are lowered into conditional jumps,
lexical scopes are removed and exception regions are converted
into an on the side exception region tree.
</p>
<p>The compiler pass which converts GENERIC into GIMPLE is referred to as
the &lsquo;<samp>gimplifier</samp>&rsquo;.  The gimplifier works recursively, generating
GIMPLE tuples out of the original GENERIC expressions.
</p>
<p>One of the early implementation strategies used for the GIMPLE
representation was to use the same internal data structures used
by front ends to represent parse trees. This simplified
implementation because we could leverage existing functionality
and interfaces. However, GIMPLE is a much more restrictive
representation than abstract syntax trees (AST), therefore it
does not require the full structural complexity provided by the
main tree data structure.
</p>
<p>The GENERIC representation of a function is stored in the
<code>DECL_SAVED_TREE</code> field of the associated <code>FUNCTION_DECL</code>
tree node.  It is converted to GIMPLE by a call to
<code>gimplify_function_tree</code>.
</p>
<p>If a front end wants to include language-specific tree codes in the tree
representation which it provides to the back end, it must provide a
definition of <code>LANG_HOOKS_GIMPLIFY_EXPR</code> which knows how to
convert the front end trees to GIMPLE.  Usually such a hook will involve
much of the same code for expanding front end trees to RTL.  This function
can return fully lowered GIMPLE, or it can return GENERIC trees and let the
main gimplifier lower them the rest of the way; this is often simpler.
GIMPLE that is not fully lowered is known as &ldquo;High GIMPLE&rdquo; and
consists of the IL before the pass <code>pass_lower_cf</code>.  High GIMPLE
contains some container statements like lexical scopes
(represented by <code>GIMPLE_BIND</code>) and nested expressions (e.g.,
<code>GIMPLE_TRY</code>), while &ldquo;Low GIMPLE&rdquo; exposes all of the
implicit jumps for control and exception expressions directly in
the IL and EH region trees.
</p>
<p>The C and C++ front ends currently convert directly from front end
trees to GIMPLE, and hand that off to the back end rather than first
converting to GENERIC.  Their gimplifier hooks know about all the
<code>_STMT</code> nodes and how to convert them to GENERIC forms.  There
was some work done on a genericization pass which would run first, but
the existence of <code>STMT_EXPR</code> meant that in order to convert all
of the C statements into GENERIC equivalents would involve walking the
entire tree anyway, so it was simpler to lower all the way.  This
might change in the future if someone writes an optimization pass
which would work better with higher-level trees, but currently the
optimizers all expect GIMPLE.
</p>
<p>You can request to dump a C-like representation of the GIMPLE form
with the flag <samp>-fdump-tree-gimple</samp>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Tuple-representation" accesskey="1">Tuple representation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE-instruction-set" accesskey="2">GIMPLE instruction set</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE-Exception-Handling" accesskey="3">GIMPLE Exception Handling</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Temporaries" accesskey="4">Temporaries</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Operands" accesskey="5">Operands</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Manipulating-GIMPLE-statements" accesskey="6">Manipulating GIMPLE statements</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tuple-specific-accessors" accesskey="7">Tuple specific accessors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE-sequences" accesskey="8">GIMPLE sequences</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Sequence-iterators" accesskey="9">Sequence iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Adding-a-new-GIMPLE-statement-code">Adding a new GIMPLE statement code</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Statement-and-operand-traversals">Statement and operand traversals</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Tuple-representation"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE-instruction-set" accesskey="n" rel="next">GIMPLE instruction set</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Tuple-representation-1"></a>
<h3 class="section">12.1 Tuple representation</h3>
<a name="index-tuples"></a>

<p>GIMPLE instructions are tuples of variable size divided in two
groups: a header describing the instruction and its locations,
and a variable length body with all the operands. Tuples are
organized into a hierarchy with 3 main classes of tuples.
</p>
<a name="gimple_005fstatement_005fbase-_0028gsbase_0029"></a>
<h4 class="subsection">12.1.1 <code>gimple_statement_base</code> (gsbase)</h4>
<a name="index-gimple_005fstatement_005fbase"></a>

<p>This is the root of the hierarchy, it holds basic information
needed by most GIMPLE statements. There are some fields that
may not be relevant to every GIMPLE statement, but those were
moved into the base structure to take advantage of holes left by
other fields (thus making the structure more compact).  The
structure takes 4 words (32 bytes) on 64 bit hosts:
</p>
<table>
<tr><td>Field</td><td>Size (bits)</td></tr>
<tr><td><code>code</code></td><td>8</td></tr>
<tr><td><code>subcode</code></td><td>16</td></tr>
<tr><td><code>no_warning</code></td><td>1</td></tr>
<tr><td><code>visited</code></td><td>1</td></tr>
<tr><td><code>nontemporal_move</code></td><td>1</td></tr>
<tr><td><code>plf</code></td><td>2</td></tr>
<tr><td><code>modified</code></td><td>1</td></tr>
<tr><td><code>has_volatile_ops</code></td><td>1</td></tr>
<tr><td><code>references_memory_p</code></td><td>1</td></tr>
<tr><td><code>uid</code></td><td>32</td></tr>
<tr><td><code>location</code></td><td>32</td></tr>
<tr><td><code>num_ops</code></td><td>32</td></tr>
<tr><td><code>bb</code></td><td>64</td></tr>
<tr><td><code>block</code></td><td>63</td></tr>
<tr><td>Total size</td><td>32 bytes</td></tr>
</table>

<ul>
<li> <code>code</code>
Main identifier for a GIMPLE instruction.

</li><li> <code>subcode</code>
Used to distinguish different variants of the same basic
instruction or provide flags applicable to a given code. The
<code>subcode</code> flags field has different uses depending on the code of
the instruction, but mostly it distinguishes instructions of the
same family. The most prominent use of this field is in
assignments, where subcode indicates the operation done on the
RHS of the assignment. For example, a = b + c is encoded as
<code>GIMPLE_ASSIGN &lt;PLUS_EXPR, a, b, c&gt;</code>.

</li><li> <code>no_warning</code>
Bitflag to indicate whether a warning has already been issued on
this statement.

</li><li> <code>visited</code>
General purpose &ldquo;visited&rdquo; marker. Set and cleared by each pass
when needed.

</li><li> <code>nontemporal_move</code>
Bitflag used in assignments that represent non-temporal moves.
Although this bitflag is only used in assignments, it was moved
into the base to take advantage of the bit holes left by the
previous fields.

</li><li> <code>plf</code>
Pass Local Flags. This 2-bit mask can be used as general purpose
markers by any pass. Passes are responsible for clearing and
setting these two flags accordingly.

</li><li> <code>modified</code>
Bitflag to indicate whether the statement has been modified.
Used mainly by the operand scanner to determine when to re-scan a
statement for operands.

</li><li> <code>has_volatile_ops</code>
Bitflag to indicate whether this statement contains operands that
have been marked volatile.

</li><li> <code>references_memory_p</code>
Bitflag to indicate whether this statement contains memory
references (i.e., its operands are either global variables, or
pointer dereferences or anything that must reside in memory).

</li><li> <code>uid</code>
This is an unsigned integer used by passes that want to assign
IDs to every statement. These IDs must be assigned and used by
each pass.

</li><li> <code>location</code>
This is a <code>location_t</code> identifier to specify source code
location for this statement. It is inherited from the front
end.

</li><li> <code>num_ops</code>
Number of operands that this statement has. This specifies the
size of the operand vector embedded in the tuple. Only used in
some tuples, but it is declared in the base tuple to take
advantage of the 32-bit hole left by the previous fields.

</li><li> <code>bb</code>
Basic block holding the instruction.

</li><li> <code>block</code>
Lexical block holding this statement.  Also used for debug
information generation.
</li></ul>

<a name="gimple_005fstatement_005fwith_005fops"></a>
<h4 class="subsection">12.1.2 <code>gimple_statement_with_ops</code></h4>
<a name="index-gimple_005fstatement_005fwith_005fops"></a>

<p>This tuple is actually split in two:
<code>gimple_statement_with_ops_base</code> and
<code>gimple_statement_with_ops</code>. This is needed to accommodate the
way the operand vector is allocated. The operand vector is
defined to be an array of 1 element. So, to allocate a dynamic
number of operands, the memory allocator (<code>gimple_alloc</code>) simply
allocates enough memory to hold the structure itself plus <code>N
- 1</code> operands which run &ldquo;off the end&rdquo; of the structure. For
example, to allocate space for a tuple with 3 operands,
<code>gimple_alloc</code> reserves <code>sizeof (struct
gimple_statement_with_ops) + 2 * sizeof (tree)</code> bytes.
</p>
<p>On the other hand, several fields in this tuple need to be shared
with the <code>gimple_statement_with_memory_ops</code> tuple. So, these
common fields are placed in <code>gimple_statement_with_ops_base</code> which
is then inherited from the other two tuples.
</p>

<table>
<tr><td><code>gsbase</code></td><td>256</td></tr>
<tr><td><code>def_ops</code></td><td>64</td></tr>
<tr><td><code>use_ops</code></td><td>64</td></tr>
<tr><td><code>op</code></td><td><code>num_ops</code> * 64</td></tr>
<tr><td>Total size</td><td>48 + 8 * <code>num_ops</code> bytes</td></tr>
</table>

<ul>
<li> <code>gsbase</code>
Inherited from <code>struct gimple_statement_base</code>.

</li><li> <code>def_ops</code>
Array of pointers into the operand array indicating all the slots that
contain a variable written-to by the statement. This array is
also used for immediate use chaining. Note that it would be
possible to not rely on this array, but the changes required to
implement this are pretty invasive.

</li><li> <code>use_ops</code>
Similar to <code>def_ops</code> but for variables read by the statement.

</li><li> <code>op</code>
Array of trees with <code>num_ops</code> slots.
</li></ul>

<a name="gimple_005fstatement_005fwith_005fmemory_005fops"></a>
<h4 class="subsection">12.1.3 <code>gimple_statement_with_memory_ops</code></h4>

<p>This tuple is essentially identical to <code>gimple_statement_with_ops</code>,
except that it contains 4 additional fields to hold vectors
related memory stores and loads.  Similar to the previous case,
the structure is split in two to accommodate for the operand
vector (<code>gimple_statement_with_memory_ops_base</code> and
<code>gimple_statement_with_memory_ops</code>).
</p>

<table>
<tr><td>Field</td><td>Size (bits)</td></tr>
<tr><td><code>gsbase</code></td><td>256</td></tr>
<tr><td><code>def_ops</code></td><td>64</td></tr>
<tr><td><code>use_ops</code></td><td>64</td></tr>
<tr><td><code>vdef_ops</code></td><td>64</td></tr>
<tr><td><code>vuse_ops</code></td><td>64</td></tr>
<tr><td><code>stores</code></td><td>64</td></tr>
<tr><td><code>loads</code></td><td>64</td></tr>
<tr><td><code>op</code></td><td><code>num_ops</code> * 64</td></tr>
<tr><td>Total size</td><td>80 + 8 * <code>num_ops</code> bytes</td></tr>
</table>

<ul>
<li> <code>vdef_ops</code>
Similar to <code>def_ops</code> but for <code>VDEF</code> operators. There is
one entry per memory symbol written by this statement. This is
used to maintain the memory SSA use-def and def-def chains.

</li><li> <code>vuse_ops</code>
Similar to <code>use_ops</code> but for <code>VUSE</code> operators. There is
one entry per memory symbol loaded by this statement. This is
used to maintain the memory SSA use-def chains.

</li><li> <code>stores</code>
Bitset with all the UIDs for the symbols written-to by the
statement.  This is different than <code>vdef_ops</code> in that all the
affected symbols are mentioned in this set.  If memory
partitioning is enabled, the <code>vdef_ops</code> vector will refer to memory
partitions. Furthermore, no SSA information is stored in this
set.

</li><li> <code>loads</code>
Similar to <code>stores</code>, but for memory loads. (Note that there
is some amount of redundancy here, it should be possible to
reduce memory utilization further by removing these sets).
</li></ul>

<p>All the other tuples are defined in terms of these three basic
ones. Each tuple will add some fields. The main gimple type
is defined to be the union of all these structures (<code>GTY</code> markers
elided for clarity):
</p>
<div class="smallexample">
<pre class="smallexample">union gimple_statement_d
{
  struct gimple_statement_base gsbase;
  struct gimple_statement_with_ops gsops;
  struct gimple_statement_with_memory_ops gsmem;
  struct gimple_statement_omp omp;
  struct gimple_statement_bind gimple_bind;
  struct gimple_statement_catch gimple_catch;
  struct gimple_statement_eh_filter gimple_eh_filter;
  struct gimple_statement_phi gimple_phi;
  struct gimple_statement_resx gimple_resx;
  struct gimple_statement_try gimple_try;
  struct gimple_statement_wce gimple_wce;
  struct gimple_statement_asm gimple_asm;
  struct gimple_statement_omp_critical gimple_omp_critical;
  struct gimple_statement_omp_for gimple_omp_for;
  struct gimple_statement_omp_parallel gimple_omp_parallel;
  struct gimple_statement_omp_task gimple_omp_task;
  struct gimple_statement_omp_sections gimple_omp_sections;
  struct gimple_statement_omp_single gimple_omp_single;
  struct gimple_statement_omp_continue gimple_omp_continue;
  struct gimple_statement_omp_atomic_load gimple_omp_atomic_load;
  struct gimple_statement_omp_atomic_store gimple_omp_atomic_store;
};
</pre></div>


<hr>
<a name="GIMPLE-instruction-set"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE-Exception-Handling" accesskey="n" rel="next">GIMPLE Exception Handling</a>, Previous: <a href="#Tuple-representation" accesskey="p" rel="prev">Tuple representation</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE-instruction-set-1"></a>
<h3 class="section">12.2 GIMPLE instruction set</h3>
<a name="index-GIMPLE-instruction-set"></a>

<p>The following table briefly describes the GIMPLE instruction set.
</p>
<table>
<tr><td>Instruction</td><td>High GIMPLE</td><td>Low GIMPLE</td></tr>
<tr><td><code>GIMPLE_ASM</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_ASSIGN</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_BIND</code></td><td>x</td><td></td></tr>
<tr><td><code>GIMPLE_CALL</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_CATCH</code></td><td>x</td><td></td></tr>
<tr><td><code>GIMPLE_COND</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_DEBUG</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_EH_FILTER</code></td><td>x</td><td></td></tr>
<tr><td><code>GIMPLE_GOTO</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_LABEL</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_NOP</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_ATOMIC_LOAD</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_ATOMIC_STORE</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_CONTINUE</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_CRITICAL</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_FOR</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_MASTER</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_ORDERED</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_PARALLEL</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_RETURN</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_SECTION</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_SECTIONS</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_SECTIONS_SWITCH</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_OMP_SINGLE</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_PHI</code></td><td></td><td>x</td></tr>
<tr><td><code>GIMPLE_RESX</code></td><td></td><td>x</td></tr>
<tr><td><code>GIMPLE_RETURN</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_SWITCH</code></td><td>x</td><td>x</td></tr>
<tr><td><code>GIMPLE_TRY</code></td><td>x</td><td></td></tr>
</table>

<hr>
<a name="GIMPLE-Exception-Handling"></a>
<div class="header">
<p>
Next: <a href="#Temporaries" accesskey="n" rel="next">Temporaries</a>, Previous: <a href="#GIMPLE-instruction-set" accesskey="p" rel="prev">GIMPLE instruction set</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Exception-Handling-1"></a>
<h3 class="section">12.3 Exception Handling</h3>
<a name="index-GIMPLE-Exception-Handling"></a>

<p>Other exception handling constructs are represented using
<code>GIMPLE_TRY_CATCH</code>.  <code>GIMPLE_TRY_CATCH</code> has two operands.  The
first operand is a sequence of statements to execute.  If executing
these statements does not throw an exception, then the second operand
is ignored.  Otherwise, if an exception is thrown, then the second
operand of the <code>GIMPLE_TRY_CATCH</code> is checked.  The second
operand may have the following forms:
</p>
<ol>
<li> A sequence of statements to execute.  When an exception occurs,
these statements are executed, and then the exception is rethrown.

</li><li> A sequence of <code>GIMPLE_CATCH</code> statements.  Each
<code>GIMPLE_CATCH</code> has a list of applicable exception types and
handler code.  If the thrown exception matches one of the caught
types, the associated handler code is executed.  If the handler
code falls off the bottom, execution continues after the original
<code>GIMPLE_TRY_CATCH</code>.

</li><li> A <code>GIMPLE_EH_FILTER</code> statement.  This has a list of
permitted exception types, and code to handle a match failure.  If the
thrown exception does not match one of the allowed types, the
associated match failure code is executed.  If the thrown exception
does match, it continues unwinding the stack looking for the next
handler.

</li></ol>

<p>Currently throwing an exception is not directly represented in
GIMPLE, since it is implemented by calling a function.  At some
point in the future we will want to add some way to express that
the call will throw an exception of a known type.
</p>
<p>Just before running the optimizers, the compiler lowers the
high-level EH constructs above into a set of &lsquo;<samp>goto</samp>&rsquo;s, magic
labels, and EH regions.  Continuing to unwind at the end of a
cleanup is represented with a <code>GIMPLE_RESX</code>.
</p>

<hr>
<a name="Temporaries"></a>
<div class="header">
<p>
Next: <a href="#Operands" accesskey="n" rel="next">Operands</a>, Previous: <a href="#GIMPLE-Exception-Handling" accesskey="p" rel="prev">GIMPLE Exception Handling</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Temporaries-1"></a>
<h3 class="section">12.4 Temporaries</h3>
<a name="index-Temporaries"></a>

<p>When gimplification encounters a subexpression that is too
complex, it creates a new temporary variable to hold the value of
the subexpression, and adds a new statement to initialize it
before the current statement. These special temporaries are known
as &lsquo;<samp>expression temporaries</samp>&rsquo;, and are allocated using
<code>get_formal_tmp_var</code>.  The compiler tries to always evaluate
identical expressions into the same temporary, to simplify
elimination of redundant calculations.
</p>
<p>We can only use expression temporaries when we know that it will
not be reevaluated before its value is used, and that it will not
be otherwise modified<a name="DOCF3" href="#FOOT3"><sup>3</sup></a>. Other temporaries can be allocated
using <code>get_initialized_tmp_var</code> or <code>create_tmp_var</code>.
</p>
<p>Currently, an expression like <code>a = b + 5</code> is not reduced any
further.  We tried converting it to something like
</p><div class="smallexample">
<pre class="smallexample">T1 = b + 5;
a = T1;
</pre></div>
<p>but this bloated the representation for minimal benefit.  However, a
variable which must live in memory cannot appear in an expression; its
value is explicitly loaded into a temporary first.  Similarly, storing
the value of an expression to a memory variable goes through a
temporary.
</p>
<hr>
<a name="Operands"></a>
<div class="header">
<p>
Next: <a href="#Manipulating-GIMPLE-statements" accesskey="n" rel="next">Manipulating GIMPLE statements</a>, Previous: <a href="#Temporaries" accesskey="p" rel="prev">Temporaries</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Operands-1"></a>
<h3 class="section">12.5 Operands</h3>
<a name="index-Operands"></a>

<p>In general, expressions in GIMPLE consist of an operation and the
appropriate number of simple operands; these operands must either be a
GIMPLE rvalue (<code>is_gimple_val</code>), i.e. a constant or a register
variable.  More complex operands are factored out into temporaries, so
that
</p><div class="smallexample">
<pre class="smallexample">a = b + c + d
</pre></div>
<p>becomes
</p><div class="smallexample">
<pre class="smallexample">T1 = b + c;
a = T1 + d;
</pre></div>

<p>The same rule holds for arguments to a <code>GIMPLE_CALL</code>.
</p>
<p>The target of an assignment is usually a variable, but can also be a
<code>MEM_REF</code> or a compound lvalue as described below.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Compound-Expressions" accesskey="1">Compound Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Compound-Lvalues" accesskey="2">Compound Lvalues</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Conditional-Expressions" accesskey="3">Conditional Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Logical-Operators" accesskey="4">Logical Operators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Compound-Expressions"></a>
<div class="header">
<p>
Next: <a href="#Compound-Lvalues" accesskey="n" rel="next">Compound Lvalues</a>, Up: <a href="#Operands" accesskey="u" rel="up">Operands</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Compound-Expressions-1"></a>
<h4 class="subsection">12.5.1 Compound Expressions</h4>
<a name="index-Compound-Expressions"></a>

<p>The left-hand side of a C comma expression is simply moved into a separate
statement.
</p>
<hr>
<a name="Compound-Lvalues"></a>
<div class="header">
<p>
Next: <a href="#Conditional-Expressions" accesskey="n" rel="next">Conditional Expressions</a>, Previous: <a href="#Compound-Expressions" accesskey="p" rel="prev">Compound Expressions</a>, Up: <a href="#Operands" accesskey="u" rel="up">Operands</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Compound-Lvalues-1"></a>
<h4 class="subsection">12.5.2 Compound Lvalues</h4>
<a name="index-Compound-Lvalues"></a>

<p>Currently compound lvalues involving array and structure field references
are not broken down; an expression like <code>a.b[2] = 42</code> is not reduced
any further (though complex array subscripts are).  This restriction is a
workaround for limitations in later optimizers; if we were to convert this
to
</p>
<div class="smallexample">
<pre class="smallexample">T1 = &amp;a.b;
T1[2] = 42;
</pre></div>

<p>alias analysis would not remember that the reference to <code>T1[2]</code> came
by way of <code>a.b</code>, so it would think that the assignment could alias
another member of <code>a</code>; this broke <code>struct-alias-1.c</code>.  Future
optimizer improvements may make this limitation unnecessary.
</p>
<hr>
<a name="Conditional-Expressions"></a>
<div class="header">
<p>
Next: <a href="#Logical-Operators" accesskey="n" rel="next">Logical Operators</a>, Previous: <a href="#Compound-Lvalues" accesskey="p" rel="prev">Compound Lvalues</a>, Up: <a href="#Operands" accesskey="u" rel="up">Operands</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Conditional-Expressions-1"></a>
<h4 class="subsection">12.5.3 Conditional Expressions</h4>
<a name="index-Conditional-Expressions"></a>

<p>A C <code>?:</code> expression is converted into an <code>if</code> statement with
each branch assigning to the same temporary.  So,
</p>
<div class="smallexample">
<pre class="smallexample">a = b ? c : d;
</pre></div>
<p>becomes
</p><div class="smallexample">
<pre class="smallexample">if (b == 1)
  T1 = c;
else
  T1 = d;
a = T1;
</pre></div>

<p>The GIMPLE level if-conversion pass re-introduces <code>?:</code>
expression, if appropriate. It is used to vectorize loops with
conditions using vector conditional operations.
</p>
<p>Note that in GIMPLE, <code>if</code> statements are represented using
<code>GIMPLE_COND</code>, as described below.
</p>
<hr>
<a name="Logical-Operators"></a>
<div class="header">
<p>
Previous: <a href="#Conditional-Expressions" accesskey="p" rel="prev">Conditional Expressions</a>, Up: <a href="#Operands" accesskey="u" rel="up">Operands</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Logical-Operators-1"></a>
<h4 class="subsection">12.5.4 Logical Operators</h4>
<a name="index-Logical-Operators"></a>

<p>Except when they appear in the condition operand of a
<code>GIMPLE_COND</code>, logical &lsquo;and&rsquo; and &lsquo;or&rsquo; operators are simplified
as follows: <code>a = b &amp;&amp; c</code> becomes
</p>
<div class="smallexample">
<pre class="smallexample">T1 = (bool)b;
if (T1 == true)
  T1 = (bool)c;
a = T1;
</pre></div>

<p>Note that <code>T1</code> in this example cannot be an expression temporary,
because it has two different assignments.
</p>
<a name="Manipulating-operands"></a>
<h4 class="subsection">12.5.5 Manipulating operands</h4>

<p>All gimple operands are of type <code>tree</code>.  But only certain
types of trees are allowed to be used as operand tuples.  Basic
validation is controlled by the function
<code>get_gimple_rhs_class</code>, which given a tree code, returns an
<code>enum</code> with the following values of type <code>enum
gimple_rhs_class</code>
</p>
<ul>
<li> <code>GIMPLE_INVALID_RHS</code>
The tree cannot be used as a GIMPLE operand.

</li><li> <code>GIMPLE_TERNARY_RHS</code>
The tree is a valid GIMPLE ternary operation.

</li><li> <code>GIMPLE_BINARY_RHS</code>
The tree is a valid GIMPLE binary operation.

</li><li> <code>GIMPLE_UNARY_RHS</code>
The tree is a valid GIMPLE unary operation.

</li><li> <code>GIMPLE_SINGLE_RHS</code>
The tree is a single object, that cannot be split into simpler
operands (for instance, <code>SSA_NAME</code>, <code>VAR_DECL</code>, <code>COMPONENT_REF</code>, etc).

<p>This operand class also acts as an escape hatch for tree nodes
that may be flattened out into the operand vector, but would need
more than two slots on the RHS.  For instance, a <code>COND_EXPR</code>
expression of the form <code>(a op b) ? x : y</code> could be flattened
out on the operand vector using 4 slots, but it would also
require additional processing to distinguish <code>c = a op b</code>
from <code>c = a op b ? x : y</code>.  Something similar occurs with
<code>ASSERT_EXPR</code>.   In time, these special case tree
expressions should be flattened into the operand vector.
</p></li></ul>

<p>For tree nodes in the categories <code>GIMPLE_TERNARY_RHS</code>,
<code>GIMPLE_BINARY_RHS</code> and <code>GIMPLE_UNARY_RHS</code>, they cannot be
stored inside tuples directly.  They first need to be flattened and
separated into individual components.  For instance, given the GENERIC
expression
</p>
<div class="smallexample">
<pre class="smallexample">a = b + c
</pre></div>

<p>its tree representation is:
</p>
<div class="smallexample">
<pre class="smallexample">MODIFY_EXPR &lt;VAR_DECL  &lt;a&gt;, PLUS_EXPR &lt;VAR_DECL &lt;b&gt;, VAR_DECL &lt;c&gt;&gt;&gt;
</pre></div>

<p>In this case, the GIMPLE form for this statement is logically
identical to its GENERIC form but in GIMPLE, the <code>PLUS_EXPR</code>
on the RHS of the assignment is not represented as a tree,
instead the two operands are taken out of the <code>PLUS_EXPR</code> sub-tree
and flattened into the GIMPLE tuple as follows:
</p>
<div class="smallexample">
<pre class="smallexample">GIMPLE_ASSIGN &lt;PLUS_EXPR, VAR_DECL &lt;a&gt;, VAR_DECL &lt;b&gt;, VAR_DECL &lt;c&gt;&gt;
</pre></div>

<a name="Operand-vector-allocation"></a>
<h4 class="subsection">12.5.6 Operand vector allocation</h4>

<p>The operand vector is stored at the bottom of the three tuple
structures that accept operands. This means, that depending on
the code of a given statement, its operand vector will be at
different offsets from the base of the structure.  To access
tuple operands use the following accessors
</p>
<dl>
<dt><a name="index-gimple_005fnum_005fops"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_num_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Returns the number of operands in statement G.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fop"></a>GIMPLE function: <em>tree</em> <strong>gimple_op</strong> <em>(gimple g, unsigned i)</em></dt>
<dd><p>Returns operand <code>I</code> from statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fops"></a>GIMPLE function: <em>tree *</em> <strong>gimple_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Returns a pointer into the operand vector for statement <code>G</code>.  This
is computed using an internal table called <code>gimple_ops_offset_</code>[].
This table is indexed by the gimple code of <code>G</code>.
</p>
<p>When the compiler is built, this table is filled-in using the
sizes of the structures used by each statement code defined in
gimple.def.  Since the operand vector is at the bottom of the
structure, for a gimple code <code>C</code> the offset is computed as sizeof
(struct-of <code>C</code>) - sizeof (tree).
</p>
<p>This mechanism adds one memory indirection to every access when
using <code>gimple_op</code>(), if this becomes a bottleneck, a pass can
choose to memoize the result from <code>gimple_ops</code>() and use that to
access the operands.
</p></dd></dl>

<a name="Operand-validation"></a>
<h4 class="subsection">12.5.7 Operand validation</h4>

<p>When adding a new operand to a gimple statement, the operand will
be validated according to what each tuple accepts in its operand
vector.  These predicates are called by the
<code>gimple_<var>name</var>_set_...()</code>.  Each tuple will use one of the
following predicates (Note, this list is not exhaustive):
</p>
<dl>
<dt><a name="index-is_005fgimple_005fval"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_val</strong> <em>(tree t)</em></dt>
<dd><p>Returns true if t is a &quot;GIMPLE value&quot;, which are all the
non-addressable stack variables (variables for which
<code>is_gimple_reg</code> returns true) and constants (expressions for which
<code>is_gimple_min_invariant</code> returns true).
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005faddressable"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_addressable</strong> <em>(tree t)</em></dt>
<dd><p>Returns true if t is a symbol or memory reference whose address
can be taken.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fasm_005fval"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_asm_val</strong> <em>(tree t)</em></dt>
<dd><p>Similar to <code>is_gimple_val</code> but it also accepts hard registers.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fcall_005faddr"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_call_addr</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is a valid expression to use as the function
called by a <code>GIMPLE_CALL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fmem_005fref_005faddr"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_mem_ref_addr</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is a valid expression to use as first operand
of a <code>MEM_REF</code> expression.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fconstant"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_constant</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is a valid gimple constant.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fmin_005finvariant"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_min_invariant</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is a valid minimal invariant.  This is different
from constants, in that the specific value of t may not be known
at compile time, but it is known that it doesn&rsquo;t change (e.g.,
the address of a function local variable).
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fip_005finvariant"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_ip_invariant</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is an interprocedural invariant.  This means that t
is a valid invariant in all functions (e.g. it can be an address of a
global variable but not of a local one).
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fip_005finvariant_005faddress"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_ip_invariant_address</strong> <em>(tree t)</em></dt>
<dd><p>Return true if t is an <code>ADDR_EXPR</code> that does not change once the
program is running (and which is valid in all functions).
</p></dd></dl>


<a name="Statement-validation"></a>
<h4 class="subsection">12.5.8 Statement validation</h4>

<dl>
<dt><a name="index-is_005fgimple_005fassign"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_assign</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if the code of g is <code>GIMPLE_ASSIGN</code>.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fcall"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_call</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if the code of g is <code>GIMPLE_CALL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fdebug"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_debug</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if the code of g is <code>GIMPLE_DEBUG</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fcast_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_assign_cast_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if g is a <code>GIMPLE_ASSIGN</code> that performs a type cast
operation.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_debug_bind_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if g is a <code>GIMPLE_DEBUG</code> that binds the value of an
expression to a variable.
</p></dd></dl>

<hr>
<a name="Manipulating-GIMPLE-statements"></a>
<div class="header">
<p>
Next: <a href="#Tuple-specific-accessors" accesskey="n" rel="next">Tuple specific accessors</a>, Previous: <a href="#Operands" accesskey="p" rel="prev">Operands</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Manipulating-GIMPLE-statements-1"></a>
<h3 class="section">12.6 Manipulating GIMPLE statements</h3>
<a name="index-Manipulating-GIMPLE-statements"></a>

<p>This section documents all the functions available to handle each
of the GIMPLE instructions.
</p>
<a name="Common-accessors"></a>
<h4 class="subsection">12.6.1 Common accessors</h4>
<p>The following are common accessors for gimple statements.
</p>
<dl>
<dt><a name="index-gimple_005fcode"></a>GIMPLE function: <em>enum gimple_code</em> <strong>gimple_code</strong> <em>(gimple g)</em></dt>
<dd><p>Return the code for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbb"></a>GIMPLE function: <em>basic_block</em> <strong>gimple_bb</strong> <em>(gimple g)</em></dt>
<dd><p>Return the basic block to which statement <code>G</code> belongs to.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fblock"></a>GIMPLE function: <em>tree</em> <strong>gimple_block</strong> <em>(gimple g)</em></dt>
<dd><p>Return the lexical scope block holding statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fexpr_005ftype"></a>GIMPLE function: <em>tree</em> <strong>gimple_expr_type</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the type of the main expression computed by <code>STMT</code>. Return
<code>void_type_node</code> if <code>STMT</code> computes nothing. This will only return
something meaningful for <code>GIMPLE_ASSIGN</code>, <code>GIMPLE_COND</code> and
<code>GIMPLE_CALL</code>.  For all other tuple codes, it will return
<code>void_type_node</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fexpr_005fcode"></a>GIMPLE function: <em>enum tree_code</em> <strong>gimple_expr_code</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the tree code for the expression computed by <code>STMT</code>.  This
is only meaningful for <code>GIMPLE_CALL</code>, <code>GIMPLE_ASSIGN</code> and
<code>GIMPLE_COND</code>.  If <code>STMT</code> is <code>GIMPLE_CALL</code>, it will return <code>CALL_EXPR</code>.
For <code>GIMPLE_COND</code>, it returns the code of the comparison predicate.
For <code>GIMPLE_ASSIGN</code> it returns the code of the operation performed
by the <code>RHS</code> of the assignment.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fblock"></a>GIMPLE function: <em>void</em> <strong>gimple_set_block</strong> <em>(gimple g, tree block)</em></dt>
<dd><p>Set the lexical scope block of <code>G</code> to <code>BLOCK</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005flocus"></a>GIMPLE function: <em>location_t</em> <strong>gimple_locus</strong> <em>(gimple g)</em></dt>
<dd><p>Return locus information for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005flocus"></a>GIMPLE function: <em>void</em> <strong>gimple_set_locus</strong> <em>(gimple g, location_t locus)</em></dt>
<dd><p>Set locus information for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005flocus_005fempty_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_locus_empty_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if <code>G</code> does not have locus information.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fno_005fwarning_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_no_warning_p</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return true if no warnings should be emitted for statement <code>STMT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fvisited"></a>GIMPLE function: <em>void</em> <strong>gimple_set_visited</strong> <em>(gimple stmt, bool visited_p)</em></dt>
<dd><p>Set the visited status on statement <code>STMT</code> to <code>VISITED_P</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fvisited_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_visited_p</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the visited status on statement <code>STMT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fplf"></a>GIMPLE function: <em>void</em> <strong>gimple_set_plf</strong> <em>(gimple stmt, enum plf_mask plf, bool val_p)</em></dt>
<dd><p>Set pass local flag <code>PLF</code> on statement <code>STMT</code> to <code>VAL_P</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fplf"></a>GIMPLE function: <em>unsigned int</em> <strong>gimple_plf</strong> <em>(gimple stmt, enum plf_mask plf)</em></dt>
<dd><p>Return the value of pass local flag <code>PLF</code> on statement <code>STMT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fhas_005fops"></a>GIMPLE function: <em>bool</em> <strong>gimple_has_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if statement <code>G</code> has register or memory operands.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fhas_005fmem_005fops"></a>GIMPLE function: <em>bool</em> <strong>gimple_has_mem_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if statement <code>G</code> has memory operands.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fnum_005fops-1"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_num_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fops-1"></a>GIMPLE function: <em>tree *</em> <strong>gimple_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the array of operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fop-1"></a>GIMPLE function: <em>tree</em> <strong>gimple_op</strong> <em>(gimple g, unsigned i)</em></dt>
<dd><p>Return operand <code>I</code> for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fop_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_op_ptr</strong> <em>(gimple g, unsigned i)</em></dt>
<dd><p>Return a pointer to operand <code>I</code> for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fop"></a>GIMPLE function: <em>void</em> <strong>gimple_set_op</strong> <em>(gimple g, unsigned i, tree op)</em></dt>
<dd><p>Set operand <code>I</code> of statement <code>G</code> to <code>OP</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005faddresses_005ftaken"></a>GIMPLE function: <em>bitmap</em> <strong>gimple_addresses_taken</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the set of symbols that have had their address taken by
<code>STMT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdef_005fops"></a>GIMPLE function: <em>struct def_optype_d *</em> <strong>gimple_def_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of <code>DEF</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fdef_005fops"></a>GIMPLE function: <em>void</em> <strong>gimple_set_def_ops</strong> <em>(gimple g, struct def_optype_d *def)</em></dt>
<dd><p>Set <code>DEF</code> to be the set of <code>DEF</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fuse_005fops"></a>GIMPLE function: <em>struct use_optype_d *</em> <strong>gimple_use_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of <code>USE</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fuse_005fops"></a>GIMPLE function: <em>void</em> <strong>gimple_set_use_ops</strong> <em>(gimple g, struct use_optype_d *use)</em></dt>
<dd><p>Set <code>USE</code> to be the set of <code>USE</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fvuse_005fops"></a>GIMPLE function: <em>struct voptype_d *</em> <strong>gimple_vuse_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of <code>VUSE</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fvuse_005fops"></a>GIMPLE function: <em>void</em> <strong>gimple_set_vuse_ops</strong> <em>(gimple g, struct voptype_d *ops)</em></dt>
<dd><p>Set <code>OPS</code> to be the set of <code>VUSE</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fvdef_005fops"></a>GIMPLE function: <em>struct voptype_d *</em> <strong>gimple_vdef_ops</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of <code>VDEF</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fvdef_005fops"></a>GIMPLE function: <em>void</em> <strong>gimple_set_vdef_ops</strong> <em>(gimple g, struct voptype_d *ops)</em></dt>
<dd><p>Set <code>OPS</code> to be the set of <code>VDEF</code> operands for statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005floaded_005fsyms"></a>GIMPLE function: <em>bitmap</em> <strong>gimple_loaded_syms</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of symbols loaded by statement <code>G</code>.  Each element of
the set is the <code>DECL_UID</code> of the corresponding symbol.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fstored_005fsyms"></a>GIMPLE function: <em>bitmap</em> <strong>gimple_stored_syms</strong> <em>(gimple g)</em></dt>
<dd><p>Return the set of symbols stored by statement <code>G</code>.  Each element of
the set is the <code>DECL_UID</code> of the corresponding symbol.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fmodified_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_modified_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if statement <code>G</code> has operands and the modified field
has been set.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fhas_005fvolatile_005fops"></a>GIMPLE function: <em>bool</em> <strong>gimple_has_volatile_ops</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return true if statement <code>STMT</code> contains volatile operands.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fset_005fhas_005fvolatile_005fops"></a>GIMPLE function: <em>void</em> <strong>gimple_set_has_volatile_ops</strong> <em>(gimple stmt, bool volatilep)</em></dt>
<dd><p>Return true if statement <code>STMT</code> contains volatile operands.
</p></dd></dl>

<dl>
<dt><a name="index-update_005fstmt"></a>GIMPLE function: <em>void</em> <strong>update_stmt</strong> <em>(gimple s)</em></dt>
<dd><p>Mark statement <code>S</code> as modified, and update it.
</p></dd></dl>

<dl>
<dt><a name="index-update_005fstmt_005fif_005fmodified"></a>GIMPLE function: <em>void</em> <strong>update_stmt_if_modified</strong> <em>(gimple s)</em></dt>
<dd><p>Update statement <code>S</code> if it has been marked modified.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcopy"></a>GIMPLE function: <em>gimple</em> <strong>gimple_copy</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return a deep copy of statement <code>STMT</code>.
</p></dd></dl>

<hr>
<a name="Tuple-specific-accessors"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE-sequences" accesskey="n" rel="next">GIMPLE sequences</a>, Previous: <a href="#Manipulating-GIMPLE-statements" accesskey="p" rel="prev">Manipulating GIMPLE statements</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Tuple-specific-accessors-1"></a>
<h3 class="section">12.7 Tuple specific accessors</h3>
<a name="index-Tuple-specific-accessors"></a>

<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fASM" accesskey="1"><code>GIMPLE_ASM</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fASSIGN" accesskey="2"><code>GIMPLE_ASSIGN</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fBIND" accesskey="3"><code>GIMPLE_BIND</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fCALL" accesskey="4"><code>GIMPLE_CALL</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fCATCH" accesskey="5"><code>GIMPLE_CATCH</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fCOND" accesskey="6"><code>GIMPLE_COND</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fDEBUG" accesskey="7"><code>GIMPLE_DEBUG</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fEH_005fFILTER" accesskey="8"><code>GIMPLE_EH_FILTER</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fLABEL" accesskey="9"><code>GIMPLE_LABEL</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fNOP"><code>GIMPLE_NOP</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fMASTER"><code>GIMPLE_OMP_MASTER</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fORDERED"><code>GIMPLE_OMP_ORDERED</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>


<hr>
<a name="GIMPLE_005fASM"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fASSIGN" accesskey="n" rel="next"><code>GIMPLE_ASSIGN</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fASM-1"></a>
<h4 class="subsection">12.7.1 <code>GIMPLE_ASM</code></h4>
<a name="index-GIMPLE_005fASM"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fasm"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_asm</strong> <em>(const char *string, ninputs, noutputs, nclobbers, ...)</em></dt>
<dd><p>Build a <code>GIMPLE_ASM</code> statement.  This statement is used for
building in-line assembly constructs.  <code>STRING</code> is the assembly
code.  <code>NINPUT</code> is the number of register inputs.  <code>NOUTPUT</code> is the
number of register outputs.  <code>NCLOBBERS</code> is the number of clobbered
registers.  The rest of the arguments trees for each input,
output, and clobbered registers.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbuild_005fasm_005fvec"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_asm_vec</strong> <em>(const char *, VEC(tree,gc) *, VEC(tree,gc) *, VEC(tree,gc) *)</em></dt>
<dd><p>Identical to gimple_build_asm, but the arguments are passed in
VECs.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fninputs"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_asm_ninputs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of input operands for <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fnoutputs"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_asm_noutputs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of output operands for <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fnclobbers"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_asm_nclobbers</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of clobber operands for <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005finput_005fop"></a>GIMPLE function: <em>tree</em> <strong>gimple_asm_input_op</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return input operand <code>INDEX</code> of <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fset_005finput_005fop"></a>GIMPLE function: <em>void</em> <strong>gimple_asm_set_input_op</strong> <em>(gimple g, unsigned index, tree in_op)</em></dt>
<dd><p>Set <code>IN_OP</code> to be input operand <code>INDEX</code> in <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005foutput_005fop"></a>GIMPLE function: <em>tree</em> <strong>gimple_asm_output_op</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return output operand <code>INDEX</code> of <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fset_005foutput_005fop"></a>GIMPLE function: <em>void</em> <strong>gimple_asm_set_output_op</strong> <em>(gimple g, unsigned index, tree out_op)</em></dt>
<dd><p>Set <code>OUT_OP</code> to be output operand <code>INDEX</code> in <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fclobber_005fop"></a>GIMPLE function: <em>tree</em> <strong>gimple_asm_clobber_op</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return clobber operand <code>INDEX</code> of <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fset_005fclobber_005fop"></a>GIMPLE function: <em>void</em> <strong>gimple_asm_set_clobber_op</strong> <em>(gimple g, unsigned index, tree clobber_op)</em></dt>
<dd><p>Set <code>CLOBBER_OP</code> to be clobber operand <code>INDEX</code> in <code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fstring"></a>GIMPLE function: <em>const char *</em> <strong>gimple_asm_string</strong> <em>(gimple g)</em></dt>
<dd><p>Return the string representing the assembly instruction in
<code>GIMPLE_ASM</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fvolatile_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_asm_volatile_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if <code>G</code> is an asm statement marked volatile.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fset_005fvolatile"></a>GIMPLE function: <em>void</em> <strong>gimple_asm_set_volatile</strong> <em>(gimple g)</em></dt>
<dd><p>Mark asm statement <code>G</code> as volatile.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fasm_005fclear_005fvolatile"></a>GIMPLE function: <em>void</em> <strong>gimple_asm_clear_volatile</strong> <em>(gimple g)</em></dt>
<dd><p>Remove volatile marker from asm statement <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fASSIGN"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fBIND" accesskey="n" rel="next"><code>GIMPLE_BIND</code></a>, Previous: <a href="#GIMPLE_005fASM" accesskey="p" rel="prev"><code>GIMPLE_ASM</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fASSIGN-1"></a>
<h4 class="subsection">12.7.2 <code>GIMPLE_ASSIGN</code></h4>
<a name="index-GIMPLE_005fASSIGN"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fassign"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_assign</strong> <em>(tree lhs, tree rhs)</em></dt>
<dd><p>Build a <code>GIMPLE_ASSIGN</code> statement.  The left-hand side is an lvalue
passed in lhs.  The right-hand side can be either a unary or
binary tree expression.  The expression tree rhs will be
flattened and its operands assigned to the corresponding operand
slots in the new statement.  This function is useful when you
already have a tree expression that you want to convert into a
tuple.  However, try to avoid building expression trees for the
sole purpose of calling this function.  If you already have the
operands in separate trees, it is better to use
<code>gimple_build_assign_with_ops</code>.
</p></dd></dl>


<dl>
<dt><a name="index-gimplify_005fassign"></a>GIMPLE function: <em>gimple</em> <strong>gimplify_assign</strong> <em>(tree dst, tree src, gimple_seq *seq_p)</em></dt>
<dd><p>Build a new <code>GIMPLE_ASSIGN</code> tuple and append it to the end of
<code>*SEQ_P</code>.
</p></dd></dl>

<p><code>DST</code>/<code>SRC</code> are the destination and source respectively.  You can
pass ungimplified trees in <code>DST</code> or <code>SRC</code>, in which
case they will be converted to a gimple operand if necessary.
</p>
<p>This function returns the newly created <code>GIMPLE_ASSIGN</code> tuple.
</p>
<dl>
<dt><a name="index-gimple_005fbuild_005fassign_005fwith_005fops"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_assign_with_ops</strong> <em>(enum tree_code subcode, tree lhs, tree op1, tree op2)</em></dt>
<dd><p>This function is similar to <code>gimple_build_assign</code>, but is used to
build a <code>GIMPLE_ASSIGN</code> statement when the operands of the
right-hand side of the assignment are already split into
different operands.
</p>
<p>The left-hand side is an lvalue passed in lhs.  Subcode is the
<code>tree_code</code> for the right-hand side of the assignment.  Op1 and op2
are the operands.  If op2 is null, subcode must be a <code>tree_code</code>
for a unary expression.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs_005fcode"></a>GIMPLE function: <em>enum tree_code</em> <strong>gimple_assign_rhs_code</strong> <em>(gimple g)</em></dt>
<dd><p>Return the code of the expression computed on the <code>RHS</code> of
assignment statement <code>G</code>.
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fassign_005frhs_005fclass"></a>GIMPLE function: <em>enum gimple_rhs_class</em> <strong>gimple_assign_rhs_class</strong> <em>(gimple g)</em></dt>
<dd><p>Return the gimple rhs class of the code for the expression
computed on the rhs of assignment statement <code>G</code>.  This will never
return <code>GIMPLE_INVALID_RHS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005flhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_assign_lhs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>LHS</code> of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005flhs_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_assign_lhs_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the <code>LHS</code> of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs1"></a>GIMPLE function: <em>tree</em> <strong>gimple_assign_rhs1</strong> <em>(gimple g)</em></dt>
<dd><p>Return the first operand on the <code>RHS</code> of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs1_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_assign_rhs1_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return the address of the first operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs2"></a>GIMPLE function: <em>tree</em> <strong>gimple_assign_rhs2</strong> <em>(gimple g)</em></dt>
<dd><p>Return the second operand on the <code>RHS</code> of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs2_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_assign_rhs2_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return the address of the second operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs3"></a>GIMPLE function: <em>tree</em> <strong>gimple_assign_rhs3</strong> <em>(gimple g)</em></dt>
<dd><p>Return the third operand on the <code>RHS</code> of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005frhs3_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_assign_rhs3_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return the address of the third operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fset_005flhs"></a>GIMPLE function: <em>void</em> <strong>gimple_assign_set_lhs</strong> <em>(gimple g, tree lhs)</em></dt>
<dd><p>Set <code>LHS</code> to be the <code>LHS</code> operand of assignment statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fset_005frhs1"></a>GIMPLE function: <em>void</em> <strong>gimple_assign_set_rhs1</strong> <em>(gimple g, tree rhs)</em></dt>
<dd><p>Set <code>RHS</code> to be the first operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fset_005frhs2"></a>GIMPLE function: <em>void</em> <strong>gimple_assign_set_rhs2</strong> <em>(gimple g, tree rhs)</em></dt>
<dd><p>Set <code>RHS</code> to be the second operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fset_005frhs3"></a>GIMPLE function: <em>void</em> <strong>gimple_assign_set_rhs3</strong> <em>(gimple g, tree rhs)</em></dt>
<dd><p>Set <code>RHS</code> to be the third operand on the <code>RHS</code> of assignment
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fassign_005fcast_005fp-1"></a>GIMPLE function: <em>bool</em> <strong>gimple_assign_cast_p</strong> <em>(gimple s)</em></dt>
<dd><p>Return true if <code>S</code> is a type-cast assignment.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fBIND"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fCALL" accesskey="n" rel="next"><code>GIMPLE_CALL</code></a>, Previous: <a href="#GIMPLE_005fASSIGN" accesskey="p" rel="prev"><code>GIMPLE_ASSIGN</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fBIND-1"></a>
<h4 class="subsection">12.7.3 <code>GIMPLE_BIND</code></h4>
<a name="index-GIMPLE_005fBIND"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fbind"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_bind</strong> <em>(tree vars, gimple_seq body)</em></dt>
<dd><p>Build a <code>GIMPLE_BIND</code> statement with a list of variables in <code>VARS</code>
and a body of statements in sequence <code>BODY</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fvars"></a>GIMPLE function: <em>tree</em> <strong>gimple_bind_vars</strong> <em>(gimple g)</em></dt>
<dd><p>Return the variables declared in the <code>GIMPLE_BIND</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fset_005fvars"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_set_vars</strong> <em>(gimple g, tree vars)</em></dt>
<dd><p>Set <code>VARS</code> to be the set of variables declared in the <code>GIMPLE_BIND</code>
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fappend_005fvars"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_append_vars</strong> <em>(gimple g, tree vars)</em></dt>
<dd><p>Append <code>VARS</code> to the set of variables declared in the <code>GIMPLE_BIND</code>
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fbody"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_bind_body</strong> <em>(gimple g)</em></dt>
<dd><p>Return the GIMPLE sequence contained in the <code>GIMPLE_BIND</code> statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fset_005fbody"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_set_body</strong> <em>(gimple g, gimple_seq seq)</em></dt>
<dd><p>Set <code>SEQ</code> to be sequence contained in the <code>GIMPLE_BIND</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fadd_005fstmt"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_add_stmt</strong> <em>(gimple gs, gimple stmt)</em></dt>
<dd><p>Append a statement to the end of a <code>GIMPLE_BIND</code>&rsquo;s body.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fadd_005fseq"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_add_seq</strong> <em>(gimple gs, gimple_seq seq)</em></dt>
<dd><p>Append a sequence of statements to the end of a <code>GIMPLE_BIND</code>&rsquo;s
body.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fblock"></a>GIMPLE function: <em>tree</em> <strong>gimple_bind_block</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>TREE_BLOCK</code> node associated with <code>GIMPLE_BIND</code> statement
<code>G</code>. This is analogous to the <code>BIND_EXPR_BLOCK</code> field in trees.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbind_005fset_005fblock"></a>GIMPLE function: <em>void</em> <strong>gimple_bind_set_block</strong> <em>(gimple g, tree block)</em></dt>
<dd><p>Set <code>BLOCK</code> to be the <code>TREE_BLOCK</code> node associated with <code>GIMPLE_BIND</code>
statement <code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fCALL"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fCATCH" accesskey="n" rel="next"><code>GIMPLE_CATCH</code></a>, Previous: <a href="#GIMPLE_005fBIND" accesskey="p" rel="prev"><code>GIMPLE_BIND</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fCALL-1"></a>
<h4 class="subsection">12.7.4 <code>GIMPLE_CALL</code></h4>
<a name="index-GIMPLE_005fCALL"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fcall"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_call</strong> <em>(tree fn, unsigned nargs, ...)</em></dt>
<dd><p>Build a <code>GIMPLE_CALL</code> statement to function <code>FN</code>.  The argument <code>FN</code>
must be either a <code>FUNCTION_DECL</code> or a gimple call address as
determined by <code>is_gimple_call_addr</code>.  <code>NARGS</code> are the number of
arguments.  The rest of the arguments follow the argument <code>NARGS</code>,
and must be trees that are valid as rvalues in gimple (i.e., each
operand is validated with <code>is_gimple_operand</code>).
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fbuild_005fcall_005ffrom_005ftree"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_call_from_tree</strong> <em>(tree call_expr)</em></dt>
<dd><p>Build a <code>GIMPLE_CALL</code> from a <code>CALL_EXPR</code> node.  The arguments and the
function are taken from the expression directly.  This routine
assumes that <code>call_expr</code> is already in GIMPLE form.  That is, its
operands are GIMPLE values and the function call needs no further
simplification.  All the call flags in <code>call_expr</code> are copied over
to the new <code>GIMPLE_CALL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fbuild_005fcall_005fvec"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_call_vec</strong> <em>(tree fn, <code>VEC</code>(tree, heap) *args)</em></dt>
<dd><p>Identical to <code>gimple_build_call</code> but the arguments are stored in a
<code>VEC</code>().
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005flhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_lhs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>LHS</code> of call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005flhs_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_call_lhs_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the <code>LHS</code> of call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005flhs"></a>GIMPLE function: <em>void</em> <strong>gimple_call_set_lhs</strong> <em>(gimple g, tree lhs)</em></dt>
<dd><p>Set <code>LHS</code> to be the <code>LHS</code> operand of call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005ffn"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_fn</strong> <em>(gimple g)</em></dt>
<dd><p>Return the tree node representing the function called by call
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005ffn"></a>GIMPLE function: <em>void</em> <strong>gimple_call_set_fn</strong> <em>(gimple g, tree fn)</em></dt>
<dd><p>Set <code>FN</code> to be the function called by call statement <code>G</code>.  This has
to be a gimple value specifying the address of the called
function.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005ffndecl"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_fndecl</strong> <em>(gimple g)</em></dt>
<dd><p>If a given <code>GIMPLE_CALL</code>&rsquo;s callee is a <code>FUNCTION_DECL</code>, return it.
Otherwise return <code>NULL</code>.  This function is analogous to
<code>get_callee_fndecl</code> in <code>GENERIC</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005ffndecl"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_set_fndecl</strong> <em>(gimple g, tree fndecl)</em></dt>
<dd><p>Set the called function to <code>FNDECL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005freturn_005ftype"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_return_type</strong> <em>(gimple g)</em></dt>
<dd><p>Return the type returned by call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fchain"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_chain</strong> <em>(gimple g)</em></dt>
<dd><p>Return the static chain for call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005fchain"></a>GIMPLE function: <em>void</em> <strong>gimple_call_set_chain</strong> <em>(gimple g, tree chain)</em></dt>
<dd><p>Set <code>CHAIN</code> to be the static chain for call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fnum_005fargs"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_call_num_args</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of arguments used by call statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005farg"></a>GIMPLE function: <em>tree</em> <strong>gimple_call_arg</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return the argument at position <code>INDEX</code> for call statement <code>G</code>.  The
first argument is 0.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005farg_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_call_arg_ptr</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return a pointer to the argument at position <code>INDEX</code> for call
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005farg"></a>GIMPLE function: <em>void</em> <strong>gimple_call_set_arg</strong> <em>(gimple g, unsigned index, tree arg)</em></dt>
<dd><p>Set <code>ARG</code> to be the argument at position <code>INDEX</code> for call statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fset_005ftail"></a>GIMPLE function: <em>void</em> <strong>gimple_call_set_tail</strong> <em>(gimple s)</em></dt>
<dd><p>Mark call statement <code>S</code> as being a tail call (i.e., a call just
before the exit of a function). These calls are candidate for
tail call optimization.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005ftail_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_call_tail_p</strong> <em>(gimple s)</em></dt>
<dd><p>Return true if <code>GIMPLE_CALL</code> <code>S</code> is marked as a tail call.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fmark_005funinlinable"></a>GIMPLE function: <em>void</em> <strong>gimple_call_mark_uninlinable</strong> <em>(gimple s)</em></dt>
<dd><p>Mark <code>GIMPLE_CALL</code> <code>S</code> as being uninlinable.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fcannot_005finline_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_call_cannot_inline_p</strong> <em>(gimple s)</em></dt>
<dd><p>Return true if <code>GIMPLE_CALL</code> <code>S</code> cannot be inlined.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fnoreturn_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_call_noreturn_p</strong> <em>(gimple s)</em></dt>
<dd><p>Return true if <code>S</code> is a noreturn call.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcall_005fcopy_005fskip_005fargs"></a>GIMPLE function: <em>gimple</em> <strong>gimple_call_copy_skip_args</strong> <em>(gimple stmt, bitmap args_to_skip)</em></dt>
<dd><p>Build a <code>GIMPLE_CALL</code> identical to <code>STMT</code> but skipping the arguments
in the positions marked by the set <code>ARGS_TO_SKIP</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fCATCH"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fCOND" accesskey="n" rel="next"><code>GIMPLE_COND</code></a>, Previous: <a href="#GIMPLE_005fCALL" accesskey="p" rel="prev"><code>GIMPLE_CALL</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fCATCH-1"></a>
<h4 class="subsection">12.7.5 <code>GIMPLE_CATCH</code></h4>
<a name="index-GIMPLE_005fCATCH"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fcatch"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_catch</strong> <em>(tree types, gimple_seq handler)</em></dt>
<dd><p>Build a <code>GIMPLE_CATCH</code> statement.  <code>TYPES</code> are the tree types this
catch handles.  <code>HANDLER</code> is a sequence of statements with the code
for the handler.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcatch_005ftypes"></a>GIMPLE function: <em>tree</em> <strong>gimple_catch_types</strong> <em>(gimple g)</em></dt>
<dd><p>Return the types handled by <code>GIMPLE_CATCH</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcatch_005ftypes_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_catch_types_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the types handled by <code>GIMPLE_CATCH</code> statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcatch_005fhandler"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_catch_handler</strong> <em>(gimple g)</em></dt>
<dd><p>Return the GIMPLE sequence representing the body of the handler
of <code>GIMPLE_CATCH</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcatch_005fset_005ftypes"></a>GIMPLE function: <em>void</em> <strong>gimple_catch_set_types</strong> <em>(gimple g, tree t)</em></dt>
<dd><p>Set <code>T</code> to be the set of types handled by <code>GIMPLE_CATCH</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcatch_005fset_005fhandler"></a>GIMPLE function: <em>void</em> <strong>gimple_catch_set_handler</strong> <em>(gimple g, gimple_seq handler)</em></dt>
<dd><p>Set <code>HANDLER</code> to be the body of <code>GIMPLE_CATCH</code> <code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fCOND"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fDEBUG" accesskey="n" rel="next"><code>GIMPLE_DEBUG</code></a>, Previous: <a href="#GIMPLE_005fCATCH" accesskey="p" rel="prev"><code>GIMPLE_CATCH</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fCOND-1"></a>
<h4 class="subsection">12.7.6 <code>GIMPLE_COND</code></h4>
<a name="index-GIMPLE_005fCOND"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fcond"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_cond</strong> <em>(enum tree_code pred_code, tree lhs, tree rhs, tree t_label, tree f_label)</em></dt>
<dd><p>Build a <code>GIMPLE_COND</code> statement.  <code>A</code> <code>GIMPLE_COND</code> statement compares
<code>LHS</code> and <code>RHS</code> and if the condition in <code>PRED_CODE</code> is true, jump to
the label in <code>t_label</code>, otherwise jump to the label in <code>f_label</code>.
<code>PRED_CODE</code> are relational operator tree codes like <code>EQ_EXPR</code>,
<code>LT_EXPR</code>, <code>LE_EXPR</code>, <code>NE_EXPR</code>, etc.
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fbuild_005fcond_005ffrom_005ftree"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_cond_from_tree</strong> <em>(tree cond, tree t_label, tree f_label)</em></dt>
<dd><p>Build a <code>GIMPLE_COND</code> statement from the conditional expression
tree <code>COND</code>.  <code>T_LABEL</code> and <code>F_LABEL</code> are as in <code>gimple_build_cond</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fcode"></a>GIMPLE function: <em>enum tree_code</em> <strong>gimple_cond_code</strong> <em>(gimple g)</em></dt>
<dd><p>Return the code of the predicate computed by conditional
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fset_005fcode"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_set_code</strong> <em>(gimple g, enum tree_code code)</em></dt>
<dd><p>Set <code>CODE</code> to be the predicate code for the conditional statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005flhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_cond_lhs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>LHS</code> of the predicate computed by conditional statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fset_005flhs"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_set_lhs</strong> <em>(gimple g, tree lhs)</em></dt>
<dd><p>Set <code>LHS</code> to be the <code>LHS</code> operand of the predicate computed by
conditional statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005frhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_cond_rhs</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>RHS</code> operand of the predicate computed by conditional
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fset_005frhs"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_set_rhs</strong> <em>(gimple g, tree rhs)</em></dt>
<dd><p>Set <code>RHS</code> to be the <code>RHS</code> operand of the predicate computed by
conditional statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005ftrue_005flabel"></a>GIMPLE function: <em>tree</em> <strong>gimple_cond_true_label</strong> <em>(gimple g)</em></dt>
<dd><p>Return the label used by conditional statement <code>G</code> when its
predicate evaluates to true.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fset_005ftrue_005flabel"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_set_true_label</strong> <em>(gimple g, tree label)</em></dt>
<dd><p>Set <code>LABEL</code> to be the label used by conditional statement <code>G</code> when
its predicate evaluates to true.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fset_005ffalse_005flabel"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_set_false_label</strong> <em>(gimple g, tree label)</em></dt>
<dd><p>Set <code>LABEL</code> to be the label used by conditional statement <code>G</code> when
its predicate evaluates to false.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005ffalse_005flabel"></a>GIMPLE function: <em>tree</em> <strong>gimple_cond_false_label</strong> <em>(gimple g)</em></dt>
<dd><p>Return the label used by conditional statement <code>G</code> when its
predicate evaluates to false.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fmake_005ffalse"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_make_false</strong> <em>(gimple g)</em></dt>
<dd><p>Set the conditional <code>COND_STMT</code> to be of the form &rsquo;if (1 == 0)&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fcond_005fmake_005ftrue"></a>GIMPLE function: <em>void</em> <strong>gimple_cond_make_true</strong> <em>(gimple g)</em></dt>
<dd><p>Set the conditional <code>COND_STMT</code> to be of the form &rsquo;if (1 == 1)&rsquo;.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fDEBUG"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fEH_005fFILTER" accesskey="n" rel="next"><code>GIMPLE_EH_FILTER</code></a>, Previous: <a href="#GIMPLE_005fCOND" accesskey="p" rel="prev"><code>GIMPLE_COND</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fDEBUG-1"></a>
<h4 class="subsection">12.7.7 <code>GIMPLE_DEBUG</code></h4>
<a name="index-GIMPLE_005fDEBUG"></a>
<a name="index-GIMPLE_005fDEBUG_005fBIND"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fdebug_005fbind"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_debug_bind</strong> <em>(tree var, tree value, gimple stmt)</em></dt>
<dd><p>Build a <code>GIMPLE_DEBUG</code> statement with <code>GIMPLE_DEBUG_BIND</code> of
<code>subcode</code>.  The effect of this statement is to tell debug
information generation machinery that the value of user variable
<code>var</code> is given by <code>value</code> at that point, and to remain with
that value until <code>var</code> runs out of scope, a
dynamically-subsequent debug bind statement overrides the binding, or
conflicting values reach a control flow merge point.  Even if
components of the <code>value</code> expression change afterwards, the
variable is supposed to retain the same value, though not necessarily
the same location.
</p>
<p>It is expected that <code>var</code> be most often a tree for automatic user
variables (<code>VAR_DECL</code> or <code>PARM_DECL</code>) that satisfy the
requirements for gimple registers, but it may also be a tree for a
scalarized component of a user variable (<code>ARRAY_REF</code>,
<code>COMPONENT_REF</code>), or a debug temporary (<code>DEBUG_EXPR_DECL</code>).
</p>
<p>As for <code>value</code>, it can be an arbitrary tree expression, but it is
recommended that it be in a suitable form for a gimple assignment
<code>RHS</code>.  It is not expected that user variables that could appear
as <code>var</code> ever appear in <code>value</code>, because in the latter we&rsquo;d
have their <code>SSA_NAME</code>s instead, but even if they were not in SSA
form, user variables appearing in <code>value</code> are to be regarded as
part of the executable code space, whereas those in <code>var</code> are to
be regarded as part of the source code space.  There is no way to
refer to the value bound to a user variable within a <code>value</code>
expression.
</p>
<p>If <code>value</code> is <code>GIMPLE_DEBUG_BIND_NOVALUE</code>, debug information
generation machinery is informed that the variable <code>var</code> is
unbound, i.e., that its value is indeterminate, which sometimes means
it is really unavailable, and other times that the compiler could not
keep track of it.
</p>
<p>Block and location information for the newly-created stmt are
taken from <code>stmt</code>, if given.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fget_005fvar"></a>GIMPLE function: <em>tree</em> <strong>gimple_debug_bind_get_var</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the user variable <var>var</var> that is bound at <code>stmt</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fget_005fvalue"></a>GIMPLE function: <em>tree</em> <strong>gimple_debug_bind_get_value</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return the value expression that is bound to a user variable at
<code>stmt</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fget_005fvalue_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_debug_bind_get_value_ptr</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return a pointer to the value expression that is bound to a user
variable at <code>stmt</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fset_005fvar"></a>GIMPLE function: <em>void</em> <strong>gimple_debug_bind_set_var</strong> <em>(gimple stmt, tree var)</em></dt>
<dd><p>Modify the user variable bound at <code>stmt</code> to <var>var</var>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fset_005fvalue"></a>GIMPLE function: <em>void</em> <strong>gimple_debug_bind_set_value</strong> <em>(gimple stmt, tree var)</em></dt>
<dd><p>Modify the value bound to the user variable bound at <code>stmt</code> to
<var>value</var>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005freset_005fvalue"></a>GIMPLE function: <em>void</em> <strong>gimple_debug_bind_reset_value</strong> <em>(gimple stmt)</em></dt>
<dd><p>Modify the value bound to the user variable bound at <code>stmt</code> so
that the variable becomes unbound.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fdebug_005fbind_005fhas_005fvalue_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_debug_bind_has_value_p</strong> <em>(gimple stmt)</em></dt>
<dd><p>Return <code>TRUE</code> if <code>stmt</code> binds a user variable to a value,
and <code>FALSE</code> if it unbinds the variable.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fEH_005fFILTER"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fLABEL" accesskey="n" rel="next"><code>GIMPLE_LABEL</code></a>, Previous: <a href="#GIMPLE_005fDEBUG" accesskey="p" rel="prev"><code>GIMPLE_DEBUG</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fEH_005fFILTER-1"></a>
<h4 class="subsection">12.7.8 <code>GIMPLE_EH_FILTER</code></h4>
<a name="index-GIMPLE_005fEH_005fFILTER"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005feh_005ffilter"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_eh_filter</strong> <em>(tree types, gimple_seq failure)</em></dt>
<dd><p>Build a <code>GIMPLE_EH_FILTER</code> statement.  <code>TYPES</code> are the filter&rsquo;s
types.  <code>FAILURE</code> is a sequence with the filter&rsquo;s failure action.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005ftypes"></a>GIMPLE function: <em>tree</em> <strong>gimple_eh_filter_types</strong> <em>(gimple g)</em></dt>
<dd><p>Return the types handled by <code>GIMPLE_EH_FILTER</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005ftypes_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_eh_filter_types_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the types handled by <code>GIMPLE_EH_FILTER</code>
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005ffailure"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_eh_filter_failure</strong> <em>(gimple g)</em></dt>
<dd><p>Return the sequence of statement to execute when <code>GIMPLE_EH_FILTER</code>
statement fails.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005fset_005ftypes"></a>GIMPLE function: <em>void</em> <strong>gimple_eh_filter_set_types</strong> <em>(gimple g, tree types)</em></dt>
<dd><p>Set <code>TYPES</code> to be the set of types handled by <code>GIMPLE_EH_FILTER</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005fset_005ffailure"></a>GIMPLE function: <em>void</em> <strong>gimple_eh_filter_set_failure</strong> <em>(gimple g, gimple_seq failure)</em></dt>
<dd><p>Set <code>FAILURE</code> to be the sequence of statements to execute on
failure for <code>GIMPLE_EH_FILTER</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005fmust_005fnot_005fthrow"></a>GIMPLE function: <em>bool</em> <strong>gimple_eh_filter_must_not_throw</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>EH_FILTER_MUST_NOT_THROW</code> flag.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005feh_005ffilter_005fset_005fmust_005fnot_005fthrow"></a>GIMPLE function: <em>void</em> <strong>gimple_eh_filter_set_must_not_throw</strong> <em>(gimple g, bool mntp)</em></dt>
<dd><p>Set the <code>EH_FILTER_MUST_NOT_THROW</code> flag.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fLABEL"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fNOP" accesskey="n" rel="next"><code>GIMPLE_NOP</code></a>, Previous: <a href="#GIMPLE_005fEH_005fFILTER" accesskey="p" rel="prev"><code>GIMPLE_EH_FILTER</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fLABEL-1"></a>
<h4 class="subsection">12.7.9 <code>GIMPLE_LABEL</code></h4>
<a name="index-GIMPLE_005fLABEL"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005flabel"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_label</strong> <em>(tree label)</em></dt>
<dd><p>Build a <code>GIMPLE_LABEL</code> statement with corresponding to the tree
label, <code>LABEL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005flabel_005flabel"></a>GIMPLE function: <em>tree</em> <strong>gimple_label_label</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>LABEL_DECL</code> node used by <code>GIMPLE_LABEL</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005flabel_005fset_005flabel"></a>GIMPLE function: <em>void</em> <strong>gimple_label_set_label</strong> <em>(gimple g, tree label)</em></dt>
<dd><p>Set <code>LABEL</code> to be the <code>LABEL_DECL</code> node used by <code>GIMPLE_LABEL</code>
statement <code>G</code>.
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fbuild_005fgoto"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_goto</strong> <em>(tree dest)</em></dt>
<dd><p>Build a <code>GIMPLE_GOTO</code> statement to label <code>DEST</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fgoto_005fdest"></a>GIMPLE function: <em>tree</em> <strong>gimple_goto_dest</strong> <em>(gimple g)</em></dt>
<dd><p>Return the destination of the unconditional jump <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fgoto_005fset_005fdest"></a>GIMPLE function: <em>void</em> <strong>gimple_goto_set_dest</strong> <em>(gimple g, tree dest)</em></dt>
<dd><p>Set <code>DEST</code> to be the destination of the unconditional jump <code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fNOP"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD" accesskey="n" rel="next"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a>, Previous: <a href="#GIMPLE_005fLABEL" accesskey="p" rel="prev"><code>GIMPLE_LABEL</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fNOP-1"></a>
<h4 class="subsection">12.7.10 <code>GIMPLE_NOP</code></h4>
<a name="index-GIMPLE_005fNOP"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fnop"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_nop</strong> <em>(void)</em></dt>
<dd><p>Build a <code>GIMPLE_NOP</code> statement.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fnop_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_nop_p</strong> <em>(gimple g)</em></dt>
<dd><p>Returns <code>TRUE</code> if statement <code>G</code> is a <code>GIMPLE_NOP</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fOMP_005fATOMIC_005fLOAD"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE" accesskey="n" rel="next"><code>GIMPLE_OMP_ATOMIC_STORE</code></a>, Previous: <a href="#GIMPLE_005fNOP" accesskey="p" rel="prev"><code>GIMPLE_NOP</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fATOMIC_005fLOAD-1"></a>
<h4 class="subsection">12.7.11 <code>GIMPLE_OMP_ATOMIC_LOAD</code></h4>
<a name="index-GIMPLE_005fOMP_005fATOMIC_005fLOAD"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fatomic_005fload"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_atomic_load</strong> <em>(tree lhs, tree rhs)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_ATOMIC_LOAD</code> statement.  <code>LHS</code> is the left-hand
side of the assignment.  <code>RHS</code> is the right-hand side of the
assignment.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fload_005fset_005flhs"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_atomic_load_set_lhs</strong> <em>(gimple g, tree lhs)</em></dt>
<dd><p>Set the <code>LHS</code> of an atomic load.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fload_005flhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_atomic_load_lhs</strong> <em>(gimple g)</em></dt>
<dd><p>Get the <code>LHS</code> of an atomic load.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fload_005fset_005frhs"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_atomic_load_set_rhs</strong> <em>(gimple g, tree rhs)</em></dt>
<dd><p>Set the <code>RHS</code> of an atomic set.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fload_005frhs"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_atomic_load_rhs</strong> <em>(gimple g)</em></dt>
<dd><p>Get the <code>RHS</code> of an atomic set.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fATOMIC_005fSTORE"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fCONTINUE" accesskey="n" rel="next"><code>GIMPLE_OMP_CONTINUE</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD" accesskey="p" rel="prev"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fATOMIC_005fSTORE-1"></a>
<h4 class="subsection">12.7.12 <code>GIMPLE_OMP_ATOMIC_STORE</code></h4>
<a name="index-GIMPLE_005fOMP_005fATOMIC_005fSTORE"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fatomic_005fstore"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_atomic_store</strong> <em>(tree val)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_ATOMIC_STORE</code> statement. <code>VAL</code> is the value to be
stored.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fstore_005fset_005fval"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_atomic_store_set_val</strong> <em>(gimple g, tree val)</em></dt>
<dd><p>Set the value being stored in an atomic store.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fatomic_005fstore_005fval"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_atomic_store_val</strong> <em>(gimple g)</em></dt>
<dd><p>Return the value being stored in an atomic store.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fOMP_005fCONTINUE"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fCRITICAL" accesskey="n" rel="next"><code>GIMPLE_OMP_CRITICAL</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE" accesskey="p" rel="prev"><code>GIMPLE_OMP_ATOMIC_STORE</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fCONTINUE-1"></a>
<h4 class="subsection">12.7.13 <code>GIMPLE_OMP_CONTINUE</code></h4>
<a name="index-GIMPLE_005fOMP_005fCONTINUE"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fcontinue"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_continue</strong> <em>(tree control_def, tree control_use)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_CONTINUE</code> statement.  <code>CONTROL_DEF</code> is the
definition of the control variable.  <code>CONTROL_USE</code> is the use of
the control variable.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fcontrol_005fdef"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_control_def</strong> <em>(gimple s)</em></dt>
<dd><p>Return the definition of the control variable on a
<code>GIMPLE_OMP_CONTINUE</code> in <code>S</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fcontrol_005fdef_005fptr"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_control_def_ptr</strong> <em>(gimple s)</em></dt>
<dd><p>Same as above, but return the pointer.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fset_005fcontrol_005fdef"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_set_control_def</strong> <em>(gimple s)</em></dt>
<dd><p>Set the control variable definition for a <code>GIMPLE_OMP_CONTINUE</code>
statement in <code>S</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fcontrol_005fuse"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_control_use</strong> <em>(gimple s)</em></dt>
<dd><p>Return the use of the control variable on a <code>GIMPLE_OMP_CONTINUE</code>
in <code>S</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fcontrol_005fuse_005fptr"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_control_use_ptr</strong> <em>(gimple s)</em></dt>
<dd><p>Same as above, but return the pointer.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcontinue_005fset_005fcontrol_005fuse"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_continue_set_control_use</strong> <em>(gimple s)</em></dt>
<dd><p>Set the control variable use for a <code>GIMPLE_OMP_CONTINUE</code> statement
in <code>S</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fCRITICAL"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fFOR" accesskey="n" rel="next"><code>GIMPLE_OMP_FOR</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fCONTINUE" accesskey="p" rel="prev"><code>GIMPLE_OMP_CONTINUE</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fCRITICAL-1"></a>
<h4 class="subsection">12.7.14 <code>GIMPLE_OMP_CRITICAL</code></h4>
<a name="index-GIMPLE_005fOMP_005fCRITICAL"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fcritical"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_critical</strong> <em>(gimple_seq body, tree name)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_CRITICAL</code> statement. <code>BODY</code> is the sequence of
statements for which only one thread can execute.  <code>NAME</code> is an
optional identifier for this critical block.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcritical_005fname"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_critical_name</strong> <em>(gimple g)</em></dt>
<dd><p>Return the name associated with <code>OMP_CRITICAL</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcritical_005fname_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_critical_name_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the name associated with <code>OMP</code> critical
statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fcritical_005fset_005fname"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_critical_set_name</strong> <em>(gimple g, tree name)</em></dt>
<dd><p>Set <code>NAME</code> to be the name associated with <code>OMP</code> critical statement <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fOMP_005fFOR"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fMASTER" accesskey="n" rel="next"><code>GIMPLE_OMP_MASTER</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fCRITICAL" accesskey="p" rel="prev"><code>GIMPLE_OMP_CRITICAL</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fFOR-1"></a>
<h4 class="subsection">12.7.15 <code>GIMPLE_OMP_FOR</code></h4>
<a name="index-GIMPLE_005fOMP_005fFOR"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005ffor"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_for</strong> <em>(gimple_seq body, tree clauses, tree index, tree initial, tree final, tree incr, gimple_seq pre_body, enum tree_code omp_for_cond)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_FOR</code> statement. <code>BODY</code> is sequence of statements
inside the for loop.  <code>CLAUSES</code>, are any of the <code>OMP</code> loop
construct&rsquo;s clauses: private, firstprivate,  lastprivate,
reductions, ordered, schedule, and nowait.  <code>PRE_BODY</code> is the
sequence of statements that are loop invariant.  <code>INDEX</code> is the
index variable.  <code>INITIAL</code> is the initial value of <code>INDEX</code>.  <code>FINAL</code> is
final value of <code>INDEX</code>.  OMP_FOR_COND is the predicate used to
compare <code>INDEX</code> and <code>FINAL</code>.  <code>INCR</code> is the increment expression.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fclauses"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_for_clauses</strong> <em>(gimple g)</em></dt>
<dd><p>Return the clauses associated with <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fclauses_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_for_clauses_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005fclauses"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_clauses</strong> <em>(gimple g, tree clauses)</em></dt>
<dd><p>Set <code>CLAUSES</code> to be the list of clauses associated with <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005findex"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_for_index</strong> <em>(gimple g)</em></dt>
<dd><p>Return the index variable for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005findex_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_for_index_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the index variable for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005findex"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_index</strong> <em>(gimple g, tree index)</em></dt>
<dd><p>Set <code>INDEX</code> to be the index variable for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005finitial"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_for_initial</strong> <em>(gimple g)</em></dt>
<dd><p>Return the initial value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005finitial_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_for_initial_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the initial value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005finitial"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_initial</strong> <em>(gimple g, tree initial)</em></dt>
<dd><p>Set <code>INITIAL</code> to be the initial value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005ffinal"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_for_final</strong> <em>(gimple g)</em></dt>
<dd><p>Return the final value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005ffinal_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_for_final_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>turn a pointer to the final value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005ffinal"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_final</strong> <em>(gimple g, tree final)</em></dt>
<dd><p>Set <code>FINAL</code> to be the final value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fincr"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_for_incr</strong> <em>(gimple g)</em></dt>
<dd><p>Return the increment value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fincr_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_for_incr_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the increment value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005fincr"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_incr</strong> <em>(gimple g, tree incr)</em></dt>
<dd><p>Set <code>INCR</code> to be the increment value for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fpre_005fbody"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_omp_for_pre_body</strong> <em>(gimple g)</em></dt>
<dd><p>Return the sequence of statements to execute before the <code>OMP_FOR</code>
statement <code>G</code> starts.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005fpre_005fbody"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_pre_body</strong> <em>(gimple g, gimple_seq pre_body)</em></dt>
<dd><p>Set <code>PRE_BODY</code> to be the sequence of statements to execute before
the <code>OMP_FOR</code> statement <code>G</code> starts.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fset_005fcond"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_for_set_cond</strong> <em>(gimple g, enum tree_code cond)</em></dt>
<dd><p>Set <code>COND</code> to be the condition code for <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005ffor_005fcond"></a>GIMPLE function: <em>enum tree_code</em> <strong>gimple_omp_for_cond</strong> <em>(gimple g)</em></dt>
<dd><p>Return the condition code associated with <code>OMP_FOR</code> <code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fMASTER"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fORDERED" accesskey="n" rel="next"><code>GIMPLE_OMP_ORDERED</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fFOR" accesskey="p" rel="prev"><code>GIMPLE_OMP_FOR</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fMASTER-1"></a>
<h4 class="subsection">12.7.16 <code>GIMPLE_OMP_MASTER</code></h4>
<a name="index-GIMPLE_005fOMP_005fMASTER"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fmaster"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_master</strong> <em>(gimple_seq body)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_MASTER</code> statement. <code>BODY</code> is the sequence of
statements to be executed by just the master.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fORDERED"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fPARALLEL" accesskey="n" rel="next"><code>GIMPLE_OMP_PARALLEL</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fMASTER" accesskey="p" rel="prev"><code>GIMPLE_OMP_MASTER</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fORDERED-1"></a>
<h4 class="subsection">12.7.17 <code>GIMPLE_OMP_ORDERED</code></h4>
<a name="index-GIMPLE_005fOMP_005fORDERED"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fordered"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_ordered</strong> <em>(gimple_seq body)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_ORDERED</code> statement.
</p></dd></dl>

<p><code>BODY</code> is the sequence of statements inside a loop that will
executed in sequence.
</p>

<hr>
<a name="GIMPLE_005fOMP_005fPARALLEL"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fRETURN" accesskey="n" rel="next"><code>GIMPLE_OMP_RETURN</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fORDERED" accesskey="p" rel="prev"><code>GIMPLE_OMP_ORDERED</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fPARALLEL-1"></a>
<h4 class="subsection">12.7.18 <code>GIMPLE_OMP_PARALLEL</code></h4>
<a name="index-GIMPLE_005fOMP_005fPARALLEL"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fparallel"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_parallel</strong> <em>(gimple_seq body, tree clauses, tree child_fn, tree data_arg)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_PARALLEL</code> statement.
</p></dd></dl>

<p><code>BODY</code> is sequence of statements which are executed in parallel.
<code>CLAUSES</code>, are the <code>OMP</code> parallel construct&rsquo;s clauses.  <code>CHILD_FN</code> is
the function created for the parallel threads to execute.
<code>DATA_ARG</code> are the shared data argument(s).
</p>
<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fcombined_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_omp_parallel_combined_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if <code>OMP</code> parallel statement <code>G</code> has the
<code>GF_OMP_PARALLEL_COMBINED</code> flag set.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fset_005fcombined_005fp"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_parallel_set_combined_p</strong> <em>(gimple g)</em></dt>
<dd><p>Set the <code>GF_OMP_PARALLEL_COMBINED</code> field in <code>OMP</code> parallel statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fbody"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_omp_body</strong> <em>(gimple g)</em></dt>
<dd><p>Return the body for the <code>OMP</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fset_005fbody"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_set_body</strong> <em>(gimple g, gimple_seq body)</em></dt>
<dd><p>Set <code>BODY</code> to be the body for the <code>OMP</code> statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fclauses"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_parallel_clauses</strong> <em>(gimple g)</em></dt>
<dd><p>Return the clauses associated with <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fclauses_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_parallel_clauses_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the clauses associated with <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fset_005fclauses"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_parallel_set_clauses</strong> <em>(gimple g, tree clauses)</em></dt>
<dd><p>Set <code>CLAUSES</code> to be the list of clauses associated with
<code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fchild_005ffn"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_parallel_child_fn</strong> <em>(gimple g)</em></dt>
<dd><p>Return the child function used to hold the body of <code>OMP_PARALLEL</code>
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fchild_005ffn_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_parallel_child_fn_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the child function used to hold the body of
<code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fset_005fchild_005ffn"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_parallel_set_child_fn</strong> <em>(gimple g, tree child_fn)</em></dt>
<dd><p>Set <code>CHILD_FN</code> to be the child function for <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fdata_005farg"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_parallel_data_arg</strong> <em>(gimple g)</em></dt>
<dd><p>Return the artificial argument used to send variables and values
from the parent to the children threads in <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fdata_005farg_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_parallel_data_arg_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the data argument for <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fparallel_005fset_005fdata_005farg"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_parallel_set_data_arg</strong> <em>(gimple g, tree data_arg)</em></dt>
<dd><p>Set <code>DATA_ARG</code> to be the data argument for <code>OMP_PARALLEL</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-is_005fgimple_005fomp"></a>GIMPLE function: <em>bool</em> <strong>is_gimple_omp</strong> <em>(gimple stmt)</em></dt>
<dd><p>Returns true when the gimple statement <code>STMT</code> is any of the OpenMP
types.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fRETURN"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fSECTION" accesskey="n" rel="next"><code>GIMPLE_OMP_SECTION</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fPARALLEL" accesskey="p" rel="prev"><code>GIMPLE_OMP_PARALLEL</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fRETURN-1"></a>
<h4 class="subsection">12.7.19 <code>GIMPLE_OMP_RETURN</code></h4>
<a name="index-GIMPLE_005fOMP_005fRETURN"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005freturn"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_return</strong> <em>(bool wait_p)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_RETURN</code> statement. <code>WAIT_P</code> is true if this is a
non-waiting return.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005freturn_005fset_005fnowait"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_return_set_nowait</strong> <em>(gimple s)</em></dt>
<dd><p>Set the nowait flag on <code>GIMPLE_OMP_RETURN</code> statement <code>S</code>.
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fomp_005freturn_005fnowait_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_omp_return_nowait_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if <code>OMP</code> return statement <code>G</code> has the
<code>GF_OMP_RETURN_NOWAIT</code> flag set.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fOMP_005fSECTION"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fSECTIONS" accesskey="n" rel="next"><code>GIMPLE_OMP_SECTIONS</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fRETURN" accesskey="p" rel="prev"><code>GIMPLE_OMP_RETURN</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fSECTION-1"></a>
<h4 class="subsection">12.7.20 <code>GIMPLE_OMP_SECTION</code></h4>
<a name="index-GIMPLE_005fOMP_005fSECTION"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fsection"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_section</strong> <em>(gimple_seq body)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_SECTION</code> statement for a sections statement.
</p></dd></dl>

<p><code>BODY</code> is the sequence of statements in the section.
</p>
<dl>
<dt><a name="index-gimple_005fomp_005fsection_005flast_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_omp_section_last_p</strong> <em>(gimple g)</em></dt>
<dd><p>Return true if <code>OMP</code> section statement <code>G</code> has the
<code>GF_OMP_SECTION_LAST</code> flag set.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsection_005fset_005flast"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_section_set_last</strong> <em>(gimple g)</em></dt>
<dd><p>Set the <code>GF_OMP_SECTION_LAST</code> flag on <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fOMP_005fSECTIONS"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fOMP_005fSINGLE" accesskey="n" rel="next"><code>GIMPLE_OMP_SINGLE</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fSECTION" accesskey="p" rel="prev"><code>GIMPLE_OMP_SECTION</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fSECTIONS-1"></a>
<h4 class="subsection">12.7.21 <code>GIMPLE_OMP_SECTIONS</code></h4>
<a name="index-GIMPLE_005fOMP_005fSECTIONS"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fsections"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_sections</strong> <em>(gimple_seq body, tree clauses)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_SECTIONS</code> statement. <code>BODY</code> is a sequence of
section statements.  <code>CLAUSES</code> are any of the <code>OMP</code> sections
construct&rsquo;s clauses: private, firstprivate, lastprivate,
reduction, and nowait.
</p></dd></dl>


<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fsections_005fswitch"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_sections_switch</strong> <em>(void)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_SECTIONS_SWITCH</code> statement.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fcontrol"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_sections_control</strong> <em>(gimple g)</em></dt>
<dd><p>Return the control variable associated with the
<code>GIMPLE_OMP_SECTIONS</code> in <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fcontrol_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_sections_control_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the clauses associated with the
<code>GIMPLE_OMP_SECTIONS</code> in <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fset_005fcontrol"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_sections_set_control</strong> <em>(gimple g, tree control)</em></dt>
<dd><p>Set <code>CONTROL</code> to be the set of clauses associated with the
<code>GIMPLE_OMP_SECTIONS</code> in <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fclauses"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_sections_clauses</strong> <em>(gimple g)</em></dt>
<dd><p>Return the clauses associated with <code>OMP_SECTIONS</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fclauses_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_sections_clauses_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the clauses associated with <code>OMP_SECTIONS</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsections_005fset_005fclauses"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_sections_set_clauses</strong> <em>(gimple g, tree clauses)</em></dt>
<dd><p>Set <code>CLAUSES</code> to be the set of clauses associated with <code>OMP_SECTIONS</code>
<code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fOMP_005fSINGLE"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fPHI" accesskey="n" rel="next"><code>GIMPLE_PHI</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fSECTIONS" accesskey="p" rel="prev"><code>GIMPLE_OMP_SECTIONS</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fOMP_005fSINGLE-1"></a>
<h4 class="subsection">12.7.22 <code>GIMPLE_OMP_SINGLE</code></h4>
<a name="index-GIMPLE_005fOMP_005fSINGLE"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fomp_005fsingle"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_omp_single</strong> <em>(gimple_seq body, tree clauses)</em></dt>
<dd><p>Build a <code>GIMPLE_OMP_SINGLE</code> statement. <code>BODY</code> is the sequence of
statements that will be executed once.  <code>CLAUSES</code> are any of the
<code>OMP</code> single construct&rsquo;s clauses: private, firstprivate,
copyprivate, nowait.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsingle_005fclauses"></a>GIMPLE function: <em>tree</em> <strong>gimple_omp_single_clauses</strong> <em>(gimple g)</em></dt>
<dd><p>Return the clauses associated with <code>OMP_SINGLE</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsingle_005fclauses_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_omp_single_clauses_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the clauses associated with <code>OMP_SINGLE</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fomp_005fsingle_005fset_005fclauses"></a>GIMPLE function: <em>void</em> <strong>gimple_omp_single_set_clauses</strong> <em>(gimple g, tree clauses)</em></dt>
<dd><p>Set <code>CLAUSES</code> to be the clauses associated with <code>OMP_SINGLE</code> <code>G</code>.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fPHI"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fRESX" accesskey="n" rel="next"><code>GIMPLE_RESX</code></a>, Previous: <a href="#GIMPLE_005fOMP_005fSINGLE" accesskey="p" rel="prev"><code>GIMPLE_OMP_SINGLE</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fPHI-1"></a>
<h4 class="subsection">12.7.23 <code>GIMPLE_PHI</code></h4>
<a name="index-GIMPLE_005fPHI"></a>

<dl>
<dt><a name="index-gimple_005fphi_005fcapacity"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_phi_capacity</strong> <em>(gimple g)</em></dt>
<dd><p>Return the maximum number of arguments supported by <code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005fnum_005fargs"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_phi_num_args</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of arguments in <code>GIMPLE_PHI</code> <code>G</code>. This must always
be exactly the number of incoming edges for the basic block
holding <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005fresult"></a>GIMPLE function: <em>tree</em> <strong>gimple_phi_result</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>SSA</code> name created by <code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005fresult_005fptr"></a>GIMPLE function: <em>tree *</em> <strong>gimple_phi_result_ptr</strong> <em>(gimple g)</em></dt>
<dd><p>Return a pointer to the <code>SSA</code> name created by <code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005fset_005fresult"></a>GIMPLE function: <em>void</em> <strong>gimple_phi_set_result</strong> <em>(gimple g, tree result)</em></dt>
<dd><p>Set <code>RESULT</code> to be the <code>SSA</code> name created by <code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005farg"></a>GIMPLE function: <em>struct phi_arg_d *</em> <strong>gimple_phi_arg</strong> <em>(gimple g, index)</em></dt>
<dd><p>Return the <code>PHI</code> argument corresponding to incoming edge <code>INDEX</code> for
<code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fphi_005fset_005farg"></a>GIMPLE function: <em>void</em> <strong>gimple_phi_set_arg</strong> <em>(gimple g, index, struct phi_arg_d * phiarg)</em></dt>
<dd><p>Set <code>PHIARG</code> to be the argument corresponding to incoming edge
<code>INDEX</code> for <code>GIMPLE_PHI</code> <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fRESX"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fRETURN" accesskey="n" rel="next"><code>GIMPLE_RETURN</code></a>, Previous: <a href="#GIMPLE_005fPHI" accesskey="p" rel="prev"><code>GIMPLE_PHI</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fRESX-1"></a>
<h4 class="subsection">12.7.24 <code>GIMPLE_RESX</code></h4>
<a name="index-GIMPLE_005fRESX"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fresx"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_resx</strong> <em>(int region)</em></dt>
<dd><p>Build a <code>GIMPLE_RESX</code> statement which is a statement.  This
statement is a placeholder for _Unwind_Resume before we know if a
function call or a branch is needed.  <code>REGION</code> is the exception
region from which control is flowing.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fresx_005fregion"></a>GIMPLE function: <em>int</em> <strong>gimple_resx_region</strong> <em>(gimple g)</em></dt>
<dd><p>Return the region number for <code>GIMPLE_RESX</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fresx_005fset_005fregion"></a>GIMPLE function: <em>void</em> <strong>gimple_resx_set_region</strong> <em>(gimple g, int region)</em></dt>
<dd><p>Set <code>REGION</code> to be the region number for <code>GIMPLE_RESX</code> <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fRETURN"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fSWITCH" accesskey="n" rel="next"><code>GIMPLE_SWITCH</code></a>, Previous: <a href="#GIMPLE_005fRESX" accesskey="p" rel="prev"><code>GIMPLE_RESX</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fRETURN-1"></a>
<h4 class="subsection">12.7.25 <code>GIMPLE_RETURN</code></h4>
<a name="index-GIMPLE_005fRETURN"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005freturn"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_return</strong> <em>(tree retval)</em></dt>
<dd><p>Build a <code>GIMPLE_RETURN</code> statement whose return value is retval.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005freturn_005fretval"></a>GIMPLE function: <em>tree</em> <strong>gimple_return_retval</strong> <em>(gimple g)</em></dt>
<dd><p>Return the return value for <code>GIMPLE_RETURN</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005freturn_005fset_005fretval"></a>GIMPLE function: <em>void</em> <strong>gimple_return_set_retval</strong> <em>(gimple g, tree retval)</em></dt>
<dd><p>Set <code>RETVAL</code> to be the return value for <code>GIMPLE_RETURN</code> <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fSWITCH"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fTRY" accesskey="n" rel="next"><code>GIMPLE_TRY</code></a>, Previous: <a href="#GIMPLE_005fRETURN" accesskey="p" rel="prev"><code>GIMPLE_RETURN</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fSWITCH-1"></a>
<h4 class="subsection">12.7.26 <code>GIMPLE_SWITCH</code></h4>
<a name="index-GIMPLE_005fSWITCH"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fswitch"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_switch</strong> <em>(tree index, tree default_label, <code>VEC</code>(tree,heap) *args)</em></dt>
<dd><p>Build a <code>GIMPLE_SWITCH</code> statement.  <code>INDEX</code> is the index variable
to switch on, and <code>DEFAULT_LABEL</code> represents the default label.
<code>ARGS</code> is a vector of <code>CASE_LABEL_EXPR</code> trees that contain the
non-default case labels.  Each label is a tree of code <code>CASE_LABEL_EXPR</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fnum_005flabels"></a>GIMPLE function: <em>unsigned</em> <strong>gimple_switch_num_labels</strong> <em>(gimple g)</em></dt>
<dd><p>Return the number of labels associated with the switch statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fset_005fnum_005flabels"></a>GIMPLE function: <em>void</em> <strong>gimple_switch_set_num_labels</strong> <em>(gimple g, unsigned nlabels)</em></dt>
<dd><p>Set <code>NLABELS</code> to be the number of labels for the switch statement
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005findex"></a>GIMPLE function: <em>tree</em> <strong>gimple_switch_index</strong> <em>(gimple g)</em></dt>
<dd><p>Return the index variable used by the switch statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fset_005findex"></a>GIMPLE function: <em>void</em> <strong>gimple_switch_set_index</strong> <em>(gimple g, tree index)</em></dt>
<dd><p>Set <code>INDEX</code> to be the index variable for switch statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005flabel"></a>GIMPLE function: <em>tree</em> <strong>gimple_switch_label</strong> <em>(gimple g, unsigned index)</em></dt>
<dd><p>Return the label numbered <code>INDEX</code>. The default label is 0, followed
by any labels in a switch statement.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fset_005flabel"></a>GIMPLE function: <em>void</em> <strong>gimple_switch_set_label</strong> <em>(gimple g, unsigned index, tree label)</em></dt>
<dd><p>Set the label number <code>INDEX</code> to <code>LABEL</code>. 0 is always the default
label.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fdefault_005flabel"></a>GIMPLE function: <em>tree</em> <strong>gimple_switch_default_label</strong> <em>(gimple g)</em></dt>
<dd><p>Return the default label for a switch statement.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fswitch_005fset_005fdefault_005flabel"></a>GIMPLE function: <em>void</em> <strong>gimple_switch_set_default_label</strong> <em>(gimple g, tree label)</em></dt>
<dd><p>Set the default label for a switch statement.
</p></dd></dl>


<hr>
<a name="GIMPLE_005fTRY"></a>
<div class="header">
<p>
Next: <a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR" accesskey="n" rel="next"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a>, Previous: <a href="#GIMPLE_005fSWITCH" accesskey="p" rel="prev"><code>GIMPLE_SWITCH</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fTRY-1"></a>
<h4 class="subsection">12.7.27 <code>GIMPLE_TRY</code></h4>
<a name="index-GIMPLE_005fTRY"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005ftry"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_try</strong> <em>(gimple_seq eval, gimple_seq cleanup, unsigned int kind)</em></dt>
<dd><p>Build a <code>GIMPLE_TRY</code> statement.  <code>EVAL</code> is a sequence with the
expression to evaluate.  <code>CLEANUP</code> is a sequence of statements to
run at clean-up time.  <code>KIND</code> is the enumeration value
<code>GIMPLE_TRY_CATCH</code> if this statement denotes a try/catch construct
or <code>GIMPLE_TRY_FINALLY</code> if this statement denotes a try/finally
construct.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fkind"></a>GIMPLE function: <em>enum gimple_try_flags</em> <strong>gimple_try_kind</strong> <em>(gimple g)</em></dt>
<dd><p>Return the kind of try block represented by <code>GIMPLE_TRY</code> <code>G</code>. This is
either <code>GIMPLE_TRY_CATCH</code> or <code>GIMPLE_TRY_FINALLY</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fcatch_005fis_005fcleanup"></a>GIMPLE function: <em>bool</em> <strong>gimple_try_catch_is_cleanup</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>GIMPLE_TRY_CATCH_IS_CLEANUP</code> flag.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005feval"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_try_eval</strong> <em>(gimple g)</em></dt>
<dd><p>Return the sequence of statements used as the body for <code>GIMPLE_TRY</code>
<code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fcleanup"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_try_cleanup</strong> <em>(gimple g)</em></dt>
<dd><p>Return the sequence of statements used as the cleanup body for
<code>GIMPLE_TRY</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fset_005fcatch_005fis_005fcleanup"></a>GIMPLE function: <em>void</em> <strong>gimple_try_set_catch_is_cleanup</strong> <em>(gimple g, bool catch_is_cleanup)</em></dt>
<dd><p>Set the <code>GIMPLE_TRY_CATCH_IS_CLEANUP</code> flag.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fset_005feval"></a>GIMPLE function: <em>void</em> <strong>gimple_try_set_eval</strong> <em>(gimple g, gimple_seq eval)</em></dt>
<dd><p>Set <code>EVAL</code> to be the sequence of statements to use as the body for
<code>GIMPLE_TRY</code> <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005ftry_005fset_005fcleanup"></a>GIMPLE function: <em>void</em> <strong>gimple_try_set_cleanup</strong> <em>(gimple g, gimple_seq cleanup)</em></dt>
<dd><p>Set <code>CLEANUP</code> to be the sequence of statements to use as the
cleanup body for <code>GIMPLE_TRY</code> <code>G</code>.
</p></dd></dl>

<hr>
<a name="GIMPLE_005fWITH_005fCLEANUP_005fEXPR"></a>
<div class="header">
<p>
Previous: <a href="#GIMPLE_005fTRY" accesskey="p" rel="prev"><code>GIMPLE_TRY</code></a>, Up: <a href="#Tuple-specific-accessors" accesskey="u" rel="up">Tuple specific accessors</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE_005fWITH_005fCLEANUP_005fEXPR-1"></a>
<h4 class="subsection">12.7.28 <code>GIMPLE_WITH_CLEANUP_EXPR</code></h4>
<a name="index-GIMPLE_005fWITH_005fCLEANUP_005fEXPR"></a>

<dl>
<dt><a name="index-gimple_005fbuild_005fwce"></a>GIMPLE function: <em>gimple</em> <strong>gimple_build_wce</strong> <em>(gimple_seq cleanup)</em></dt>
<dd><p>Build a <code>GIMPLE_WITH_CLEANUP_EXPR</code> statement.  <code>CLEANUP</code> is the
clean-up expression.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fwce_005fcleanup"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_wce_cleanup</strong> <em>(gimple g)</em></dt>
<dd><p>Return the cleanup sequence for cleanup statement <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fwce_005fset_005fcleanup"></a>GIMPLE function: <em>void</em> <strong>gimple_wce_set_cleanup</strong> <em>(gimple g, gimple_seq cleanup)</em></dt>
<dd><p>Set <code>CLEANUP</code> to be the cleanup sequence for <code>G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fwce_005fcleanup_005feh_005fonly"></a>GIMPLE function: <em>bool</em> <strong>gimple_wce_cleanup_eh_only</strong> <em>(gimple g)</em></dt>
<dd><p>Return the <code>CLEANUP_EH_ONLY</code> flag for a <code>WCE</code> tuple.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fwce_005fset_005fcleanup_005feh_005fonly"></a>GIMPLE function: <em>void</em> <strong>gimple_wce_set_cleanup_eh_only</strong> <em>(gimple g, bool eh_only_p)</em></dt>
<dd><p>Set the <code>CLEANUP_EH_ONLY</code> flag for a <code>WCE</code> tuple.
</p></dd></dl>


<hr>
<a name="GIMPLE-sequences"></a>
<div class="header">
<p>
Next: <a href="#Sequence-iterators" accesskey="n" rel="next">Sequence iterators</a>, Previous: <a href="#Tuple-specific-accessors" accesskey="p" rel="prev">Tuple specific accessors</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GIMPLE-sequences-1"></a>
<h3 class="section">12.8 GIMPLE sequences</h3>
<a name="index-GIMPLE-sequences"></a>

<p>GIMPLE sequences are the tuple equivalent of <code>STATEMENT_LIST</code>&rsquo;s
used in <code>GENERIC</code>.  They are used to chain statements together, and
when used in conjunction with sequence iterators, provide a
framework for iterating through statements.
</p>
<p>GIMPLE sequences are of type struct <code>gimple_sequence</code>, but are more
commonly passed by reference to functions dealing with sequences.
The type for a sequence pointer is <code>gimple_seq</code> which is the same
as struct <code>gimple_sequence</code> *.  When declaring a local sequence,
you can define a local variable of type struct <code>gimple_sequence</code>.
When declaring a sequence allocated on the garbage collected
heap, use the function <code>gimple_seq_alloc</code> documented below.
</p>
<p>There are convenience functions for iterating through sequences
in the section entitled Sequence Iterators.
</p>
<p>Below is a list of functions to manipulate and query sequences.
</p>
<dl>
<dt><a name="index-gimple_005fseq_005fadd_005fstmt"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_add_stmt</strong> <em>(gimple_seq *seq, gimple g)</em></dt>
<dd><p>Link a gimple statement to the end of the sequence *<code>SEQ</code> if <code>G</code> is
not <code>NULL</code>.  If *<code>SEQ</code> is <code>NULL</code>, allocate a sequence before linking.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fadd_005fseq"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_add_seq</strong> <em>(gimple_seq *dest, gimple_seq src)</em></dt>
<dd><p>Append sequence <code>SRC</code> to the end of sequence *<code>DEST</code> if <code>SRC</code> is not
<code>NULL</code>.  If *<code>DEST</code> is <code>NULL</code>, allocate a new sequence before
appending.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fdeep_005fcopy"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_seq_deep_copy</strong> <em>(gimple_seq src)</em></dt>
<dd><p>Perform a deep copy of sequence <code>SRC</code> and return the result.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005freverse"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_seq_reverse</strong> <em>(gimple_seq seq)</em></dt>
<dd><p>Reverse the order of the statements in the sequence <code>SEQ</code>.  Return
<code>SEQ</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005ffirst"></a>GIMPLE function: <em>gimple</em> <strong>gimple_seq_first</strong> <em>(gimple_seq s)</em></dt>
<dd><p>Return the first statement in sequence <code>S</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005flast"></a>GIMPLE function: <em>gimple</em> <strong>gimple_seq_last</strong> <em>(gimple_seq s)</em></dt>
<dd><p>Return the last statement in sequence <code>S</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fset_005flast"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_set_last</strong> <em>(gimple_seq s, gimple last)</em></dt>
<dd><p>Set the last statement in sequence <code>S</code> to the statement in <code>LAST</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fset_005ffirst"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_set_first</strong> <em>(gimple_seq s, gimple first)</em></dt>
<dd><p>Set the first statement in sequence <code>S</code> to the statement in <code>FIRST</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005finit"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_init</strong> <em>(gimple_seq s)</em></dt>
<dd><p>Initialize sequence <code>S</code> to an empty sequence.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005falloc"></a>GIMPLE function: <em>gimple_seq</em> <strong>gimple_seq_alloc</strong> <em>(void)</em></dt>
<dd><p>Allocate a new sequence in the garbage collected store and return
it.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fcopy"></a>GIMPLE function: <em>void</em> <strong>gimple_seq_copy</strong> <em>(gimple_seq dest, gimple_seq src)</em></dt>
<dd><p>Copy the sequence <code>SRC</code> into the sequence <code>DEST</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fempty_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_seq_empty_p</strong> <em>(gimple_seq s)</em></dt>
<dd><p>Return true if the sequence <code>S</code> is empty.
</p></dd></dl>

<dl>
<dt><a name="index-bb_005fseq"></a>GIMPLE function: <em>gimple_seq</em> <strong>bb_seq</strong> <em>(basic_block bb)</em></dt>
<dd><p>Returns the sequence of statements in <code>BB</code>.
</p></dd></dl>

<dl>
<dt><a name="index-set_005fbb_005fseq"></a>GIMPLE function: <em>void</em> <strong>set_bb_seq</strong> <em>(basic_block bb, gimple_seq seq)</em></dt>
<dd><p>Sets the sequence of statements in <code>BB</code> to <code>SEQ</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gimple_005fseq_005fsingleton_005fp"></a>GIMPLE function: <em>bool</em> <strong>gimple_seq_singleton_p</strong> <em>(gimple_seq seq)</em></dt>
<dd><p>Determine whether <code>SEQ</code> contains exactly one statement.
</p></dd></dl>

<hr>
<a name="Sequence-iterators"></a>
<div class="header">
<p>
Next: <a href="#Adding-a-new-GIMPLE-statement-code" accesskey="n" rel="next">Adding a new GIMPLE statement code</a>, Previous: <a href="#GIMPLE-sequences" accesskey="p" rel="prev">GIMPLE sequences</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Sequence-iterators-1"></a>
<h3 class="section">12.9 Sequence iterators</h3>
<a name="index-Sequence-iterators"></a>

<p>Sequence iterators are convenience constructs for iterating
through statements in a sequence.  Given a sequence <code>SEQ</code>, here is
a typical use of gimple sequence iterators:
</p>
<div class="smallexample">
<pre class="smallexample">gimple_stmt_iterator gsi;

for (gsi = gsi_start (seq); !gsi_end_p (gsi); gsi_next (&amp;gsi))
  {
    gimple g = gsi_stmt (gsi);
    /* Do something with gimple statement <code>G</code>.  */
  }
</pre></div>

<p>Backward iterations are possible:
</p>
<div class="smallexample">
<pre class="smallexample">        for (gsi = gsi_last (seq); !gsi_end_p (gsi); gsi_prev (&amp;gsi))
</pre></div>

<p>Forward and backward iterations on basic blocks are possible with
<code>gsi_start_bb</code> and <code>gsi_last_bb</code>.
</p>
<p>In the documentation below we sometimes refer to enum
<code>gsi_iterator_update</code>.  The valid options for this enumeration are:
</p>
<ul>
<li> <code>GSI_NEW_STMT</code>
Only valid when a single statement is added.  Move the iterator to it.

</li><li> <code>GSI_SAME_STMT</code>
Leave the iterator at the same statement.

</li><li> <code>GSI_CONTINUE_LINKING</code>
Move iterator to whatever position is suitable for linking other
statements in the same direction.
</li></ul>

<p>Below is a list of the functions used to manipulate and use
statement iterators.
</p>
<dl>
<dt><a name="index-gsi_005fstart"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_start</strong> <em>(gimple_seq seq)</em></dt>
<dd><p>Return a new iterator pointing to the sequence <code>SEQ</code>&rsquo;s first
statement.  If <code>SEQ</code> is empty, the iterator&rsquo;s basic block is <code>NULL</code>.
Use <code>gsi_start_bb</code> instead when the iterator needs to always have
the correct basic block set.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fstart_005fbb"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_start_bb</strong> <em>(basic_block bb)</em></dt>
<dd><p>Return a new iterator pointing to the first statement in basic
block <code>BB</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flast"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_last</strong> <em>(gimple_seq seq)</em></dt>
<dd><p>Return a new iterator initially pointing to the last statement of
sequence <code>SEQ</code>.  If <code>SEQ</code> is empty, the iterator&rsquo;s basic block is
<code>NULL</code>.  Use <code>gsi_last_bb</code> instead when the iterator needs to always
have the correct basic block set.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flast_005fbb"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_last_bb</strong> <em>(basic_block bb)</em></dt>
<dd><p>Return a new iterator pointing to the last statement in basic
block <code>BB</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fend_005fp"></a>GIMPLE function: <em>bool</em> <strong>gsi_end_p</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Return <code>TRUE</code> if at the end of <code>I</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fone_005fbefore_005fend_005fp"></a>GIMPLE function: <em>bool</em> <strong>gsi_one_before_end_p</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Return <code>TRUE</code> if we&rsquo;re one statement before the end of <code>I</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fnext"></a>GIMPLE function: <em>void</em> <strong>gsi_next</strong> <em>(gimple_stmt_iterator *i)</em></dt>
<dd><p>Advance the iterator to the next gimple statement.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fprev"></a>GIMPLE function: <em>void</em> <strong>gsi_prev</strong> <em>(gimple_stmt_iterator *i)</em></dt>
<dd><p>Advance the iterator to the previous gimple statement.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fstmt"></a>GIMPLE function: <em>gimple</em> <strong>gsi_stmt</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Return the current stmt.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fafter_005flabels"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_after_labels</strong> <em>(basic_block bb)</em></dt>
<dd><p>Return a block statement iterator that points to the first
non-label statement in block <code>BB</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fstmt_005fptr"></a>GIMPLE function: <em>gimple *</em> <strong>gsi_stmt_ptr</strong> <em>(gimple_stmt_iterator *i)</em></dt>
<dd><p>Return a pointer to the current stmt.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fbb"></a>GIMPLE function: <em>basic_block</em> <strong>gsi_bb</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Return the basic block associated with this iterator.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fseq"></a>GIMPLE function: <em>gimple_seq</em> <strong>gsi_seq</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Return the sequence associated with this iterator.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fremove"></a>GIMPLE function: <em>void</em> <strong>gsi_remove</strong> <em>(gimple_stmt_iterator *i, bool remove_eh_info)</em></dt>
<dd><p>Remove the current stmt from the sequence.  The iterator is
updated to point to the next statement.  When <code>REMOVE_EH_INFO</code> is
true we remove the statement pointed to by iterator <code>I</code> from the <code>EH</code>
tables.  Otherwise we do not modify the <code>EH</code> tables.  Generally,
<code>REMOVE_EH_INFO</code> should be true when the statement is going to be
removed from the <code>IL</code> and not reinserted elsewhere.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flink_005fseq_005fbefore"></a>GIMPLE function: <em>void</em> <strong>gsi_link_seq_before</strong> <em>(gimple_stmt_iterator *i, gimple_seq seq, enum gsi_iterator_update mode)</em></dt>
<dd><p>Links the sequence of statements <code>SEQ</code> before the statement pointed
by iterator <code>I</code>.  <code>MODE</code> indicates what to do with the iterator
after insertion (see <code>enum gsi_iterator_update</code> above).
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flink_005fbefore"></a>GIMPLE function: <em>void</em> <strong>gsi_link_before</strong> <em>(gimple_stmt_iterator *i, gimple g, enum gsi_iterator_update mode)</em></dt>
<dd><p>Links statement <code>G</code> before the statement pointed-to by iterator <code>I</code>.
Updates iterator <code>I</code> according to <code>MODE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flink_005fseq_005fafter"></a>GIMPLE function: <em>void</em> <strong>gsi_link_seq_after</strong> <em>(gimple_stmt_iterator *i, gimple_seq seq, enum gsi_iterator_update mode)</em></dt>
<dd><p>Links sequence <code>SEQ</code> after the statement pointed-to by iterator <code>I</code>.
<code>MODE</code> is as in <code>gsi_insert_after</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005flink_005fafter"></a>GIMPLE function: <em>void</em> <strong>gsi_link_after</strong> <em>(gimple_stmt_iterator *i, gimple g, enum gsi_iterator_update mode)</em></dt>
<dd><p>Links statement <code>G</code> after the statement pointed-to by iterator <code>I</code>.
<code>MODE</code> is as in <code>gsi_insert_after</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fsplit_005fseq_005fafter"></a>GIMPLE function: <em>gimple_seq</em> <strong>gsi_split_seq_after</strong> <em>(gimple_stmt_iterator i)</em></dt>
<dd><p>Move all statements in the sequence after <code>I</code> to a new sequence.
Return this new sequence.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fsplit_005fseq_005fbefore"></a>GIMPLE function: <em>gimple_seq</em> <strong>gsi_split_seq_before</strong> <em>(gimple_stmt_iterator *i)</em></dt>
<dd><p>Move all statements in the sequence before <code>I</code> to a new sequence.
Return this new sequence.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005freplace"></a>GIMPLE function: <em>void</em> <strong>gsi_replace</strong> <em>(gimple_stmt_iterator *i, gimple stmt, bool update_eh_info)</em></dt>
<dd><p>Replace the statement pointed-to by <code>I</code> to <code>STMT</code>.  If <code>UPDATE_EH_INFO</code>
is true, the exception handling information of the original
statement is moved to the new statement.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fbefore"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_before</strong> <em>(gimple_stmt_iterator *i, gimple stmt, enum gsi_iterator_update mode)</em></dt>
<dd><p>Insert statement <code>STMT</code> before the statement pointed-to by iterator
<code>I</code>, update <code>STMT</code>&rsquo;s basic block and scan it for new operands.  <code>MODE</code>
specifies how to update iterator <code>I</code> after insertion (see enum
<code>gsi_iterator_update</code>).
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fseq_005fbefore"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_seq_before</strong> <em>(gimple_stmt_iterator *i, gimple_seq seq, enum gsi_iterator_update mode)</em></dt>
<dd><p>Like <code>gsi_insert_before</code>, but for all the statements in <code>SEQ</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fafter"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_after</strong> <em>(gimple_stmt_iterator *i, gimple stmt, enum gsi_iterator_update mode)</em></dt>
<dd><p>Insert statement <code>STMT</code> after the statement pointed-to by iterator
<code>I</code>, update <code>STMT</code>&rsquo;s basic block and scan it for new operands.  <code>MODE</code>
specifies how to update iterator <code>I</code> after insertion (see enum
<code>gsi_iterator_update</code>).
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fseq_005fafter"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_seq_after</strong> <em>(gimple_stmt_iterator *i, gimple_seq seq, enum gsi_iterator_update mode)</em></dt>
<dd><p>Like <code>gsi_insert_after</code>, but for all the statements in <code>SEQ</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005ffor_005fstmt"></a>GIMPLE function: <em>gimple_stmt_iterator</em> <strong>gsi_for_stmt</strong> <em>(gimple stmt)</em></dt>
<dd><p>Finds iterator for <code>STMT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fmove_005fafter"></a>GIMPLE function: <em>void</em> <strong>gsi_move_after</strong> <em>(gimple_stmt_iterator *from, gimple_stmt_iterator *to)</em></dt>
<dd><p>Move the statement at <code>FROM</code> so it comes right after the statement
at <code>TO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fmove_005fbefore"></a>GIMPLE function: <em>void</em> <strong>gsi_move_before</strong> <em>(gimple_stmt_iterator *from, gimple_stmt_iterator *to)</em></dt>
<dd><p>Move the statement at <code>FROM</code> so it comes right before the statement
at <code>TO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fmove_005fto_005fbb_005fend"></a>GIMPLE function: <em>void</em> <strong>gsi_move_to_bb_end</strong> <em>(gimple_stmt_iterator *from, basic_block bb)</em></dt>
<dd><p>Move the statement at <code>FROM</code> to the end of basic block <code>BB</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fon_005fedge"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_on_edge</strong> <em>(edge e, gimple stmt)</em></dt>
<dd><p>Add <code>STMT</code> to the pending list of edge <code>E</code>.  No actual insertion is
made until a call to <code>gsi_commit_edge_inserts</code>() is made.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fseq_005fon_005fedge"></a>GIMPLE function: <em>void</em> <strong>gsi_insert_seq_on_edge</strong> <em>(edge e, gimple_seq seq)</em></dt>
<dd><p>Add the sequence of statements in <code>SEQ</code> to the pending list of edge
<code>E</code>.  No actual insertion is made until a call to
<code>gsi_commit_edge_inserts</code>() is made.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005finsert_005fon_005fedge_005fimmediate"></a>GIMPLE function: <em>basic_block</em> <strong>gsi_insert_on_edge_immediate</strong> <em>(edge e, gimple stmt)</em></dt>
<dd><p>Similar to <code>gsi_insert_on_edge</code>+<code>gsi_commit_edge_inserts</code>.  If a new
block has to be created, it is returned.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fcommit_005fone_005fedge_005finsert"></a>GIMPLE function: <em>void</em> <strong>gsi_commit_one_edge_insert</strong> <em>(edge e, basic_block *new_bb)</em></dt>
<dd><p>Commit insertions pending at edge <code>E</code>.  If a new block is created,
set <code>NEW_BB</code> to this block, otherwise set it to <code>NULL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-gsi_005fcommit_005fedge_005finserts"></a>GIMPLE function: <em>void</em> <strong>gsi_commit_edge_inserts</strong> <em>(void)</em></dt>
<dd><p>This routine will commit all pending edge insertions, creating
any new basic blocks which are necessary.
</p></dd></dl>


<hr>
<a name="Adding-a-new-GIMPLE-statement-code"></a>
<div class="header">
<p>
Next: <a href="#Statement-and-operand-traversals" accesskey="n" rel="next">Statement and operand traversals</a>, Previous: <a href="#Sequence-iterators" accesskey="p" rel="prev">Sequence iterators</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Adding-a-new-GIMPLE-statement-code-1"></a>
<h3 class="section">12.10 Adding a new GIMPLE statement code</h3>
<a name="index-Adding-a-new-GIMPLE-statement-code"></a>

<p>The first step in adding a new GIMPLE statement code, is
modifying the file <code>gimple.def</code>, which contains all the GIMPLE
codes.  Then you must add a corresponding structure, and an entry
in <code>union gimple_statement_d</code>, both of which are located in
<code>gimple.h</code>.  This in turn, will require you to add a corresponding
<code>GTY</code> tag in <code>gsstruct.def</code>, and code to handle this tag in
<code>gss_for_code</code> which is located in <code>gimple.c</code>.
</p>
<p>In order for the garbage collector to know the size of the
structure you created in <code>gimple.h</code>, you need to add a case to
handle your new GIMPLE statement in <code>gimple_size</code> which is located
in <code>gimple.c</code>.
</p>
<p>You will probably want to create a function to build the new
gimple statement in <code>gimple.c</code>.  The function should be called
<code>gimple_build_<var>new-tuple-name</var></code>, and should return the new tuple
of type gimple.
</p>
<p>If your new statement requires accessors for any members or
operands it may have, put simple inline accessors in
<code>gimple.h</code> and any non-trivial accessors in <code>gimple.c</code> with a
corresponding prototype in <code>gimple.h</code>.
</p>

<hr>
<a name="Statement-and-operand-traversals"></a>
<div class="header">
<p>
Previous: <a href="#Adding-a-new-GIMPLE-statement-code" accesskey="p" rel="prev">Adding a new GIMPLE statement code</a>, Up: <a href="#GIMPLE" accesskey="u" rel="up">GIMPLE</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Statement-and-operand-traversals-1"></a>
<h3 class="section">12.11 Statement and operand traversals</h3>
<a name="index-Statement-and-operand-traversals"></a>

<p>There are two functions available for walking statements and
sequences: <code>walk_gimple_stmt</code> and <code>walk_gimple_seq</code>,
accordingly, and a third function for walking the operands in a
statement: <code>walk_gimple_op</code>.
</p>
<dl>
<dt><a name="index-walk_005fgimple_005fstmt"></a>GIMPLE function: <em>tree</em> <strong>walk_gimple_stmt</strong> <em>(gimple_stmt_iterator *gsi,   walk_stmt_fn callback_stmt, walk_tree_fn callback_op, struct walk_stmt_info *wi)</em></dt>
<dd><p>This function is used to walk the current statement in <code>GSI</code>,
optionally using traversal state stored in <code>WI</code>.  If <code>WI</code> is <code>NULL</code>, no
state is kept during the traversal.
</p>
<p>The callback <code>CALLBACK_STMT</code> is called.  If <code>CALLBACK_STMT</code> returns
true, it means that the callback function has handled all the
operands of the statement and it is not necessary to walk its
operands.
</p>
<p>If <code>CALLBACK_STMT</code> is <code>NULL</code> or it returns false, <code>CALLBACK_OP</code> is
called on each operand of the statement via <code>walk_gimple_op</code>.  If
<code>walk_gimple_op</code> returns non-<code>NULL</code> for any operand, the remaining
operands are not scanned.
</p>
<p>The return value is that returned by the last call to
<code>walk_gimple_op</code>, or <code>NULL_TREE</code> if no <code>CALLBACK_OP</code> is specified.
</p></dd></dl>


<dl>
<dt><a name="index-walk_005fgimple_005fop"></a>GIMPLE function: <em>tree</em> <strong>walk_gimple_op</strong> <em>(gimple stmt,   walk_tree_fn callback_op, struct walk_stmt_info *wi)</em></dt>
<dd><p>Use this function to walk the operands of statement <code>STMT</code>.  Every
operand is walked via <code>walk_tree</code> with optional state information
in <code>WI</code>.
</p>
<p><code>CALLBACK_OP</code> is called on each operand of <code>STMT</code> via <code>walk_tree</code>.
Additional parameters to <code>walk_tree</code> must be stored in <code>WI</code>.  For
each operand <code>OP</code>, <code>walk_tree</code> is called as:
</p>
<div class="smallexample">
<pre class="smallexample">walk_tree (&amp;<code>OP</code>, <code>CALLBACK_OP</code>, <code>WI</code>, <code>PSET</code>)
</pre></div>

<p>If <code>CALLBACK_OP</code> returns non-<code>NULL</code> for an operand, the remaining
operands are not scanned.  The return value is that returned by
the last call to <code>walk_tree</code>, or <code>NULL_TREE</code> if no <code>CALLBACK_OP</code> is
specified.
</p></dd></dl>


<dl>
<dt><a name="index-walk_005fgimple_005fseq"></a>GIMPLE function: <em>tree</em> <strong>walk_gimple_seq</strong> <em>(gimple_seq seq,   walk_stmt_fn callback_stmt, walk_tree_fn callback_op, struct walk_stmt_info *wi)</em></dt>
<dd><p>This function walks all the statements in the sequence <code>SEQ</code>
calling <code>walk_gimple_stmt</code> on each one.  <code>WI</code> is as in
<code>walk_gimple_stmt</code>.  If <code>walk_gimple_stmt</code> returns non-<code>NULL</code>, the walk
is stopped and the value returned.  Otherwise, all the statements
are walked and <code>NULL_TREE</code> returned.
</p></dd></dl>


<hr>
<a name="Tree-SSA"></a>
<div class="header">
<p>
Next: <a href="#Loop-Analysis-and-Representation" accesskey="n" rel="next">Loop Analysis and Representation</a>, Previous: <a href="#GIMPLE" accesskey="p" rel="prev">GIMPLE</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Analysis-and-Optimization-of-GIMPLE-tuples"></a>
<h2 class="chapter">13 Analysis and Optimization of GIMPLE tuples</h2>
<a name="index-Tree-SSA"></a>
<a name="index-Optimization-infrastructure-for-GIMPLE"></a>

<p>GCC uses three main intermediate languages to represent the program
during compilation: GENERIC, GIMPLE and RTL.  GENERIC is a
language-independent representation generated by each front end.  It
is used to serve as an interface between the parser and optimizer.
GENERIC is a common representation that is able to represent programs
written in all the languages supported by GCC.
</p>
<p>GIMPLE and RTL are used to optimize the program.  GIMPLE is used for
target and language independent optimizations (e.g., inlining,
constant propagation, tail call elimination, redundancy elimination,
etc).  Much like GENERIC, GIMPLE is a language independent, tree based
representation.  However, it differs from GENERIC in that the GIMPLE
grammar is more restrictive: expressions contain no more than 3
operands (except function calls), it has no control flow structures
and expressions with side-effects are only allowed on the right hand
side of assignments.  See the chapter describing GENERIC and GIMPLE
for more details.
</p>
<p>This chapter describes the data structures and functions used in the
GIMPLE optimizers (also known as &ldquo;tree optimizers&rdquo; or &ldquo;middle
end&rdquo;).  In particular, it focuses on all the macros, data structures,
functions and programming constructs needed to implement optimization
passes for GIMPLE.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Annotations" accesskey="1">Annotations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Attributes for variables.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#SSA-Operands" accesskey="2">SSA Operands</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">SSA names referenced by GIMPLE statements.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#SSA" accesskey="3">SSA</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Static Single Assignment representation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Alias-analysis" accesskey="4">Alias analysis</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Representing aliased loads and stores.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Memory-model" accesskey="5">Memory model</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Memory model used by the middle-end.
</td></tr>
</table>

<hr>
<a name="Annotations"></a>
<div class="header">
<p>
Next: <a href="#SSA-Operands" accesskey="n" rel="next">SSA Operands</a>, Up: <a href="#Tree-SSA" accesskey="u" rel="up">Tree SSA</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Annotations-1"></a>
<h3 class="section">13.1 Annotations</h3>
<a name="index-annotations"></a>

<p>The optimizers need to associate attributes with variables during the
optimization process.  For instance, we need to know whether a
variable has aliases.  All these attributes are stored in data
structures called annotations which are then linked to the field
<code>ann</code> in <code>struct tree_common</code>.
</p>
<p>Presently, we define annotations for variables (<code>var_ann_t</code>).
Annotations are defined and documented in <samp>tree-flow.h</samp>.
</p>

<hr>
<a name="SSA-Operands"></a>
<div class="header">
<p>
Next: <a href="#SSA" accesskey="n" rel="next">SSA</a>, Previous: <a href="#Annotations" accesskey="p" rel="prev">Annotations</a>, Up: <a href="#Tree-SSA" accesskey="u" rel="up">Tree SSA</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="SSA-Operands-1"></a>
<h3 class="section">13.2 SSA Operands</h3>
<a name="index-operands"></a>
<a name="index-virtual-operands"></a>
<a name="index-real-operands"></a>
<a name="index-update_005fstmt-1"></a>

<p>Almost every GIMPLE statement will contain a reference to a variable
or memory location.  Since statements come in different shapes and
sizes, their operands are going to be located at various spots inside
the statement&rsquo;s tree.  To facilitate access to the statement&rsquo;s
operands, they are organized into lists associated inside each
statement&rsquo;s annotation.  Each element in an operand list is a pointer
to a <code>VAR_DECL</code>, <code>PARM_DECL</code> or <code>SSA_NAME</code> tree node.
This provides a very convenient way of examining and replacing
operands.
</p>
<p>Data flow analysis and optimization is done on all tree nodes
representing variables.  Any node for which <code>SSA_VAR_P</code> returns
nonzero is considered when scanning statement operands.  However, not
all <code>SSA_VAR_P</code> variables are processed in the same way.  For the
purposes of optimization, we need to distinguish between references to
local scalar variables and references to globals, statics, structures,
arrays, aliased variables, etc.  The reason is simple, the compiler
can gather complete data flow information for a local scalar.  On the
other hand, a global variable may be modified by a function call, it
may not be possible to keep track of all the elements of an array or
the fields of a structure, etc.
</p>
<p>The operand scanner gathers two kinds of operands: <em>real</em> and
<em>virtual</em>.  An operand for which <code>is_gimple_reg</code> returns true
is considered real, otherwise it is a virtual operand.  We also
distinguish between uses and definitions.  An operand is used if its
value is loaded by the statement (e.g., the operand at the RHS of an
assignment).  If the statement assigns a new value to the operand, the
operand is considered a definition (e.g., the operand at the LHS of
an assignment).
</p>
<p>Virtual and real operands also have very different data flow
properties.  Real operands are unambiguous references to the
full object that they represent.  For instance, given
</p>
<div class="smallexample">
<pre class="smallexample">{
  int a, b;
  a = b
}
</pre></div>

<p>Since <code>a</code> and <code>b</code> are non-aliased locals, the statement
<code>a = b</code> will have one real definition and one real use because
variable <code>a</code> is completely modified with the contents of
variable <code>b</code>.  Real definition are also known as <em>killing
definitions</em>.  Similarly, the use of <code>b</code> reads all its bits.
</p>
<p>In contrast, virtual operands are used with variables that can have
a partial or ambiguous reference.  This includes structures, arrays,
globals, and aliased variables.  In these cases, we have two types of
definitions.  For globals, structures, and arrays, we can determine from
a statement whether a variable of these types has a killing definition.
If the variable does, then the statement is marked as having a
<em>must definition</em> of that variable.  However, if a statement is only
defining a part of the variable (i.e. a field in a structure), or if we
know that a statement might define the variable but we cannot say for sure,
then we mark that statement as having a <em>may definition</em>.  For
instance, given
</p>
<div class="smallexample">
<pre class="smallexample">{
  int a, b, *p;

  if (&hellip;)
    p = &amp;a;
  else
    p = &amp;b;
  *p = 5;
  return *p;
}
</pre></div>

<p>The assignment <code>*p = 5</code> may be a definition of <code>a</code> or
<code>b</code>.  If we cannot determine statically where <code>p</code> is
pointing to at the time of the store operation, we create virtual
definitions to mark that statement as a potential definition site for
<code>a</code> and <code>b</code>.  Memory loads are similarly marked with virtual
use operands.  Virtual operands are shown in tree dumps right before
the statement that contains them.  To request a tree dump with virtual
operands, use the <samp>-vops</samp> option to <samp>-fdump-tree</samp>:
</p>
<div class="smallexample">
<pre class="smallexample">{
  int a, b, *p;

  if (&hellip;)
    p = &amp;a;
  else
    p = &amp;b;
  # a = VDEF &lt;a&gt;
  # b = VDEF &lt;b&gt;
  *p = 5;

  # VUSE &lt;a&gt;
  # VUSE &lt;b&gt;
  return *p;
}
</pre></div>

<p>Notice that <code>VDEF</code> operands have two copies of the referenced
variable.  This indicates that this is not a killing definition of
that variable.  In this case we refer to it as a <em>may definition</em>
or <em>aliased store</em>.  The presence of the second copy of the
variable in the <code>VDEF</code> operand will become important when the
function is converted into SSA form.  This will be used to link all
the non-killing definitions to prevent optimizations from making
incorrect assumptions about them.
</p>
<p>Operands are updated as soon as the statement is finished via a call
to <code>update_stmt</code>.  If statement elements are changed via
<code>SET_USE</code> or <code>SET_DEF</code>, then no further action is required
(i.e., those macros take care of updating the statement).  If changes
are made by manipulating the statement&rsquo;s tree directly, then a call
must be made to <code>update_stmt</code> when complete.  Calling one of the
<code>bsi_insert</code> routines or <code>bsi_replace</code> performs an implicit
call to <code>update_stmt</code>.
</p>
<a name="Operand-Iterators-And-Access-Routines"></a>
<h4 class="subsection">13.2.1 Operand Iterators And Access Routines</h4>
<a name="index-Operand-Iterators"></a>
<a name="index-Operand-Access-Routines"></a>

<p>Operands are collected by <samp>tree-ssa-operands.c</samp>.  They are stored
inside each statement&rsquo;s annotation and can be accessed through either the
operand iterators or an access routine.
</p>
<p>The following access routines are available for examining operands:
</p>
<ol>
<li> <code>SINGLE_SSA_{USE,DEF,TREE}_OPERAND</code>: These accessors will return
NULL unless there is exactly one operand matching the specified flags.  If
there is exactly one operand, the operand is returned as either a <code>tree</code>,
<code>def_operand_p</code>, or <code>use_operand_p</code>.

<div class="smallexample">
<pre class="smallexample">tree t = SINGLE_SSA_TREE_OPERAND (stmt, flags);
use_operand_p u = SINGLE_SSA_USE_OPERAND (stmt, SSA_ALL_VIRTUAL_USES);
def_operand_p d = SINGLE_SSA_DEF_OPERAND (stmt, SSA_OP_ALL_DEFS);
</pre></div>

</li><li> <code>ZERO_SSA_OPERANDS</code>: This macro returns true if there are no
operands matching the specified flags.

<div class="smallexample">
<pre class="smallexample">if (ZERO_SSA_OPERANDS (stmt, SSA_OP_ALL_VIRTUALS))
  return;
</pre></div>

</li><li> <code>NUM_SSA_OPERANDS</code>: This macro Returns the number of operands
matching &rsquo;flags&rsquo;.  This actually executes a loop to perform the count, so
only use this if it is really needed.

<div class="smallexample">
<pre class="smallexample">int count = NUM_SSA_OPERANDS (stmt, flags)
</pre></div>
</li></ol>


<p>If you wish to iterate over some or all operands, use the
<code>FOR_EACH_SSA_{USE,DEF,TREE}_OPERAND</code> iterator.  For example, to print
all the operands for a statement:
</p>
<div class="smallexample">
<pre class="smallexample">void
print_ops (tree stmt)
{
  ssa_op_iter;
  tree var;

  FOR_EACH_SSA_TREE_OPERAND (var, stmt, iter, SSA_OP_ALL_OPERANDS)
    print_generic_expr (stderr, var, TDF_SLIM);
}
</pre></div>


<p>How to choose the appropriate iterator:
</p>
<ol>
<li> Determine whether you are need to see the operand pointers, or just the
trees, and choose the appropriate macro:

<div class="smallexample">
<pre class="smallexample">Need            Macro:
----            -------
use_operand_p   FOR_EACH_SSA_USE_OPERAND
def_operand_p   FOR_EACH_SSA_DEF_OPERAND
tree            FOR_EACH_SSA_TREE_OPERAND
</pre></div>

</li><li> You need to declare a variable of the type you are interested
in, and an ssa_op_iter structure which serves as the loop controlling
variable.

</li><li> Determine which operands you wish to use, and specify the flags of
those you are interested in.  They are documented in
<samp>tree-ssa-operands.h</samp>:

<div class="smallexample">
<pre class="smallexample">#define SSA_OP_USE              0x01    /* <span class="roman">Real USE operands.</span>  */
#define SSA_OP_DEF              0x02    /* <span class="roman">Real DEF operands.</span>  */
#define SSA_OP_VUSE             0x04    /* <span class="roman">VUSE operands.</span>  */
#define SSA_OP_VMAYUSE          0x08    /* <span class="roman">USE portion of VDEFS.</span>  */
#define SSA_OP_VDEF             0x10    /* <span class="roman">DEF portion of VDEFS.</span>  */

/* <span class="roman">These are commonly grouped operand flags.</span>  */
#define SSA_OP_VIRTUAL_USES     (SSA_OP_VUSE | SSA_OP_VMAYUSE)
#define SSA_OP_VIRTUAL_DEFS     (SSA_OP_VDEF)
#define SSA_OP_ALL_USES         (SSA_OP_VIRTUAL_USES | SSA_OP_USE)
#define SSA_OP_ALL_DEFS         (SSA_OP_VIRTUAL_DEFS | SSA_OP_DEF)
#define SSA_OP_ALL_OPERANDS     (SSA_OP_ALL_USES | SSA_OP_ALL_DEFS)
</pre></div>
</li></ol>

<p>So if you want to look at the use pointers for all the <code>USE</code> and
<code>VUSE</code> operands, you would do something like:
</p>
<div class="smallexample">
<pre class="smallexample">  use_operand_p use_p;
  ssa_op_iter iter;

  FOR_EACH_SSA_USE_OPERAND (use_p, stmt, iter, (SSA_OP_USE | SSA_OP_VUSE))
    {
      process_use_ptr (use_p);
    }
</pre></div>

<p>The <code>TREE</code> macro is basically the same as the <code>USE</code> and
<code>DEF</code> macros, only with the use or def dereferenced via
<code>USE_FROM_PTR (use_p)</code> and <code>DEF_FROM_PTR (def_p)</code>.  Since we
aren&rsquo;t using operand pointers, use and defs flags can be mixed.
</p>
<div class="smallexample">
<pre class="smallexample">  tree var;
  ssa_op_iter iter;

  FOR_EACH_SSA_TREE_OPERAND (var, stmt, iter, SSA_OP_VUSE)
    {
       print_generic_expr (stderr, var, TDF_SLIM);
    }
</pre></div>

<p><code>VDEF</code>s are broken into two flags, one for the
<code>DEF</code> portion (<code>SSA_OP_VDEF</code>) and one for the USE portion
(<code>SSA_OP_VMAYUSE</code>).  If all you want to look at are the
<code>VDEF</code>s together, there is a fourth iterator macro for this,
which returns both a def_operand_p and a use_operand_p for each
<code>VDEF</code> in the statement.  Note that you don&rsquo;t need any flags for
this one.
</p>
<div class="smallexample">
<pre class="smallexample">  use_operand_p use_p;
  def_operand_p def_p;
  ssa_op_iter iter;

  FOR_EACH_SSA_MAYDEF_OPERAND (def_p, use_p, stmt, iter)
    {
      my_code;
    }
</pre></div>

<p>There are many examples in the code as well, as well as the
documentation in <samp>tree-ssa-operands.h</samp>.
</p>
<p>There are also a couple of variants on the stmt iterators regarding PHI
nodes.
</p>
<p><code>FOR_EACH_PHI_ARG</code> Works exactly like
<code>FOR_EACH_SSA_USE_OPERAND</code>, except it works over <code>PHI</code> arguments
instead of statement operands.
</p>
<div class="smallexample">
<pre class="smallexample">/* Look at every virtual PHI use.  */
FOR_EACH_PHI_ARG (use_p, phi_stmt, iter, SSA_OP_VIRTUAL_USES)
{
   my_code;
}

/* Look at every real PHI use.  */
FOR_EACH_PHI_ARG (use_p, phi_stmt, iter, SSA_OP_USES)
  my_code;

/* Look at every PHI use.  */
FOR_EACH_PHI_ARG (use_p, phi_stmt, iter, SSA_OP_ALL_USES)
  my_code;
</pre></div>

<p><code>FOR_EACH_PHI_OR_STMT_{USE,DEF}</code> works exactly like
<code>FOR_EACH_SSA_{USE,DEF}_OPERAND</code>, except it will function on
either a statement or a <code>PHI</code> node.  These should be used when it is
appropriate but they are not quite as efficient as the individual
<code>FOR_EACH_PHI</code> and <code>FOR_EACH_SSA</code> routines.
</p>
<div class="smallexample">
<pre class="smallexample">FOR_EACH_PHI_OR_STMT_USE (use_operand_p, stmt, iter, flags)
  {
     my_code;
  }

FOR_EACH_PHI_OR_STMT_DEF (def_operand_p, phi, iter, flags)
  {
     my_code;
  }
</pre></div>

<a name="Immediate-Uses"></a>
<h4 class="subsection">13.2.2 Immediate Uses</h4>
<a name="index-Immediate-Uses"></a>

<p>Immediate use information is now always available.  Using the immediate use
iterators, you may examine every use of any <code>SSA_NAME</code>. For instance,
to change each use of <code>ssa_var</code> to <code>ssa_var2</code> and call fold_stmt on
each stmt after that is done:
</p>
<div class="smallexample">
<pre class="smallexample">  use_operand_p imm_use_p;
  imm_use_iterator iterator;
  tree ssa_var, stmt;


  FOR_EACH_IMM_USE_STMT (stmt, iterator, ssa_var)
    {
      FOR_EACH_IMM_USE_ON_STMT (imm_use_p, iterator)
        SET_USE (imm_use_p, ssa_var_2);
      fold_stmt (stmt);
    }
</pre></div>

<p>There are 2 iterators which can be used. <code>FOR_EACH_IMM_USE_FAST</code> is
used when the immediate uses are not changed, i.e., you are looking at the
uses, but not setting them.
</p>
<p>If they do get changed, then care must be taken that things are not changed
under the iterators, so use the <code>FOR_EACH_IMM_USE_STMT</code> and
<code>FOR_EACH_IMM_USE_ON_STMT</code> iterators.  They attempt to preserve the
sanity of the use list by moving all the uses for a statement into
a controlled position, and then iterating over those uses.  Then the
optimization can manipulate the stmt when all the uses have been
processed.  This is a little slower than the FAST version since it adds a
placeholder element and must sort through the list a bit for each statement.
This placeholder element must be also be removed if the loop is
terminated early.  The macro <code>BREAK_FROM_IMM_USE_SAFE</code> is provided
to do this :
</p>
<div class="smallexample">
<pre class="smallexample">  FOR_EACH_IMM_USE_STMT (stmt, iterator, ssa_var)
    {
      if (stmt == last_stmt)
        BREAK_FROM_SAFE_IMM_USE (iter);

      FOR_EACH_IMM_USE_ON_STMT (imm_use_p, iterator)
        SET_USE (imm_use_p, ssa_var_2);
      fold_stmt (stmt);
    }
</pre></div>

<p>There are checks in <code>verify_ssa</code> which verify that the immediate use list
is up to date, as well as checking that an optimization didn&rsquo;t break from the
loop without using this macro.  It is safe to simply &rsquo;break&rsquo;; from a
<code>FOR_EACH_IMM_USE_FAST</code> traverse.
</p>
<p>Some useful functions and macros:
</p><ol>
<li> <code>has_zero_uses (ssa_var)</code> : Returns true if there are no uses of
<code>ssa_var</code>.
</li><li> <code>has_single_use (ssa_var)</code> : Returns true if there is only a
single use of <code>ssa_var</code>.
</li><li> <code>single_imm_use (ssa_var, use_operand_p *ptr, tree *stmt)</code> :
Returns true if there is only a single use of <code>ssa_var</code>, and also returns
the use pointer and statement it occurs in, in the second and third parameters.
</li><li> <code>num_imm_uses (ssa_var)</code> : Returns the number of immediate uses of
<code>ssa_var</code>. It is better not to use this if possible since it simply
utilizes a loop to count the uses.
</li><li> <code>PHI_ARG_INDEX_FROM_USE (use_p)</code> : Given a use within a <code>PHI</code>
node, return the index number for the use.  An assert is triggered if the use
isn&rsquo;t located in a <code>PHI</code> node.
</li><li> <code>USE_STMT (use_p)</code> : Return the statement a use occurs in.
</li></ol>

<p>Note that uses are not put into an immediate use list until their statement is
actually inserted into the instruction stream via a <code>bsi_*</code> routine.
</p>
<p>It is also still possible to utilize lazy updating of statements, but this
should be used only when absolutely required.  Both alias analysis and the
dominator optimizations currently do this.
</p>
<p>When lazy updating is being used, the immediate use information is out of date
and cannot be used reliably.  Lazy updating is achieved by simply marking
statements modified via calls to <code>mark_stmt_modified</code> instead of
<code>update_stmt</code>.  When lazy updating is no longer required, all the
modified statements must have <code>update_stmt</code> called in order to bring them
up to date.  This must be done before the optimization is finished, or
<code>verify_ssa</code> will trigger an abort.
</p>
<p>This is done with a simple loop over the instruction stream:
</p><div class="smallexample">
<pre class="smallexample">  block_stmt_iterator bsi;
  basic_block bb;
  FOR_EACH_BB (bb)
    {
      for (bsi = bsi_start (bb); !bsi_end_p (bsi); bsi_next (&amp;bsi))
        update_stmt_if_modified (bsi_stmt (bsi));
    }
</pre></div>

<hr>
<a name="SSA"></a>
<div class="header">
<p>
Next: <a href="#Alias-analysis" accesskey="n" rel="next">Alias analysis</a>, Previous: <a href="#SSA-Operands" accesskey="p" rel="prev">SSA Operands</a>, Up: <a href="#Tree-SSA" accesskey="u" rel="up">Tree SSA</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Static-Single-Assignment"></a>
<h3 class="section">13.3 Static Single Assignment</h3>
<a name="index-SSA"></a>
<a name="index-static-single-assignment"></a>

<p>Most of the tree optimizers rely on the data flow information provided
by the Static Single Assignment (SSA) form.  We implement the SSA form
as described in <cite>R. Cytron, J. Ferrante, B. Rosen, M. Wegman, and
K. Zadeck.  Efficiently Computing Static Single Assignment Form and the
Control Dependence Graph.  ACM Transactions on Programming Languages
and Systems, 13(4):451-490, October 1991</cite>.
</p>
<p>The SSA form is based on the premise that program variables are
assigned in exactly one location in the program.  Multiple assignments
to the same variable create new versions of that variable.  Naturally,
actual programs are seldom in SSA form initially because variables
tend to be assigned multiple times.  The compiler modifies the program
representation so that every time a variable is assigned in the code,
a new version of the variable is created.  Different versions of the
same variable are distinguished by subscripting the variable name with
its version number.  Variables used in the right-hand side of
expressions are renamed so that their version number matches that of
the most recent assignment.
</p>
<p>We represent variable versions using <code>SSA_NAME</code> nodes.  The
renaming process in <samp>tree-ssa.c</samp> wraps every real and
virtual operand with an <code>SSA_NAME</code> node which contains
the version number and the statement that created the
<code>SSA_NAME</code>.  Only definitions and virtual definitions may
create new <code>SSA_NAME</code> nodes.
</p>
<a name="index-PHI-nodes"></a>
<p>Sometimes, flow of control makes it impossible to determine the
most recent version of a variable.  In these cases, the compiler
inserts an artificial definition for that variable called
<em>PHI function</em> or <em>PHI node</em>.  This new definition merges
all the incoming versions of the variable to create a new name
for it.  For instance,
</p>
<div class="smallexample">
<pre class="smallexample">if (&hellip;)
  a_1 = 5;
else if (&hellip;)
  a_2 = 2;
else
  a_3 = 13;

# a_4 = PHI &lt;a_1, a_2, a_3&gt;
return a_4;
</pre></div>

<p>Since it is not possible to determine which of the three branches
will be taken at runtime, we don&rsquo;t know which of <code>a_1</code>,
<code>a_2</code> or <code>a_3</code> to use at the return statement.  So, the
SSA renamer creates a new version <code>a_4</code> which is assigned
the result of &ldquo;merging&rdquo; <code>a_1</code>, <code>a_2</code> and <code>a_3</code>.
Hence, PHI nodes mean &ldquo;one of these operands.  I don&rsquo;t know
which&rdquo;.
</p>
<p>The following macros can be used to examine PHI nodes
</p>
<dl>
<dt><a name="index-PHI_005fRESULT"></a>Macro: <strong>PHI_RESULT</strong> <em>(<var>phi</var>)</em></dt>
<dd><p>Returns the <code>SSA_NAME</code> created by PHI node <var>phi</var> (i.e.,
<var>phi</var>&rsquo;s LHS).
</p></dd></dl>

<dl>
<dt><a name="index-PHI_005fNUM_005fARGS"></a>Macro: <strong>PHI_NUM_ARGS</strong> <em>(<var>phi</var>)</em></dt>
<dd><p>Returns the number of arguments in <var>phi</var>.  This number is exactly
the number of incoming edges to the basic block holding <var>phi</var>.
</p></dd></dl>

<dl>
<dt><a name="index-PHI_005fARG_005fELT"></a>Macro: <strong>PHI_ARG_ELT</strong> <em>(<var>phi</var>, <var>i</var>)</em></dt>
<dd><p>Returns a tuple representing the <var>i</var>th argument of <var>phi</var>.
Each element of this tuple contains an <code>SSA_NAME</code> <var>var</var> and
the incoming edge through which <var>var</var> flows.
</p></dd></dl>

<dl>
<dt><a name="index-PHI_005fARG_005fEDGE"></a>Macro: <strong>PHI_ARG_EDGE</strong> <em>(<var>phi</var>, <var>i</var>)</em></dt>
<dd><p>Returns the incoming edge for the <var>i</var>th argument of <var>phi</var>.
</p></dd></dl>

<dl>
<dt><a name="index-PHI_005fARG_005fDEF"></a>Macro: <strong>PHI_ARG_DEF</strong> <em>(<var>phi</var>, <var>i</var>)</em></dt>
<dd><p>Returns the <code>SSA_NAME</code> for the <var>i</var>th argument of <var>phi</var>.
</p></dd></dl>


<a name="Preserving-the-SSA-form"></a>
<h4 class="subsection">13.3.1 Preserving the SSA form</h4>
<a name="index-update_005fssa"></a>
<a name="index-preserving-SSA-form"></a>
<p>Some optimization passes make changes to the function that
invalidate the SSA property.  This can happen when a pass has
added new symbols or changed the program so that variables that
were previously aliased aren&rsquo;t anymore.  Whenever something like this
happens, the affected symbols must be renamed into SSA form again.
Transformations that emit new code or replicate existing statements
will also need to update the SSA form.
</p>
<p>Since GCC implements two different SSA forms for register and virtual
variables, keeping the SSA form up to date depends on whether you are
updating register or virtual names.  In both cases, the general idea
behind incremental SSA updates is similar: when new SSA names are
created, they typically are meant to replace other existing names in
the program.
</p>
<p>For instance, given the following code:
</p>
<div class="smallexample">
<pre class="smallexample">     1  L0:
     2  x_1 = PHI (0, x_5)
     3  if (x_1 &lt; 10)
     4    if (x_1 &gt; 7)
     5      y_2 = 0
     6    else
     7      y_3 = x_1 + x_7
     8    endif
     9    x_5 = x_1 + 1
     10   goto L0;
     11 endif
</pre></div>

<p>Suppose that we insert new names <code>x_10</code> and <code>x_11</code> (lines
<code>4</code> and <code>8</code>).
</p>
<div class="smallexample">
<pre class="smallexample">     1  L0:
     2  x_1 = PHI (0, x_5)
     3  if (x_1 &lt; 10)
     4    x_10 = &hellip;
     5    if (x_1 &gt; 7)
     6      y_2 = 0
     7    else
     8      x_11 = &hellip;
     9      y_3 = x_1 + x_7
     10   endif
     11   x_5 = x_1 + 1
     12   goto L0;
     13 endif
</pre></div>

<p>We want to replace all the uses of <code>x_1</code> with the new definitions
of <code>x_10</code> and <code>x_11</code>.  Note that the only uses that should
be replaced are those at lines <code>5</code>, <code>9</code> and <code>11</code>.
Also, the use of <code>x_7</code> at line <code>9</code> should <em>not</em> be
replaced (this is why we cannot just mark symbol <code>x</code> for
renaming).
</p>
<p>Additionally, we may need to insert a PHI node at line <code>11</code>
because that is a merge point for <code>x_10</code> and <code>x_11</code>.  So the
use of <code>x_1</code> at line <code>11</code> will be replaced with the new PHI
node.  The insertion of PHI nodes is optional.  They are not strictly
necessary to preserve the SSA form, and depending on what the caller
inserted, they may not even be useful for the optimizers.
</p>
<p>Updating the SSA form is a two step process.  First, the pass has to
identify which names need to be updated and/or which symbols need to
be renamed into SSA form for the first time.  When new names are
introduced to replace existing names in the program, the mapping
between the old and the new names are registered by calling
<code>register_new_name_mapping</code> (note that if your pass creates new
code by duplicating basic blocks, the call to <code>tree_duplicate_bb</code>
will set up the necessary mappings automatically).
</p>
<p>After the replacement mappings have been registered and new symbols
marked for renaming, a call to <code>update_ssa</code> makes the registered
changes.  This can be done with an explicit call or by creating
<code>TODO</code> flags in the <code>tree_opt_pass</code> structure for your pass.
There are several <code>TODO</code> flags that control the behavior of
<code>update_ssa</code>:
</p>
<ul>
<li> <code>TODO_update_ssa</code>.  Update the SSA form inserting PHI nodes
for newly exposed symbols and virtual names marked for updating.
When updating real names, only insert PHI nodes for a real name
<code>O_j</code> in blocks reached by all the new and old definitions for
<code>O_j</code>.  If the iterated dominance frontier for <code>O_j</code>
is not pruned, we may end up inserting PHI nodes in blocks that
have one or more edges with no incoming definition for
<code>O_j</code>.  This would lead to uninitialized warnings for
<code>O_j</code>&rsquo;s symbol.

</li><li> <code>TODO_update_ssa_no_phi</code>.  Update the SSA form without
inserting any new PHI nodes at all.  This is used by passes that
have either inserted all the PHI nodes themselves or passes that
need only to patch use-def and def-def chains for virtuals
(e.g., DCE).


</li><li> <code>TODO_update_ssa_full_phi</code>.  Insert PHI nodes everywhere
they are needed.  No pruning of the IDF is done.  This is used
by passes that need the PHI nodes for <code>O_j</code> even if it
means that some arguments will come from the default definition
of <code>O_j</code>&rsquo;s symbol (e.g., <code>pass_linear_transform</code>).

<p>WARNING: If you need to use this flag, chances are that your
pass may be doing something wrong.  Inserting PHI nodes for an
old name where not all edges carry a new replacement may lead to
silent codegen errors or spurious uninitialized warnings.
</p>
</li><li> <code>TODO_update_ssa_only_virtuals</code>.  Passes that update the
SSA form on their own may want to delegate the updating of
virtual names to the generic updater.  Since FUD chains are
easier to maintain, this simplifies the work they need to do.
NOTE: If this flag is used, any OLD-&gt;NEW mappings for real names
are explicitly destroyed and only the symbols marked for
renaming are processed.
</li></ul>

<a name="Preserving-the-virtual-SSA-form"></a>
<h4 class="subsection">13.3.2 Preserving the virtual SSA form</h4>
<a name="index-preserving-virtual-SSA-form"></a>

<p>The virtual SSA form is harder to preserve than the non-virtual SSA form
mainly because the set of virtual operands for a statement may change at
what some would consider unexpected times.  In general, statement
modifications should be bracketed between calls to
<code>push_stmt_changes</code> and <code>pop_stmt_changes</code>.  For example,
</p>
<div class="smallexample">
<pre class="smallexample">    munge_stmt (tree stmt)
    {
       push_stmt_changes (&amp;stmt);
       &hellip; rewrite STMT &hellip;
       pop_stmt_changes (&amp;stmt);
    }
</pre></div>

<p>The call to <code>push_stmt_changes</code> saves the current state of the
statement operands and the call to <code>pop_stmt_changes</code> compares
the saved state with the current one and does the appropriate symbol
marking for the SSA renamer.
</p>
<p>It is possible to modify several statements at a time, provided that
<code>push_stmt_changes</code> and <code>pop_stmt_changes</code> are called in
LIFO order, as when processing a stack of statements.
</p>
<p>Additionally, if the pass discovers that it did not need to make
changes to the statement after calling <code>push_stmt_changes</code>, it
can simply discard the topmost change buffer by calling
<code>discard_stmt_changes</code>.  This will avoid the expensive operand
re-scan operation and the buffer comparison that determines if symbols
need to be marked for renaming.
</p>
<a name="Examining-SSA_005fNAME-nodes"></a>
<h4 class="subsection">13.3.3 Examining <code>SSA_NAME</code> nodes</h4>
<a name="index-examining-SSA_005fNAMEs"></a>

<p>The following macros can be used to examine <code>SSA_NAME</code> nodes
</p>
<dl>
<dt><a name="index-SSA_005fNAME_005fDEF_005fSTMT"></a>Macro: <strong>SSA_NAME_DEF_STMT</strong> <em>(<var>var</var>)</em></dt>
<dd><p>Returns the statement <var>s</var> that creates the <code>SSA_NAME</code>
<var>var</var>.  If <var>s</var> is an empty statement (i.e., <code>IS_EMPTY_STMT
(<var>s</var>)</code> returns <code>true</code>), it means that the first reference to
this variable is a USE or a VUSE.
</p></dd></dl>

<dl>
<dt><a name="index-SSA_005fNAME_005fVERSION"></a>Macro: <strong>SSA_NAME_VERSION</strong> <em>(<var>var</var>)</em></dt>
<dd><p>Returns the version number of the <code>SSA_NAME</code> object <var>var</var>.
</p></dd></dl>


<a name="Walking-use_002ddef-chains"></a>
<h4 class="subsection">13.3.4 Walking use-def chains</h4>

<dl>
<dt><a name="index-walk_005fuse_005fdef_005fchains"></a>Tree SSA function: <em>void</em> <strong>walk_use_def_chains</strong> <em>(<var>var</var>, <var>fn</var>, <var>data</var>)</em></dt>
<dd>
<p>Walks use-def chains starting at the <code>SSA_NAME</code> node <var>var</var>.
Calls function <var>fn</var> at each reaching definition found.  Function
<var>FN</var> takes three arguments: <var>var</var>, its defining statement
(<var>def_stmt</var>) and a generic pointer to whatever state information
that <var>fn</var> may want to maintain (<var>data</var>).  Function <var>fn</var> is
able to stop the walk by returning <code>true</code>, otherwise in order to
continue the walk, <var>fn</var> should return <code>false</code>.
</p>
<p>Note, that if <var>def_stmt</var> is a <code>PHI</code> node, the semantics are
slightly different.  For each argument <var>arg</var> of the PHI node, this
function will:
</p>
<ol>
<li> Walk the use-def chains for <var>arg</var>.
</li><li> Call <code>FN (<var>arg</var>, <var>phi</var>, <var>data</var>)</code>.
</li></ol>

<p>Note how the first argument to <var>fn</var> is no longer the original
variable <var>var</var>, but the PHI argument currently being examined.
If <var>fn</var> wants to get at <var>var</var>, it should call
<code>PHI_RESULT</code> (<var>phi</var>).
</p></dd></dl>

<a name="Walking-the-dominator-tree"></a>
<h4 class="subsection">13.3.5 Walking the dominator tree</h4>

<dl>
<dt><a name="index-walk_005fdominator_005ftree"></a>Tree SSA function: <em>void</em> <strong>walk_dominator_tree</strong> <em>(<var>walk_data</var>, <var>bb</var>)</em></dt>
<dd>
<p>This function walks the dominator tree for the current CFG calling a
set of callback functions defined in <var>struct dom_walk_data</var> in
<samp>domwalk.h</samp>.  The call back functions you need to define give you
hooks to execute custom code at various points during traversal:
</p>
<ol>
<li> Once to initialize any local data needed while processing
<var>bb</var> and its children.  This local data is pushed into an
internal stack which is automatically pushed and popped as the
walker traverses the dominator tree.

</li><li> Once before traversing all the statements in the <var>bb</var>.

</li><li> Once for every statement inside <var>bb</var>.

</li><li> Once after traversing all the statements and before recursing
into <var>bb</var>&rsquo;s dominator children.

</li><li> It then recurses into all the dominator children of <var>bb</var>.

</li><li> After recursing into all the dominator children of <var>bb</var> it
can, optionally, traverse every statement in <var>bb</var> again
(i.e., repeating steps 2 and 3).

</li><li> Once after walking the statements in <var>bb</var> and <var>bb</var>&rsquo;s
dominator children.  At this stage, the block local data stack
is popped.
</li></ol>
</dd></dl>

<hr>
<a name="Alias-analysis"></a>
<div class="header">
<p>
Next: <a href="#Memory-model" accesskey="n" rel="next">Memory model</a>, Previous: <a href="#SSA" accesskey="p" rel="prev">SSA</a>, Up: <a href="#Tree-SSA" accesskey="u" rel="up">Tree SSA</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Alias-analysis-1"></a>
<h3 class="section">13.4 Alias analysis</h3>
<a name="index-alias"></a>
<a name="index-flow_002dsensitive-alias-analysis"></a>
<a name="index-flow_002dinsensitive-alias-analysis"></a>

<p>Alias analysis in GIMPLE SSA form consists of two pieces.  First
the virtual SSA web ties conflicting memory accesses and provides
a SSA use-def chain and SSA immediate-use chains for walking
possibly dependent memory accesses.  Second an alias-oracle can
be queried to disambiguate explicit and implicit memory references.
</p>
<ol>
<li> Memory SSA form.

<p>All statements that may use memory have exactly one accompanied use of
a virtual SSA name that represents the state of memory at the
given point in the IL.
</p>
<p>All statements that may define memory have exactly one accompanied
definition of a virtual SSA name using the previous state of memory
and defining the new state of memory after the given point in the IL.
</p>
<div class="smallexample">
<pre class="smallexample">int i;
int foo (void)
{
  # .MEM_3 = VDEF &lt;.MEM_2(D)&gt;
  i = 1;
  # VUSE &lt;.MEM_3&gt;
  return i;
}
</pre></div>

<p>The virtual SSA names in this case are <code>.MEM_2(D)</code> and
<code>.MEM_3</code>.  The store to the global variable <code>i</code>
defines <code>.MEM_3</code> invalidating <code>.MEM_2(D)</code>.  The
load from <code>i</code> uses that new state <code>.MEM_3</code>.
</p>
<p>The virtual SSA web serves as constraints to SSA optimizers
preventing illegitimate code-motion and optimization.  It
also provides a way to walk related memory statements.
</p>
</li><li> Points-to and escape analysis.

<p>Points-to analysis builds a set of constraints from the GIMPLE
SSA IL representing all pointer operations and facts we do
or do not know about pointers.  Solving this set of constraints
yields a conservatively correct solution for each pointer
variable in the program (though we are only interested in
SSA name pointers) as to what it may possibly point to.
</p>
<p>This points-to solution for a given SSA name pointer is stored
in the <code>pt_solution</code> sub-structure of the
<code>SSA_NAME_PTR_INFO</code> record.  The following accessor
functions are available:
</p>
<ul>
<li> <code>pt_solution_includes</code>
</li><li> <code>pt_solutions_intersect</code>
</li></ul>

<p>Points-to analysis also computes the solution for two special
set of pointers, <code>ESCAPED</code> and <code>CALLUSED</code>.  Those
represent all memory that has escaped the scope of analysis
or that is used by pure or nested const calls.
</p>
</li><li> Type-based alias analysis

<p>Type-based alias analysis is frontend dependent though generic
support is provided by the middle-end in <code>alias.c</code>.  TBAA
code is used by both tree optimizers and RTL optimizers.
</p>
<p>Every language that wishes to perform language-specific alias analysis
should define a function that computes, given a <code>tree</code>
node, an alias set for the node.  Nodes in different alias sets are not
allowed to alias.  For an example, see the C front-end function
<code>c_get_alias_set</code>.
</p>
</li><li> Tree alias-oracle

<p>The tree alias-oracle provides means to disambiguate two memory
references and memory references against statements.  The following
queries are available:
</p>
<ul>
<li> <code>refs_may_alias_p</code>
</li><li> <code>ref_maybe_used_by_stmt_p</code>
</li><li> <code>stmt_may_clobber_ref_p</code>
</li></ul>

<p>In addition to those two kind of statement walkers are available
walking statements related to a reference ref.
<code>walk_non_aliased_vuses</code> walks over dominating memory defining
statements and calls back if the statement does not clobber ref
providing the non-aliased VUSE.  The walk stops at
the first clobbering statement or if asked to.
<code>walk_aliased_vdefs</code> walks over dominating memory defining
statements and calls back on each statement clobbering ref
providing its aliasing VDEF.  The walk stops if asked to.
</p>
</li></ol>


<hr>
<a name="Memory-model"></a>
<div class="header">
<p>
Previous: <a href="#Alias-analysis" accesskey="p" rel="prev">Alias analysis</a>, Up: <a href="#Tree-SSA" accesskey="u" rel="up">Tree SSA</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Memory-model-1"></a>
<h3 class="section">13.5 Memory model</h3>
<a name="index-memory-model"></a>

<p>The memory model used by the middle-end models that of the C/C++
languages.  The middle-end has the notion of an effective type
of a memory region which is used for type-based alias analysis.
</p>
<p>The following is a refinement of ISO C99 6.5/6, clarifying the block copy case
to follow common sense and extending the concept of a dynamic effective
type to objects with a declared type as required for C++.
</p>
<div class="smallexample">
<pre class="smallexample">The effective type of an object for an access to its stored value is
the declared type of the object or the effective type determined by
a previous store to it.  If a value is stored into an object through
an lvalue having a type that is not a character type, then the
type of the lvalue becomes the effective type of the object for that
access and for subsequent accesses that do not modify the stored value.
If a value is copied into an object using <code>memcpy</code> or <code>memmove</code>,
or is copied as an array of character type, then the effective type
of the modified object for that access and for subsequent accesses that
do not modify the value is undetermined.  For all other accesses to an
object, the effective type of the object is simply the type of the
lvalue used for the access.
</pre></div>



<hr>
<a name="Loop-Analysis-and-Representation"></a>
<div class="header">
<p>
Next: <a href="#Control-Flow" accesskey="n" rel="next">Control Flow</a>, Previous: <a href="#Tree-SSA" accesskey="p" rel="prev">Tree SSA</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Analysis-and-Representation-of-Loops"></a>
<h2 class="chapter">14 Analysis and Representation of Loops</h2>

<p>GCC provides extensive infrastructure for work with natural loops, i.e.,
strongly connected components of CFG with only one entry block.  This
chapter describes representation of loops in GCC, both on GIMPLE and in
RTL, as well as the interfaces to loop-related analyses (induction
variable analysis and number of iterations analysis).
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Loop-representation" accesskey="1">Loop representation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Representation and analysis of loops.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Loop-querying" accesskey="2">Loop querying</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Getting information about loops.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Loop-manipulation" accesskey="3">Loop manipulation</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Loop manipulation functions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#LCSSA" accesskey="4">LCSSA</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Loop-closed SSA form.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Scalar-evolutions" accesskey="5">Scalar evolutions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Induction variables on GIMPLE.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#loop_002div" accesskey="6">loop-iv</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Induction variables on RTL.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Number-of-iterations" accesskey="7">Number of iterations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Number of iterations analysis.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Dependency-analysis" accesskey="8">Dependency analysis</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Data dependency analysis.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Lambda" accesskey="9">Lambda</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Linear loop transformations framework.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Omega">Omega</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">A solver for linear programming problems.
</td></tr>
</table>

<hr>
<a name="Loop-representation"></a>
<div class="header">
<p>
Next: <a href="#Loop-querying" accesskey="n" rel="next">Loop querying</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Loop-representation-1"></a>
<h3 class="section">14.1 Loop representation</h3>
<a name="index-Loop-representation"></a>
<a name="index-Loop-analysis"></a>

<p>This chapter describes the representation of loops in GCC, and functions
that can be used to build, modify and analyze this representation.  Most
of the interfaces and data structures are declared in <samp>cfgloop.h</samp>.
At the moment, loop structures are analyzed and this information is
updated only by the optimization passes that deal with loops, but some
efforts are being made to make it available throughout most of the
optimization passes.
</p>
<p>In general, a natural loop has one entry block (header) and possibly
several back edges (latches) leading to the header from the inside of
the loop.  Loops with several latches may appear if several loops share
a single header, or if there is a branching in the middle of the loop.
The representation of loops in GCC however allows only loops with a
single latch.  During loop analysis, headers of such loops are split and
forwarder blocks are created in order to disambiguate their structures.
Heuristic based on profile information and structure of the induction
variables in the loops is used to determine whether the latches
correspond to sub-loops or to control flow in a single loop.  This means
that the analysis sometimes changes the CFG, and if you run it in the
middle of an optimization pass, you must be able to deal with the new
blocks.  You may avoid CFG changes by passing
<code>LOOPS_MAY_HAVE_MULTIPLE_LATCHES</code> flag to the loop discovery,
note however that most other loop manipulation functions will not work
correctly for loops with multiple latch edges (the functions that only
query membership of blocks to loops and subloop relationships, or
enumerate and test loop exits, can be expected to work).
</p>
<p>Body of the loop is the set of blocks that are dominated by its header,
and reachable from its latch against the direction of edges in CFG.  The
loops are organized in a containment hierarchy (tree) such that all the
loops immediately contained inside loop L are the children of L in the
tree.  This tree is represented by the <code>struct loops</code> structure.
The root of this tree is a fake loop that contains all blocks in the
function.  Each of the loops is represented in a <code>struct loop</code>
structure.  Each loop is assigned an index (<code>num</code> field of the
<code>struct loop</code> structure), and the pointer to the loop is stored in
the corresponding field of the <code>larray</code> vector in the loops
structure.  The indices do not have to be continuous, there may be
empty (<code>NULL</code>) entries in the <code>larray</code> created by deleting
loops.  Also, there is no guarantee on the relative order of a loop
and its subloops in the numbering.  The index of a loop never changes.
</p>
<p>The entries of the <code>larray</code> field should not be accessed directly.
The function <code>get_loop</code> returns the loop description for a loop with
the given index.  <code>number_of_loops</code> function returns number of
loops in the function.  To traverse all loops, use <code>FOR_EACH_LOOP</code>
macro.  The <code>flags</code> argument of the macro is used to determine
the direction of traversal and the set of loops visited.  Each loop is
guaranteed to be visited exactly once, regardless of the changes to the
loop tree, and the loops may be removed during the traversal.  The newly
created loops are never traversed, if they need to be visited, this
must be done separately after their creation.  The <code>FOR_EACH_LOOP</code>
macro allocates temporary variables.  If the <code>FOR_EACH_LOOP</code> loop
were ended using break or goto, they would not be released;
<code>FOR_EACH_LOOP_BREAK</code> macro must be used instead.
</p>
<p>Each basic block contains the reference to the innermost loop it belongs
to (<code>loop_father</code>).  For this reason, it is only possible to have
one <code>struct loops</code> structure initialized at the same time for each
CFG.  The global variable <code>current_loops</code> contains the
<code>struct loops</code> structure.  Many of the loop manipulation functions
assume that dominance information is up-to-date.
</p>
<p>The loops are analyzed through <code>loop_optimizer_init</code> function.  The
argument of this function is a set of flags represented in an integer
bitmask.  These flags specify what other properties of the loop
structures should be calculated/enforced and preserved later:
</p>
<ul>
<li> <code>LOOPS_MAY_HAVE_MULTIPLE_LATCHES</code>: If this flag is set, no
changes to CFG will be performed in the loop analysis, in particular,
loops with multiple latch edges will not be disambiguated.  If a loop
has multiple latches, its latch block is set to NULL.  Most of
the loop manipulation functions will not work for loops in this shape.
No other flags that require CFG changes can be passed to
loop_optimizer_init.
</li><li> <code>LOOPS_HAVE_PREHEADERS</code>: Forwarder blocks are created in such
a way that each loop has only one entry edge, and additionally, the
source block of this entry edge has only one successor.  This creates a
natural place where the code can be moved out of the loop, and ensures
that the entry edge of the loop leads from its immediate super-loop.
</li><li> <code>LOOPS_HAVE_SIMPLE_LATCHES</code>: Forwarder blocks are created to
force the latch block of each loop to have only one successor.  This
ensures that the latch of the loop does not belong to any of its
sub-loops, and makes manipulation with the loops significantly easier.
Most of the loop manipulation functions assume that the loops are in
this shape.  Note that with this flag, the &ldquo;normal&rdquo; loop without any
control flow inside and with one exit consists of two basic blocks.
</li><li> <code>LOOPS_HAVE_MARKED_IRREDUCIBLE_REGIONS</code>: Basic blocks and
edges in the strongly connected components that are not natural loops
(have more than one entry block) are marked with
<code>BB_IRREDUCIBLE_LOOP</code> and <code>EDGE_IRREDUCIBLE_LOOP</code> flags.  The
flag is not set for blocks and edges that belong to natural loops that
are in such an irreducible region (but it is set for the entry and exit
edges of such a loop, if they lead to/from this region).
</li><li> <code>LOOPS_HAVE_RECORDED_EXITS</code>: The lists of exits are recorded
and updated for each loop.  This makes some functions (e.g.,
<code>get_loop_exit_edges</code>) more efficient.  Some functions (e.g.,
<code>single_exit</code>) can be used only if the lists of exits are
recorded.
</li></ul>

<p>These properties may also be computed/enforced later, using functions
<code>create_preheaders</code>, <code>force_single_succ_latches</code>,
<code>mark_irreducible_loops</code> and <code>record_loop_exits</code>.
</p>
<p>The memory occupied by the loops structures should be freed with
<code>loop_optimizer_finalize</code> function.
</p>
<p>The CFG manipulation functions in general do not update loop structures.
Specialized versions that additionally do so are provided for the most
common tasks.  On GIMPLE, <code>cleanup_tree_cfg_loop</code> function can be
used to cleanup CFG while updating the loops structures if
<code>current_loops</code> is set.
</p>
<hr>
<a name="Loop-querying"></a>
<div class="header">
<p>
Next: <a href="#Loop-manipulation" accesskey="n" rel="next">Loop manipulation</a>, Previous: <a href="#Loop-representation" accesskey="p" rel="prev">Loop representation</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Loop-querying-1"></a>
<h3 class="section">14.2 Loop querying</h3>
<a name="index-Loop-querying"></a>

<p>The functions to query the information about loops are declared in
<samp>cfgloop.h</samp>.  Some of the information can be taken directly from
the structures.  <code>loop_father</code> field of each basic block contains
the innermost loop to that the block belongs.  The most useful fields of
loop structure (that are kept up-to-date at all times) are:
</p>
<ul>
<li> <code>header</code>, <code>latch</code>: Header and latch basic blocks of the
loop.
</li><li> <code>num_nodes</code>: Number of basic blocks in the loop (including
the basic blocks of the sub-loops).
</li><li> <code>depth</code>: The depth of the loop in the loops tree, i.e., the
number of super-loops of the loop.
</li><li> <code>outer</code>, <code>inner</code>, <code>next</code>: The super-loop, the first
sub-loop, and the sibling of the loop in the loops tree.
</li></ul>

<p>There are other fields in the loop structures, many of them used only by
some of the passes, or not updated during CFG changes; in general, they
should not be accessed directly.
</p>
<p>The most important functions to query loop structures are:
</p>
<ul>
<li> <code>flow_loops_dump</code>: Dumps the information about loops to a
file.
</li><li> <code>verify_loop_structure</code>: Checks consistency of the loop
structures.
</li><li> <code>loop_latch_edge</code>: Returns the latch edge of a loop.
</li><li> <code>loop_preheader_edge</code>: If loops have preheaders, returns
the preheader edge of a loop.
</li><li> <code>flow_loop_nested_p</code>: Tests whether loop is a sub-loop of
another loop.
</li><li> <code>flow_bb_inside_loop_p</code>: Tests whether a basic block belongs
to a loop (including its sub-loops).
</li><li> <code>find_common_loop</code>: Finds the common super-loop of two loops.
</li><li> <code>superloop_at_depth</code>: Returns the super-loop of a loop with
the given depth.
</li><li> <code>tree_num_loop_insns</code>, <code>num_loop_insns</code>: Estimates the
number of insns in the loop, on GIMPLE and on RTL.
</li><li> <code>loop_exit_edge_p</code>: Tests whether edge is an exit from a
loop.
</li><li> <code>mark_loop_exit_edges</code>: Marks all exit edges of all loops
with <code>EDGE_LOOP_EXIT</code> flag.
</li><li> <code>get_loop_body</code>, <code>get_loop_body_in_dom_order</code>,
<code>get_loop_body_in_bfs_order</code>: Enumerates the basic blocks in the
loop in depth-first search order in reversed CFG, ordered by dominance
relation, and breath-first search order, respectively.
</li><li> <code>single_exit</code>: Returns the single exit edge of the loop, or
<code>NULL</code> if the loop has more than one exit.  You can only use this
function if LOOPS_HAVE_MARKED_SINGLE_EXITS property is used.
</li><li> <code>get_loop_exit_edges</code>: Enumerates the exit edges of a loop.
</li><li> <code>just_once_each_iteration_p</code>: Returns true if the basic block
is executed exactly once during each iteration of a loop (that is, it
does not belong to a sub-loop, and it dominates the latch of the loop).
</li></ul>

<hr>
<a name="Loop-manipulation"></a>
<div class="header">
<p>
Next: <a href="#LCSSA" accesskey="n" rel="next">LCSSA</a>, Previous: <a href="#Loop-querying" accesskey="p" rel="prev">Loop querying</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Loop-manipulation-1"></a>
<h3 class="section">14.3 Loop manipulation</h3>
<a name="index-Loop-manipulation"></a>

<p>The loops tree can be manipulated using the following functions:
</p>
<ul>
<li> <code>flow_loop_tree_node_add</code>: Adds a node to the tree.
</li><li> <code>flow_loop_tree_node_remove</code>: Removes a node from the tree.
</li><li> <code>add_bb_to_loop</code>: Adds a basic block to a loop.
</li><li> <code>remove_bb_from_loops</code>: Removes a basic block from loops.
</li></ul>

<p>Most low-level CFG functions update loops automatically.  The following
functions handle some more complicated cases of CFG manipulations:
</p>
<ul>
<li> <code>remove_path</code>: Removes an edge and all blocks it dominates.
</li><li> <code>split_loop_exit_edge</code>: Splits exit edge of the loop,
ensuring that PHI node arguments remain in the loop (this ensures that
loop-closed SSA form is preserved).  Only useful on GIMPLE.
</li></ul>

<p>Finally, there are some higher-level loop transformations implemented.
While some of them are written so that they should work on non-innermost
loops, they are mostly untested in that case, and at the moment, they
are only reliable for the innermost loops:
</p>
<ul>
<li> <code>create_iv</code>: Creates a new induction variable.  Only works on
GIMPLE.  <code>standard_iv_increment_position</code> can be used to find a
suitable place for the iv increment.
</li><li> <code>duplicate_loop_to_header_edge</code>,
<code>tree_duplicate_loop_to_header_edge</code>: These functions (on RTL and
on GIMPLE) duplicate the body of the loop prescribed number of times on
one of the edges entering loop header, thus performing either loop
unrolling or loop peeling.  <code>can_duplicate_loop_p</code>
(<code>can_unroll_loop_p</code> on GIMPLE) must be true for the duplicated
loop.
</li><li> <code>loop_version</code>, <code>tree_ssa_loop_version</code>: These function
create a copy of a loop, and a branch before them that selects one of
them depending on the prescribed condition.  This is useful for
optimizations that need to verify some assumptions in runtime (one of
the copies of the loop is usually left unchanged, while the other one is
transformed in some way).
</li><li> <code>tree_unroll_loop</code>: Unrolls the loop, including peeling the
extra iterations to make the number of iterations divisible by unroll
factor, updating the exit condition, and removing the exits that now
cannot be taken.  Works only on GIMPLE.
</li></ul>

<hr>
<a name="LCSSA"></a>
<div class="header">
<p>
Next: <a href="#Scalar-evolutions" accesskey="n" rel="next">Scalar evolutions</a>, Previous: <a href="#Loop-manipulation" accesskey="p" rel="prev">Loop manipulation</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Loop_002dclosed-SSA-form"></a>
<h3 class="section">14.4 Loop-closed SSA form</h3>
<a name="index-LCSSA"></a>
<a name="index-Loop_002dclosed-SSA-form"></a>

<p>Throughout the loop optimizations on tree level, one extra condition is
enforced on the SSA form:  No SSA name is used outside of the loop in
that it is defined.  The SSA form satisfying this condition is called
&ldquo;loop-closed SSA form&rdquo; &ndash; LCSSA.  To enforce LCSSA, PHI nodes must be
created at the exits of the loops for the SSA names that are used
outside of them.  Only the real operands (not virtual SSA names) are
held in LCSSA, in order to save memory.
</p>
<p>There are various benefits of LCSSA:
</p>
<ul>
<li> Many optimizations (value range analysis, final value
replacement) are interested in the values that are defined in the loop
and used outside of it, i.e., exactly those for that we create new PHI
nodes.
</li><li> In induction variable analysis, it is not necessary to specify the
loop in that the analysis should be performed &ndash; the scalar evolution
analysis always returns the results with respect to the loop in that the
SSA name is defined.
</li><li> It makes updating of SSA form during loop transformations simpler.
Without LCSSA, operations like loop unrolling may force creation of PHI
nodes arbitrarily far from the loop, while in LCSSA, the SSA form can be
updated locally.  However, since we only keep real operands in LCSSA, we
cannot use this advantage (we could have local updating of real
operands, but it is not much more efficient than to use generic SSA form
updating for it as well; the amount of changes to SSA is the same).
</li></ul>

<p>However, it also means LCSSA must be updated.  This is usually
straightforward, unless you create a new value in loop and use it
outside, or unless you manipulate loop exit edges (functions are
provided to make these manipulations simple).
<code>rewrite_into_loop_closed_ssa</code> is used to rewrite SSA form to
LCSSA, and <code>verify_loop_closed_ssa</code> to check that the invariant of
LCSSA is preserved.
</p>
<hr>
<a name="Scalar-evolutions"></a>
<div class="header">
<p>
Next: <a href="#loop_002div" accesskey="n" rel="next">loop-iv</a>, Previous: <a href="#LCSSA" accesskey="p" rel="prev">LCSSA</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Scalar-evolutions-1"></a>
<h3 class="section">14.5 Scalar evolutions</h3>
<a name="index-Scalar-evolutions"></a>
<a name="index-IV-analysis-on-GIMPLE"></a>

<p>Scalar evolutions (SCEV) are used to represent results of induction
variable analysis on GIMPLE.  They enable us to represent variables with
complicated behavior in a simple and consistent way (we only use it to
express values of polynomial induction variables, but it is possible to
extend it).  The interfaces to SCEV analysis are declared in
<samp>tree-scalar-evolution.h</samp>.  To use scalar evolutions analysis,
<code>scev_initialize</code> must be used.  To stop using SCEV,
<code>scev_finalize</code> should be used.  SCEV analysis caches results in
order to save time and memory.  This cache however is made invalid by
most of the loop transformations, including removal of code.  If such a
transformation is performed, <code>scev_reset</code> must be called to clean
the caches.
</p>
<p>Given an SSA name, its behavior in loops can be analyzed using the
<code>analyze_scalar_evolution</code> function.  The returned SCEV however
does not have to be fully analyzed and it may contain references to
other SSA names defined in the loop.  To resolve these (potentially
recursive) references, <code>instantiate_parameters</code> or
<code>resolve_mixers</code> functions must be used.
<code>instantiate_parameters</code> is useful when you use the results of SCEV
only for some analysis, and when you work with whole nest of loops at
once.  It will try replacing all SSA names by their SCEV in all loops,
including the super-loops of the current loop, thus providing a complete
information about the behavior of the variable in the loop nest.
<code>resolve_mixers</code> is useful if you work with only one loop at a
time, and if you possibly need to create code based on the value of the
induction variable.  It will only resolve the SSA names defined in the
current loop, leaving the SSA names defined outside unchanged, even if
their evolution in the outer loops is known.
</p>
<p>The SCEV is a normal tree expression, except for the fact that it may
contain several special tree nodes.  One of them is
<code>SCEV_NOT_KNOWN</code>, used for SSA names whose value cannot be
expressed.  The other one is <code>POLYNOMIAL_CHREC</code>.  Polynomial chrec
has three arguments &ndash; base, step and loop (both base and step may
contain further polynomial chrecs).  Type of the expression and of base
and step must be the same.  A variable has evolution
<code>POLYNOMIAL_CHREC(base, step, loop)</code> if it is (in the specified
loop) equivalent to <code>x_1</code> in the following example
</p>
<div class="smallexample">
<pre class="smallexample">while (&hellip;)
  {
    x_1 = phi (base, x_2);
    x_2 = x_1 + step;
  }
</pre></div>

<p>Note that this includes the language restrictions on the operations.
For example, if we compile C code and <code>x</code> has signed type, then the
overflow in addition would cause undefined behavior, and we may assume
that this does not happen.  Hence, the value with this SCEV cannot
overflow (which restricts the number of iterations of such a loop).
</p>
<p>In many cases, one wants to restrict the attention just to affine
induction variables.  In this case, the extra expressive power of SCEV
is not useful, and may complicate the optimizations.  In this case,
<code>simple_iv</code> function may be used to analyze a value &ndash; the result
is a loop-invariant base and step.
</p>
<hr>
<a name="loop_002div"></a>
<div class="header">
<p>
Next: <a href="#Number-of-iterations" accesskey="n" rel="next">Number of iterations</a>, Previous: <a href="#Scalar-evolutions" accesskey="p" rel="prev">Scalar evolutions</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="IV-analysis-on-RTL"></a>
<h3 class="section">14.6 IV analysis on RTL</h3>
<a name="index-IV-analysis-on-RTL"></a>

<p>The induction variable on RTL is simple and only allows analysis of
affine induction variables, and only in one loop at once.  The interface
is declared in <samp>cfgloop.h</samp>.  Before analyzing induction variables
in a loop L, <code>iv_analysis_loop_init</code> function must be called on L.
After the analysis (possibly calling <code>iv_analysis_loop_init</code> for
several loops) is finished, <code>iv_analysis_done</code> should be called.
The following functions can be used to access the results of the
analysis:
</p>
<ul>
<li> <code>iv_analyze</code>: Analyzes a single register used in the given
insn.  If no use of the register in this insn is found, the following
insns are scanned, so that this function can be called on the insn
returned by get_condition.
</li><li> <code>iv_analyze_result</code>: Analyzes result of the assignment in the
given insn.
</li><li> <code>iv_analyze_expr</code>: Analyzes a more complicated expression.
All its operands are analyzed by <code>iv_analyze</code>, and hence they must
be used in the specified insn or one of the following insns.
</li></ul>

<p>The description of the induction variable is provided in <code>struct
rtx_iv</code>.  In order to handle subregs, the representation is a bit
complicated; if the value of the <code>extend</code> field is not
<code>UNKNOWN</code>, the value of the induction variable in the i-th
iteration is
</p>
<div class="smallexample">
<pre class="smallexample">delta + mult * extend_{extend_mode} (subreg_{mode} (base + i * step)),
</pre></div>

<p>with the following exception:  if <code>first_special</code> is true, then the
value in the first iteration (when <code>i</code> is zero) is <code>delta +
mult * base</code>.  However, if <code>extend</code> is equal to <code>UNKNOWN</code>,
then <code>first_special</code> must be false, <code>delta</code> 0, <code>mult</code> 1
and the value in the i-th iteration is
</p>
<div class="smallexample">
<pre class="smallexample">subreg_{mode} (base + i * step)
</pre></div>

<p>The function <code>get_iv_value</code> can be used to perform these
calculations.
</p>
<hr>
<a name="Number-of-iterations"></a>
<div class="header">
<p>
Next: <a href="#Dependency-analysis" accesskey="n" rel="next">Dependency analysis</a>, Previous: <a href="#loop_002div" accesskey="p" rel="prev">loop-iv</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Number-of-iterations-analysis"></a>
<h3 class="section">14.7 Number of iterations analysis</h3>
<a name="index-Number-of-iterations-analysis"></a>

<p>Both on GIMPLE and on RTL, there are functions available to determine
the number of iterations of a loop, with a similar interface.  The
number of iterations of a loop in GCC is defined as the number of
executions of the loop latch.  In many cases, it is not possible to
determine the number of iterations unconditionally &ndash; the determined
number is correct only if some assumptions are satisfied.  The analysis
tries to verify these conditions using the information contained in the
program; if it fails, the conditions are returned together with the
result.  The following information and conditions are provided by the
analysis:
</p>
<ul>
<li> <code>assumptions</code>: If this condition is false, the rest of
the information is invalid.
</li><li> <code>noloop_assumptions</code> on RTL, <code>may_be_zero</code> on GIMPLE: If
this condition is true, the loop exits in the first iteration.
</li><li> <code>infinite</code>: If this condition is true, the loop is infinite.
This condition is only available on RTL.  On GIMPLE, conditions for
finiteness of the loop are included in <code>assumptions</code>.
</li><li> <code>niter_expr</code> on RTL, <code>niter</code> on GIMPLE: The expression
that gives number of iterations.  The number of iterations is defined as
the number of executions of the loop latch.
</li></ul>

<p>Both on GIMPLE and on RTL, it necessary for the induction variable
analysis framework to be initialized (SCEV on GIMPLE, loop-iv on RTL).
On GIMPLE, the results are stored to <code>struct tree_niter_desc</code>
structure.  Number of iterations before the loop is exited through a
given exit can be determined using <code>number_of_iterations_exit</code>
function.  On RTL, the results are returned in <code>struct niter_desc</code>
structure.  The corresponding function is named
<code>check_simple_exit</code>.  There are also functions that pass through
all the exits of a loop and try to find one with easy to determine
number of iterations &ndash; <code>find_loop_niter</code> on GIMPLE and
<code>find_simple_exit</code> on RTL.  Finally, there are functions that
provide the same information, but additionally cache it, so that
repeated calls to number of iterations are not so costly &ndash;
<code>number_of_latch_executions</code> on GIMPLE and <code>get_simple_loop_desc</code>
on RTL.
</p>
<p>Note that some of these functions may behave slightly differently than
others &ndash; some of them return only the expression for the number of
iterations, and fail if there are some assumptions.  The function
<code>number_of_latch_executions</code> works only for single-exit loops.
The function <code>number_of_cond_exit_executions</code> can be used to
determine number of executions of the exit condition of a single-exit
loop (i.e., the <code>number_of_latch_executions</code> increased by one).
</p>
<hr>
<a name="Dependency-analysis"></a>
<div class="header">
<p>
Next: <a href="#Lambda" accesskey="n" rel="next">Lambda</a>, Previous: <a href="#Number-of-iterations" accesskey="p" rel="prev">Number of iterations</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Data-Dependency-Analysis"></a>
<h3 class="section">14.8 Data Dependency Analysis</h3>
<a name="index-Data-Dependency-Analysis"></a>

<p>The code for the data dependence analysis can be found in
<samp>tree-data-ref.c</samp> and its interface and data structures are
described in <samp>tree-data-ref.h</samp>.  The function that computes the
data dependences for all the array and pointer references for a given
loop is <code>compute_data_dependences_for_loop</code>.  This function is
currently used by the linear loop transform and the vectorization
passes.  Before calling this function, one has to allocate two vectors:
a first vector will contain the set of data references that are
contained in the analyzed loop body, and the second vector will contain
the dependence relations between the data references.  Thus if the
vector of data references is of size <code>n</code>, the vector containing the
dependence relations will contain <code>n*n</code> elements.  However if the
analyzed loop contains side effects, such as calls that potentially can
interfere with the data references in the current analyzed loop, the
analysis stops while scanning the loop body for data references, and
inserts a single <code>chrec_dont_know</code> in the dependence relation
array.
</p>
<p>The data references are discovered in a particular order during the
scanning of the loop body: the loop body is analyzed in execution order,
and the data references of each statement are pushed at the end of the
data reference array.  Two data references syntactically occur in the
program in the same order as in the array of data references.  This
syntactic order is important in some classical data dependence tests,
and mapping this order to the elements of this array avoids costly
queries to the loop body representation.
</p>
<p>Three types of data references are currently handled: ARRAY_REF,
INDIRECT_REF and COMPONENT_REF. The data structure for the data reference
is <code>data_reference</code>, where <code>data_reference_p</code> is a name of a
pointer to the data reference structure. The structure contains the
following elements:
</p>
<ul>
<li> <code>base_object_info</code>: Provides information about the base object
of the data reference and its access functions. These access functions
represent the evolution of the data reference in the loop relative to
its base, in keeping with the classical meaning of the data reference
access function for the support of arrays. For example, for a reference
<code>a.b[i][j]</code>, the base object is <code>a.b</code> and the access functions,
one for each array subscript, are:
<code>{i_init, + i_step}_1, {j_init, +, j_step}_2</code>.

</li><li> <code>first_location_in_loop</code>: Provides information about the first
location accessed by the data reference in the loop and about the access
function used to represent evolution relative to this location. This data
is used to support pointers, and is not used for arrays (for which we
have base objects). Pointer accesses are represented as a one-dimensional
access that starts from the first location accessed in the loop. For
example:

<div class="smallexample">
<pre class="smallexample">      for1 i
         for2 j
          *((int *)p + i + j) = a[i][j];
</pre></div>

<p>The access function of the pointer access is <code>{0, + 4B}_for2</code>
relative to <code>p + i</code>. The access functions of the array are
<code>{i_init, + i_step}_for1</code> and <code>{j_init, +, j_step}_for2</code>
relative to <code>a</code>.
</p>
<p>Usually, the object the pointer refers to is either unknown, or we can&rsquo;t
prove that the access is confined to the boundaries of a certain object.
</p>
<p>Two data references can be compared only if at least one of these two
representations has all its fields filled for both data references.
</p>
<p>The current strategy for data dependence tests is as follows:
If both <code>a</code> and <code>b</code> are represented as arrays, compare
<code>a.base_object</code> and <code>b.base_object</code>;
if they are equal, apply dependence tests (use access functions based on
base_objects).
Else if both <code>a</code> and <code>b</code> are represented as pointers, compare
<code>a.first_location</code> and <code>b.first_location</code>;
if they are equal, apply dependence tests (use access functions based on
first location).
However, if <code>a</code> and <code>b</code> are represented differently, only try
to prove that the bases are definitely different.
</p>
</li><li> Aliasing information.
</li><li> Alignment information.
</li></ul>

<p>The structure describing the relation between two data references is
<code>data_dependence_relation</code> and the shorter name for a pointer to
such a structure is <code>ddr_p</code>.  This structure contains:
</p>
<ul>
<li> a pointer to each data reference,
</li><li> a tree node <code>are_dependent</code> that is set to <code>chrec_known</code>
if the analysis has proved that there is no dependence between these two
data references, <code>chrec_dont_know</code> if the analysis was not able to
determine any useful result and potentially there could exist a
dependence between these data references, and <code>are_dependent</code> is
set to <code>NULL_TREE</code> if there exist a dependence relation between the
data references, and the description of this dependence relation is
given in the <code>subscripts</code>, <code>dir_vects</code>, and <code>dist_vects</code>
arrays,
</li><li> a boolean that determines whether the dependence relation can be
represented by a classical distance vector,
</li><li> an array <code>subscripts</code> that contains a description of each
subscript of the data references.  Given two array accesses a
subscript is the tuple composed of the access functions for a given
dimension.  For example, given <code>A[f1][f2][f3]</code> and
<code>B[g1][g2][g3]</code>, there are three subscripts: <code>(f1, g1), (f2,
g2), (f3, g3)</code>.
</li><li> two arrays <code>dir_vects</code> and <code>dist_vects</code> that contain
classical representations of the data dependences under the form of
direction and distance dependence vectors,
</li><li> an array of loops <code>loop_nest</code> that contains the loops to
which the distance and direction vectors refer to.
</li></ul>

<p>Several functions for pretty printing the information extracted by the
data dependence analysis are available: <code>dump_ddrs</code> prints with a
maximum verbosity the details of a data dependence relations array,
<code>dump_dist_dir_vectors</code> prints only the classical distance and
direction vectors for a data dependence relations array, and
<code>dump_data_references</code> prints the details of the data references
contained in a data reference array.
</p>
<hr>
<a name="Lambda"></a>
<div class="header">
<p>
Next: <a href="#Omega" accesskey="n" rel="next">Omega</a>, Previous: <a href="#Dependency-analysis" accesskey="p" rel="prev">Dependency analysis</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Linear-loop-transformations-framework"></a>
<h3 class="section">14.9 Linear loop transformations framework</h3>
<a name="index-Linear-loop-transformations-framework"></a>

<p>Lambda is a framework that allows transformations of loops using
non-singular matrix based transformations of the iteration space and
loop bounds. This allows compositions of skewing, scaling, interchange,
and reversal transformations.  These transformations are often used to
improve cache behavior or remove inner loop dependencies to allow
parallelization and vectorization to take place.
</p>
<p>To perform these transformations, Lambda requires that the loopnest be
converted into an internal form that can be matrix transformed easily.
To do this conversion, the function
<code>gcc_loopnest_to_lambda_loopnest</code> is provided.  If the loop cannot
be transformed using lambda, this function will return NULL.
</p>
<p>Once a <code>lambda_loopnest</code> is obtained from the conversion function,
it can be transformed by using <code>lambda_loopnest_transform</code>, which
takes a transformation matrix to apply.  Note that it is up to the
caller to verify that the transformation matrix is legal to apply to the
loop (dependence respecting, etc).  Lambda simply applies whatever
matrix it is told to provide.  It can be extended to make legal matrices
out of any non-singular matrix, but this is not currently implemented.
Legality of a matrix for a given loopnest can be verified using
<code>lambda_transform_legal_p</code>.
</p>
<p>Given a transformed loopnest, conversion back into gcc IR is done by
<code>lambda_loopnest_to_gcc_loopnest</code>.  This function will modify the
loops so that they match the transformed loopnest.
</p>

<hr>
<a name="Omega"></a>
<div class="header">
<p>
Previous: <a href="#Lambda" accesskey="p" rel="prev">Lambda</a>, Up: <a href="#Loop-Analysis-and-Representation" accesskey="u" rel="up">Loop Analysis and Representation</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Omega-a-solver-for-linear-programming-problems"></a>
<h3 class="section">14.10 Omega a solver for linear programming problems</h3>
<a name="index-Omega-a-solver-for-linear-programming-problems"></a>

<p>The data dependence analysis contains several solvers triggered
sequentially from the less complex ones to the more sophisticated.
For ensuring the consistency of the results of these solvers, a data
dependence check pass has been implemented based on two different
solvers.  The second method that has been integrated to GCC is based
on the Omega dependence solver, written in the 1990&rsquo;s by William Pugh
and David Wonnacott.  Data dependence tests can be formulated using a
subset of the Presburger arithmetics that can be translated to linear
constraint systems.  These linear constraint systems can then be
solved using the Omega solver.
</p>
<p>The Omega solver is using Fourier-Motzkin&rsquo;s algorithm for variable
elimination: a linear constraint system containing <code>n</code> variables
is reduced to a linear constraint system with <code>n-1</code> variables.
The Omega solver can also be used for solving other problems that can
be expressed under the form of a system of linear equalities and
inequalities.  The Omega solver is known to have an exponential worst
case, also known under the name of &ldquo;omega nightmare&rdquo; in the
literature, but in practice, the omega test is known to be efficient
for the common data dependence tests.
</p>
<p>The interface used by the Omega solver for describing the linear
programming problems is described in <samp>omega.h</samp>, and the solver is
<code>omega_solve_problem</code>.
</p>

<hr>
<a name="Control-Flow"></a>
<div class="header">
<p>
Next: <a href="#Machine-Desc" accesskey="n" rel="next">Machine Desc</a>, Previous: <a href="#Loop-Analysis-and-Representation" accesskey="p" rel="prev">Loop Analysis and Representation</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Control-Flow-Graph"></a>
<h2 class="chapter">15 Control Flow Graph</h2>
<a name="index-CFG_002c-Control-Flow-Graph"></a>
<a name="index-basic_002dblock_002eh"></a>

<p>A control flow graph (CFG) is a data structure built on top of the
intermediate code representation (the RTL or <code>GIMPLE</code> instruction
stream) abstracting the control flow behavior of a function that is
being compiled.  The CFG is a directed graph where the vertices
represent basic blocks and edges represent possible transfer of
control flow from one basic block to another.  The data structures
used to represent the control flow graph are defined in
<samp>basic-block.h</samp>.
</p>
<p>In GCC, the representation of control flow is maintained throughout
the compilation process, from constructing the CFG early in 
<code>pass_build_cfg</code> to <code>pass_free_cfg</code> (see <samp>passes.c</samp>).
The CFG takes various different modes and may undergo extensive
manipulations, but the graph is always valid between its construction
and its release.  This way, transfer of information such as data flow,
a measured profile, or the loop tree, can be propagated through the
passes pipeline, and even from <code>GIMPLE</code> to <code>RTL</code>.
</p>
<p>Often the CFG may be better viewed as integral part of instruction
chain, than structure built on the top of it.  Updating the compiler&rsquo;s
intermediate representation for instructions can not be easily done
without proper maintenance of the CFG simultaneously.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Basic-Blocks" accesskey="1">Basic Blocks</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The definition and representation of basic blocks.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Edges" accesskey="2">Edges</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Types of edges and their representation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Profile-information" accesskey="3">Profile information</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Representation of frequencies and probabilities.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Maintaining-the-CFG" accesskey="4">Maintaining the CFG</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Keeping the control flow graph and up to date.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Liveness-information" accesskey="5">Liveness information</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Using and maintaining liveness information.
</td></tr>
</table>


<hr>
<a name="Basic-Blocks"></a>
<div class="header">
<p>
Next: <a href="#Edges" accesskey="n" rel="next">Edges</a>, Up: <a href="#Control-Flow" accesskey="u" rel="up">Control Flow</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Basic-Blocks-1"></a>
<h3 class="section">15.1 Basic Blocks</h3>

<a name="index-basic-block"></a>
<a name="index-basic_005fblock"></a>
<p>A basic block is a straight-line sequence of code with only one entry
point and only one exit.  In GCC, basic blocks are represented using
the <code>basic_block</code> data type.
</p>
<a name="index-ENTRY_005fBLOCK_005fPTR_002c-EXIT_005fBLOCK_005fPTR"></a>
<p>Special basic blocks represent possible entry and exit points of a
function.  These blocks are called <code>ENTRY_BLOCK_PTR</code> and
<code>EXIT_BLOCK_PTR</code>.  These blocks do not contain any code.
</p>
<a name="index-BASIC_005fBLOCK"></a>
<p>The <code>BASIC_BLOCK</code> array contains all basic blocks in an
unspecified order.  Each <code>basic_block</code> structure has a field
that holds a unique integer identifier <code>index</code> that is the
index of the block in the <code>BASIC_BLOCK</code> array.
The total number of basic blocks in the function is
<code>n_basic_blocks</code>.  Both the basic block indices and
the total number of basic blocks may vary during the compilation
process, as passes reorder, create, duplicate, and destroy basic
blocks.  The index for any block should never be greater than
<code>last_basic_block</code>.  The indices 0 and 1 are special codes
reserved for <code>ENTRY_BLOCK</code> and <code>EXIT_BLOCK</code>, the
indices of <code>ENTRY_BLOCK_PTR</code> and <code>EXIT_BLOCK_PTR</code>.
</p>
<a name="index-next_005fbb_002c-prev_005fbb_002c-FOR_005fEACH_005fBB_002c-FOR_005fALL_005fBB"></a>
<p>Two pointer members of the <code>basic_block</code> structure are the
pointers <code>next_bb</code> and <code>prev_bb</code>.  These are used to keep
doubly linked chain of basic blocks in the same order as the
underlying instruction stream.  The chain of basic blocks is updated
transparently by the provided API for manipulating the CFG.  The macro
<code>FOR_EACH_BB</code> can be used to visit all the basic blocks in
lexicographical order, except <code>ENTRY_BLOCK</code> and <code>EXIT_BLOCK</code>.
The macro <code>FOR_ALL_BB</code> also visits all basic blocks in
lexicographical order, including <code>ENTRY_BLOCK</code> and <code>EXIT_BLOCK</code>.
</p>
<a name="index-post_005forder_005fcompute_002c-inverted_005fpost_005forder_005fcompute_002c-walk_005fdominator_005ftree"></a>
<p>The functions <code>post_order_compute</code> and <code>inverted_post_order_compute</code>
can be used to compute topological orders of the CFG.  The orders are
stored as vectors of basic block indices.  The <code>BASIC_BLOCK</code> array
can be used to iterate each basic block by index.
Dominator traversals are also possible using
<code>walk_dominator_tree</code>.  Given two basic blocks A and B, block A
dominates block B if A is <em>always</em> executed before B.
</p>
<p>Each <code>basic_block</code> also contains pointers to the first
instruction (the <em>head</em>) and the last instruction (the <em>tail</em>)
or <em>end</em> of the instruction stream contained in a basic block.  In
fact, since the <code>basic_block</code> data type is used to represent
blocks in both major intermediate representations of GCC (<code>GIMPLE</code>
and RTL), there are pointers to the head and end of a basic block for
both representations, stored in intermediate representation specific
data in the <code>il</code> field of <code>struct basic_block_def</code>.
</p>
<a name="index-CODE_005fLABEL"></a>
<a name="index-NOTE_005fINSN_005fBASIC_005fBLOCK"></a>
<p>For RTL, these pointers are <code>BB_HEAD</code> and <code>BB_END</code>.
</p>
<a name="index-insn-notes_002c-notes"></a>
<a name="index-NOTE_005fINSN_005fBASIC_005fBLOCK-1"></a>
<p>In the RTL representation of a function, the instruction stream
contains not only the &ldquo;real&rdquo; instructions, but also <em>notes</em>
or <em>insn notes</em> (to distinguish them from <em>reg notes</em>).
Any function that moves or duplicates the basic blocks needs
to take care of updating of these notes.  Many of these notes expect
that the instruction stream consists of linear regions, so updating
can sometimes be tedious.  All types of insn notes are defined
in <samp>insn-notes.def</samp>.
</p>
<p>In the RTL function representation, the instructions contained in a
basic block always follow a <code>NOTE_INSN_BASIC_BLOCK</code>, but zero
or more <code>CODE_LABEL</code> nodes can precede the block note.
A basic block ends with a control flow instruction or with the last
instruction before the next <code>CODE_LABEL</code> or
<code>NOTE_INSN_BASIC_BLOCK</code>.
By definition, a <code>CODE_LABEL</code> cannot appear in the middle of
the instruction stream of a basic block.
</p>
<a name="index-can_005ffallthru"></a>
<a name="index-table-jump"></a>
<p>In addition to notes, the jump table vectors are also represented as
&ldquo;pseudo-instructions&rdquo; inside the insn stream.  These vectors never
appear in the basic block and should always be placed just after the
table jump instructions referencing them.  After removing the
table-jump it is often difficult to eliminate the code computing the
address and referencing the vector, so cleaning up these vectors is
postponed until after liveness analysis.   Thus the jump table vectors
may appear in the insn stream unreferenced and without any purpose.
Before any edge is made <em>fall-thru</em>, the existence of such
construct in the way needs to be checked by calling
<code>can_fallthru</code> function.
</p>
<a name="index-GIMPLE-statement-iterators"></a>
<p>For the <code>GIMPLE</code> representation, the PHI nodes and statements
contained in a basic block are in a <code>gimple_seq</code> pointed to by
the basic block intermediate language specific pointers.
Abstract containers and iterators are used to access the PHI nodes
and statements in a basic blocks.  These iterators are called
<em>GIMPLE statement iterators</em> (GSIs).  Grep for <code>^gsi</code>
in the various <samp>gimple-*</samp> and <samp>tree-*</samp> files.
The following snippet will pretty-print all PHI nodes the statements
of the current function in the GIMPLE representation.
</p>
<div class="smallexample">
<pre class="smallexample">basic_block bb;

FOR_EACH_BB (bb)
  {
   gimple_stmt_iterator si;

   for (si = gsi_start_phis (bb); !gsi_end_p (si); gsi_next (&amp;si))
     {
       gimple phi = gsi_stmt (si);
       print_gimple_stmt (dump_file, phi, 0, TDF_SLIM);
     }
   for (si = gsi_start_bb (bb); !gsi_end_p (si); gsi_next (&amp;si))
     {
       gimple stmt = gsi_stmt (si);
       print_gimple_stmt (dump_file, stmt, 0, TDF_SLIM);
     }
  }
</pre></div>


<hr>
<a name="Edges"></a>
<div class="header">
<p>
Next: <a href="#Profile-information" accesskey="n" rel="next">Profile information</a>, Previous: <a href="#Basic-Blocks" accesskey="p" rel="prev">Basic Blocks</a>, Up: <a href="#Control-Flow" accesskey="u" rel="up">Control Flow</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Edges-1"></a>
<h3 class="section">15.2 Edges</h3>

<a name="index-edge-in-the-flow-graph"></a>
<a name="index-edge"></a>
<p>Edges represent possible control flow transfers from the end of some
basic block A to the head of another basic block B.  We say that A is
a predecessor of B, and B is a successor of A.  Edges are represented
in GCC with the <code>edge</code> data type.  Each <code>edge</code> acts as a
link between two basic blocks: The <code>src</code> member of an edge
points to the predecessor basic block of the <code>dest</code> basic block.
The members <code>preds</code> and <code>succs</code> of the <code>basic_block</code> data
type point to type-safe vectors of edges to the predecessors and
successors of the block.
</p>
<a name="index-edge-iterators"></a>
<p>When walking the edges in an edge vector, <em>edge iterators</em> should
be used.  Edge iterators are constructed using the
<code>edge_iterator</code> data structure and several methods are available
to operate on them:
</p>
<dl compact="compact">
<dt><code>ei_start</code>
<a name="index-ei_005fstart"></a>
</dt>
<dd><p>This function initializes an <code>edge_iterator</code> that points to the
first edge in a vector of edges.
</p>
</dd>
<dt><code>ei_last</code>
<a name="index-ei_005flast"></a>
</dt>
<dd><p>This function initializes an <code>edge_iterator</code> that points to the
last edge in a vector of edges.
</p>
</dd>
<dt><code>ei_end_p</code>
<a name="index-ei_005fend_005fp"></a>
</dt>
<dd><p>This predicate is <code>true</code> if an <code>edge_iterator</code> represents
the last edge in an edge vector.
</p>
</dd>
<dt><code>ei_one_before_end_p</code>
<a name="index-ei_005fone_005fbefore_005fend_005fp"></a>
</dt>
<dd><p>This predicate is <code>true</code> if an <code>edge_iterator</code> represents
the second last edge in an edge vector.
</p>
</dd>
<dt><code>ei_next</code>
<a name="index-ei_005fnext"></a>
</dt>
<dd><p>This function takes a pointer to an <code>edge_iterator</code> and makes it
point to the next edge in the sequence.
</p>
</dd>
<dt><code>ei_prev</code>
<a name="index-ei_005fprev"></a>
</dt>
<dd><p>This function takes a pointer to an <code>edge_iterator</code> and makes it
point to the previous edge in the sequence.
</p>
</dd>
<dt><code>ei_edge</code>
<a name="index-ei_005fedge"></a>
</dt>
<dd><p>This function returns the <code>edge</code> currently pointed to by an
<code>edge_iterator</code>.
</p>
</dd>
<dt><code>ei_safe_safe</code>
<a name="index-ei_005fsafe_005fsafe"></a>
</dt>
<dd><p>This function returns the <code>edge</code> currently pointed to by an
<code>edge_iterator</code>, but returns <code>NULL</code> if the iterator is
pointing at the end of the sequence.  This function has been provided
for existing code makes the assumption that a <code>NULL</code> edge
indicates the end of the sequence.
</p>
</dd>
</dl>

<p>The convenience macro <code>FOR_EACH_EDGE</code> can be used to visit all of
the edges in a sequence of predecessor or successor edges.  It must
not be used when an element might be removed during the traversal,
otherwise elements will be missed.  Here is an example of how to use
the macro:
</p>
<div class="smallexample">
<pre class="smallexample">edge e;
edge_iterator ei;

FOR_EACH_EDGE (e, ei, bb-&gt;succs)
  {
     if (e-&gt;flags &amp; EDGE_FALLTHRU)
       break;
  }
</pre></div>

<a name="index-fall_002dthru"></a>
<p>There are various reasons why control flow may transfer from one block
to another.  One possibility is that some instruction, for example a
<code>CODE_LABEL</code>, in a linearized instruction stream just always
starts a new basic block.  In this case a <em>fall-thru</em> edge links
the basic block to the first following basic block.  But there are
several other reasons why edges may be created.  The <code>flags</code>
field of the <code>edge</code> data type is used to store information
about the type of edge we are dealing with.  Each edge is of one of
the following types:
</p>
<dl compact="compact">
<dt><em>jump</em></dt>
<dd><p>No type flags are set for edges corresponding to jump instructions.
These edges are used for unconditional or conditional jumps and in
RTL also for table jumps.  They are the easiest to manipulate as they
may be freely redirected when the flow graph is not in SSA form.
</p>
</dd>
<dt><em>fall-thru</em></dt>
<dd><a name="index-EDGE_005fFALLTHRU_002c-force_005fnonfallthru"></a>
<p>Fall-thru edges are present in case where the basic block may continue
execution to the following one without branching.  These edges have
the <code>EDGE_FALLTHRU</code> flag set.  Unlike other types of edges, these
edges must come into the basic block immediately following in the
instruction stream.  The function <code>force_nonfallthru</code> is
available to insert an unconditional jump in the case that redirection
is needed.  Note that this may require creation of a new basic block.
</p>
</dd>
<dt><em>exception handling</em></dt>
<dd><a name="index-exception-handling"></a>
<a name="index-EDGE_005fABNORMAL_002c-EDGE_005fEH"></a>
<p>Exception handling edges represent possible control transfers from a
trapping instruction to an exception handler.  The definition of
&ldquo;trapping&rdquo; varies.  In C++, only function calls can throw, but for
Java and Ada, exceptions like division by zero or segmentation fault are
defined and thus each instruction possibly throwing this kind of
exception needs to be handled as control flow instruction.  Exception
edges have the <code>EDGE_ABNORMAL</code> and <code>EDGE_EH</code> flags set.
</p>
<a name="index-purge_005fdead_005fedges"></a>
<p>When updating the instruction stream it is easy to change possibly
trapping instruction to non-trapping, by simply removing the exception
edge.  The opposite conversion is difficult, but should not happen
anyway.  The edges can be eliminated via <code>purge_dead_edges</code> call.
</p>
<a name="index-REG_005fEH_005fREGION_002c-EDGE_005fABNORMAL_005fCALL"></a>
<p>In the RTL representation, the destination of an exception edge is
specified by <code>REG_EH_REGION</code> note attached to the insn.
In case of a trapping call the <code>EDGE_ABNORMAL_CALL</code> flag is set
too.  In the <code>GIMPLE</code> representation, this extra flag is not set.
</p>
<a name="index-may_005ftrap_005fp_002c-tree_005fcould_005ftrap_005fp"></a>
<p>In the RTL representation, the predicate <code>may_trap_p</code> may be used
to check whether instruction still may trap or not.  For the tree
representation, the <code>tree_could_trap_p</code> predicate is available,
but this predicate only checks for possible memory traps, as in
dereferencing an invalid pointer location.
</p>

</dd>
<dt><em>sibling calls</em></dt>
<dd><a name="index-sibling-call"></a>
<a name="index-EDGE_005fABNORMAL_002c-EDGE_005fSIBCALL"></a>
<p>Sibling calls or tail calls terminate the function in a non-standard
way and thus an edge to the exit must be present.
<code>EDGE_SIBCALL</code> and <code>EDGE_ABNORMAL</code> are set in such case.
These edges only exist in the RTL representation.
</p>
</dd>
<dt><em>computed jumps</em></dt>
<dd><a name="index-computed-jump"></a>
<a name="index-EDGE_005fABNORMAL"></a>
<p>Computed jumps contain edges to all labels in the function referenced
from the code.  All those edges have <code>EDGE_ABNORMAL</code> flag set.
The edges used to represent computed jumps often cause compile time
performance problems, since functions consisting of many taken labels
and many computed jumps may have <em>very</em> dense flow graphs, so
these edges need to be handled with special care.  During the earlier
stages of the compilation process, GCC tries to avoid such dense flow
graphs by factoring computed jumps.  For example, given the following
series of jumps,
</p>
<div class="smallexample">
<pre class="smallexample">  goto *x;
  [ &hellip; ]

  goto *x;
  [ &hellip; ]

  goto *x;
  [ &hellip; ]
</pre></div>

<p>factoring the computed jumps results in the following code sequence
which has a much simpler flow graph:
</p>
<div class="smallexample">
<pre class="smallexample">  goto y;
  [ &hellip; ]

  goto y;
  [ &hellip; ]

  goto y;
  [ &hellip; ]

y:
  goto *x;
</pre></div>

<a name="index-pass_005fduplicate_005fcomputed_005fgotos"></a>
<p>However, the classic problem with this transformation is that it has a
runtime cost in there resulting code: An extra jump.  Therefore, the
computed jumps are un-factored in the later passes of the compiler
(in the pass called <code>pass_duplicate_computed_gotos</code>).
Be aware of that when you work on passes in that area.  There have
been numerous examples already where the compile time for code with
unfactored computed jumps caused some serious headaches.
</p>
</dd>
<dt><em>nonlocal goto handlers</em></dt>
<dd><a name="index-nonlocal-goto-handler"></a>
<a name="index-EDGE_005fABNORMAL_002c-EDGE_005fABNORMAL_005fCALL"></a>
<p>GCC allows nested functions to return into caller using a <code>goto</code>
to a label passed to as an argument to the callee.  The labels passed
to nested functions contain special code to cleanup after function
call.  Such sections of code are referred to as &ldquo;nonlocal goto
receivers&rdquo;.  If a function contains such nonlocal goto receivers, an
edge from the call to the label is created with the
<code>EDGE_ABNORMAL</code> and <code>EDGE_ABNORMAL_CALL</code> flags set.
</p>
</dd>
<dt><em>function entry points</em></dt>
<dd><a name="index-function-entry-point_002c-alternate-function-entry-point"></a>
<a name="index-LABEL_005fALTERNATE_005fNAME"></a>
<p>By definition, execution of function starts at basic block 0, so there
is always an edge from the <code>ENTRY_BLOCK_PTR</code> to basic block 0.
There is no <code>GIMPLE</code> representation for alternate entry points at
this moment.  In RTL, alternate entry points are specified by
<code>CODE_LABEL</code> with <code>LABEL_ALTERNATE_NAME</code> defined.  This
feature is currently used for multiple entry point prologues and is
limited to post-reload passes only.  This can be used by back-ends to
emit alternate prologues for functions called from different contexts.
In future full support for multiple entry functions defined by Fortran
90 needs to be implemented.
</p>
</dd>
<dt><em>function exits</em></dt>
<dd><p>In the pre-reload representation a function terminates after the last
instruction in the insn chain and no explicit return instructions are
used.  This corresponds to the fall-thru edge into exit block.  After
reload, optimal RTL epilogues are used that use explicit (conditional)
return instructions that are represented by edges with no flags set.
</p>
</dd>
</dl>


<hr>
<a name="Profile-information"></a>
<div class="header">
<p>
Next: <a href="#Maintaining-the-CFG" accesskey="n" rel="next">Maintaining the CFG</a>, Previous: <a href="#Edges" accesskey="p" rel="prev">Edges</a>, Up: <a href="#Control-Flow" accesskey="u" rel="up">Control Flow</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Profile-information-1"></a>
<h3 class="section">15.3 Profile information</h3>

<a name="index-profile-representation"></a>
<p>In many cases a compiler must make a choice whether to trade speed in
one part of code for speed in another, or to trade code size for code
speed.  In such cases it is useful to know information about how often
some given block will be executed.  That is the purpose for
maintaining profile within the flow graph.
GCC can handle profile information obtained through <em>profile
feedback</em>, but it can also estimate branch probabilities based on
statics and heuristics.
</p>
<a name="index-profile-feedback"></a>
<p>The feedback based profile is produced by compiling the program with
instrumentation, executing it on a train run and reading the numbers
of executions of basic blocks and edges back to the compiler while
re-compiling the program to produce the final executable.  This method
provides very accurate information about where a program spends most
of its time on the train run.  Whether it matches the average run of
course depends on the choice of train data set, but several studies
have shown that the behavior of a program usually changes just
marginally over different data sets.
</p>
<a name="index-Static-profile-estimation"></a>
<a name="index-branch-prediction"></a>
<a name="index-predict_002edef"></a>
<p>When profile feedback is not available, the compiler may be asked to
attempt to predict the behavior of each branch in the program using a
set of heuristics (see <samp>predict.def</samp> for details) and compute
estimated frequencies of each basic block by propagating the
probabilities over the graph.
</p>
<a name="index-frequency_002c-count_002c-BB_005fFREQ_005fBASE"></a>
<p>Each <code>basic_block</code> contains two integer fields to represent
profile information: <code>frequency</code> and <code>count</code>.  The
<code>frequency</code> is an estimation how often is basic block executed
within a function.  It is represented as an integer scaled in the
range from 0 to <code>BB_FREQ_BASE</code>.  The most frequently executed
basic block in function is initially set to <code>BB_FREQ_BASE</code> and
the rest of frequencies are scaled accordingly.  During optimization,
the frequency of the most frequent basic block can both decrease (for
instance by loop unrolling) or grow (for instance by cross-jumping
optimization), so scaling sometimes has to be performed multiple
times.
</p>
<a name="index-gcov_005ftype"></a>
<p>The <code>count</code> contains hard-counted numbers of execution measured
during training runs and is nonzero only when profile feedback is
available.  This value is represented as the host&rsquo;s widest integer
(typically a 64 bit integer) of the special type <code>gcov_type</code>.
</p>
<p>Most optimization passes can use only the frequency information of a
basic block, but a few passes may want to know hard execution counts.
The frequencies should always match the counts after scaling, however
during updating of the profile information numerical error may
accumulate into quite large errors.
</p>
<a name="index-REG_005fBR_005fPROB_005fBASE_002c-EDGE_005fFREQUENCY"></a>
<p>Each edge also contains a branch probability field: an integer in the
range from 0 to <code>REG_BR_PROB_BASE</code>.  It represents probability of
passing control from the end of the <code>src</code> basic block to the
<code>dest</code> basic block, i.e. the probability that control will flow
along this edge.  The <code>EDGE_FREQUENCY</code> macro is available to
compute how frequently a given edge is taken.  There is a <code>count</code>
field for each edge as well, representing same information as for a
basic block.
</p>
<p>The basic block frequencies are not represented in the instruction
stream, but in the RTL representation the edge frequencies are
represented for conditional jumps (via the <code>REG_BR_PROB</code>
macro) since they are used when instructions are output to the
assembly file and the flow graph is no longer maintained.
</p>
<a name="index-reverse-probability"></a>
<p>The probability that control flow arrives via a given edge to its
destination basic block is called <em>reverse probability</em> and is not
directly represented, but it may be easily computed from frequencies
of basic blocks.
</p>
<a name="index-redirect_005fedge_005fand_005fbranch"></a>
<p>Updating profile information is a delicate task that can unfortunately
not be easily integrated with the CFG manipulation API.  Many of the
functions and hooks to modify the CFG, such as
<code>redirect_edge_and_branch</code>, do not have enough information to
easily update the profile, so updating it is in the majority of cases
left up to the caller.  It is difficult to uncover bugs in the profile
updating code, because they manifest themselves only by producing
worse code, and checking profile consistency is not possible because
of numeric error accumulation.  Hence special attention needs to be
given to this issue in each pass that modifies the CFG.
</p>
<a name="index-REG_005fBR_005fPROB_005fBASE_002c-BB_005fFREQ_005fBASE_002c-count"></a>
<p>It is important to point out that <code>REG_BR_PROB_BASE</code> and
<code>BB_FREQ_BASE</code> are both set low enough to be possible to compute
second power of any frequency or probability in the flow graph, it is
not possible to even square the <code>count</code> field, as modern CPUs are
fast enough to execute $2^32$ operations quickly.
</p>

<hr>
<a name="Maintaining-the-CFG"></a>
<div class="header">
<p>
Next: <a href="#Liveness-information" accesskey="n" rel="next">Liveness information</a>, Previous: <a href="#Profile-information" accesskey="p" rel="prev">Profile information</a>, Up: <a href="#Control-Flow" accesskey="u" rel="up">Control Flow</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Maintaining-the-CFG-1"></a>
<h3 class="section">15.4 Maintaining the CFG</h3>
<a name="index-cfghooks_002eh"></a>

<p>An important task of each compiler pass is to keep both the control
flow graph and all profile information up-to-date.  Reconstruction of
the control flow graph after each pass is not an option, since it may be
very expensive and lost profile information cannot be reconstructed at
all.
</p>
<p>GCC has two major intermediate representations, and both use the
<code>basic_block</code> and <code>edge</code> data types to represent control
flow.  Both representations share as much of the CFG maintenance code
as possible.  For each representation, a set of <em>hooks</em> is defined
so that each representation can provide its own implementation of CFG
manipulation routines when necessary.  These hooks are defined in
<samp>cfghooks.h</samp>.  There are hooks for almost all common CFG
manipulations, including block splitting and merging, edge redirection
and creating and deleting basic blocks.  These hooks should provide
everything you need to maintain and manipulate the CFG in both the RTL
and <code>GIMPLE</code> representation.
</p>
<p>At the moment, the basic block boundaries are maintained transparently
when modifying instructions, so there rarely is a need to move them
manually (such as in case someone wants to output instruction outside
basic block explicitly).
</p>
<a name="index-BLOCK_005fFOR_005fINSN_002c-gimple_005fbb"></a>
<p>In the RTL representation, each instruction has a
<code>BLOCK_FOR_INSN</code> value that represents pointer to the basic block
that contains the instruction.  In the <code>GIMPLE</code> representation, the
function <code>gimple_bb</code> returns a pointer to the basic block
containing the queried statement.
</p>
<a name="index-GIMPLE-statement-iterators-1"></a>
<p>When changes need to be applied to a function in its <code>GIMPLE</code>
representation, <em>GIMPLE statement iterators</em> should be used.  These
iterators provide an integrated abstraction of the flow graph and the
instruction stream.  Block statement iterators are constructed using
the <code>gimple_stmt_iterator</code> data structure and several modifier are
available, including the following:
</p>
<dl compact="compact">
<dt><code>gsi_start</code>
<a name="index-gsi_005fstart-1"></a>
</dt>
<dd><p>This function initializes a <code>gimple_stmt_iterator</code> that points to
the first non-empty statement in a basic block.
</p>
</dd>
<dt><code>gsi_last</code>
<a name="index-gsi_005flast-1"></a>
</dt>
<dd><p>This function initializes a <code>gimple_stmt_iterator</code> that points to
the last statement in a basic block.
</p>
</dd>
<dt><code>gsi_end_p</code>
<a name="index-gsi_005fend_005fp-1"></a>
</dt>
<dd><p>This predicate is <code>true</code> if a <code>gimple_stmt_iterator</code>
represents the end of a basic block.
</p>
</dd>
<dt><code>gsi_next</code>
<a name="index-gsi_005fnext-1"></a>
</dt>
<dd><p>This function takes a <code>gimple_stmt_iterator</code> and makes it point to
its successor.
</p>
</dd>
<dt><code>gsi_prev</code>
<a name="index-gsi_005fprev-1"></a>
</dt>
<dd><p>This function takes a <code>gimple_stmt_iterator</code> and makes it point to
its predecessor.
</p>
</dd>
<dt><code>gsi_insert_after</code>
<a name="index-gsi_005finsert_005fafter-1"></a>
</dt>
<dd><p>This function inserts a statement after the <code>gimple_stmt_iterator</code>
passed in.  The final parameter determines whether the statement
iterator is updated to point to the newly inserted statement, or left
pointing to the original statement.
</p>
</dd>
<dt><code>gsi_insert_before</code>
<a name="index-gsi_005finsert_005fbefore-1"></a>
</dt>
<dd><p>This function inserts a statement before the <code>gimple_stmt_iterator</code>
passed in.  The final parameter determines whether the statement
iterator is updated to point to the newly inserted statement, or left
pointing to the original  statement.
</p>
</dd>
<dt><code>gsi_remove</code>
<a name="index-gsi_005fremove-1"></a>
</dt>
<dd><p>This function removes the <code>gimple_stmt_iterator</code> passed in and
rechains the remaining statements in a basic block, if any.
</p></dd>
</dl>

<a name="index-BB_005fHEAD_002c-BB_005fEND"></a>
<p>In the RTL representation, the macros <code>BB_HEAD</code> and <code>BB_END</code>
may be used to get the head and end <code>rtx</code> of a basic block.  No
abstract iterators are defined for traversing the insn chain, but you
can just use <code>NEXT_INSN</code> and <code>PREV_INSN</code> instead.  See <a href="#Insns">Insns</a>.
</p>
<a name="index-purge_005fdead_005fedges-1"></a>
<p>Usually a code manipulating pass simplifies the instruction stream and
the flow of control, possibly eliminating some edges.  This may for
example happen when a conditional jump is replaced with an
unconditional jump, but also when simplifying possibly trapping
instruction to non-trapping while compiling Java.  Updating of edges
is not transparent and each optimization pass is required to do so
manually.  However only few cases occur in practice.  The pass may
call <code>purge_dead_edges</code> on a given basic block to remove
superfluous edges, if any.
</p>
<a name="index-redirect_005fedge_005fand_005fbranch_002c-redirect_005fjump"></a>
<p>Another common scenario is redirection of branch instructions, but
this is best modeled as redirection of edges in the control flow graph
and thus use of <code>redirect_edge_and_branch</code> is preferred over more
low level functions, such as <code>redirect_jump</code> that operate on RTL
chain only.  The CFG hooks defined in <samp>cfghooks.h</samp> should provide
the complete API required for manipulating and maintaining the CFG.
</p>
<a name="index-split_005fblock"></a>
<p>It is also possible that a pass has to insert control flow instruction
into the middle of a basic block, thus creating an entry point in the
middle of the basic block, which is impossible by definition: The
block must be split to make sure it only has one entry point, i.e. the
head of the basic block.  The CFG hook <code>split_block</code> may be used
when an instruction in the middle of a basic block has to become the
target of a jump or branch instruction.
</p>
<a name="index-insert_005finsn_005fon_005fedge"></a>
<a name="index-commit_005fedge_005finsertions"></a>
<a name="index-gsi_005finsert_005fon_005fedge-1"></a>
<a name="index-gsi_005fcommit_005fedge_005finserts-1"></a>
<a name="index-edge-splitting"></a>
<p>For a global optimizer, a common operation is to split edges in the
flow graph and insert instructions on them.  In the RTL
representation, this can be easily done using the
<code>insert_insn_on_edge</code> function that emits an instruction
&ldquo;on the edge&rdquo;, caching it for a later <code>commit_edge_insertions</code>
call that will take care of moving the inserted instructions off the
edge into the instruction stream contained in a basic block.  This
includes the creation of new basic blocks where needed.  In the
<code>GIMPLE</code> representation, the equivalent functions are
<code>gsi_insert_on_edge</code> which inserts a block statement
iterator on an edge, and <code>gsi_commit_edge_inserts</code> which flushes
the instruction to actual instruction stream.
</p>
<a name="index-verify_005fflow_005finfo"></a>
<a name="index-CFG-verification"></a>
<p>While debugging the optimization pass, the <code>verify_flow_info</code>
function may be useful to find bugs in the control flow graph updating
code.
</p>

<hr>
<a name="Liveness-information"></a>
<div class="header">
<p>
Previous: <a href="#Maintaining-the-CFG" accesskey="p" rel="prev">Maintaining the CFG</a>, Up: <a href="#Control-Flow" accesskey="u" rel="up">Control Flow</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Liveness-information-1"></a>
<h3 class="section">15.5 Liveness information</h3>
<a name="index-Liveness-representation"></a>
<p>Liveness information is useful to determine whether some register is
&ldquo;live&rdquo; at given point of program, i.e. that it contains a value that
may be used at a later point in the program.  This information is
used, for instance, during register allocation, as the pseudo
registers only need to be assigned to a unique hard register or to a
stack slot if they are live.  The hard registers and stack slots may
be freely reused for other values when a register is dead.
</p>
<p>Liveness information is available in the back end starting with
<code>pass_df_initialize</code> and ending with <code>pass_df_finish</code>.  Three
flavors of live analysis are available: With <code>LR</code>, it is possible
to determine at any point <code>P</code> in the function if the register may be
used on some path from <code>P</code> to the end of the function.  With
<code>UR</code>, it is possible to determine if there is a path from the
beginning of the function to <code>P</code> that defines the variable.
<code>LIVE</code> is the intersection of the <code>LR</code> and <code>UR</code> and a
variable is live at <code>P</code> if there is both an assignment that reaches
it from the beginning of the function and a use that can be reached on
some path from <code>P</code> to the end of the function.
</p>
<p>In general <code>LIVE</code> is the most useful of the three.  The macros
<code>DF_[LR,UR,LIVE]_[IN,OUT]</code> can be used to access this information.
The macros take a basic block number and return a bitmap that is indexed
by the register number.  This information is only guaranteed to be up to
date after calls are made to <code>df_analyze</code>.  See the file
<code>df-core.c</code> for details on using the dataflow.
</p>

<a name="index-REG_005fDEAD_002c-REG_005fUNUSED"></a>
<p>The liveness information is stored partly in the RTL instruction stream
and partly in the flow graph.  Local information is stored in the
instruction stream: Each instruction may contain <code>REG_DEAD</code> notes
representing that the value of a given register is no longer needed, or
<code>REG_UNUSED</code> notes representing that the value computed by the
instruction is never used.  The second is useful for instructions
computing multiple values at once.
</p>

<hr>
<a name="Machine-Desc"></a>
<div class="header">
<p>
Next: <a href="#Target-Macros" accesskey="n" rel="next">Target Macros</a>, Previous: <a href="#Control-Flow" accesskey="p" rel="prev">Control Flow</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Machine-Descriptions"></a>
<h2 class="chapter">16 Machine Descriptions</h2>
<a name="index-machine-descriptions"></a>

<p>A machine description has two parts: a file of instruction patterns
(<samp>.md</samp> file) and a C header file of macro definitions.
</p>
<p>The <samp>.md</samp> file for a target machine contains a pattern for each
instruction that the target machine supports (or at least each instruction
that is worth telling the compiler about).  It may also contain comments.
A semicolon causes the rest of the line to be a comment, unless the semicolon
is inside a quoted string.
</p>
<p>See the next chapter for information on the C header file.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Overview" accesskey="1">Overview</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How the machine description is used.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Patterns" accesskey="2">Patterns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to write instruction patterns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Example" accesskey="3">Example</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">An explained example of a <code>define_insn</code> pattern.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#RTL-Template" accesskey="4">RTL Template</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The RTL template defines what insns match a pattern.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Output-Template" accesskey="5">Output Template</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The output template says how to make assembler code
                        from such an insn.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Output-Statement" accesskey="6">Output Statement</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">For more generality, write C code to output
                        the assembler code.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Predicates" accesskey="7">Predicates</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Controlling what kinds of operands can be used
                        for an insn.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Constraints" accesskey="8">Constraints</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Fine-tuning operand selection.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Standard-Names" accesskey="9">Standard Names</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Names mark patterns to use for code generation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Pattern-Ordering">Pattern Ordering</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">When the order of patterns makes a difference.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Dependent-Patterns">Dependent Patterns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Having one pattern may make you need another.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Jump-Patterns">Jump Patterns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Special considerations for patterns for jump insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Looping-Patterns">Looping Patterns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to define patterns for special looping insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Insn-Canonicalizations">Insn Canonicalizations</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Canonicalization of Instructions
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Expander-Definitions">Expander Definitions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating a sequence of several RTL insns
                        for a standard operation.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Insn-Splitting">Insn Splitting</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Splitting Instructions into Multiple Instructions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Including-Patterns">Including Patterns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Including Patterns in Machine Descriptions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Peephole-Definitions">Peephole Definitions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining machine-specific peephole optimizations.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Insn-Attributes">Insn Attributes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Specifying the value of attributes for generated insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Conditional-Execution">Conditional Execution</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating <code>define_insn</code> patterns for
                         predication.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Define-Subst">Define Subst</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating <code>define_insn</code> and <code>define_expand</code>
			patterns from other patterns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Constant-Definitions">Constant Definitions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining symbolic constants that can be used in the
                        md file.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Iterators">Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Using iterators to generate patterns from a template.
</td></tr>
</table>

<hr>
<a name="Overview"></a>
<div class="header">
<p>
Next: <a href="#Patterns" accesskey="n" rel="next">Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Overview-of-How-the-Machine-Description-is-Used"></a>
<h3 class="section">16.1 Overview of How the Machine Description is Used</h3>

<p>There are three main conversions that happen in the compiler:
</p>
<ol>
<li> The front end reads the source code and builds a parse tree.

</li><li> The parse tree is used to generate an RTL insn list based on named
instruction patterns.

</li><li> The insn list is matched against the RTL templates to produce assembler
code.

</li></ol>

<p>For the generate pass, only the names of the insns matter, from either a
named <code>define_insn</code> or a <code>define_expand</code>.  The compiler will
choose the pattern with the right name and apply the operands according
to the documentation later in this chapter, without regard for the RTL
template or operand constraints.  Note that the names the compiler looks
for are hard-coded in the compiler&mdash;it will ignore unnamed patterns and
patterns with names it doesn&rsquo;t know about, but if you don&rsquo;t provide a
named pattern it needs, it will abort.
</p>
<p>If a <code>define_insn</code> is used, the template given is inserted into the
insn list.  If a <code>define_expand</code> is used, one of three things
happens, based on the condition logic.  The condition logic may manually
create new insns for the insn list, say via <code>emit_insn()</code>, and
invoke <code>DONE</code>.  For certain named patterns, it may invoke <code>FAIL</code> to tell the
compiler to use an alternate way of performing that task.  If it invokes
neither <code>DONE</code> nor <code>FAIL</code>, the template given in the pattern
is inserted, as if the <code>define_expand</code> were a <code>define_insn</code>.
</p>
<p>Once the insn list is generated, various optimization passes convert,
replace, and rearrange the insns in the insn list.  This is where the
<code>define_split</code> and <code>define_peephole</code> patterns get used, for
example.
</p>
<p>Finally, the insn list&rsquo;s RTL is matched up with the RTL templates in the
<code>define_insn</code> patterns, and those patterns are used to emit the
final assembly code.  For this purpose, each named <code>define_insn</code>
acts like it&rsquo;s unnamed, since the names are ignored.
</p>
<hr>
<a name="Patterns"></a>
<div class="header">
<p>
Next: <a href="#Example" accesskey="n" rel="next">Example</a>, Previous: <a href="#Overview" accesskey="p" rel="prev">Overview</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Everything-about-Instruction-Patterns"></a>
<h3 class="section">16.2 Everything about Instruction Patterns</h3>
<a name="index-patterns"></a>
<a name="index-instruction-patterns"></a>

<a name="index-define_005finsn"></a>
<p>Each instruction pattern contains an incomplete RTL expression, with pieces
to be filled in later, operand constraints that restrict how the pieces can
be filled in, and an output pattern or C code to generate the assembler
output, all wrapped up in a <code>define_insn</code> expression.
</p>
<p>A <code>define_insn</code> is an RTL expression containing four or five operands:
</p>
<ol>
<li> An optional name.  The presence of a name indicate that this instruction
pattern can perform a certain standard job for the RTL-generation
pass of the compiler.  This pass knows certain names and will use
the instruction patterns with those names, if the names are defined
in the machine description.

<p>The absence of a name is indicated by writing an empty string
where the name should go.  Nameless instruction patterns are never
used for generating RTL code, but they may permit several simpler insns
to be combined later on.
</p>
<p>Names that are not thus known and used in RTL-generation have no
effect; they are equivalent to no name at all.
</p>
<p>For the purpose of debugging the compiler, you may also specify a
name beginning with the &lsquo;<samp>*</samp>&rsquo; character.  Such a name is used only
for identifying the instruction in RTL dumps; it is entirely equivalent
to having a nameless pattern for all other purposes.
</p>
</li><li> The <em>RTL template</em> (see <a href="#RTL-Template">RTL Template</a>) is a vector of incomplete
RTL expressions which show what the instruction should look like.  It is
incomplete because it may contain <code>match_operand</code>,
<code>match_operator</code>, and <code>match_dup</code> expressions that stand for
operands of the instruction.

<p>If the vector has only one element, that element is the template for the
instruction pattern.  If the vector has multiple elements, then the
instruction pattern is a <code>parallel</code> expression containing the
elements described.
</p>
</li><li> <a name="index-pattern-conditions"></a>
<a name="index-conditions_002c-in-patterns"></a>
A condition.  This is a string which contains a C expression that is
the final test to decide whether an insn body matches this pattern.

<a name="index-named-patterns-and-conditions"></a>
<p>For a named pattern, the condition (if present) may not depend on
the data in the insn being matched, but only the target-machine-type
flags.  The compiler needs to test these conditions during
initialization in order to learn exactly which named instructions are
available in a particular run.
</p>
<a name="index-operands-1"></a>
<p>For nameless patterns, the condition is applied only when matching an
individual insn, and only after the insn has matched the pattern&rsquo;s
recognition template.  The insn&rsquo;s operands may be found in the vector
<code>operands</code>.  For an insn where the condition has once matched, it
can&rsquo;t be used to control register allocation, for example by excluding
certain hard registers or hard register combinations.
</p>
</li><li> The <em>output template</em>: a string that says how to output matching
insns as assembler code.  &lsquo;<samp>%</samp>&rsquo; in this string specifies where
to substitute the value of an operand.  See <a href="#Output-Template">Output Template</a>.

<p>When simple substitution isn&rsquo;t general enough, you can specify a piece
of C code to compute the output.  See <a href="#Output-Statement">Output Statement</a>.
</p>
</li><li> Optionally, a vector containing the values of attributes for insns matching
this pattern.  See <a href="#Insn-Attributes">Insn Attributes</a>.
</li></ol>

<hr>
<a name="Example"></a>
<div class="header">
<p>
Next: <a href="#RTL-Template" accesskey="n" rel="next">RTL Template</a>, Previous: <a href="#Patterns" accesskey="p" rel="prev">Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Example-of-define_005finsn"></a>
<h3 class="section">16.3 Example of <code>define_insn</code></h3>
<a name="index-define_005finsn-example"></a>

<p>Here is an actual example of an instruction pattern, for the 68000/68020.
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;tstsi&quot;
  [(set (cc0)
        (match_operand:SI 0 &quot;general_operand&quot; &quot;rm&quot;))]
  &quot;&quot;
  &quot;*
{
  if (TARGET_68020 || ! ADDRESS_REG_P (operands[0]))
    return \&quot;tstl %0\&quot;;
  return \&quot;cmpl #0,%0\&quot;;
}&quot;)
</pre></div>

<p>This can also be written using braced strings:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;tstsi&quot;
  [(set (cc0)
        (match_operand:SI 0 &quot;general_operand&quot; &quot;rm&quot;))]
  &quot;&quot;
{
  if (TARGET_68020 || ! ADDRESS_REG_P (operands[0]))
    return &quot;tstl %0&quot;;
  return &quot;cmpl #0,%0&quot;;
})
</pre></div>

<p>This is an instruction that sets the condition codes based on the value of
a general operand.  It has no condition, so any insn whose RTL description
has the form shown may be handled according to this pattern.  The name
&lsquo;<samp>tstsi</samp>&rsquo; means &ldquo;test a <code>SImode</code> value&rdquo; and tells the RTL generation
pass that, when it is necessary to test such a value, an insn to do so
can be constructed using this pattern.
</p>
<p>The output control string is a piece of C code which chooses which
output template to return based on the kind of operand and the specific
type of CPU for which code is being generated.
</p>
<p>&lsquo;<samp>&quot;rm&quot;</samp>&rsquo; is an operand constraint.  Its meaning is explained below.
</p>
<hr>
<a name="RTL-Template"></a>
<div class="header">
<p>
Next: <a href="#Output-Template" accesskey="n" rel="next">Output Template</a>, Previous: <a href="#Example" accesskey="p" rel="prev">Example</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Template-1"></a>
<h3 class="section">16.4 RTL Template</h3>
<a name="index-RTL-insn-template"></a>
<a name="index-generating-insns"></a>
<a name="index-insns_002c-generating"></a>
<a name="index-recognizing-insns"></a>
<a name="index-insns_002c-recognizing"></a>

<p>The RTL template is used to define which insns match the particular pattern
and how to find their operands.  For named patterns, the RTL template also
says how to construct an insn from specified operands.
</p>
<p>Construction involves substituting specified operands into a copy of the
template.  Matching involves determining the values that serve as the
operands in the insn being matched.  Both of these activities are
controlled by special expression types that direct matching and
substitution of the operands.
</p>
<dl compact="compact">
<dd><a name="index-match_005foperand"></a>
</dd>
<dt><code>(match_operand:<var>m</var> <var>n</var> <var>predicate</var> <var>constraint</var>)</code></dt>
<dd><p>This expression is a placeholder for operand number <var>n</var> of
the insn.  When constructing an insn, operand number <var>n</var>
will be substituted at this point.  When matching an insn, whatever
appears at this position in the insn will be taken as operand
number <var>n</var>; but it must satisfy <var>predicate</var> or this instruction
pattern will not match at all.
</p>
<p>Operand numbers must be chosen consecutively counting from zero in
each instruction pattern.  There may be only one <code>match_operand</code>
expression in the pattern for each operand number.  Usually operands
are numbered in the order of appearance in <code>match_operand</code>
expressions.  In the case of a <code>define_expand</code>, any operand numbers
used only in <code>match_dup</code> expressions have higher values than all
other operand numbers.
</p>
<p><var>predicate</var> is a string that is the name of a function that
accepts two arguments, an expression and a machine mode.
See <a href="#Predicates">Predicates</a>.  During matching, the function will be called with
the putative operand as the expression and <var>m</var> as the mode
argument (if <var>m</var> is not specified, <code>VOIDmode</code> will be used,
which normally causes <var>predicate</var> to accept any mode).  If it
returns zero, this instruction pattern fails to match.
<var>predicate</var> may be an empty string; then it means no test is to be
done on the operand, so anything which occurs in this position is
valid.
</p>
<p>Most of the time, <var>predicate</var> will reject modes other than <var>m</var>&mdash;but
not always.  For example, the predicate <code>address_operand</code> uses
<var>m</var> as the mode of memory ref that the address should be valid for.
Many predicates accept <code>const_int</code> nodes even though their mode is
<code>VOIDmode</code>.
</p>
<p><var>constraint</var> controls reloading and the choice of the best register
class to use for a value, as explained later (see <a href="#Constraints">Constraints</a>).
If the constraint would be an empty string, it can be omitted.
</p>
<p>People are often unclear on the difference between the constraint and the
predicate.  The predicate helps decide whether a given insn matches the
pattern.  The constraint plays no role in this decision; instead, it
controls various decisions in the case of an insn which does match.
</p>
<a name="index-match_005fscratch"></a>
</dd>
<dt><code>(match_scratch:<var>m</var> <var>n</var> <var>constraint</var>)</code></dt>
<dd><p>This expression is also a placeholder for operand number <var>n</var>
and indicates that operand must be a <code>scratch</code> or <code>reg</code>
expression.
</p>
<p>When matching patterns, this is equivalent to
</p>
<div class="smallexample">
<pre class="smallexample">(match_operand:<var>m</var> <var>n</var> &quot;scratch_operand&quot; <var>pred</var>)
</pre></div>

<p>but, when generating RTL, it produces a (<code>scratch</code>:<var>m</var>)
expression.
</p>
<p>If the last few expressions in a <code>parallel</code> are <code>clobber</code>
expressions whose operands are either a hard register or
<code>match_scratch</code>, the combiner can add or delete them when
necessary.  See <a href="#Side-Effects">Side Effects</a>.
</p>
<a name="index-match_005fdup"></a>
</dd>
<dt><code>(match_dup <var>n</var>)</code></dt>
<dd><p>This expression is also a placeholder for operand number <var>n</var>.
It is used when the operand needs to appear more than once in the
insn.
</p>
<p>In construction, <code>match_dup</code> acts just like <code>match_operand</code>:
the operand is substituted into the insn being constructed.  But in
matching, <code>match_dup</code> behaves differently.  It assumes that operand
number <var>n</var> has already been determined by a <code>match_operand</code>
appearing earlier in the recognition template, and it matches only an
identical-looking expression.
</p>
<p>Note that <code>match_dup</code> should not be used to tell the compiler that
a particular register is being used for two operands (example:
<code>add</code> that adds one register to another; the second register is
both an input operand and the output operand).  Use a matching
constraint (see <a href="#Simple-Constraints">Simple Constraints</a>) for those.  <code>match_dup</code> is for the cases where one
operand is used in two places in the template, such as an instruction
that computes both a quotient and a remainder, where the opcode takes
two input operands but the RTL template has to refer to each of those
twice; once for the quotient pattern and once for the remainder pattern.
</p>
<a name="index-match_005foperator"></a>
</dd>
<dt><code>(match_operator:<var>m</var> <var>n</var> <var>predicate</var> [<var>operands</var>&hellip;])</code></dt>
<dd><p>This pattern is a kind of placeholder for a variable RTL expression
code.
</p>
<p>When constructing an insn, it stands for an RTL expression whose
expression code is taken from that of operand <var>n</var>, and whose
operands are constructed from the patterns <var>operands</var>.
</p>
<p>When matching an expression, it matches an expression if the function
<var>predicate</var> returns nonzero on that expression <em>and</em> the
patterns <var>operands</var> match the operands of the expression.
</p>
<p>Suppose that the function <code>commutative_operator</code> is defined as
follows, to match any expression whose operator is one of the
commutative arithmetic operators of RTL and whose mode is <var>mode</var>:
</p>
<div class="smallexample">
<pre class="smallexample">int
commutative_integer_operator (x, mode)
     rtx x;
     enum machine_mode mode;
{
  enum rtx_code code = GET_CODE (x);
  if (GET_MODE (x) != mode)
    return 0;
  return (GET_RTX_CLASS (code) == RTX_COMM_ARITH
          || code == EQ || code == NE);
}
</pre></div>

<p>Then the following pattern will match any RTL expression consisting
of a commutative operator applied to two general operands:
</p>
<div class="smallexample">
<pre class="smallexample">(match_operator:SI 3 &quot;commutative_operator&quot;
  [(match_operand:SI 1 &quot;general_operand&quot; &quot;g&quot;)
   (match_operand:SI 2 &quot;general_operand&quot; &quot;g&quot;)])
</pre></div>

<p>Here the vector <code>[<var>operands</var>&hellip;]</code> contains two patterns
because the expressions to be matched all contain two operands.
</p>
<p>When this pattern does match, the two operands of the commutative
operator are recorded as operands 1 and 2 of the insn.  (This is done
by the two instances of <code>match_operand</code>.)  Operand 3 of the insn
will be the entire commutative expression: use <code>GET_CODE
(operands[3])</code> to see which commutative operator was used.
</p>
<p>The machine mode <var>m</var> of <code>match_operator</code> works like that of
<code>match_operand</code>: it is passed as the second argument to the
predicate function, and that function is solely responsible for
deciding whether the expression to be matched &ldquo;has&rdquo; that mode.
</p>
<p>When constructing an insn, argument 3 of the gen-function will specify
the operation (i.e. the expression code) for the expression to be
made.  It should be an RTL expression, whose expression code is copied
into a new expression whose operands are arguments 1 and 2 of the
gen-function.  The subexpressions of argument 3 are not used;
only its expression code matters.
</p>
<p>When <code>match_operator</code> is used in a pattern for matching an insn,
it usually best if the operand number of the <code>match_operator</code>
is higher than that of the actual operands of the insn.  This improves
register allocation because the register allocator often looks at
operands 1 and 2 of insns to see if it can do register tying.
</p>
<p>There is no way to specify constraints in <code>match_operator</code>.  The
operand of the insn which corresponds to the <code>match_operator</code>
never has any constraints because it is never reloaded as a whole.
However, if parts of its <var>operands</var> are matched by
<code>match_operand</code> patterns, those parts may have constraints of
their own.
</p>
<a name="index-match_005fop_005fdup"></a>
</dd>
<dt><code>(match_op_dup:<var>m</var> <var>n</var>[<var>operands</var>&hellip;])</code></dt>
<dd><p>Like <code>match_dup</code>, except that it applies to operators instead of
operands.  When constructing an insn, operand number <var>n</var> will be
substituted at this point.  But in matching, <code>match_op_dup</code> behaves
differently.  It assumes that operand number <var>n</var> has already been
determined by a <code>match_operator</code> appearing earlier in the
recognition template, and it matches only an identical-looking
expression.
</p>
<a name="index-match_005fparallel"></a>
</dd>
<dt><code>(match_parallel <var>n</var> <var>predicate</var> [<var>subpat</var>&hellip;])</code></dt>
<dd><p>This pattern is a placeholder for an insn that consists of a
<code>parallel</code> expression with a variable number of elements.  This
expression should only appear at the top level of an insn pattern.
</p>
<p>When constructing an insn, operand number <var>n</var> will be substituted at
this point.  When matching an insn, it matches if the body of the insn
is a <code>parallel</code> expression with at least as many elements as the
vector of <var>subpat</var> expressions in the <code>match_parallel</code>, if each
<var>subpat</var> matches the corresponding element of the <code>parallel</code>,
<em>and</em> the function <var>predicate</var> returns nonzero on the
<code>parallel</code> that is the body of the insn.  It is the responsibility
of the predicate to validate elements of the <code>parallel</code> beyond
those listed in the <code>match_parallel</code>.
</p>
<p>A typical use of <code>match_parallel</code> is to match load and store
multiple expressions, which can contain a variable number of elements
in a <code>parallel</code>.  For example,
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(match_parallel 0 &quot;load_multiple_operation&quot;
     [(set (match_operand:SI 1 &quot;gpc_reg_operand&quot; &quot;=r&quot;)
           (match_operand:SI 2 &quot;memory_operand&quot; &quot;m&quot;))
      (use (reg:SI 179))
      (clobber (reg:SI 179))])]
  &quot;&quot;
  &quot;loadm 0,0,%1,%2&quot;)
</pre></div>

<p>This example comes from <samp>a29k.md</samp>.  The function
<code>load_multiple_operation</code> is defined in <samp>a29k.c</samp> and checks
that subsequent elements in the <code>parallel</code> are the same as the
<code>set</code> in the pattern, except that they are referencing subsequent
registers and memory locations.
</p>
<p>An insn that matches this pattern might look like:
</p>
<div class="smallexample">
<pre class="smallexample">(parallel
 [(set (reg:SI 20) (mem:SI (reg:SI 100)))
  (use (reg:SI 179))
  (clobber (reg:SI 179))
  (set (reg:SI 21)
       (mem:SI (plus:SI (reg:SI 100)
                        (const_int 4))))
  (set (reg:SI 22)
       (mem:SI (plus:SI (reg:SI 100)
                        (const_int 8))))])
</pre></div>

<a name="index-match_005fpar_005fdup"></a>
</dd>
<dt><code>(match_par_dup <var>n</var> [<var>subpat</var>&hellip;])</code></dt>
<dd><p>Like <code>match_op_dup</code>, but for <code>match_parallel</code> instead of
<code>match_operator</code>.
</p>
</dd>
</dl>

<hr>
<a name="Output-Template"></a>
<div class="header">
<p>
Next: <a href="#Output-Statement" accesskey="n" rel="next">Output Statement</a>, Previous: <a href="#RTL-Template" accesskey="p" rel="prev">RTL Template</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-Templates-and-Operand-Substitution"></a>
<h3 class="section">16.5 Output Templates and Operand Substitution</h3>
<a name="index-output-templates"></a>
<a name="index-operand-substitution"></a>

<a name="index-_0025-in-template"></a>
<a name="index-percent-sign"></a>
<p>The <em>output template</em> is a string which specifies how to output the
assembler code for an instruction pattern.  Most of the template is a
fixed string which is output literally.  The character &lsquo;<samp>%</samp>&rsquo; is used
to specify where to substitute an operand; it can also be used to
identify places where different variants of the assembler require
different syntax.
</p>
<p>In the simplest case, a &lsquo;<samp>%</samp>&rsquo; followed by a digit <var>n</var> says to output
operand <var>n</var> at that point in the string.
</p>
<p>&lsquo;<samp>%</samp>&rsquo; followed by a letter and a digit says to output an operand in an
alternate fashion.  Four letters have standard, built-in meanings described
below.  The machine description macro <code>PRINT_OPERAND</code> can define
additional letters with nonstandard meanings.
</p>
<p>&lsquo;<samp>%c<var>digit</var></samp>&rsquo; can be used to substitute an operand that is a
constant value without the syntax that normally indicates an immediate
operand.
</p>
<p>&lsquo;<samp>%n<var>digit</var></samp>&rsquo; is like &lsquo;<samp>%c<var>digit</var></samp>&rsquo; except that the value of
the constant is negated before printing.
</p>
<p>&lsquo;<samp>%a<var>digit</var></samp>&rsquo; can be used to substitute an operand as if it were a
memory reference, with the actual operand treated as the address.  This may
be useful when outputting a &ldquo;load address&rdquo; instruction, because often the
assembler syntax for such an instruction requires you to write the operand
as if it were a memory reference.
</p>
<p>&lsquo;<samp>%l<var>digit</var></samp>&rsquo; is used to substitute a <code>label_ref</code> into a jump
instruction.
</p>
<p>&lsquo;<samp>%=</samp>&rsquo; outputs a number which is unique to each instruction in the
entire compilation.  This is useful for making local labels to be
referred to more than once in a single template that generates multiple
assembler instructions.
</p>
<p>&lsquo;<samp>%</samp>&rsquo; followed by a punctuation character specifies a substitution that
does not use an operand.  Only one case is standard: &lsquo;<samp>%%</samp>&rsquo; outputs a
&lsquo;<samp>%</samp>&rsquo; into the assembler code.  Other nonstandard cases can be
defined in the <code>PRINT_OPERAND</code> macro.  You must also define
which punctuation characters are valid with the
<code>PRINT_OPERAND_PUNCT_VALID_P</code> macro.
</p>
<a name="index-_005c"></a>
<a name="index-backslash"></a>
<p>The template may generate multiple assembler instructions.  Write the text
for the instructions, with &lsquo;<samp>\;</samp>&rsquo; between them.
</p>
<a name="index-matching-operands"></a>
<p>When the RTL contains two operands which are required by constraint to match
each other, the output template must refer only to the lower-numbered operand.
Matching operands are not always identical, and the rest of the compiler
arranges to put the proper RTL expression for printing into the lower-numbered
operand.
</p>
<p>One use of nonstandard letters or punctuation following &lsquo;<samp>%</samp>&rsquo; is to
distinguish between different assembler languages for the same machine; for
example, Motorola syntax versus MIT syntax for the 68000.  Motorola syntax
requires periods in most opcode names, while MIT syntax does not.  For
example, the opcode &lsquo;<samp>movel</samp>&rsquo; in MIT syntax is &lsquo;<samp>move.l</samp>&rsquo; in Motorola
syntax.  The same file of patterns is used for both kinds of output syntax,
but the character sequence &lsquo;<samp>%.</samp>&rsquo; is used in each place where Motorola
syntax wants a period.  The <code>PRINT_OPERAND</code> macro for Motorola syntax
defines the sequence to output a period; the macro for MIT syntax defines
it to do nothing.
</p>
<a name="index-_0023-in-template"></a>
<p>As a special case, a template consisting of the single character <code>#</code>
instructs the compiler to first split the insn, and then output the
resulting instructions separately.  This helps eliminate redundancy in the
output templates.   If you have a <code>define_insn</code> that needs to emit
multiple assembler instructions, and there is a matching <code>define_split</code>
already defined, then you can simply use <code>#</code> as the output template
instead of writing an output template that emits the multiple assembler
instructions.
</p>
<p>If the macro <code>ASSEMBLER_DIALECT</code> is defined, you can use construct
of the form &lsquo;<samp>{option0|option1|option2}</samp>&rsquo; in the templates.  These
describe multiple variants of assembler language syntax.
See <a href="#Instruction-Output">Instruction Output</a>.
</p>
<hr>
<a name="Output-Statement"></a>
<div class="header">
<p>
Next: <a href="#Predicates" accesskey="n" rel="next">Predicates</a>, Previous: <a href="#Output-Template" accesskey="p" rel="prev">Output Template</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="C-Statements-for-Assembler-Output"></a>
<h3 class="section">16.6 C Statements for Assembler Output</h3>
<a name="index-output-statements"></a>
<a name="index-C-statements-for-assembler-output"></a>
<a name="index-generating-assembler-output"></a>

<p>Often a single fixed template string cannot produce correct and efficient
assembler code for all the cases that are recognized by a single
instruction pattern.  For example, the opcodes may depend on the kinds of
operands; or some unfortunate combinations of operands may require extra
machine instructions.
</p>
<p>If the output control string starts with a &lsquo;<samp>@</samp>&rsquo;, then it is actually
a series of templates, each on a separate line.  (Blank lines and
leading spaces and tabs are ignored.)  The templates correspond to the
pattern&rsquo;s constraint alternatives (see <a href="#Multi_002dAlternative">Multi-Alternative</a>).  For example,
if a target machine has a two-address add instruction &lsquo;<samp>addr</samp>&rsquo; to add
into a register and another &lsquo;<samp>addm</samp>&rsquo; to add a register to memory, you
might write this pattern:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;addsi3&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r,m&quot;)
        (plus:SI (match_operand:SI 1 &quot;general_operand&quot; &quot;0,0&quot;)
                 (match_operand:SI 2 &quot;general_operand&quot; &quot;g,r&quot;)))]
  &quot;&quot;
  &quot;@
   addr %2,%0
   addm %2,%0&quot;)
</pre></div>

<a name="index-_002a-in-template"></a>
<a name="index-asterisk-in-template"></a>
<p>If the output control string starts with a &lsquo;<samp>*</samp>&rsquo;, then it is not an
output template but rather a piece of C program that should compute a
template.  It should execute a <code>return</code> statement to return the
template-string you want.  Most such templates use C string literals, which
require doublequote characters to delimit them.  To include these
doublequote characters in the string, prefix each one with &lsquo;<samp>\</samp>&rsquo;.
</p>
<p>If the output control string is written as a brace block instead of a
double-quoted string, it is automatically assumed to be C code.  In that
case, it is not necessary to put in a leading asterisk, or to escape the
doublequotes surrounding C string literals.
</p>
<p>The operands may be found in the array <code>operands</code>, whose C data type
is <code>rtx []</code>.
</p>
<p>It is very common to select different ways of generating assembler code
based on whether an immediate operand is within a certain range.  Be
careful when doing this, because the result of <code>INTVAL</code> is an
integer on the host machine.  If the host machine has more bits in an
<code>int</code> than the target machine has in the mode in which the constant
will be used, then some of the bits you get from <code>INTVAL</code> will be
superfluous.  For proper results, you must carefully disregard the
values of those bits.
</p>
<a name="index-output_005fasm_005finsn"></a>
<p>It is possible to output an assembler instruction and then go on to output
or compute more of them, using the subroutine <code>output_asm_insn</code>.  This
receives two arguments: a template-string and a vector of operands.  The
vector may be <code>operands</code>, or it may be another array of <code>rtx</code>
that you declare locally and initialize yourself.
</p>
<a name="index-which_005falternative"></a>
<p>When an insn pattern has multiple alternatives in its constraints, often
the appearance of the assembler code is determined mostly by which alternative
was matched.  When this is so, the C code can test the variable
<code>which_alternative</code>, which is the ordinal number of the alternative
that was actually satisfied (0 for the first, 1 for the second alternative,
etc.).
</p>
<p>For example, suppose there are two opcodes for storing zero, &lsquo;<samp>clrreg</samp>&rsquo;
for registers and &lsquo;<samp>clrmem</samp>&rsquo; for memory locations.  Here is how
a pattern could use <code>which_alternative</code> to choose between them:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r,m&quot;)
        (const_int 0))]
  &quot;&quot;
  {
  return (which_alternative == 0
          ? &quot;clrreg %0&quot; : &quot;clrmem %0&quot;);
  })
</pre></div>

<p>The example above, where the assembler code to generate was
<em>solely</em> determined by the alternative, could also have been specified
as follows, having the output control string start with a &lsquo;<samp>@</samp>&rsquo;:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r,m&quot;)
        (const_int 0))]
  &quot;&quot;
  &quot;@
   clrreg %0
   clrmem %0&quot;)
</pre></div>

<p>If you just need a little bit of C code in one (or a few) alternatives,
you can use &lsquo;<samp>*</samp>&rsquo; inside of a &lsquo;<samp>@</samp>&rsquo; multi-alternative template:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r,&lt;,m&quot;)
        (const_int 0))]
  &quot;&quot;
  &quot;@
   clrreg %0
   * return stack_mem_p (operands[0]) ? \&quot;push 0\&quot; : \&quot;clrmem %0\&quot;;
   clrmem %0&quot;)
</pre></div>

<hr>
<a name="Predicates"></a>
<div class="header">
<p>
Next: <a href="#Constraints" accesskey="n" rel="next">Constraints</a>, Previous: <a href="#Output-Statement" accesskey="p" rel="prev">Output Statement</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Predicates-1"></a>
<h3 class="section">16.7 Predicates</h3>
<a name="index-predicates"></a>
<a name="index-operand-predicates"></a>
<a name="index-operator-predicates"></a>

<p>A predicate determines whether a <code>match_operand</code> or
<code>match_operator</code> expression matches, and therefore whether the
surrounding instruction pattern will be used for that combination of
operands.  GCC has a number of machine-independent predicates, and you
can define machine-specific predicates as needed.  By convention,
predicates used with <code>match_operand</code> have names that end in
&lsquo;<samp>_operand</samp>&rsquo;, and those used with <code>match_operator</code> have names
that end in &lsquo;<samp>_operator</samp>&rsquo;.
</p>
<p>All predicates are Boolean functions (in the mathematical sense) of
two arguments: the RTL expression that is being considered at that
position in the instruction pattern, and the machine mode that the
<code>match_operand</code> or <code>match_operator</code> specifies.  In this
section, the first argument is called <var>op</var> and the second argument
<var>mode</var>.  Predicates can be called from C as ordinary two-argument
functions; this can be useful in output templates or other
machine-specific code.
</p>
<p>Operand predicates can allow operands that are not actually acceptable
to the hardware, as long as the constraints give reload the ability to
fix them up (see <a href="#Constraints">Constraints</a>).  However, GCC will usually generate
better code if the predicates specify the requirements of the machine
instructions as closely as possible.  Reload cannot fix up operands
that must be constants (&ldquo;immediate operands&rdquo;); you must use a
predicate that allows only constants, or else enforce the requirement
in the extra condition.
</p>
<a name="index-predicates-and-machine-modes"></a>
<a name="index-normal-predicates"></a>
<a name="index-special-predicates"></a>
<p>Most predicates handle their <var>mode</var> argument in a uniform manner.
If <var>mode</var> is <code>VOIDmode</code> (unspecified), then <var>op</var> can have
any mode.  If <var>mode</var> is anything else, then <var>op</var> must have the
same mode, unless <var>op</var> is a <code>CONST_INT</code> or integer
<code>CONST_DOUBLE</code>.  These RTL expressions always have
<code>VOIDmode</code>, so it would be counterproductive to check that their
mode matches.  Instead, predicates that accept <code>CONST_INT</code> and/or
integer <code>CONST_DOUBLE</code> check that the value stored in the
constant will fit in the requested mode.
</p>
<p>Predicates with this behavior are called <em>normal</em>.
<code>genrecog</code> can optimize the instruction recognizer based on
knowledge of how normal predicates treat modes.  It can also diagnose
certain kinds of common errors in the use of normal predicates; for
instance, it is almost always an error to use a normal predicate
without specifying a mode.
</p>
<p>Predicates that do something different with their <var>mode</var> argument
are called <em>special</em>.  The generic predicates
<code>address_operand</code> and <code>pmode_register_operand</code> are special
predicates.  <code>genrecog</code> does not do any optimizations or
diagnosis when special predicates are used.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Machine_002dIndependent-Predicates" accesskey="1">Machine-Independent Predicates</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Predicates available to all back ends.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Defining-Predicates" accesskey="2">Defining Predicates</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to write machine-specific predicate
                                    functions.
</td></tr>
</table>

<hr>
<a name="Machine_002dIndependent-Predicates"></a>
<div class="header">
<p>
Next: <a href="#Defining-Predicates" accesskey="n" rel="next">Defining Predicates</a>, Up: <a href="#Predicates" accesskey="u" rel="up">Predicates</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Machine_002dIndependent-Predicates-1"></a>
<h4 class="subsection">16.7.1 Machine-Independent Predicates</h4>
<a name="index-machine_002dindependent-predicates"></a>
<a name="index-generic-predicates"></a>

<p>These are the generic predicates available to all back ends.  They are
defined in <samp>recog.c</samp>.  The first category of predicates allow
only constant, or <em>immediate</em>, operands.
</p>
<dl>
<dt><a name="index-immediate_005foperand"></a>Function: <strong>immediate_operand</strong></dt>
<dd><p>This predicate allows any sort of constant that fits in <var>mode</var>.
It is an appropriate choice for instructions that take operands that
must be constant.
</p></dd></dl>

<dl>
<dt><a name="index-const_005fint_005foperand"></a>Function: <strong>const_int_operand</strong></dt>
<dd><p>This predicate allows any <code>CONST_INT</code> expression that fits in
<var>mode</var>.  It is an appropriate choice for an immediate operand that
does not allow a symbol or label.
</p></dd></dl>

<dl>
<dt><a name="index-const_005fdouble_005foperand"></a>Function: <strong>const_double_operand</strong></dt>
<dd><p>This predicate accepts any <code>CONST_DOUBLE</code> expression that has
exactly <var>mode</var>.  If <var>mode</var> is <code>VOIDmode</code>, it will also
accept <code>CONST_INT</code>.  It is intended for immediate floating point
constants.
</p></dd></dl>

<p>The second category of predicates allow only some kind of machine
register.
</p>
<dl>
<dt><a name="index-register_005foperand"></a>Function: <strong>register_operand</strong></dt>
<dd><p>This predicate allows any <code>REG</code> or <code>SUBREG</code> expression that
is valid for <var>mode</var>.  It is often suitable for arithmetic
instruction operands on a RISC machine.
</p></dd></dl>

<dl>
<dt><a name="index-pmode_005fregister_005foperand"></a>Function: <strong>pmode_register_operand</strong></dt>
<dd><p>This is a slight variant on <code>register_operand</code> which works around
a limitation in the machine-description reader.
</p>
<div class="smallexample">
<pre class="smallexample">(match_operand <var>n</var> &quot;pmode_register_operand&quot; <var>constraint</var>)
</pre></div>

<p>means exactly what
</p>
<div class="smallexample">
<pre class="smallexample">(match_operand:P <var>n</var> &quot;register_operand&quot; <var>constraint</var>)
</pre></div>

<p>would mean, if the machine-description reader accepted &lsquo;<samp>:P</samp>&rsquo;
mode suffixes.  Unfortunately, it cannot, because <code>Pmode</code> is an
alias for some other mode, and might vary with machine-specific
options.  See <a href="#Misc">Misc</a>.
</p></dd></dl>

<dl>
<dt><a name="index-scratch_005foperand"></a>Function: <strong>scratch_operand</strong></dt>
<dd><p>This predicate allows hard registers and <code>SCRATCH</code> expressions,
but not pseudo-registers.  It is used internally by <code>match_scratch</code>;
it should not be used directly.
</p></dd></dl>

<p>The third category of predicates allow only some kind of memory reference.
</p>
<dl>
<dt><a name="index-memory_005foperand"></a>Function: <strong>memory_operand</strong></dt>
<dd><p>This predicate allows any valid reference to a quantity of mode
<var>mode</var> in memory, as determined by the weak form of
<code>GO_IF_LEGITIMATE_ADDRESS</code> (see <a href="#Addressing-Modes">Addressing Modes</a>).
</p></dd></dl>

<dl>
<dt><a name="index-address_005foperand"></a>Function: <strong>address_operand</strong></dt>
<dd><p>This predicate is a little unusual; it allows any operand that is a
valid expression for the <em>address</em> of a quantity of mode
<var>mode</var>, again determined by the weak form of
<code>GO_IF_LEGITIMATE_ADDRESS</code>.  To first order, if
&lsquo;<samp>(mem:<var>mode</var>&nbsp;(<var>exp</var>))<!-- /@w --></samp>&rsquo; is acceptable to
<code>memory_operand</code>, then <var>exp</var> is acceptable to
<code>address_operand</code>.  Note that <var>exp</var> does not necessarily have
the mode <var>mode</var>.
</p></dd></dl>

<dl>
<dt><a name="index-indirect_005foperand"></a>Function: <strong>indirect_operand</strong></dt>
<dd><p>This is a stricter form of <code>memory_operand</code> which allows only
memory references with a <code>general_operand</code> as the address
expression.  New uses of this predicate are discouraged, because
<code>general_operand</code> is very permissive, so it&rsquo;s hard to tell what
an <code>indirect_operand</code> does or does not allow.  If a target has
different requirements for memory operands for different instructions,
it is better to define target-specific predicates which enforce the
hardware&rsquo;s requirements explicitly.
</p></dd></dl>

<dl>
<dt><a name="index-push_005foperand"></a>Function: <strong>push_operand</strong></dt>
<dd><p>This predicate allows a memory reference suitable for pushing a value
onto the stack.  This will be a <code>MEM</code> which refers to
<code>stack_pointer_rtx</code>, with a side-effect in its address expression
(see <a href="#Incdec">Incdec</a>); which one is determined by the
<code>STACK_PUSH_CODE</code> macro (see <a href="#Frame-Layout">Frame Layout</a>).
</p></dd></dl>

<dl>
<dt><a name="index-pop_005foperand"></a>Function: <strong>pop_operand</strong></dt>
<dd><p>This predicate allows a memory reference suitable for popping a value
off the stack.  Again, this will be a <code>MEM</code> referring to
<code>stack_pointer_rtx</code>, with a side-effect in its address
expression.  However, this time <code>STACK_POP_CODE</code> is expected.
</p></dd></dl>

<p>The fourth category of predicates allow some combination of the above
operands.
</p>
<dl>
<dt><a name="index-nonmemory_005foperand"></a>Function: <strong>nonmemory_operand</strong></dt>
<dd><p>This predicate allows any immediate or register operand valid for <var>mode</var>.
</p></dd></dl>

<dl>
<dt><a name="index-nonimmediate_005foperand"></a>Function: <strong>nonimmediate_operand</strong></dt>
<dd><p>This predicate allows any register or memory operand valid for <var>mode</var>.
</p></dd></dl>

<dl>
<dt><a name="index-general_005foperand"></a>Function: <strong>general_operand</strong></dt>
<dd><p>This predicate allows any immediate, register, or memory operand
valid for <var>mode</var>.
</p></dd></dl>

<p>Finally, there are two generic operator predicates.
</p>
<dl>
<dt><a name="index-comparison_005foperator"></a>Function: <strong>comparison_operator</strong></dt>
<dd><p>This predicate matches any expression which performs an arithmetic
comparison in <var>mode</var>; that is, <code>COMPARISON_P</code> is true for the
expression code.
</p></dd></dl>

<dl>
<dt><a name="index-ordered_005fcomparison_005foperator"></a>Function: <strong>ordered_comparison_operator</strong></dt>
<dd><p>This predicate matches any expression which performs an arithmetic
comparison in <var>mode</var> and whose expression code is valid for integer
modes; that is, the expression code will be one of <code>eq</code>, <code>ne</code>,
<code>lt</code>, <code>ltu</code>, <code>le</code>, <code>leu</code>, <code>gt</code>, <code>gtu</code>,
<code>ge</code>, <code>geu</code>.
</p></dd></dl>

<hr>
<a name="Defining-Predicates"></a>
<div class="header">
<p>
Previous: <a href="#Machine_002dIndependent-Predicates" accesskey="p" rel="prev">Machine-Independent Predicates</a>, Up: <a href="#Predicates" accesskey="u" rel="up">Predicates</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Machine_002dSpecific-Predicates"></a>
<h4 class="subsection">16.7.2 Defining Machine-Specific Predicates</h4>
<a name="index-defining-predicates"></a>
<a name="index-define_005fpredicate"></a>
<a name="index-define_005fspecial_005fpredicate"></a>

<p>Many machines have requirements for their operands that cannot be
expressed precisely using the generic predicates.  You can define
additional predicates using <code>define_predicate</code> and
<code>define_special_predicate</code> expressions.  These expressions have
three operands:
</p>
<ul>
<li> The name of the predicate, as it will be referred to in
<code>match_operand</code> or <code>match_operator</code> expressions.

</li><li> An RTL expression which evaluates to true if the predicate allows the
operand <var>op</var>, false if it does not.  This expression can only use
the following RTL codes:

<dl compact="compact">
<dt><code>MATCH_OPERAND</code></dt>
<dd><p>When written inside a predicate expression, a <code>MATCH_OPERAND</code>
expression evaluates to true if the predicate it names would allow
<var>op</var>.  The operand number and constraint are ignored.  Due to
limitations in <code>genrecog</code>, you can only refer to generic
predicates and predicates that have already been defined.
</p>
</dd>
<dt><code>MATCH_CODE</code></dt>
<dd><p>This expression evaluates to true if <var>op</var> or a specified
subexpression of <var>op</var> has one of a given list of RTX codes.
</p>
<p>The first operand of this expression is a string constant containing a
comma-separated list of RTX code names (in lower case).  These are the
codes for which the <code>MATCH_CODE</code> will be true.
</p>
<p>The second operand is a string constant which indicates what
subexpression of <var>op</var> to examine.  If it is absent or the empty
string, <var>op</var> itself is examined.  Otherwise, the string constant
must be a sequence of digits and/or lowercase letters.  Each character
indicates a subexpression to extract from the current expression; for
the first character this is <var>op</var>, for the second and subsequent
characters it is the result of the previous character.  A digit
<var>n</var> extracts &lsquo;<samp>XEXP&nbsp;(<var>e</var>,&nbsp;<var>n</var>)<!-- /@w --></samp>&rsquo;; a letter <var>l</var>
extracts &lsquo;<samp>XVECEXP&nbsp;(<var>e</var>,&nbsp;0,&nbsp;<var>n</var>)<!-- /@w --></samp>&rsquo; where <var>n</var> is the
alphabetic ordinal of <var>l</var> (0 for &lsquo;a&rsquo;, 1 for &rsquo;b&rsquo;, and so on).  The
<code>MATCH_CODE</code> then examines the RTX code of the subexpression
extracted by the complete string.  It is not possible to extract
components of an <code>rtvec</code> that is not at position 0 within its RTX
object.
</p>
</dd>
<dt><code>MATCH_TEST</code></dt>
<dd><p>This expression has one operand, a string constant containing a C
expression.  The predicate&rsquo;s arguments, <var>op</var> and <var>mode</var>, are
available with those names in the C expression.  The <code>MATCH_TEST</code>
evaluates to true if the C expression evaluates to a nonzero value.
<code>MATCH_TEST</code> expressions must not have side effects.
</p>
</dd>
<dt><code>AND</code></dt>
<dt><code>IOR</code></dt>
<dt><code>NOT</code></dt>
<dt><code>IF_THEN_ELSE</code></dt>
<dd><p>The basic &lsquo;<samp>MATCH_</samp>&rsquo; expressions can be combined using these
logical operators, which have the semantics of the C operators
&lsquo;<samp>&amp;&amp;</samp>&rsquo;, &lsquo;<samp>||</samp>&rsquo;, &lsquo;<samp>!</samp>&rsquo;, and &lsquo;<samp>?&nbsp;:<!-- /@w --></samp>&rsquo; respectively.  As
in Common Lisp, you may give an <code>AND</code> or <code>IOR</code> expression an
arbitrary number of arguments; this has exactly the same effect as
writing a chain of two-argument <code>AND</code> or <code>IOR</code> expressions.
</p></dd>
</dl>

</li><li> An optional block of C code, which should execute
&lsquo;<samp>return&nbsp;true<!-- /@w --></samp>&rsquo; if the predicate is found to match and
&lsquo;<samp>return&nbsp;false<!-- /@w --></samp>&rsquo; if it does not.  It must not have any side
effects.  The predicate arguments, <var>op</var> and <var>mode</var>, are
available with those names.

<p>If a code block is present in a predicate definition, then the RTL
expression must evaluate to true <em>and</em> the code block must
execute &lsquo;<samp>return&nbsp;true<!-- /@w --></samp>&rsquo; for the predicate to allow the operand.
The RTL expression is evaluated first; do not re-check anything in the
code block that was checked in the RTL expression.
</p></li></ul>

<p>The program <code>genrecog</code> scans <code>define_predicate</code> and
<code>define_special_predicate</code> expressions to determine which RTX
codes are possibly allowed.  You should always make this explicit in
the RTL predicate expression, using <code>MATCH_OPERAND</code> and
<code>MATCH_CODE</code>.
</p>
<p>Here is an example of a simple predicate definition, from the IA64
machine description:
</p>
<div class="smallexample">
<pre class="smallexample">;; <span class="roman">True if <var>op</var> is a <code>SYMBOL_REF</code> which refers to the sdata section.</span>
(define_predicate &quot;small_addr_symbolic_operand&quot;
  (and (match_code &quot;symbol_ref&quot;)
       (match_test &quot;SYMBOL_REF_SMALL_ADDR_P (op)&quot;)))
</pre></div>

<p>And here is another, showing the use of the C block.
</p>
<div class="smallexample">
<pre class="smallexample">;; <span class="roman">True if <var>op</var> is a register operand that is (or could be) a GR reg.</span>
(define_predicate &quot;gr_register_operand&quot;
  (match_operand 0 &quot;register_operand&quot;)
{
  unsigned int regno;
  if (GET_CODE (op) == SUBREG)
    op = SUBREG_REG (op);

  regno = REGNO (op);
  return (regno &gt;= FIRST_PSEUDO_REGISTER || GENERAL_REGNO_P (regno));
})
</pre></div>

<p>Predicates written with <code>define_predicate</code> automatically include
a test that <var>mode</var> is <code>VOIDmode</code>, or <var>op</var> has the same
mode as <var>mode</var>, or <var>op</var> is a <code>CONST_INT</code> or
<code>CONST_DOUBLE</code>.  They do <em>not</em> check specifically for
integer <code>CONST_DOUBLE</code>, nor do they test that the value of either
kind of constant fits in the requested mode.  This is because
target-specific predicates that take constants usually have to do more
stringent value checks anyway.  If you need the exact same treatment
of <code>CONST_INT</code> or <code>CONST_DOUBLE</code> that the generic predicates
provide, use a <code>MATCH_OPERAND</code> subexpression to call
<code>const_int_operand</code>, <code>const_double_operand</code>, or
<code>immediate_operand</code>.
</p>
<p>Predicates written with <code>define_special_predicate</code> do not get any
automatic mode checks, and are treated as having special mode handling
by <code>genrecog</code>.
</p>
<p>The program <code>genpreds</code> is responsible for generating code to
test predicates.  It also writes a header file containing function
declarations for all machine-specific predicates.  It is not necessary
to declare these predicates in <samp><var>cpu</var>-protos.h</samp>.
</p>
<hr>
<a name="Constraints"></a>
<div class="header">
<p>
Next: <a href="#Standard-Names" accesskey="n" rel="next">Standard Names</a>, Previous: <a href="#Predicates" accesskey="p" rel="prev">Predicates</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Operand-Constraints"></a>
<h3 class="section">16.8 Operand Constraints</h3>
<a name="index-operand-constraints"></a>
<a name="index-constraints"></a>

<p>Each <code>match_operand</code> in an instruction pattern can specify
constraints for the operands allowed.  The constraints allow you to
fine-tune matching within the set of operands allowed by the
predicate.
</p>
<p>Constraints can say whether
an operand may be in a register, and which kinds of register; whether the
operand can be a memory reference, and which kinds of address; whether the
operand may be an immediate constant, and which possible values it may
have.  Constraints can also require two operands to match.
Side-effects aren&rsquo;t allowed in operands of inline <code>asm</code>, unless
&lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo; constraints are used, because there is no guarantee
that the side-effects will happen exactly once in an instruction that can update
the addressing register.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Simple-Constraints" accesskey="1">Simple Constraints</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Basic use of constraints.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Multi_002dAlternative" accesskey="2">Multi-Alternative</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">When an insn has two alternative constraint-patterns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Class-Preferences" accesskey="3">Class Preferences</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Constraints guide which hard register to put things in.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Modifiers" accesskey="4">Modifiers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">More precise control over effects of constraints.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Machine-Constraints" accesskey="5">Machine Constraints</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Existing constraints for some particular machines.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Disable-Insn-Alternatives" accesskey="6">Disable Insn Alternatives</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Disable insn alternatives using the <code>enabled</code> attribute.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Define-Constraints" accesskey="7">Define Constraints</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to define machine-specific constraints.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#C-Constraint-Interface" accesskey="8">C Constraint Interface</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to test constraints from C code.
</td></tr>
</table>


<hr>
<a name="Simple-Constraints"></a>
<div class="header">
<p>
Next: <a href="#Multi_002dAlternative" accesskey="n" rel="next">Multi-Alternative</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Simple-Constraints-1"></a>
<h4 class="subsection">16.8.1 Simple Constraints</h4>
<a name="index-simple-constraints"></a>

<p>The simplest kind of constraint is a string full of letters, each of
which describes one kind of operand that is permitted.  Here are
the letters that are allowed:
</p>
<dl compact="compact">
<dt>whitespace</dt>
<dd><p>Whitespace characters are ignored and can be inserted at any position
except the first.  This enables each alternative for different operands to
be visually aligned in the machine description even if they have different
number of constraints and modifiers.
</p>
<a name="index-m-in-constraint"></a>
<a name="index-memory-references-in-constraints"></a>
</dd>
<dt>&lsquo;<samp>m</samp>&rsquo;</dt>
<dd><p>A memory operand is allowed, with any kind of address that the machine
supports in general.
Note that the letter used for the general memory constraint can be
re-defined by a back end using the <code>TARGET_MEM_CONSTRAINT</code> macro.
</p>
<a name="index-offsettable-address"></a>
<a name="index-o-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>o</samp>&rsquo;</dt>
<dd><p>A memory operand is allowed, but only if the address is
<em>offsettable</em>.  This means that adding a small integer (actually,
the width in bytes of the operand, as determined by its machine mode)
may be added to the address and the result is also a valid memory
address.
</p>
<a name="index-autoincrement_002fdecrement-addressing"></a>
<p>For example, an address which is constant is offsettable; so is an
address that is the sum of a register and a constant (as long as a
slightly larger constant is also within the range of address-offsets
supported by the machine); but an autoincrement or autodecrement
address is not offsettable.  More complicated indirect/indexed
addresses may or may not be offsettable depending on the other
addressing modes that the machine supports.
</p>
<p>Note that in an output operand which can be matched by another
operand, the constraint letter &lsquo;<samp>o</samp>&rsquo; is valid only when accompanied
by both &lsquo;<samp>&lt;</samp>&rsquo; (if the target machine has predecrement addressing)
and &lsquo;<samp>&gt;</samp>&rsquo; (if the target machine has preincrement addressing).
</p>
<a name="index-V-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>V</samp>&rsquo;</dt>
<dd><p>A memory operand that is not offsettable.  In other words, anything that
would fit the &lsquo;<samp>m</samp>&rsquo; constraint but not the &lsquo;<samp>o</samp>&rsquo; constraint.
</p>
<a name="index-_003c-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>&lt;</samp>&rsquo;</dt>
<dd><p>A memory operand with autodecrement addressing (either predecrement or
postdecrement) is allowed.  In inline <code>asm</code> this constraint is only
allowed if the operand is used exactly once in an instruction that can
handle the side-effects.  Not using an operand with &lsquo;<samp>&lt;</samp>&rsquo; in constraint
string in the inline <code>asm</code> pattern at all or using it in multiple
instructions isn&rsquo;t valid, because the side-effects wouldn&rsquo;t be performed
or would be performed more than once.  Furthermore, on some targets
the operand with &lsquo;<samp>&lt;</samp>&rsquo; in constraint string must be accompanied by
special instruction suffixes like <code>%U0</code> instruction suffix on PowerPC
or <code>%P0</code> on IA-64.
</p>
<a name="index-_003e-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>&gt;</samp>&rsquo;</dt>
<dd><p>A memory operand with autoincrement addressing (either preincrement or
postincrement) is allowed.  In inline <code>asm</code> the same restrictions
as for &lsquo;<samp>&lt;</samp>&rsquo; apply.
</p>
<a name="index-r-in-constraint"></a>
<a name="index-registers-in-constraints"></a>
</dd>
<dt>&lsquo;<samp>r</samp>&rsquo;</dt>
<dd><p>A register operand is allowed provided that it is in a general
register.
</p>
<a name="index-constants-in-constraints"></a>
<a name="index-i-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>i</samp>&rsquo;</dt>
<dd><p>An immediate integer operand (one with constant value) is allowed.
This includes symbolic constants whose values will be known only at
assembly time or later.
</p>
<a name="index-n-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>n</samp>&rsquo;</dt>
<dd><p>An immediate integer operand with a known numeric value is allowed.
Many systems cannot support assembly-time constants for operands less
than a word wide.  Constraints for these operands should use &lsquo;<samp>n</samp>&rsquo;
rather than &lsquo;<samp>i</samp>&rsquo;.
</p>
<a name="index-I-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>I</samp>&rsquo;, &lsquo;<samp>J</samp>&rsquo;, &lsquo;<samp>K</samp>&rsquo;, &hellip; &lsquo;<samp>P</samp>&rsquo;</dt>
<dd><p>Other letters in the range &lsquo;<samp>I</samp>&rsquo; through &lsquo;<samp>P</samp>&rsquo; may be defined in
a machine-dependent fashion to permit immediate integer operands with
explicit integer values in specified ranges.  For example, on the
68000, &lsquo;<samp>I</samp>&rsquo; is defined to stand for the range of values 1 to 8.
This is the range permitted as a shift count in the shift
instructions.
</p>
<a name="index-E-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>E</samp>&rsquo;</dt>
<dd><p>An immediate floating operand (expression code <code>const_double</code>) is
allowed, but only if the target floating point format is the same as
that of the host machine (on which the compiler is running).
</p>
<a name="index-F-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>F</samp>&rsquo;</dt>
<dd><p>An immediate floating operand (expression code <code>const_double</code> or
<code>const_vector</code>) is allowed.
</p>
<a name="index-G-in-constraint"></a>
<a name="index-H-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>G</samp>&rsquo;, &lsquo;<samp>H</samp>&rsquo;</dt>
<dd><p>&lsquo;<samp>G</samp>&rsquo; and &lsquo;<samp>H</samp>&rsquo; may be defined in a machine-dependent fashion to
permit immediate floating operands in particular ranges of values.
</p>
<a name="index-s-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>s</samp>&rsquo;</dt>
<dd><p>An immediate integer operand whose value is not an explicit integer is
allowed.
</p>
<p>This might appear strange; if an insn allows a constant operand with a
value not known at compile time, it certainly must allow any known
value.  So why use &lsquo;<samp>s</samp>&rsquo; instead of &lsquo;<samp>i</samp>&rsquo;?  Sometimes it allows
better code to be generated.
</p>
<p>For example, on the 68000 in a fullword instruction it is possible to
use an immediate operand; but if the immediate value is between -128
and 127, better code results from loading the value into a register and
using the register.  This is because the load into the register can be
done with a &lsquo;<samp>moveq</samp>&rsquo; instruction.  We arrange for this to happen
by defining the letter &lsquo;<samp>K</samp>&rsquo; to mean &ldquo;any integer outside the
range -128 to 127&rdquo;, and then specifying &lsquo;<samp>Ks</samp>&rsquo; in the operand
constraints.
</p>
<a name="index-g-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>g</samp>&rsquo;</dt>
<dd><p>Any register, memory or immediate integer operand is allowed, except for
registers that are not general registers.
</p>
<a name="index-X-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>X</samp>&rsquo;</dt>
<dd><p>Any operand whatsoever is allowed, even if it does not satisfy
<code>general_operand</code>.  This is normally used in the constraint of
a <code>match_scratch</code> when certain alternatives will not actually
require a scratch register.
</p>
<a name="index-0-in-constraint"></a>
<a name="index-digits-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>0</samp>&rsquo;, &lsquo;<samp>1</samp>&rsquo;, &lsquo;<samp>2</samp>&rsquo;, &hellip; &lsquo;<samp>9</samp>&rsquo;</dt>
<dd><p>An operand that matches the specified operand number is allowed.  If a
digit is used together with letters within the same alternative, the
digit should come last.
</p>
<p>This number is allowed to be more than a single digit.  If multiple
digits are encountered consecutively, they are interpreted as a single
decimal integer.  There is scant chance for ambiguity, since to-date
it has never been desirable that &lsquo;<samp>10</samp>&rsquo; be interpreted as matching
either operand 1 <em>or</em> operand 0.  Should this be desired, one
can use multiple alternatives instead.
</p>
<a name="index-matching-constraint"></a>
<a name="index-constraint_002c-matching"></a>
<p>This is called a <em>matching constraint</em> and what it really means is
that the assembler has only a single operand that fills two roles
considered separate in the RTL insn.  For example, an add insn has two
input operands and one output operand in the RTL, but on most CISC
machines an add instruction really has only two operands, one of them an
input-output operand:
</p>
<div class="smallexample">
<pre class="smallexample">addl #35,r12
</pre></div>

<p>Matching constraints are used in these circumstances.
More precisely, the two operands that match must include one input-only
operand and one output-only operand.  Moreover, the digit must be a
smaller number than the number of the operand that uses it in the
constraint.
</p>
<p>For operands to match in a particular case usually means that they
are identical-looking RTL expressions.  But in a few special cases
specific kinds of dissimilarity are allowed.  For example, <code>*x</code>
as an input operand will match <code>*x++</code> as an output operand.
For proper results in such cases, the output template should always
use the output-operand&rsquo;s number when printing the operand.
</p>
<a name="index-load-address-instruction"></a>
<a name="index-push-address-instruction"></a>
<a name="index-address-constraints"></a>
<a name="index-p-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>p</samp>&rsquo;</dt>
<dd><p>An operand that is a valid memory address is allowed.  This is
for &ldquo;load address&rdquo; and &ldquo;push address&rdquo; instructions.
</p>
<a name="index-address_005foperand-1"></a>
<p>&lsquo;<samp>p</samp>&rsquo; in the constraint must be accompanied by <code>address_operand</code>
as the predicate in the <code>match_operand</code>.  This predicate interprets
the mode specified in the <code>match_operand</code> as the mode of the memory
reference for which the address would be valid.
</p>
<a name="index-other-register-constraints"></a>
<a name="index-extensible-constraints"></a>
</dd>
<dt><var>other-letters</var></dt>
<dd><p>Other letters can be defined in machine-dependent fashion to stand for
particular classes of registers or other arbitrary operand types.
&lsquo;<samp>d</samp>&rsquo;, &lsquo;<samp>a</samp>&rsquo; and &lsquo;<samp>f</samp>&rsquo; are defined on the 68000/68020 to stand
for data, address and floating point registers.
</p></dd>
</dl>

<p>In order to have valid assembler code, each operand must satisfy
its constraint.  But a failure to do so does not prevent the pattern
from applying to an insn.  Instead, it directs the compiler to modify
the code so that the constraint will be satisfied.  Usually this is
done by copying an operand into a register.
</p>
<p>Contrast, therefore, the two instruction patterns that follow:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r&quot;)
        (plus:SI (match_dup 0)
                 (match_operand:SI 1 &quot;general_operand&quot; &quot;r&quot;)))]
  &quot;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>which has two operands, one of which must appear in two places, and
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r&quot;)
        (plus:SI (match_operand:SI 1 &quot;general_operand&quot; &quot;0&quot;)
                 (match_operand:SI 2 &quot;general_operand&quot; &quot;r&quot;)))]
  &quot;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>which has three operands, two of which are required by a constraint to be
identical.  If we are considering an insn of the form
</p>
<div class="smallexample">
<pre class="smallexample">(insn <var>n</var> <var>prev</var> <var>next</var>
  (set (reg:SI 3)
       (plus:SI (reg:SI 6) (reg:SI 109)))
  &hellip;)
</pre></div>

<p>the first pattern would not apply at all, because this insn does not
contain two identical subexpressions in the right place.  The pattern would
say, &ldquo;That does not look like an add instruction; try other patterns&rdquo;.
The second pattern would say, &ldquo;Yes, that&rsquo;s an add instruction, but there
is something wrong with it&rdquo;.  It would direct the reload pass of the
compiler to generate additional insns to make the constraint true.  The
results might look like this:
</p>
<div class="smallexample">
<pre class="smallexample">(insn <var>n2</var> <var>prev</var> <var>n</var>
  (set (reg:SI 3) (reg:SI 6))
  &hellip;)

(insn <var>n</var> <var>n2</var> <var>next</var>
  (set (reg:SI 3)
       (plus:SI (reg:SI 3) (reg:SI 109)))
  &hellip;)
</pre></div>

<p>It is up to you to make sure that each operand, in each pattern, has
constraints that can handle any RTL expression that could be present for
that operand.  (When multiple alternatives are in use, each pattern must,
for each possible combination of operand expressions, have at least one
alternative which can handle that combination of operands.)  The
constraints don&rsquo;t need to <em>allow</em> any possible operand&mdash;when this is
the case, they do not constrain&mdash;but they must at least point the way to
reloading any possible operand so that it will fit.
</p>
<ul>
<li> If the constraint accepts whatever operands the predicate permits,
there is no problem: reloading is never necessary for this operand.

<p>For example, an operand whose constraints permit everything except
registers is safe provided its predicate rejects registers.
</p>
<p>An operand whose predicate accepts only constant values is safe
provided its constraints include the letter &lsquo;<samp>i</samp>&rsquo;.  If any possible
constant value is accepted, then nothing less than &lsquo;<samp>i</samp>&rsquo; will do;
if the predicate is more selective, then the constraints may also be
more selective.
</p>
</li><li> Any operand expression can be reloaded by copying it into a register.
So if an operand&rsquo;s constraints allow some kind of register, it is
certain to be safe.  It need not permit all classes of registers; the
compiler knows how to copy a register into another register of the
proper class in order to make an instruction valid.

</li><li> <a name="index-nonoffsettable-memory-reference"></a>
<a name="index-memory-reference_002c-nonoffsettable"></a>
A nonoffsettable memory reference can be reloaded by copying the
address into a register.  So if the constraint uses the letter
&lsquo;<samp>o</samp>&rsquo;, all memory references are taken care of.

</li><li> A constant operand can be reloaded by allocating space in memory to
hold it as preinitialized data.  Then the memory reference can be used
in place of the constant.  So if the constraint uses the letters
&lsquo;<samp>o</samp>&rsquo; or &lsquo;<samp>m</samp>&rsquo;, constant operands are not a problem.

</li><li> If the constraint permits a constant and a pseudo register used in an insn
was not allocated to a hard register and is equivalent to a constant,
the register will be replaced with the constant.  If the predicate does
not permit a constant and the insn is re-recognized for some reason, the
compiler will crash.  Thus the predicate must always recognize any
objects allowed by the constraint.
</li></ul>

<p>If the operand&rsquo;s predicate can recognize registers, but the constraint does
not permit them, it can make the compiler crash.  When this operand happens
to be a register, the reload pass will be stymied, because it does not know
how to copy a register temporarily into memory.
</p>
<p>If the predicate accepts a unary operator, the constraint applies to the
operand.  For example, the MIPS processor at ISA level 3 supports an
instruction which adds two registers in <code>SImode</code> to produce a
<code>DImode</code> result, but only if the registers are correctly sign
extended.  This predicate for the input operands accepts a
<code>sign_extend</code> of an <code>SImode</code> register.  Write the constraint
to indicate the type of register that is required for the operand of the
<code>sign_extend</code>.
</p>
<hr>
<a name="Multi_002dAlternative"></a>
<div class="header">
<p>
Next: <a href="#Class-Preferences" accesskey="n" rel="next">Class Preferences</a>, Previous: <a href="#Simple-Constraints" accesskey="p" rel="prev">Simple Constraints</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Multiple-Alternative-Constraints"></a>
<h4 class="subsection">16.8.2 Multiple Alternative Constraints</h4>
<a name="index-multiple-alternative-constraints"></a>

<p>Sometimes a single instruction has multiple alternative sets of possible
operands.  For example, on the 68000, a logical-or instruction can combine
register or an immediate value into memory, or it can combine any kind of
operand into a register; but it cannot combine one memory location into
another.
</p>
<p>These constraints are represented as multiple alternatives.  An alternative
can be described by a series of letters for each operand.  The overall
constraint for an operand is made from the letters for this operand
from the first alternative, a comma, the letters for this operand from
the second alternative, a comma, and so on until the last alternative.
Here is how it is done for fullword logical-or on the 68000:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;iorsi3&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=m,d&quot;)
        (ior:SI (match_operand:SI 1 &quot;general_operand&quot; &quot;%0,0&quot;)
                (match_operand:SI 2 &quot;general_operand&quot; &quot;dKs,dmKs&quot;)))]
  &hellip;)
</pre></div>

<p>The first alternative has &lsquo;<samp>m</samp>&rsquo; (memory) for operand 0, &lsquo;<samp>0</samp>&rsquo; for
operand 1 (meaning it must match operand 0), and &lsquo;<samp>dKs</samp>&rsquo; for operand
2.  The second alternative has &lsquo;<samp>d</samp>&rsquo; (data register) for operand 0,
&lsquo;<samp>0</samp>&rsquo; for operand 1, and &lsquo;<samp>dmKs</samp>&rsquo; for operand 2.  The &lsquo;<samp>=</samp>&rsquo; and
&lsquo;<samp>%</samp>&rsquo; in the constraints apply to all the alternatives; their
meaning is explained in the next section (see <a href="#Class-Preferences">Class Preferences</a>).
</p>
<p>If all the operands fit any one alternative, the instruction is valid.
Otherwise, for each alternative, the compiler counts how many instructions
must be added to copy the operands so that that alternative applies.
The alternative requiring the least copying is chosen.  If two alternatives
need the same amount of copying, the one that comes first is chosen.
These choices can be altered with the &lsquo;<samp>?</samp>&rsquo; and &lsquo;<samp>!</samp>&rsquo; characters:
</p>
<dl compact="compact">
<dd><a name="index-_003f-in-constraint"></a>
<a name="index-question-mark"></a>
</dd>
<dt><code>?</code></dt>
<dd><p>Disparage slightly the alternative that the &lsquo;<samp>?</samp>&rsquo; appears in,
as a choice when no alternative applies exactly.  The compiler regards
this alternative as one unit more costly for each &lsquo;<samp>?</samp>&rsquo; that appears
in it.
</p>
<a name="index-_0021-in-constraint"></a>
<a name="index-exclamation-point"></a>
</dd>
<dt><code>!</code></dt>
<dd><p>Disparage severely the alternative that the &lsquo;<samp>!</samp>&rsquo; appears in.
This alternative can still be used if it fits without reloading,
but if reloading is needed, some other alternative will be used.
</p></dd>
</dl>

<p>When an insn pattern has multiple alternatives in its constraints, often
the appearance of the assembler code is determined mostly by which
alternative was matched.  When this is so, the C code for writing the
assembler code can use the variable <code>which_alternative</code>, which is
the ordinal number of the alternative that was actually satisfied (0 for
the first, 1 for the second alternative, etc.).  See <a href="#Output-Statement">Output Statement</a>.
</p>
<hr>
<a name="Class-Preferences"></a>
<div class="header">
<p>
Next: <a href="#Modifiers" accesskey="n" rel="next">Modifiers</a>, Previous: <a href="#Multi_002dAlternative" accesskey="p" rel="prev">Multi-Alternative</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Register-Class-Preferences"></a>
<h4 class="subsection">16.8.3 Register Class Preferences</h4>
<a name="index-class-preference-constraints"></a>
<a name="index-register-class-preference-constraints"></a>

<a name="index-voting-between-constraint-alternatives"></a>
<p>The operand constraints have another function: they enable the compiler
to decide which kind of hardware register a pseudo register is best
allocated to.  The compiler examines the constraints that apply to the
insns that use the pseudo register, looking for the machine-dependent
letters such as &lsquo;<samp>d</samp>&rsquo; and &lsquo;<samp>a</samp>&rsquo; that specify classes of registers.
The pseudo register is put in whichever class gets the most &ldquo;votes&rdquo;.
The constraint letters &lsquo;<samp>g</samp>&rsquo; and &lsquo;<samp>r</samp>&rsquo; also vote: they vote in
favor of a general register.  The machine description says which registers
are considered general.
</p>
<p>Of course, on some machines all registers are equivalent, and no register
classes are defined.  Then none of this complexity is relevant.
</p>
<hr>
<a name="Modifiers"></a>
<div class="header">
<p>
Next: <a href="#Machine-Constraints" accesskey="n" rel="next">Machine Constraints</a>, Previous: <a href="#Class-Preferences" accesskey="p" rel="prev">Class Preferences</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constraint-Modifier-Characters"></a>
<h4 class="subsection">16.8.4 Constraint Modifier Characters</h4>
<a name="index-modifiers-in-constraints"></a>
<a name="index-constraint-modifier-characters"></a>

<p>Here are constraint modifier characters.
</p>
<dl compact="compact">
<dd><a name="index-_003d-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>=</samp>&rsquo;</dt>
<dd><p>Means that this operand is write-only for this instruction: the previous
value is discarded and replaced by output data.
</p>
<a name="index-_002b-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>+</samp>&rsquo;</dt>
<dd><p>Means that this operand is both read and written by the instruction.
</p>
<p>When the compiler fixes up the operands to satisfy the constraints,
it needs to know which operands are inputs to the instruction and
which are outputs from it.  &lsquo;<samp>=</samp>&rsquo; identifies an output; &lsquo;<samp>+</samp>&rsquo;
identifies an operand that is both input and output; all other operands
are assumed to be input only.
</p>
<p>If you specify &lsquo;<samp>=</samp>&rsquo; or &lsquo;<samp>+</samp>&rsquo; in a constraint, you put it in the
first character of the constraint string.
</p>
<a name="index-_0026-in-constraint"></a>
<a name="index-earlyclobber-operand"></a>
</dd>
<dt>&lsquo;<samp>&amp;</samp>&rsquo;</dt>
<dd><p>Means (in a particular alternative) that this operand is an
<em>earlyclobber</em> operand, which is modified before the instruction is
finished using the input operands.  Therefore, this operand may not lie
in a register that is used as an input operand or as part of any memory
address.
</p>
<p>&lsquo;<samp>&amp;</samp>&rsquo; applies only to the alternative in which it is written.  In
constraints with multiple alternatives, sometimes one alternative
requires &lsquo;<samp>&amp;</samp>&rsquo; while others do not.  See, for example, the
&lsquo;<samp>movdf</samp>&rsquo; insn of the 68000.
</p>
<p>An input operand can be tied to an earlyclobber operand if its only
use as an input occurs before the early result is written.  Adding
alternatives of this form often allows GCC to produce better code
when only some of the inputs can be affected by the earlyclobber.
See, for example, the &lsquo;<samp>mulsi3</samp>&rsquo; insn of the ARM.
</p>
<p>&lsquo;<samp>&amp;</samp>&rsquo; does not obviate the need to write &lsquo;<samp>=</samp>&rsquo;.
</p>
<a name="index-_0025-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>%</samp>&rsquo;</dt>
<dd><p>Declares the instruction to be commutative for this operand and the
following operand.  This means that the compiler may interchange the
two operands if that is the cheapest way to make all operands fit the
constraints.
This is often used in patterns for addition instructions
that really have only two operands: the result must go in one of the
arguments.  Here for example, is how the 68000 halfword-add
instruction is defined:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;addhi3&quot;
  [(set (match_operand:HI 0 &quot;general_operand&quot; &quot;=m,r&quot;)
     (plus:HI (match_operand:HI 1 &quot;general_operand&quot; &quot;%0,0&quot;)
              (match_operand:HI 2 &quot;general_operand&quot; &quot;di,g&quot;)))]
  &hellip;)
</pre></div>
<p>GCC can only handle one commutative pair in an asm; if you use more,
the compiler may fail.  Note that you need not use the modifier if
the two alternatives are strictly identical; this would only waste
time in the reload pass.  The modifier is not operational after
register allocation, so the result of <code>define_peephole2</code>
and <code>define_split</code>s performed after reload cannot rely on
&lsquo;<samp>%</samp>&rsquo; to make the intended insn match.
</p>
<a name="index-_0023-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>#</samp>&rsquo;</dt>
<dd><p>Says that all following characters, up to the next comma, are to be
ignored as a constraint.  They are significant only for choosing
register preferences.
</p>
<a name="index-_002a-in-constraint"></a>
</dd>
<dt>&lsquo;<samp>*</samp>&rsquo;</dt>
<dd><p>Says that the following character should be ignored when choosing
register preferences.  &lsquo;<samp>*</samp>&rsquo; has no effect on the meaning of the
constraint as a constraint, and no effect on reloading.  For LRA
&lsquo;<samp>*</samp>&rsquo; additionally disparages slightly the alternative if the
following character matches the operand.
</p>
<p>Here is an example: the 68000 has an instruction to sign-extend a
halfword in a data register, and can also sign-extend a value by
copying it into an address register.  While either kind of register is
acceptable, the constraints on an address-register destination are
less strict, so it is best if register allocation makes an address
register its goal.  Therefore, &lsquo;<samp>*</samp>&rsquo; is used so that the &lsquo;<samp>d</samp>&rsquo;
constraint letter (for data register) is ignored when computing
register preferences.
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;extendhisi2&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=*d,a&quot;)
        (sign_extend:SI
         (match_operand:HI 1 &quot;general_operand&quot; &quot;0,g&quot;)))]
  &hellip;)
</pre></div>
</dd>
</dl>

<hr>
<a name="Machine-Constraints"></a>
<div class="header">
<p>
Next: <a href="#Disable-Insn-Alternatives" accesskey="n" rel="next">Disable Insn Alternatives</a>, Previous: <a href="#Modifiers" accesskey="p" rel="prev">Modifiers</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constraints-for-Particular-Machines"></a>
<h4 class="subsection">16.8.5 Constraints for Particular Machines</h4>
<a name="index-machine-specific-constraints"></a>
<a name="index-constraints_002c-machine-specific"></a>

<p>Whenever possible, you should use the general-purpose constraint letters
in <code>asm</code> arguments, since they will convey meaning more readily to
people reading your code.  Failing that, use the constraint letters
that usually have very similar meanings across architectures.  The most
commonly used constraints are &lsquo;<samp>m</samp>&rsquo; and &lsquo;<samp>r</samp>&rsquo; (for memory and
general-purpose registers respectively; see <a href="#Simple-Constraints">Simple Constraints</a>), and
&lsquo;<samp>I</samp>&rsquo;, usually the letter indicating the most common
immediate-constant format.
</p>
<p>Each architecture defines additional constraints.  These constraints
are used by the compiler itself for instruction generation, as well as
for <code>asm</code> statements; therefore, some of the constraints are not
particularly useful for <code>asm</code>.  Here is a summary of some of the
machine-dependent constraints available on some particular machines;
it includes both constraints that are useful for <code>asm</code> and
constraints that aren&rsquo;t.  The compiler source file mentioned in the
table heading for each architecture is the definitive reference for
the meanings of that architecture&rsquo;s constraints.
</p>
<dl compact="compact">
<dt><em>AArch64 family&mdash;<samp>config/aarch64/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>k</code></dt>
<dd><p>The stack pointer register (<code>SP</code>)
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>Floating point or SIMD vector register
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Integer constant that is valid as an immediate operand in an <code>ADD</code>
instruction
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Integer constant that is valid as an immediate operand in a <code>SUB</code>
instruction (once negated)
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Integer constant that can be used with a 32-bit logical instruction
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Integer constant that can be used with a 64-bit logical instruction
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Integer constant that is valid as an immediate operand in a 32-bit <code>MOV</code>
pseudo instruction. The <code>MOV</code> may be assembled to one of several different
machine instructions depending on the value
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Integer constant that is valid as an immediate operand in a 64-bit <code>MOV</code>
pseudo instruction
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>An absolute symbolic address or a label reference
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>Floating point constant zero
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Integer constant zero
</p>
</dd>
<dt><code>Usa</code></dt>
<dd><p>An absolute symbolic address
</p>
</dd>
<dt><code>Ush</code></dt>
<dd><p>The high part (bits 12 and upwards) of the pc-relative address of a symbol
within 4GB of the instruction
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory address which uses a single base register with no offset
</p>
</dd>
<dt><code>Ump</code></dt>
<dd><p>A memory address suitable for a load/store pair instruction in SI, DI, SF and
DF modes
</p>
</dd>
</dl>


</dd>
<dt><em>ARM family&mdash;<samp>config/arm/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>w</code></dt>
<dd><p>VFP floating-point register
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>The floating-point constant 0.0
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Integer that is valid as an immediate operand in a data processing
instruction.  That is, an integer in the range 0 to 255 rotated by a
multiple of 2
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Integer in the range -4095 to 4095
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Integer that satisfies constraint &lsquo;<samp>I</samp>&rsquo; when inverted (ones complement)
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Integer that satisfies constraint &lsquo;<samp>I</samp>&rsquo; when negated (twos complement)
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Integer in the range 0 to 32
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory reference where the exact address is in a single register
(&lsquo;&lsquo;<samp>m</samp>&rsquo;&rsquo; is preferable for <code>asm</code> statements)
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>An item in the constant pool
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>A symbol in the text segment of the current file
</p>
</dd>
<dt><code>Uv</code></dt>
<dd><p>A memory reference suitable for VFP load/store insns (reg+constant offset)
</p>
</dd>
<dt><code>Uy</code></dt>
<dd><p>A memory reference suitable for iWMMXt load/store instructions.
</p>
</dd>
<dt><code>Uq</code></dt>
<dd><p>A memory reference suitable for the ARMv4 ldrsb instruction.
</p></dd>
</dl>

</dd>
<dt><em>AVR family&mdash;<samp>config/avr/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>l</code></dt>
<dd><p>Registers from r0 to r15
</p>
</dd>
<dt><code>a</code></dt>
<dd><p>Registers from r16 to r23
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Registers from r16 to r31
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>Registers from r24 to r31.  These registers can be used in &lsquo;<samp>adiw</samp>&rsquo; command
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Pointer register (r26&ndash;r31)
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Base pointer register (r28&ndash;r31)
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>Stack pointer register (SPH:SPL)
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Temporary register r0
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>Register pair X (r27:r26)
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>Register pair Y (r29:r28)
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>Register pair Z (r31:r30)
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Constant greater than -1, less than 64
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Constant greater than -64, less than 1
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Constant integer 2
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Constant integer 0
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Constant that fits in 8 bits
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Constant integer -1
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Constant integer 8, 16, or 24
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Constant integer 1
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>A floating point constant 0.0
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory address based on Y or Z pointer with displacement.
</p></dd>
</dl>

</dd>
<dt><em>Epiphany&mdash;<samp>config/epiphany/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>U16</code></dt>
<dd><p>An unsigned 16-bit constant.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>An unsigned 5-bit constant.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>A signed 11-bit constant.
</p>
</dd>
<dt><code>Cm1</code></dt>
<dd><p>A signed 11-bit constant added to -1.
Can only match when the <samp>-m1reg-<var>reg</var></samp> option is active.
</p>
</dd>
<dt><code>Cl1</code></dt>
<dd><p>Left-shift of -1, i.e., a bit mask with a block of leading ones, the rest
being a block of trailing zeroes.
Can only match when the <samp>-m1reg-<var>reg</var></samp> option is active.
</p>
</dd>
<dt><code>Cr1</code></dt>
<dd><p>Right-shift of -1, i.e., a bit mask with a trailing block of ones, the
rest being zeroes.  Or to put it another way, one less than a power of two.
Can only match when the <samp>-m1reg-<var>reg</var></samp> option is active.
</p>
</dd>
<dt><code>Cal</code></dt>
<dd><p>Constant for arithmetic/logical operations.
This is like <code>i</code>, except that for position independent code,
no symbols / expressions needing relocations are allowed.
</p>
</dd>
<dt><code>Csy</code></dt>
<dd><p>Symbolic constant for call/jump instruction.
</p>
</dd>
<dt><code>Rcs</code></dt>
<dd><p>The register class usable in short insns.  This is a register class
constraint, and can thus drive register allocation.
This constraint won&rsquo;t match unless <samp>-mprefer-short-insn-regs</samp> is
in effect.
</p>
</dd>
<dt><code>Rsc</code></dt>
<dd><p>The the register class of registers that can be used to hold a
sibcall call address.  I.e., a caller-saved register.
</p>
</dd>
<dt><code>Rct</code></dt>
<dd><p>Core control register class.
</p>
</dd>
<dt><code>Rgs</code></dt>
<dd><p>The register group usable in short insns.
This constraint does not use a register class, so that it only
passively matches suitable registers, and doesn&rsquo;t drive register allocation.
</p>
</dd>
<dt><code>Car</code></dt>
<dd><p>Constant suitable for the addsi3_r pattern.  This is a valid offset
For byte, halfword, or word addressing.
</p>
</dd>
<dt><code>Rra</code></dt>
<dd><p>Matches the return address if it can be replaced with the link register.
</p>
</dd>
<dt><code>Rcc</code></dt>
<dd><p>Matches the integer condition code register.
</p>
</dd>
<dt><code>Sra</code></dt>
<dd><p>Matches the return address if it is in a stack slot.
</p>
</dd>
<dt><code>Cfm</code></dt>
<dd><p>Matches control register values to switch fp mode, which are encapsulated in
<code>UNSPEC_FP_MODE</code>.
</p></dd>
</dl>

</dd>
<dt><em>CR16 Architecture&mdash;<samp>config/cr16/cr16.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>b</code></dt>
<dd><p>Registers from r0 to r14 (registers without stack pointer)
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Register from r0 to r11 (all 16-bit registers)
</p>
</dd>
<dt><code>p</code></dt>
<dd><p>Register from r12 to r15 (all 32-bit registers)
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed constant that fits in 4 bits
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Signed constant that fits in 5 bits
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Signed constant that fits in 6 bits
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Unsigned constant that fits in 4 bits
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Signed constant that fits in 32 bits
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Check for 64 bits wide constants for add/sub instructions
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating point constant that is legal for store immediate
</p></dd>
</dl>

</dd>
<dt><em>Hewlett-Packard PA-RISC&mdash;<samp>config/pa/pa.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>General register 1
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Floating point register
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>Shift amount register
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>Floating point register (deprecated)
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>Upper floating point register (32-bit), floating point register (64-bit)
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Any register
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 11-bit integer constant
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Signed 14-bit integer constant
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Integer constant that can be deposited with a <code>zdepi</code> instruction
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Signed 5-bit integer constant
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Integer constant 0
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Integer constant that can be loaded with a <code>ldil</code> instruction
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Integer constant whose value plus one is a power of 2
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Integer constant that can be used for <code>and</code> operations in <code>depi</code>
and <code>extru</code> instructions
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Integer constant 31
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Integer constant 63
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating-point constant 0.0
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>A <code>lo_sum</code> data-linkage-table memory operand
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory operand that can be used as the destination operand of an
integer store instruction
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>A scaled or unscaled indexed memory operand
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>A memory operand for floating-point loads and stores
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>A register indirect memory operand
</p></dd>
</dl>

</dd>
<dt><em>picoChip family&mdash;<samp>picochip.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>k</code></dt>
<dd><p>Stack register.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Pointer register.  A register which can be used to access memory without
supplying an offset.  Any other register can be used to access memory,
but will need a constant offset.  In the case of the offset being zero,
it is more efficient to use a pointer register, since this reduces code
size.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>A twin register.  A register which may be paired with an adjacent
register to create a 32-bit register.
</p>
</dd>
<dt><code>a</code></dt>
<dd><p>Any absolute memory address (e.g., symbolic constant, symbolic
constant + offset).
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>4-bit signed integer.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>4-bit unsigned integer.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>8-bit signed integer.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Any constant whose absolute value is no greater than 4-bits.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>10-bit signed integer
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>16-bit signed integer.
</p>
</dd>
</dl>

</dd>
<dt><em>PowerPC and IBM RS6000&mdash;<samp>config/rs6000/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>b</code></dt>
<dd><p>Address base register
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Floating point register (containing 64-bit value)
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Floating point register (containing 32-bit value)
</p>
</dd>
<dt><code>v</code></dt>
<dd><p>Altivec vector register
</p>
</dd>
<dt><code>wa</code></dt>
<dd><p>Any VSX register if the -mvsx option was used or NO_REGS.
</p>
</dd>
<dt><code>wd</code></dt>
<dd><p>VSX vector register to hold vector double data or NO_REGS.
</p>
</dd>
<dt><code>wf</code></dt>
<dd><p>VSX vector register to hold vector float data or NO_REGS.
</p>
</dd>
<dt><code>wg</code></dt>
<dd><p>If <samp>-mmfpgpr</samp> was used, a floating point register or NO_REGS.
</p>
</dd>
<dt><code>wh</code></dt>
<dd><p>Floating point register if direct moves are available, or NO_REGS.
</p>
</dd>
<dt><code>wi</code></dt>
<dd><p>FP or VSX register to hold 64-bit integers for VSX insns or NO_REGS.
</p>
</dd>
<dt><code>wj</code></dt>
<dd><p>FP or VSX register to hold 64-bit integers for direct moves or NO_REGS.
</p>
</dd>
<dt><code>wk</code></dt>
<dd><p>FP or VSX register to hold 64-bit doubles for direct moves or NO_REGS.
</p>
</dd>
<dt><code>wl</code></dt>
<dd><p>Floating point register if the LFIWAX instruction is enabled or NO_REGS.
</p>
</dd>
<dt><code>wm</code></dt>
<dd><p>VSX register if direct move instructions are enabled, or NO_REGS.
</p>
</dd>
<dt><code>wn</code></dt>
<dd><p>No register (NO_REGS).
</p>
</dd>
<dt><code>wr</code></dt>
<dd><p>General purpose register if 64-bit instructions are enabled or NO_REGS.
</p>
</dd>
<dt><code>ws</code></dt>
<dd><p>VSX vector register to hold scalar double values or NO_REGS.
</p>
</dd>
<dt><code>wt</code></dt>
<dd><p>VSX vector register to hold 128 bit integer or NO_REGS.
</p>
</dd>
<dt><code>wu</code></dt>
<dd><p>Altivec register to use for float/32-bit int loads/stores  or NO_REGS.
</p>
</dd>
<dt><code>wv</code></dt>
<dd><p>Altivec register to use for double loads/stores  or NO_REGS.
</p>
</dd>
<dt><code>ww</code></dt>
<dd><p>FP or VSX register to perform float operations under <samp>-mvsx</samp> or NO_REGS.
</p>
</dd>
<dt><code>wx</code></dt>
<dd><p>Floating point register if the STFIWX instruction is enabled or NO_REGS.
</p>
</dd>
<dt><code>wy</code></dt>
<dd><p>FP or VSX register to perform ISA 2.07 float ops or NO_REGS.
</p>
</dd>
<dt><code>wz</code></dt>
<dd><p>Floating point register if the LFIWZX instruction is enabled or NO_REGS.
</p>
</dd>
<dt><code>wQ</code></dt>
<dd><p>A memory address that will work with the <code>lq</code> and <code>stq</code>
instructions.
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>&lsquo;<samp>MQ</samp>&rsquo;, &lsquo;<samp>CTR</samp>&rsquo;, or &lsquo;<samp>LINK</samp>&rsquo; register
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>&lsquo;<samp>MQ</samp>&rsquo; register
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>&lsquo;<samp>CTR</samp>&rsquo; register
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>&lsquo;<samp>LINK</samp>&rsquo; register
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>&lsquo;<samp>CR</samp>&rsquo; register (condition register) number 0
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>&lsquo;<samp>CR</samp>&rsquo; register (condition register)
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>&lsquo;<samp>XER[CA]</samp>&rsquo; carry bit (part of the XER register)
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 16-bit constant
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Unsigned 16-bit constant shifted left 16 bits (use &lsquo;<samp>L</samp>&rsquo; instead for
<code>SImode</code> constants)
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Unsigned 16-bit constant
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Signed 16-bit constant shifted left 16 bits
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Constant larger than 31
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Exact power of 2
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Zero
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Constant whose negation is a signed 16-bit constant
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating point constant that can be loaded into a register with one
instruction per word
</p>
</dd>
<dt><code>H</code></dt>
<dd><p>Integer/Floating point constant that can be loaded into a register using
three instructions
</p>
</dd>
<dt><code>m</code></dt>
<dd><p>Memory operand.
Normally, <code>m</code> does not allow addresses that update the base register.
If &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo; constraint is also used, they are allowed and
therefore on PowerPC targets in that case it is only safe
to use &lsquo;<samp>m&lt;&gt;</samp>&rsquo; in an <code>asm</code> statement if that <code>asm</code> statement
accesses the operand exactly once.  The <code>asm</code> statement must also
use &lsquo;<samp>%U<var>&lt;opno&gt;</var></samp>&rsquo; as a placeholder for the &ldquo;update&rdquo; flag in the
corresponding load or store instruction.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">asm (&quot;st%U0 %1,%0&quot; : &quot;=m&lt;&gt;&quot; (mem) : &quot;r&quot; (val));
</pre></div>

<p>is correct but:
</p>
<div class="smallexample">
<pre class="smallexample">asm (&quot;st %1,%0&quot; : &quot;=m&lt;&gt;&quot; (mem) : &quot;r&quot; (val));
</pre></div>

<p>is not.
</p>
</dd>
<dt><code>es</code></dt>
<dd><p>A &ldquo;stable&rdquo; memory operand; that is, one which does not include any
automodification of the base register.  This used to be useful when
&lsquo;<samp>m</samp>&rsquo; allowed automodification of the base register, but as those are now only
allowed when &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo; is used, &lsquo;<samp>es</samp>&rsquo; is basically the same
as &lsquo;<samp>m</samp>&rsquo; without &lsquo;<samp>&lt;</samp>&rsquo; and &lsquo;<samp>&gt;</samp>&rsquo;.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Memory operand that is an offset from a register (it is usually better
to use &lsquo;<samp>m</samp>&rsquo; or &lsquo;<samp>es</samp>&rsquo; in <code>asm</code> statements)
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Memory operand that is an indexed or indirect from a register (it is
usually better to use &lsquo;<samp>m</samp>&rsquo; or &lsquo;<samp>es</samp>&rsquo; in <code>asm</code> statements)
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>AIX TOC entry
</p>
</dd>
<dt><code>a</code></dt>
<dd><p>Address operand that is an indexed or indirect from a register (&lsquo;<samp>p</samp>&rsquo; is
preferable for <code>asm</code> statements)
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Constant suitable as a 64-bit mask operand
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Constant suitable as a 32-bit mask operand
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>System V Release 4 small data area reference
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>AND masks that can be performed by two rldic{l, r} instructions
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>Vector constant that does not require memory
</p>
</dd>
<dt><code>j</code></dt>
<dd><p>Vector constant that is all zeros.
</p>
</dd>
</dl>

</dd>
<dt><em>Intel 386&mdash;<samp>config/i386/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>R</code></dt>
<dd><p>Legacy register&mdash;the eight integer registers available on all
i386 processors (<code>a</code>, <code>b</code>, <code>c</code>, <code>d</code>,
<code>si</code>, <code>di</code>, <code>bp</code>, <code>sp</code>).
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>Any register accessible as <code><var>r</var>l</code>.  In 32-bit mode, <code>a</code>,
<code>b</code>, <code>c</code>, and <code>d</code>; in 64-bit mode, any integer register.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Any register accessible as <code><var>r</var>h</code>: <code>a</code>, <code>b</code>,
<code>c</code>, and <code>d</code>.
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>Any register that can be used as the index in a base+index memory
access: that is, any general register except the stack pointer.
</p>
</dd>
<dt><code>a</code></dt>
<dd><p>The <code>a</code> register.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>The <code>b</code> register.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>The <code>c</code> register.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>The <code>d</code> register.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>The <code>si</code> register.
</p>
</dd>
<dt><code>D</code></dt>
<dd><p>The <code>di</code> register.
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>The <code>a</code> and <code>d</code> registers.  This class is used for instructions
that return double word results in the <code>ax:dx</code> register pair.  Single
word values will be allocated either in <code>ax</code> or <code>dx</code>.
For example on i386 the following implements <code>rdtsc</code>:
</p>
<div class="smallexample">
<pre class="smallexample">unsigned long long rdtsc (void)
{
  unsigned long long tick;
  __asm__ __volatile__(&quot;rdtsc&quot;:&quot;=A&quot;(tick));
  return tick;
}
</pre></div>

<p>This is not correct on x86_64 as it would allocate tick in either <code>ax</code>
or <code>dx</code>.  You have to use the following variant instead:
</p>
<div class="smallexample">
<pre class="smallexample">unsigned long long rdtsc (void)
{
  unsigned int tickl, tickh;
  __asm__ __volatile__(&quot;rdtsc&quot;:&quot;=a&quot;(tickl),&quot;=d&quot;(tickh));
  return ((unsigned long long)tickh &lt;&lt; 32)|tickl;
}
</pre></div>


</dd>
<dt><code>f</code></dt>
<dd><p>Any 80387 floating-point (stack) register.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Top of 80387 floating-point stack (<code>%st(0)</code>).
</p>
</dd>
<dt><code>u</code></dt>
<dd><p>Second from top of 80387 floating-point stack (<code>%st(1)</code>).
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>Any MMX register.
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>Any SSE register.
</p>
</dd>
<dt><code>Yz</code></dt>
<dd><p>First SSE register (<code>%xmm0</code>).
</p>
</dd>
<dt><code>Y2</code></dt>
<dd><p>Any SSE register, when SSE2 is enabled.
</p>
</dd>
<dt><code>Yi</code></dt>
<dd><p>Any SSE register, when SSE2 and inter-unit moves are enabled.
</p>
</dd>
<dt><code>Ym</code></dt>
<dd><p>Any MMX register, when inter-unit moves are enabled.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 31, for 32-bit shifts.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 63, for 64-bit shifts.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Signed 8-bit integer constant.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p><code>0xFF</code> or <code>0xFFFF</code>, for andsi as a zero-extending move.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>0, 1, 2, or 3 (shifts for the <code>lea</code> instruction).
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Unsigned 8-bit integer constant (for <code>in</code> and <code>out</code>
instructions).
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 127, for 128-bit shifts.
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Standard 80387 floating point constant.
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>Standard SSE floating point constant.
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>32-bit signed integer constant, or a symbolic reference known
to fit that range (for immediate operands in sign-extending x86-64
instructions).
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>32-bit unsigned integer constant, or a symbolic reference known
to fit that range (for immediate operands in zero-extending x86-64
instructions).
</p>
</dd>
</dl>

</dd>
<dt><em>Intel IA-64&mdash;<samp>config/ia64/ia64.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>General register <code>r0</code> to <code>r3</code> for <code>addl</code> instruction
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Branch register
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Predicate register (&lsquo;<samp>c</samp>&rsquo; as in &ldquo;conditional&rdquo;)
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Application register residing in M-unit
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Application register residing in I-unit
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Floating-point register
</p>
</dd>
<dt><code>m</code></dt>
<dd><p>Memory operand.  If used together with &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo;,
the operand can have postincrement and postdecrement which
require printing with &lsquo;<samp>%Pn</samp>&rsquo; on IA-64.
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating-point constant 0.0 or 1.0
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>14-bit signed integer constant
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>22-bit signed integer constant
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>8-bit signed integer constant for logical instructions
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>8-bit adjusted signed integer constant for compare pseudo-ops
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>6-bit unsigned integer constant for shift counts
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>9-bit signed integer constant for load and store postincrements
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>The constant zero
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>0 or -1 for <code>dep</code> instruction
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Non-volatile memory for floating-point loads and stores
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Integer constant in the range 1 to 4 for <code>shladd</code> instruction
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Memory operand except postincrement and postdecrement.  This is
now roughly the same as &lsquo;<samp>m</samp>&rsquo; when not used together with &lsquo;<samp>&lt;</samp>&rsquo;
or &lsquo;<samp>&gt;</samp>&rsquo;.
</p></dd>
</dl>

</dd>
<dt><em>FRV&mdash;<samp>config/frv/frv.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Register in the class <code>ACC_REGS</code> (<code>acc0</code> to <code>acc7</code>).
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Register in the class <code>EVEN_ACC_REGS</code> (<code>acc0</code> to <code>acc7</code>).
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Register in the class <code>CC_REGS</code> (<code>fcc0</code> to <code>fcc3</code> and
<code>icc0</code> to <code>icc3</code>).
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Register in the class <code>GPR_REGS</code> (<code>gr0</code> to <code>gr63</code>).
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Register in the class <code>EVEN_REGS</code> (<code>gr0</code> to <code>gr63</code>).
Odd registers are excluded not in the class but through the use of a machine
mode larger than 4 bytes.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Register in the class <code>FPR_REGS</code> (<code>fr0</code> to <code>fr63</code>).
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>Register in the class <code>FEVEN_REGS</code> (<code>fr0</code> to <code>fr63</code>).
Odd registers are excluded not in the class but through the use of a machine
mode larger than 4 bytes.
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>Register in the class <code>LR_REG</code> (the <code>lr</code> register).
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>Register in the class <code>QUAD_REGS</code> (<code>gr2</code> to <code>gr63</code>).
Register numbers not divisible by 4 are excluded not in the class but through
the use of a machine mode larger than 8 bytes.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Register in the class <code>ICC_REGS</code> (<code>icc0</code> to <code>icc3</code>).
</p>
</dd>
<dt><code>u</code></dt>
<dd><p>Register in the class <code>FCC_REGS</code> (<code>fcc0</code> to <code>fcc3</code>).
</p>
</dd>
<dt><code>v</code></dt>
<dd><p>Register in the class <code>ICR_REGS</code> (<code>cc4</code> to <code>cc7</code>).
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>Register in the class <code>FCR_REGS</code> (<code>cc0</code> to <code>cc3</code>).
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>Register in the class <code>QUAD_FPR_REGS</code> (<code>fr0</code> to <code>fr63</code>).
Register numbers not divisible by 4 are excluded not in the class but through
the use of a machine mode larger than 8 bytes.
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>Register in the class <code>SPR_REGS</code> (<code>lcr</code> and <code>lr</code>).
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>Register in the class <code>QUAD_ACC_REGS</code> (<code>acc0</code> to <code>acc7</code>).
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>Register in the class <code>ACCG_REGS</code> (<code>accg0</code> to <code>accg7</code>).
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>Register in the class <code>CR_REGS</code> (<code>cc0</code> to <code>cc7</code>).
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating point constant zero
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>6-bit signed integer constant
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>10-bit signed integer constant
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>16-bit signed integer constant
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>16-bit unsigned integer constant
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>12-bit signed integer constant that is negative&mdash;i.e. in the
range of -2048 to -1
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Constant zero
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>12-bit signed integer constant that is greater than zero&mdash;i.e. in the
range of 1 to 2047.
</p>
</dd>
</dl>

</dd>
<dt><em>Blackfin family&mdash;<samp>config/bfin/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>P register
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>D register
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>A call clobbered P register.
</p>
</dd>
<dt><code>q<var>n</var></code></dt>
<dd><p>A single register.  If <var>n</var> is in the range 0 to 7, the corresponding D
register.  If it is <code>A</code>, then the register P0.
</p>
</dd>
<dt><code>D</code></dt>
<dd><p>Even-numbered D register
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>Odd-numbered D register
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Accumulator register.
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>Even-numbered accumulator register.
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>Odd-numbered accumulator register.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>I register
</p>
</dd>
<dt><code>v</code></dt>
<dd><p>B register
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>M register
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Registers used for circular buffering, i.e. I, B, or L registers.
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>The CC register.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>LT0 or LT1.
</p>
</dd>
<dt><code>k</code></dt>
<dd><p>LC0 or LC1.
</p>
</dd>
<dt><code>u</code></dt>
<dd><p>LB0 or LB1.
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>Any D, P, B, M, I or L register.
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>Additional registers typically used only in prologues and epilogues: RETS,
RETN, RETI, RETX, RETE, ASTAT, SEQSTAT and USP.
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>Any register except accumulators or CC.
</p>
</dd>
<dt><code>Ksh</code></dt>
<dd><p>Signed 16 bit integer (in the range -32768 to 32767)
</p>
</dd>
<dt><code>Kuh</code></dt>
<dd><p>Unsigned 16 bit integer (in the range 0 to 65535)
</p>
</dd>
<dt><code>Ks7</code></dt>
<dd><p>Signed 7 bit integer (in the range -64 to 63)
</p>
</dd>
<dt><code>Ku7</code></dt>
<dd><p>Unsigned 7 bit integer (in the range 0 to 127)
</p>
</dd>
<dt><code>Ku5</code></dt>
<dd><p>Unsigned 5 bit integer (in the range 0 to 31)
</p>
</dd>
<dt><code>Ks4</code></dt>
<dd><p>Signed 4 bit integer (in the range -8 to 7)
</p>
</dd>
<dt><code>Ks3</code></dt>
<dd><p>Signed 3 bit integer (in the range -3 to 4)
</p>
</dd>
<dt><code>Ku3</code></dt>
<dd><p>Unsigned 3 bit integer (in the range 0 to 7)
</p>
</dd>
<dt><code>P<var>n</var></code></dt>
<dd><p>Constant <var>n</var>, where <var>n</var> is a single-digit constant in the range 0 to 4.
</p>
</dd>
<dt><code>PA</code></dt>
<dd><p>An integer equal to one of the MACFLAG_XXX constants that is suitable for
use with either accumulator.
</p>
</dd>
<dt><code>PB</code></dt>
<dd><p>An integer equal to one of the MACFLAG_XXX constants that is suitable for
use only with accumulator A1.
</p>
</dd>
<dt><code>M1</code></dt>
<dd><p>Constant 255.
</p>
</dd>
<dt><code>M2</code></dt>
<dd><p>Constant 65535.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>An integer constant with exactly a single bit set.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>An integer constant with all bits set except exactly one.
</p>
</dd>
<dt><code>H</code></dt>
<dt><code>Q</code></dt>
<dd><p>Any SYMBOL_REF.
</p></dd>
</dl>

</dd>
<dt><em>M32C&mdash;<samp>config/m32c/m32c.c</samp></em></dt>
<dd><dl compact="compact">
<dt><code>Rsp</code></dt>
<dt><code>Rfb</code></dt>
<dt><code>Rsb</code></dt>
<dd><p>&lsquo;<samp>$sp</samp>&rsquo;, &lsquo;<samp>$fb</samp>&rsquo;, &lsquo;<samp>$sb</samp>&rsquo;.
</p>
</dd>
<dt><code>Rcr</code></dt>
<dd><p>Any control register, when they&rsquo;re 16 bits wide (nothing if control
registers are 24 bits wide)
</p>
</dd>
<dt><code>Rcl</code></dt>
<dd><p>Any control register, when they&rsquo;re 24 bits wide.
</p>
</dd>
<dt><code>R0w</code></dt>
<dt><code>R1w</code></dt>
<dt><code>R2w</code></dt>
<dt><code>R3w</code></dt>
<dd><p>$r0, $r1, $r2, $r3.
</p>
</dd>
<dt><code>R02</code></dt>
<dd><p>$r0 or $r2, or $r2r0 for 32 bit values.
</p>
</dd>
<dt><code>R13</code></dt>
<dd><p>$r1 or $r3, or $r3r1 for 32 bit values.
</p>
</dd>
<dt><code>Rdi</code></dt>
<dd><p>A register that can hold a 64 bit value.
</p>
</dd>
<dt><code>Rhl</code></dt>
<dd><p>$r0 or $r1 (registers with addressable high/low bytes)
</p>
</dd>
<dt><code>R23</code></dt>
<dd><p>$r2 or $r3
</p>
</dd>
<dt><code>Raa</code></dt>
<dd><p>Address registers
</p>
</dd>
<dt><code>Raw</code></dt>
<dd><p>Address registers when they&rsquo;re 16 bits wide.
</p>
</dd>
<dt><code>Ral</code></dt>
<dd><p>Address registers when they&rsquo;re 24 bits wide.
</p>
</dd>
<dt><code>Rqi</code></dt>
<dd><p>Registers that can hold QI values.
</p>
</dd>
<dt><code>Rad</code></dt>
<dd><p>Registers that can be used with displacements ($a0, $a1, $sb).
</p>
</dd>
<dt><code>Rsi</code></dt>
<dd><p>Registers that can hold 32 bit values.
</p>
</dd>
<dt><code>Rhi</code></dt>
<dd><p>Registers that can hold 16 bit values.
</p>
</dd>
<dt><code>Rhc</code></dt>
<dd><p>Registers chat can hold 16 bit values, including all control
registers.
</p>
</dd>
<dt><code>Rra</code></dt>
<dd><p>$r0 through R1, plus $a0 and $a1.
</p>
</dd>
<dt><code>Rfl</code></dt>
<dd><p>The flags register.
</p>
</dd>
<dt><code>Rmm</code></dt>
<dd><p>The memory-based pseudo-registers $mem0 through $mem15.
</p>
</dd>
<dt><code>Rpi</code></dt>
<dd><p>Registers that can hold pointers (16 bit registers for r8c, m16c; 24
bit registers for m32cm, m32c).
</p>
</dd>
<dt><code>Rpa</code></dt>
<dd><p>Matches multiple registers in a PARALLEL to form a larger register.
Used to match function return values.
</p>
</dd>
<dt><code>Is3</code></dt>
<dd><p>-8 &hellip; 7
</p>
</dd>
<dt><code>IS1</code></dt>
<dd><p>-128 &hellip; 127
</p>
</dd>
<dt><code>IS2</code></dt>
<dd><p>-32768 &hellip; 32767
</p>
</dd>
<dt><code>IU2</code></dt>
<dd><p>0 &hellip; 65535
</p>
</dd>
<dt><code>In4</code></dt>
<dd><p>-8 &hellip; -1 or 1 &hellip; 8
</p>
</dd>
<dt><code>In5</code></dt>
<dd><p>-16 &hellip; -1 or 1 &hellip; 16
</p>
</dd>
<dt><code>In6</code></dt>
<dd><p>-32 &hellip; -1 or 1 &hellip; 32
</p>
</dd>
<dt><code>IM2</code></dt>
<dd><p>-65536 &hellip; -1
</p>
</dd>
<dt><code>Ilb</code></dt>
<dd><p>An 8 bit value with exactly one bit set.
</p>
</dd>
<dt><code>Ilw</code></dt>
<dd><p>A 16 bit value with exactly one bit set.
</p>
</dd>
<dt><code>Sd</code></dt>
<dd><p>The common src/dest memory addressing modes.
</p>
</dd>
<dt><code>Sa</code></dt>
<dd><p>Memory addressed using $a0 or $a1.
</p>
</dd>
<dt><code>Si</code></dt>
<dd><p>Memory addressed with immediate addresses.
</p>
</dd>
<dt><code>Ss</code></dt>
<dd><p>Memory addressed using the stack pointer ($sp).
</p>
</dd>
<dt><code>Sf</code></dt>
<dd><p>Memory addressed using the frame base register ($fb).
</p>
</dd>
<dt><code>Ss</code></dt>
<dd><p>Memory addressed using the small base register ($sb).
</p>
</dd>
<dt><code>S1</code></dt>
<dd><p>$r1h
</p></dd>
</dl>

</dd>
<dt><em>MeP&mdash;<samp>config/mep/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>The $sp register.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>The $tp register.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Any control register.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Either the $hi or the $lo register.
</p>
</dd>
<dt><code>em</code></dt>
<dd><p>Coprocessor registers that can be directly loaded ($c0-$c15).
</p>
</dd>
<dt><code>ex</code></dt>
<dd><p>Coprocessor registers that can be moved to each other.
</p>
</dd>
<dt><code>er</code></dt>
<dd><p>Coprocessor registers that can be moved to core registers.
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>The $hi register.
</p>
</dd>
<dt><code>j</code></dt>
<dd><p>The $rpc register.
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>The $lo register.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Registers which can be used in $tp-relative addressing.
</p>
</dd>
<dt><code>v</code></dt>
<dd><p>The $gp register.
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>The coprocessor registers.
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>The coprocessor control registers.
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>The $0 register.
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>User-defined register set A.
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>User-defined register set B.
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>User-defined register set C.
</p>
</dd>
<dt><code>D</code></dt>
<dd><p>User-defined register set D.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Offsets for $gp-rel addressing.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Constants that can be used directly with boolean insns.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Constants that can be moved directly to registers.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Small constants that can be added to registers.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Long shift counts.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Small constants that can be compared to registers.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Constants that can be loaded into the top half of registers.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Signed 8-bit immediates.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Symbols encoded for $tp-rel or $gp-rel addressing.
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Non-constant addresses for loading/saving coprocessor registers.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>The top half of a symbol&rsquo;s value.
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>A register indirect address without offset.
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Symbolic references to the control bus.
</p>
</dd>
</dl>

</dd>
<dt><em>MicroBlaze&mdash;<samp>config/microblaze/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>d</code></dt>
<dd><p>A general register (<code>r0</code> to <code>r31</code>).
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>A status register (<code>rmsr</code>, <code>$fcc1</code> to <code>$fcc7</code>).
</p>
</dd>
</dl>

</dd>
<dt><em>MIPS&mdash;<samp>config/mips/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>d</code></dt>
<dd><p>An address register.  This is equivalent to <code>r</code> unless
generating MIPS16 code.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>A floating-point register (if available).
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>Formerly the <code>hi</code> register.  This constraint is no longer supported.
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>The <code>lo</code> register.  Use this register to store values that are
no bigger than a word.
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>The concatenated <code>hi</code> and <code>lo</code> registers.  Use this register
to store doubleword values.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>A register suitable for use in an indirect jump.  This will always be
<code>$25</code> for <samp>-mabicalls</samp>.
</p>
</dd>
<dt><code>v</code></dt>
<dd><p>Register <code>$3</code>.  Do not use this constraint in new code;
it is retained only for compatibility with glibc.
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>Equivalent to <code>r</code>; retained for backwards compatibility.
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>A floating-point condition code register.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>A signed 16-bit constant (for arithmetic instructions).
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Integer zero.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>An unsigned 16-bit constant (for logic instructions).
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>A signed 32-bit constant in which the lower 16 bits are zero.
Such constants can be loaded using <code>lui</code>.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>A constant that cannot be loaded using <code>lui</code>, <code>addiu</code>
or <code>ori</code>.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>A constant in the range -65535 to -1 (inclusive).
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>A signed 15-bit constant.
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>A constant in the range 1 to 65535 (inclusive).
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating-point zero.
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>An address that can be used in a non-macro load or store.
</p></dd>
</dl>

</dd>
<dt><em>Motorola 680x0&mdash;<samp>config/m68k/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Address register
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Data register
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>68881 floating-point register, if available
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Integer in the range 1 to 8
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>16-bit signed number
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Signed number whose magnitude is greater than 0x80
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Integer in the range -8 to -1
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Signed number whose magnitude is greater than 0x100
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Range 24 to 31, rotatert:SI 8 to 1 expressed as rotate
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>16 (for rotate using swap)
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Range 8 to 15, rotatert:HI 8 to 1 expressed as rotate
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Numbers that mov3q can handle
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating point constant that is not a 68881 constant
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Operands that satisfy &rsquo;m&rsquo; when -mpcrel is in effect
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Operands that satisfy &rsquo;s&rsquo; when -mpcrel is not in effect
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Address register indirect addressing mode
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Register offset addressing
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>const_call_operand
</p>
</dd>
<dt><code>Cs</code></dt>
<dd><p>symbol_ref or const
</p>
</dd>
<dt><code>Ci</code></dt>
<dd><p>const_int
</p>
</dd>
<dt><code>C0</code></dt>
<dd><p>const_int 0
</p>
</dd>
<dt><code>Cj</code></dt>
<dd><p>Range of signed numbers that don&rsquo;t fit in 16 bits
</p>
</dd>
<dt><code>Cmvq</code></dt>
<dd><p>Integers valid for mvq
</p>
</dd>
<dt><code>Capsw</code></dt>
<dd><p>Integers valid for a moveq followed by a swap
</p>
</dd>
<dt><code>Cmvz</code></dt>
<dd><p>Integers valid for mvz
</p>
</dd>
<dt><code>Cmvs</code></dt>
<dd><p>Integers valid for mvs
</p>
</dd>
<dt><code>Ap</code></dt>
<dd><p>push_operand
</p>
</dd>
<dt><code>Ac</code></dt>
<dd><p>Non-register operands allowed in clr
</p>
</dd>
</dl>

</dd>
<dt><em>Moxie&mdash;<samp>config/moxie/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>A</code></dt>
<dd><p>An absolute address
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>An offset address
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>A register indirect memory operand
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>A constant in the range of 0 to 255.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>A constant in the range of 0 to -255.
</p>
</dd>
</dl>

</dd>
<dt><em>PDP-11&mdash;<samp>config/pdp11/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Floating point registers AC0 through AC3.  These can be loaded from/to
memory with a single instruction.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Odd numbered general registers (R1, R3, R5).  These are used for
16-bit multiply operations.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Any of the floating point registers (AC0 through AC5).
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating point constant 0.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>An integer constant that fits in 16 bits.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>An integer constant whose low order 16 bits are zero.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>An integer constant that does not meet the constraints for codes
&lsquo;<samp>I</samp>&rsquo; or &lsquo;<samp>J</samp>&rsquo;.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>The integer constant 1.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>The integer constant -1.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>The integer constant 0.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>Integer constants -4 through -1 and 1 through 4; shifts by these
amounts are handled as multiple single-bit shifts rather than a single
variable-length shift.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory reference which requires an additional word (address or
offset) after the opcode.
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>A memory reference that is encoded within the opcode.
</p>
</dd>
</dl>

</dd>
<dt><em>RL78&mdash;<samp>config/rl78/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>Int3</code></dt>
<dd><p>An integer constant in the range 1 &hellip; 7.
</p></dd>
<dt><code>Int8</code></dt>
<dd><p>An integer constant in the range 0 &hellip; 255.
</p></dd>
<dt><code>J</code></dt>
<dd><p>An integer constant in the range -255 &hellip; 0
</p></dd>
<dt><code>K</code></dt>
<dd><p>The integer constant 1.
</p></dd>
<dt><code>L</code></dt>
<dd><p>The integer constant -1.
</p></dd>
<dt><code>M</code></dt>
<dd><p>The integer constant 0.
</p></dd>
<dt><code>N</code></dt>
<dd><p>The integer constant 2.
</p></dd>
<dt><code>O</code></dt>
<dd><p>The integer constant -2.
</p></dd>
<dt><code>P</code></dt>
<dd><p>An integer constant in the range 1 &hellip; 15.
</p></dd>
<dt><code>Qbi</code></dt>
<dd><p>The built-in compare types&ndash;eq, ne, gtu, ltu, geu, and leu.
</p></dd>
<dt><code>Qsc</code></dt>
<dd><p>The synthetic compare types&ndash;gt, lt, ge, and le.
</p></dd>
<dt><code>Wab</code></dt>
<dd><p>A memory reference with an absolute address.
</p></dd>
<dt><code>Wbc</code></dt>
<dd><p>A memory reference using <code>BC</code> as a base register, with an optional offset.
</p></dd>
<dt><code>Wca</code></dt>
<dd><p>A memory reference using <code>AX</code>, <code>BC</code>, <code>DE</code>, or <code>HL</code> for the address, for calls.
</p></dd>
<dt><code>Wcv</code></dt>
<dd><p>A memory reference using any 16-bit register pair for the address, for calls.
</p></dd>
<dt><code>Wd2</code></dt>
<dd><p>A memory reference using <code>DE</code> as a base register, with an optional offset.
</p></dd>
<dt><code>Wde</code></dt>
<dd><p>A memory reference using <code>DE</code> as a base register, without any offset.
</p></dd>
<dt><code>Wfr</code></dt>
<dd><p>Any memory reference to an address in the far address space.
</p></dd>
<dt><code>Wh1</code></dt>
<dd><p>A memory reference using <code>HL</code> as a base register, with an optional one-byte offset.
</p></dd>
<dt><code>Whb</code></dt>
<dd><p>A memory reference using <code>HL</code> as a base register, with <code>B</code> or <code>C</code> as the index register.
</p></dd>
<dt><code>Whl</code></dt>
<dd><p>A memory reference using <code>HL</code> as a base register, without any offset.
</p></dd>
<dt><code>Ws1</code></dt>
<dd><p>A memory reference using <code>SP</code> as a base register, with an optional one-byte offset.
</p></dd>
<dt><code>Y</code></dt>
<dd><p>Any memory reference to an address in the near address space.
</p></dd>
<dt><code>A</code></dt>
<dd><p>The <code>AX</code> register.
</p></dd>
<dt><code>B</code></dt>
<dd><p>The <code>BC</code> register.
</p></dd>
<dt><code>D</code></dt>
<dd><p>The <code>DE</code> register.
</p></dd>
<dt><code>R</code></dt>
<dd><p><code>A</code> through <code>L</code> registers.
</p></dd>
<dt><code>S</code></dt>
<dd><p>The <code>SP</code> register.
</p></dd>
<dt><code>T</code></dt>
<dd><p>The <code>HL</code> register.
</p></dd>
<dt><code>Z08W</code></dt>
<dd><p>The 16-bit <code>R8</code> register.
</p></dd>
<dt><code>Z10W</code></dt>
<dd><p>The 16-bit <code>R10</code> register.
</p></dd>
<dt><code>Zint</code></dt>
<dd><p>The registers reserved for interrupts (<code>R24</code> to <code>R31</code>).
</p></dd>
<dt><code>a</code></dt>
<dd><p>The <code>A</code> register.
</p></dd>
<dt><code>b</code></dt>
<dd><p>The <code>B</code> register.
</p></dd>
<dt><code>c</code></dt>
<dd><p>The <code>C</code> register.
</p></dd>
<dt><code>d</code></dt>
<dd><p>The <code>D</code> register.
</p></dd>
<dt><code>e</code></dt>
<dd><p>The <code>E</code> register.
</p></dd>
<dt><code>h</code></dt>
<dd><p>The <code>H</code> register.
</p></dd>
<dt><code>l</code></dt>
<dd><p>The <code>L</code> register.
</p></dd>
<dt><code>v</code></dt>
<dd><p>The virtual registers.
</p></dd>
<dt><code>w</code></dt>
<dd><p>The <code>PSW</code> register.
</p></dd>
<dt><code>x</code></dt>
<dd><p>The <code>X</code> register.
</p>
</dd>
</dl>

</dd>
<dt><em>RX&mdash;<samp>config/rx/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>Q</code></dt>
<dd><p>An address which does not involve register indirect addressing or
pre/post increment/decrement addressing.
</p>
</dd>
<dt><code>Symbol</code></dt>
<dd><p>A symbol reference.
</p>
</dd>
<dt><code>Int08</code></dt>
<dd><p>A constant in the range -256 to 255, inclusive.
</p>
</dd>
<dt><code>Sint08</code></dt>
<dd><p>A constant in the range -128 to 127, inclusive.
</p>
</dd>
<dt><code>Sint16</code></dt>
<dd><p>A constant in the range -32768 to 32767, inclusive.
</p>
</dd>
<dt><code>Sint24</code></dt>
<dd><p>A constant in the range -8388608 to 8388607, inclusive.
</p>
</dd>
<dt><code>Uint04</code></dt>
<dd><p>A constant in the range 0 to 15, inclusive.
</p>
</dd>
</dl>

</dd>
<dt><em>SPARC&mdash;<samp>config/sparc/sparc.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>f</code></dt>
<dd><p>Floating-point register on the SPARC-V8 architecture and
lower floating-point register on the SPARC-V9 architecture.
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Floating-point register.  It is equivalent to &lsquo;<samp>f</samp>&rsquo; on the
SPARC-V8 architecture and contains both lower and upper
floating-point registers on the SPARC-V9 architecture.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Floating-point condition code register.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Lower floating-point register.  It is only valid on the SPARC-V9
architecture when the Visual Instruction Set is available.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Floating-point register.  It is only valid on the SPARC-V9 architecture
when the Visual Instruction Set is available.
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>64-bit global or out register for the SPARC-V8+ architecture.
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>The constant all-ones, for floating-point.
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>Signed 5-bit constant
</p>
</dd>
<dt><code>D</code></dt>
<dd><p>A vector constant
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 13-bit constant
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Zero
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>32-bit constant with the low 12 bits clear (a constant that can be
loaded with the <code>sethi</code> instruction)
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>A constant in the range supported by <code>movcc</code> instructions (11-bit
signed immediate)
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>A constant in the range supported by <code>movrcc</code> instructions (10-bit
signed immediate)
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Same as &lsquo;<samp>K</samp>&rsquo;, except that it verifies that bits that are not in the
lower 32-bit range are all zero.  Must be used instead of &lsquo;<samp>K</samp>&rsquo; for
modes wider than <code>SImode</code>
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>The constant 4096
</p>
</dd>
<dt><code>G</code></dt>
<dd><p>Floating-point zero
</p>
</dd>
<dt><code>H</code></dt>
<dd><p>Signed 13-bit constant, sign-extended to 32 or 64 bits
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>The constant -1
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Floating-point constant whose integral representation can
be moved into an integer register using a single sethi
instruction
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Floating-point constant whose integral representation can
be moved into an integer register using a single mov
instruction
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Floating-point constant whose integral representation can
be moved into an integer register using a high/lo_sum
instruction sequence
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Memory address aligned to an 8-byte boundary
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Even register
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>Memory address for &lsquo;<samp>e</samp>&rsquo; constraint registers
</p>
</dd>
<dt><code>w</code></dt>
<dd><p>Memory address with only a base register
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>Vector zero
</p>
</dd>
</dl>

</dd>
<dt><em>SPU&mdash;<samp>config/spu/spu.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is treated as a 64 bit value.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>An immediate for and/xor/or instructions.  const_int is treated as a 64 bit value.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>An immediate for the <code>iohl</code> instruction.  const_int is treated as a 64 bit value.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>An immediate which can be loaded with <code>fsmbi</code>.
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is treated as a 32 bit value.
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>An immediate for most arithmetic instructions.  const_int is treated as a 32 bit value.
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>An immediate for and/xor/or instructions.  const_int is treated as a 32 bit value.
</p>
</dd>
<dt><code>D</code></dt>
<dd><p>An immediate for the <code>iohl</code> instruction.  const_int is treated as a 32 bit value.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>A constant in the range [-64, 63] for shift/rotate instructions.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>An unsigned 7-bit constant for conversion/nop/channel instructions.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>A signed 10-bit constant for most arithmetic instructions.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>A signed 16 bit immediate for <code>stop</code>.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>An unsigned 16-bit constant for <code>iohl</code> and <code>fsmbi</code>.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>An unsigned 7-bit constant whose 3 least significant bits are 0.
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>An unsigned 3-bit constant for 16-byte rotates and shifts
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Call operand, reg, for indirect calls
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Call operand, symbol, for relative calls.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Call operand, const_int, for absolute calls.
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>An immediate which can be loaded with the il/ila/ilh/ilhu instructions.  const_int is sign extended to 128 bit.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>An immediate for shift and rotate instructions.  const_int is treated as a 32 bit value.
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>An immediate for and/xor/or instructions.  const_int is sign extended as a 128 bit.
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>An immediate for the <code>iohl</code> instruction.  const_int is sign extended to 128 bit.
</p>
</dd>
</dl>

</dd>
<dt><em>S/390 and zSeries&mdash;<samp>config/s390/s390.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Address register (general purpose register except r0)
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Condition code register
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Data register (arbitrary general purpose register)
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>Floating-point register
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Unsigned 8-bit constant (0&ndash;255)
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Unsigned 12-bit constant (0&ndash;4095)
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Signed 16-bit constant (-32768&ndash;32767)
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Value appropriate as displacement.
</p><dl compact="compact">
<dt><code>(0..4095)</code></dt>
<dd><p>for short displacement
</p></dd>
<dt><code>(-524288..524287)</code></dt>
<dd><p>for long displacement
</p></dd>
</dl>

</dd>
<dt><code>M</code></dt>
<dd><p>Constant integer with a value of 0x7fffffff.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Multiple letter constraint followed by 4 parameter letters.
</p><dl compact="compact">
<dt><code>0..9:</code></dt>
<dd><p>number of the part counting from most to least significant
</p></dd>
<dt><code>H,Q:</code></dt>
<dd><p>mode of the part
</p></dd>
<dt><code>D,S,H:</code></dt>
<dd><p>mode of the containing operand
</p></dd>
<dt><code>0,F:</code></dt>
<dd><p>value of the other parts (F&mdash;all bits set)
</p></dd>
</dl>
<p>The constraint matches if the specified part of a constant
has a value different from its other parts.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Memory reference without index register and with short displacement.
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Memory reference with index register and short displacement.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Memory reference without index register but with long displacement.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Memory reference with index register and long displacement.
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Pointer with short displacement.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>Pointer with long displacement.
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>Shift count operand.
</p>
</dd>
</dl>

</dd>
<dt><em>Score family&mdash;<samp>config/score/score.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>d</code></dt>
<dd><p>Registers from r0 to r32.
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Registers from r0 to r16.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>r8&mdash;r11 or r22&mdash;r27 registers.
</p>
</dd>
<dt><code>h</code></dt>
<dd><p>hi register.
</p>
</dd>
<dt><code>l</code></dt>
<dd><p>lo register.
</p>
</dd>
<dt><code>x</code></dt>
<dd><p>hi + lo register.
</p>
</dd>
<dt><code>q</code></dt>
<dd><p>cnt register.
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>lcb register.
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>scb register.
</p>
</dd>
<dt><code>a</code></dt>
<dd><p>cnt + lcb + scb register.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>cr0&mdash;cr15 register.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>cp1 registers.
</p>
</dd>
<dt><code>f</code></dt>
<dd><p>cp2 registers.
</p>
</dd>
<dt><code>i</code></dt>
<dd><p>cp3 registers.
</p>
</dd>
<dt><code>j</code></dt>
<dd><p>cp1 + cp2 + cp3 registers.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>High 16-bit constant (32-bit constant with 16 LSBs zero).
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Unsigned 5 bit integer (in the range 0 to 31).
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Unsigned 16 bit integer (in the range 0 to 65535).
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Signed 16 bit integer (in the range -32768 to 32767).
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>Unsigned 14 bit integer (in the range 0 to 16383).
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Signed 14 bit integer (in the range -8192 to 8191).
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Any SYMBOL_REF.
</p></dd>
</dl>

</dd>
<dt><em>Xstormy16&mdash;<samp>config/stormy16/stormy16.h</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Register r0.
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Register r1.
</p>
</dd>
<dt><code>c</code></dt>
<dd><p>Register r2.
</p>
</dd>
<dt><code>d</code></dt>
<dd><p>Register r8.
</p>
</dd>
<dt><code>e</code></dt>
<dd><p>Registers r0 through r7.
</p>
</dd>
<dt><code>t</code></dt>
<dd><p>Registers r0 and r1.
</p>
</dd>
<dt><code>y</code></dt>
<dd><p>The carry register.
</p>
</dd>
<dt><code>z</code></dt>
<dd><p>Registers r8 and r9.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>A constant between 0 and 3 inclusive.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>A constant that has exactly one bit set.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>A constant that has exactly one bit clear.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>A constant between 0 and 255 inclusive.
</p>
</dd>
<dt><code>M</code></dt>
<dd><p>A constant between -255 and 0 inclusive.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>A constant between -3 and 0 inclusive.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>A constant between 1 and 4 inclusive.
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>A constant between -4 and -1 inclusive.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>A memory reference that is a stack push.
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>A memory reference that is a stack pop.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>A memory reference that refers to a constant address of known value.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>The register indicated by Rx (not implemented yet).
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>A constant that is not between 2 and 15 inclusive.
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>The constant 0.
</p>
</dd>
</dl>

</dd>
<dt><em>TI C6X family&mdash;<samp>config/c6x/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>Register file A (A0&ndash;A31).
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>Register file B (B0&ndash;B31).
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>Predicate registers in register file A (A0&ndash;A2 on C64X and
higher, A1 and A2 otherwise).
</p>
</dd>
<dt><code>B</code></dt>
<dd><p>Predicate registers in register file B (B0&ndash;B2).
</p>
</dd>
<dt><code>C</code></dt>
<dd><p>A call-used register in register file B (B0&ndash;B9, B16&ndash;B31).
</p>
</dd>
<dt><code>Da</code></dt>
<dd><p>Register file A, excluding predicate registers (A3&ndash;A31,
plus A0 if not C64X or higher).
</p>
</dd>
<dt><code>Db</code></dt>
<dd><p>Register file B, excluding predicate registers (B3&ndash;B31).
</p>
</dd>
<dt><code>Iu4</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 15.
</p>
</dd>
<dt><code>Iu5</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 31.
</p>
</dd>
<dt><code>In5</code></dt>
<dd><p>Integer constant in the range -31 &hellip; 0.
</p>
</dd>
<dt><code>Is5</code></dt>
<dd><p>Integer constant in the range -16 &hellip; 15.
</p>
</dd>
<dt><code>I5x</code></dt>
<dd><p>Integer constant that can be the operand of an ADDA or a SUBA insn.
</p>
</dd>
<dt><code>IuB</code></dt>
<dd><p>Integer constant in the range 0 &hellip; 65535.
</p>
</dd>
<dt><code>IsB</code></dt>
<dd><p>Integer constant in the range -32768 &hellip; 32767.
</p>
</dd>
<dt><code>IsC</code></dt>
<dd><p>Integer constant in the range <em>-2^{20}</em> &hellip; <em>2^{20} - 1</em>.
</p>
</dd>
<dt><code>Jc</code></dt>
<dd><p>Integer constant that is a valid mask for the clr instruction.
</p>
</dd>
<dt><code>Js</code></dt>
<dd><p>Integer constant that is a valid mask for the set instruction.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Memory location with A base register.
</p>
</dd>
<dt><code>R</code></dt>
<dd><p>Memory location with B base register.
</p>
</dd>
<dt><code>S0</code></dt>
<dd><p>On C64x+ targets, a GP-relative small data reference.
</p>
</dd>
<dt><code>S1</code></dt>
<dd><p>Any kind of <code>SYMBOL_REF</code>, for use in a call address.
</p>
</dd>
<dt><code>Si</code></dt>
<dd><p>Any kind of immediate operand, unless it matches the S0 constraint.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>Memory location with B base register, but not using a long offset.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>A memory operand with an address that can&rsquo;t be used in an unaligned access.
</p>
</dd>
<dt><code>Z</code></dt>
<dd><p>Register B14 (aka DP).
</p>
</dd>
</dl>

</dd>
<dt><em>TILE-Gx&mdash;<samp>config/tilegx/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>R00</code></dt>
<dt><code>R01</code></dt>
<dt><code>R02</code></dt>
<dt><code>R03</code></dt>
<dt><code>R04</code></dt>
<dt><code>R05</code></dt>
<dt><code>R06</code></dt>
<dt><code>R07</code></dt>
<dt><code>R08</code></dt>
<dt><code>R09</code></dt>
<dt><code>R10</code></dt>
<dd><p>Each of these represents a register constraint for an individual
register, from r0 to r10.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 8-bit integer constant.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Signed 16-bit integer constant.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Unsigned 16-bit integer constant.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Integer constant that fits in one signed byte when incremented by one
(-129 &hellip; 126).
</p>
</dd>
<dt><code>m</code></dt>
<dd><p>Memory operand.  If used together with &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo;, the
operand can have postincrement which requires printing with &lsquo;<samp>%In</samp>&rsquo;
and &lsquo;<samp>%in</samp>&rsquo; on TILE-Gx.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">asm (&quot;st_add %I0,%1,%i0&quot; : &quot;=m&lt;&gt;&quot; (*mem) : &quot;r&quot; (val));
</pre></div>

</dd>
<dt><code>M</code></dt>
<dd><p>A bit mask suitable for the BFINS instruction.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Integer constant that is a byte tiled out eight times.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>The integer zero constant.
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Integer constant that is a sign-extended byte tiled out as four shorts.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Integer constant that fits in one signed byte when incremented
(-129 &hellip; 126), but excluding -1.
</p>
</dd>
<dt><code>S</code></dt>
<dd><p>Integer constant that has all 1 bits consecutive and starting at bit 0.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>A 16-bit fragment of a got, tls, or pc-relative reference.
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Memory operand except postincrement.  This is roughly the same as
&lsquo;<samp>m</samp>&rsquo; when not used together with &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo;.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>An 8-element vector constant with identical elements.
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>A 4-element vector constant with identical elements.
</p>
</dd>
<dt><code>Z0</code></dt>
<dd><p>The integer constant 0xffffffff.
</p>
</dd>
<dt><code>Z1</code></dt>
<dd><p>The integer constant 0xffffffff00000000.
</p>
</dd>
</dl>

</dd>
<dt><em>TILEPro&mdash;<samp>config/tilepro/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>R00</code></dt>
<dt><code>R01</code></dt>
<dt><code>R02</code></dt>
<dt><code>R03</code></dt>
<dt><code>R04</code></dt>
<dt><code>R05</code></dt>
<dt><code>R06</code></dt>
<dt><code>R07</code></dt>
<dt><code>R08</code></dt>
<dt><code>R09</code></dt>
<dt><code>R10</code></dt>
<dd><p>Each of these represents a register constraint for an individual
register, from r0 to r10.
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 8-bit integer constant.
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Signed 16-bit integer constant.
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Nonzero integer constant with low 16 bits zero.
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Integer constant that fits in one signed byte when incremented by one
(-129 &hellip; 126).
</p>
</dd>
<dt><code>m</code></dt>
<dd><p>Memory operand.  If used together with &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo;, the
operand can have postincrement which requires printing with &lsquo;<samp>%In</samp>&rsquo;
and &lsquo;<samp>%in</samp>&rsquo; on TILEPro.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">asm (&quot;swadd %I0,%1,%i0&quot; : &quot;=m&lt;&gt;&quot; (mem) : &quot;r&quot; (val));
</pre></div>

</dd>
<dt><code>M</code></dt>
<dd><p>A bit mask suitable for the MM instruction.
</p>
</dd>
<dt><code>N</code></dt>
<dd><p>Integer constant that is a byte tiled out four times.
</p>
</dd>
<dt><code>O</code></dt>
<dd><p>The integer zero constant.
</p>
</dd>
<dt><code>P</code></dt>
<dd><p>Integer constant that is a sign-extended byte tiled out as two shorts.
</p>
</dd>
<dt><code>Q</code></dt>
<dd><p>Integer constant that fits in one signed byte when incremented
(-129 &hellip; 126), but excluding -1.
</p>
</dd>
<dt><code>T</code></dt>
<dd><p>A symbolic operand, or a 16-bit fragment of a got, tls, or pc-relative
reference.
</p>
</dd>
<dt><code>U</code></dt>
<dd><p>Memory operand except postincrement.  This is roughly the same as
&lsquo;<samp>m</samp>&rsquo; when not used together with &lsquo;<samp>&lt;</samp>&rsquo; or &lsquo;<samp>&gt;</samp>&rsquo;.
</p>
</dd>
<dt><code>W</code></dt>
<dd><p>A 4-element vector constant with identical elements.
</p>
</dd>
<dt><code>Y</code></dt>
<dd><p>A 2-element vector constant with identical elements.
</p>
</dd>
</dl>

</dd>
<dt><em>Xtensa&mdash;<samp>config/xtensa/constraints.md</samp></em></dt>
<dd><dl compact="compact">
<dt><code>a</code></dt>
<dd><p>General-purpose 32-bit register
</p>
</dd>
<dt><code>b</code></dt>
<dd><p>One-bit boolean register
</p>
</dd>
<dt><code>A</code></dt>
<dd><p>MAC16 40-bit accumulator register
</p>
</dd>
<dt><code>I</code></dt>
<dd><p>Signed 12-bit integer constant, for use in MOVI instructions
</p>
</dd>
<dt><code>J</code></dt>
<dd><p>Signed 8-bit integer constant, for use in ADDI instructions
</p>
</dd>
<dt><code>K</code></dt>
<dd><p>Integer constant valid for BccI instructions
</p>
</dd>
<dt><code>L</code></dt>
<dd><p>Unsigned constant valid for BccUI instructions
</p>
</dd>
</dl>

</dd>
</dl>

<hr>
<a name="Disable-Insn-Alternatives"></a>
<div class="header">
<p>
Next: <a href="#Define-Constraints" accesskey="n" rel="next">Define Constraints</a>, Previous: <a href="#Machine-Constraints" accesskey="p" rel="prev">Machine Constraints</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Disable-insn-alternatives-using-the-enabled-attribute"></a>
<h4 class="subsection">16.8.6 Disable insn alternatives using the <code>enabled</code> attribute</h4>
<a name="index-enabled"></a>

<p>The <code>enabled</code> insn attribute may be used to disable certain insn
alternatives for machine-specific reasons.  This is useful when adding
new instructions to an existing pattern which are only available for
certain cpu architecture levels as specified with the <code>-march=</code>
option.
</p>
<p>If an insn alternative is disabled, then it will never be used.  The
compiler treats the constraints for the disabled alternative as
unsatisfiable.
</p>
<p>In order to make use of the <code>enabled</code> attribute a back end has to add
in the machine description files:
</p>
<ol>
<li> A definition of the <code>enabled</code> insn attribute.  The attribute is
defined as usual using the <code>define_attr</code> command.  This
definition should be based on other insn attributes and/or target flags.
The <code>enabled</code> attribute is a numeric attribute and should evaluate to
<code>(const_int 1)</code> for an enabled alternative and to
<code>(const_int 0)</code> otherwise.
</li><li> A definition of another insn attribute used to describe for what
reason an insn alternative might be available or
not.  E.g. <code>cpu_facility</code> as in the example below.
</li><li> An assignment for the second attribute to each insn definition
combining instructions which are not all available under the same
circumstances.  (Note: It obviously only makes sense for definitions
with more than one alternative.  Otherwise the insn pattern should be
disabled or enabled using the insn condition.)
</li></ol>

<p>E.g. the following two patterns could easily be merged using the <code>enabled</code>
attribute:
</p>
<div class="smallexample">
<pre class="smallexample">
(define_insn &quot;*movdi_old&quot;
  [(set (match_operand:DI 0 &quot;register_operand&quot; &quot;=d&quot;)
        (match_operand:DI 1 &quot;register_operand&quot; &quot; d&quot;))]
  &quot;!TARGET_NEW&quot;
  &quot;lgr %0,%1&quot;)

(define_insn &quot;*movdi_new&quot;
  [(set (match_operand:DI 0 &quot;register_operand&quot; &quot;=d,f,d&quot;)
        (match_operand:DI 1 &quot;register_operand&quot; &quot; d,d,f&quot;))]
  &quot;TARGET_NEW&quot;
  &quot;@
   lgr  %0,%1
   ldgr %0,%1
   lgdr %0,%1&quot;)

</pre></div>

<p>to:
</p>
<div class="smallexample">
<pre class="smallexample">
(define_insn &quot;*movdi_combined&quot;
  [(set (match_operand:DI 0 &quot;register_operand&quot; &quot;=d,f,d&quot;)
        (match_operand:DI 1 &quot;register_operand&quot; &quot; d,d,f&quot;))]
  &quot;&quot;
  &quot;@
   lgr  %0,%1
   ldgr %0,%1
   lgdr %0,%1&quot;
  [(set_attr &quot;cpu_facility&quot; &quot;*,new,new&quot;)])

</pre></div>

<p>with the <code>enabled</code> attribute defined like this:
</p>
<div class="smallexample">
<pre class="smallexample">
(define_attr &quot;cpu_facility&quot; &quot;standard,new&quot; (const_string &quot;standard&quot;))

(define_attr &quot;enabled&quot; &quot;&quot;
  (cond [(eq_attr &quot;cpu_facility&quot; &quot;standard&quot;) (const_int 1)
         (and (eq_attr &quot;cpu_facility&quot; &quot;new&quot;)
              (ne (symbol_ref &quot;TARGET_NEW&quot;) (const_int 0)))
         (const_int 1)]
        (const_int 0)))

</pre></div>


<hr>
<a name="Define-Constraints"></a>
<div class="header">
<p>
Next: <a href="#C-Constraint-Interface" accesskey="n" rel="next">C Constraint Interface</a>, Previous: <a href="#Disable-Insn-Alternatives" accesskey="p" rel="prev">Disable Insn Alternatives</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Machine_002dSpecific-Constraints"></a>
<h4 class="subsection">16.8.7 Defining Machine-Specific Constraints</h4>
<a name="index-defining-constraints"></a>
<a name="index-constraints_002c-defining"></a>

<p>Machine-specific constraints fall into two categories: register and
non-register constraints.  Within the latter category, constraints
which allow subsets of all possible memory or address operands should
be specially marked, to give <code>reload</code> more information.
</p>
<p>Machine-specific constraints can be given names of arbitrary length,
but they must be entirely composed of letters, digits, underscores
(&lsquo;<samp>_</samp>&rsquo;), and angle brackets (&lsquo;<samp>&lt; &gt;</samp>&rsquo;).  Like C identifiers, they
must begin with a letter or underscore.
</p>
<p>In order to avoid ambiguity in operand constraint strings, no
constraint can have a name that begins with any other constraint&rsquo;s
name.  For example, if <code>x</code> is defined as a constraint name,
<code>xy</code> may not be, and vice versa.  As a consequence of this rule,
no constraint may begin with one of the generic constraint letters:
&lsquo;<samp>E F V X g i m n o p r s</samp>&rsquo;.
</p>
<p>Register constraints correspond directly to register classes.
See <a href="#Register-Classes">Register Classes</a>.  There is thus not much flexibility in their
definitions.
</p>
<dl>
<dt><a name="index-define_005fregister_005fconstraint"></a>MD Expression: <strong>define_register_constraint</strong> <em>name regclass docstring</em></dt>
<dd><p>All three arguments are string constants.
<var>name</var> is the name of the constraint, as it will appear in
<code>match_operand</code> expressions.  If <var>name</var> is a multi-letter
constraint its length shall be the same for all constraints starting
with the same letter.  <var>regclass</var> can be either the
name of the corresponding register class (see <a href="#Register-Classes">Register Classes</a>),
or a C expression which evaluates to the appropriate register class.
If it is an expression, it must have no side effects, and it cannot
look at the operand.  The usual use of expressions is to map some
register constraints to <code>NO_REGS</code> when the register class
is not available on a given subarchitecture.
</p>
<p><var>docstring</var> is a sentence documenting the meaning of the
constraint.  Docstrings are explained further below.
</p></dd></dl>

<p>Non-register constraints are more like predicates: the constraint
definition gives a Boolean expression which indicates whether the
constraint matches.
</p>
<dl>
<dt><a name="index-define_005fconstraint"></a>MD Expression: <strong>define_constraint</strong> <em>name docstring exp</em></dt>
<dd><p>The <var>name</var> and <var>docstring</var> arguments are the same as for
<code>define_register_constraint</code>, but note that the docstring comes
immediately after the name for these expressions.  <var>exp</var> is an RTL
expression, obeying the same rules as the RTL expressions in predicate
definitions.  See <a href="#Defining-Predicates">Defining Predicates</a>, for details.  If it
evaluates true, the constraint matches; if it evaluates false, it
doesn&rsquo;t. Constraint expressions should indicate which RTL codes they
might match, just like predicate expressions.
</p>
<p><code>match_test</code> C expressions have access to the
following variables:
</p>
<dl compact="compact">
<dt><var>op</var></dt>
<dd><p>The RTL object defining the operand.
</p></dd>
<dt><var>mode</var></dt>
<dd><p>The machine mode of <var>op</var>.
</p></dd>
<dt><var>ival</var></dt>
<dd><p>&lsquo;<samp>INTVAL (<var>op</var>)</samp>&rsquo;, if <var>op</var> is a <code>const_int</code>.
</p></dd>
<dt><var>hval</var></dt>
<dd><p>&lsquo;<samp>CONST_DOUBLE_HIGH (<var>op</var>)</samp>&rsquo;, if <var>op</var> is an integer
<code>const_double</code>.
</p></dd>
<dt><var>lval</var></dt>
<dd><p>&lsquo;<samp>CONST_DOUBLE_LOW (<var>op</var>)</samp>&rsquo;, if <var>op</var> is an integer
<code>const_double</code>.
</p></dd>
<dt><var>rval</var></dt>
<dd><p>&lsquo;<samp>CONST_DOUBLE_REAL_VALUE (<var>op</var>)</samp>&rsquo;, if <var>op</var> is a floating-point
<code>const_double</code>.
</p></dd>
</dl>

<p>The <var>*val</var> variables should only be used once another piece of the
expression has verified that <var>op</var> is the appropriate kind of RTL
object.
</p></dd></dl>

<p>Most non-register constraints should be defined with
<code>define_constraint</code>.  The remaining two definition expressions
are only appropriate for constraints that should be handled specially
by <code>reload</code> if they fail to match.
</p>
<dl>
<dt><a name="index-define_005fmemory_005fconstraint"></a>MD Expression: <strong>define_memory_constraint</strong> <em>name docstring exp</em></dt>
<dd><p>Use this expression for constraints that match a subset of all memory
operands: that is, <code>reload</code> can make them match by converting the
operand to the form &lsquo;<samp>(mem&nbsp;(reg&nbsp;<var>X</var>))<!-- /@w --></samp>&rsquo;, where <var>X</var> is a
base register (from the register class specified by
<code>BASE_REG_CLASS</code>, see <a href="#Register-Classes">Register Classes</a>).
</p>
<p>For example, on the S/390, some instructions do not accept arbitrary
memory references, but only those that do not make use of an index
register.  The constraint letter &lsquo;<samp>Q</samp>&rsquo; is defined to represent a
memory address of this type.  If &lsquo;<samp>Q</samp>&rsquo; is defined with
<code>define_memory_constraint</code>, a &lsquo;<samp>Q</samp>&rsquo; constraint can handle any
memory operand, because <code>reload</code> knows it can simply copy the
memory address into a base register if required.  This is analogous to
the way an &lsquo;<samp>o</samp>&rsquo; constraint can handle any memory operand.
</p>
<p>The syntax and semantics are otherwise identical to
<code>define_constraint</code>.
</p></dd></dl>

<dl>
<dt><a name="index-define_005faddress_005fconstraint"></a>MD Expression: <strong>define_address_constraint</strong> <em>name docstring exp</em></dt>
<dd><p>Use this expression for constraints that match a subset of all address
operands: that is, <code>reload</code> can make the constraint match by
converting the operand to the form &lsquo;<samp>(reg&nbsp;<var>X</var>)<!-- /@w --></samp>&rsquo;, again
with <var>X</var> a base register.
</p>
<p>Constraints defined with <code>define_address_constraint</code> can only be
used with the <code>address_operand</code> predicate, or machine-specific
predicates that work the same way.  They are treated analogously to
the generic &lsquo;<samp>p</samp>&rsquo; constraint.
</p>
<p>The syntax and semantics are otherwise identical to
<code>define_constraint</code>.
</p></dd></dl>

<p>For historical reasons, names beginning with the letters &lsquo;<samp>G H</samp>&rsquo;
are reserved for constraints that match only <code>const_double</code>s, and
names beginning with the letters &lsquo;<samp>I J K L M N O P</samp>&rsquo; are reserved
for constraints that match only <code>const_int</code>s.  This may change in
the future.  For the time being, constraints with these names must be
written in a stylized form, so that <code>genpreds</code> can tell you did
it correctly:
</p>
<div class="smallexample">
<pre class="smallexample">(define_constraint &quot;[<var>GHIJKLMNOP</var>]&hellip;&quot;
  &quot;<var>doc</var>&hellip;&quot;
  (and (match_code &quot;const_int&quot;)  ; <span class="roman"><code>const_double</code> for G/H</span>
       <var>condition</var>&hellip;))            ; <span class="roman">usually a <code>match_test</code></span>
</pre></div>

<p>It is fine to use names beginning with other letters for constraints
that match <code>const_double</code>s or <code>const_int</code>s.
</p>
<p>Each docstring in a constraint definition should be one or more complete
sentences, marked up in Texinfo format.  <em>They are currently unused.</em>
In the future they will be copied into the GCC manual, in <a href="#Machine-Constraints">Machine Constraints</a>, replacing the hand-maintained tables currently found in
that section.  Also, in the future the compiler may use this to give
more helpful diagnostics when poor choice of <code>asm</code> constraints
causes a reload failure.
</p>
<p>If you put the pseudo-Texinfo directive &lsquo;<samp>@internal</samp>&rsquo; at the
beginning of a docstring, then (in the future) it will appear only in
the internals manual&rsquo;s version of the machine-specific constraint tables.
Use this for constraints that should not appear in <code>asm</code> statements.
</p>
<hr>
<a name="C-Constraint-Interface"></a>
<div class="header">
<p>
Previous: <a href="#Define-Constraints" accesskey="p" rel="prev">Define Constraints</a>, Up: <a href="#Constraints" accesskey="u" rel="up">Constraints</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Testing-constraints-from-C"></a>
<h4 class="subsection">16.8.8 Testing constraints from C</h4>
<a name="index-testing-constraints"></a>
<a name="index-constraints_002c-testing"></a>

<p>It is occasionally useful to test a constraint from C code rather than
implicitly via the constraint string in a <code>match_operand</code>.  The
generated file <samp>tm_p.h</samp> declares a few interfaces for working
with machine-specific constraints.  None of these interfaces work with
the generic constraints described in <a href="#Simple-Constraints">Simple Constraints</a>.  This
may change in the future.
</p>
<p><strong>Warning:</strong> <samp>tm_p.h</samp> may declare other functions that
operate on constraints, besides the ones documented here.  Do not use
those functions from machine-dependent code.  They exist to implement
the old constraint interface that machine-independent components of
the compiler still expect.  They will change or disappear in the
future.
</p>
<p>Some valid constraint names are not valid C identifiers, so there is a
mangling scheme for referring to them from C.  Constraint names that
do not contain angle brackets or underscores are left unchanged.
Underscores are doubled, each &lsquo;<samp>&lt;</samp>&rsquo; is replaced with &lsquo;<samp>_l</samp>&rsquo;, and
each &lsquo;<samp>&gt;</samp>&rsquo; with &lsquo;<samp>_g</samp>&rsquo;.  Here are some examples:
</p>
<div class="example">
<table>
<tr><td><pre class="example"><strong>Original</strong></pre></td><td><pre class="example"><strong>Mangled</strong></pre></td></tr>
<tr><td><pre class="example"><code>x</code></pre></td><td><pre class="example"><code>x</code></pre></td></tr>
<tr><td><pre class="example"><code>P42x</code></pre></td><td><pre class="example"><code>P42x</code></pre></td></tr>
<tr><td><pre class="example"><code>P4_x</code></pre></td><td><pre class="example"><code>P4__x</code></pre></td></tr>
<tr><td><pre class="example"><code>P4&gt;x</code></pre></td><td><pre class="example"><code>P4_gx</code></pre></td></tr>
<tr><td><pre class="example"><code>P4&gt;&gt;</code></pre></td><td><pre class="example"><code>P4_g_g</code></pre></td></tr>
<tr><td><pre class="example"><code>P4_g&gt;</code></pre></td><td><pre class="example"><code>P4__g_g</code></pre></td></tr>
</table>
</div>

<p>Throughout this section, the variable <var>c</var> is either a constraint
in the abstract sense, or a constant from <code>enum constraint_num</code>;
the variable <var>m</var> is a mangled constraint name (usually as part of
a larger identifier).
</p>
<dl>
<dt><a name="index-constraint_005fnum"></a>Enum: <strong>constraint_num</strong></dt>
<dd><p>For each machine-specific constraint, there is a corresponding
enumeration constant: &lsquo;<samp>CONSTRAINT_</samp>&rsquo; plus the mangled name of the
constraint.  Functions that take an <code>enum constraint_num</code> as an
argument expect one of these constants.
</p>
<p>Machine-independent constraints do not have associated constants.
This may change in the future.
</p></dd></dl>

<dl>
<dt><a name="index-satisfies_005fconstraint_005f"></a>Function: <em>inline bool</em> <strong>satisfies_constraint_</strong> <em><var>m</var> (rtx <var>exp</var>)</em></dt>
<dd><p>For each machine-specific, non-register constraint <var>m</var>, there is
one of these functions; it returns <code>true</code> if <var>exp</var> satisfies the
constraint.  These functions are only visible if <samp>rtl.h</samp> was included
before <samp>tm_p.h</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-constraint_005fsatisfied_005fp"></a>Function: <em>bool</em> <strong>constraint_satisfied_p</strong> <em>(rtx <var>exp</var>, enum constraint_num <var>c</var>)</em></dt>
<dd><p>Like the <code>satisfies_constraint_<var>m</var></code> functions, but the
constraint to test is given as an argument, <var>c</var>.  If <var>c</var>
specifies a register constraint, this function will always return
<code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-regclass_005ffor_005fconstraint"></a>Function: <em>enum reg_class</em> <strong>regclass_for_constraint</strong> <em>(enum constraint_num <var>c</var>)</em></dt>
<dd><p>Returns the register class associated with <var>c</var>.  If <var>c</var> is not
a register constraint, or those registers are not available for the
currently selected subtarget, returns <code>NO_REGS</code>.
</p></dd></dl>

<p>Here is an example use of <code>satisfies_constraint_<var>m</var></code>.  In
peephole optimizations (see <a href="#Peephole-Definitions">Peephole Definitions</a>), operand
constraint strings are ignored, so if there are relevant constraints,
they must be tested in the C condition.  In the example, the
optimization is applied if operand 2 does <em>not</em> satisfy the
&lsquo;<samp>K</samp>&rsquo; constraint.  (This is a simplified version of a peephole
definition from the i386 machine description.)
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole2
  [(match_scratch:SI 3 &quot;r&quot;)
   (set (match_operand:SI 0 &quot;register_operand&quot; &quot;&quot;)
        (mult:SI (match_operand:SI 1 &quot;memory_operand&quot; &quot;&quot;)
                 (match_operand:SI 2 &quot;immediate_operand&quot; &quot;&quot;)))]

  &quot;!satisfies_constraint_K (operands[2])&quot;

  [(set (match_dup 3) (match_dup 1))
   (set (match_dup 0) (mult:SI (match_dup 3) (match_dup 2)))]

  &quot;&quot;)
</pre></div>

<hr>
<a name="Standard-Names"></a>
<div class="header">
<p>
Next: <a href="#Pattern-Ordering" accesskey="n" rel="next">Pattern Ordering</a>, Previous: <a href="#Constraints" accesskey="p" rel="prev">Constraints</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Standard-Pattern-Names-For-Generation"></a>
<h3 class="section">16.9 Standard Pattern Names For Generation</h3>
<a name="index-standard-pattern-names"></a>
<a name="index-pattern-names"></a>
<a name="index-names_002c-pattern"></a>

<p>Here is a table of the instruction names that are meaningful in the RTL
generation pass of the compiler.  Giving one of these names to an
instruction pattern tells the RTL generation pass that it can use the
pattern to accomplish a certain task.
</p>
<dl compact="compact">
<dd><a name="index-movm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mov<var>m</var></samp>&rsquo;</dt>
<dd><p>Here <var>m</var> stands for a two-letter machine mode name, in lowercase.
This instruction pattern moves data with that machine mode from operand
1 to operand 0.  For example, &lsquo;<samp>movsi</samp>&rsquo; moves full-word data.
</p>
<p>If operand 0 is a <code>subreg</code> with mode <var>m</var> of a register whose
own mode is wider than <var>m</var>, the effect of this instruction is
to store the specified value in the part of the register that corresponds
to mode <var>m</var>.  Bits outside of <var>m</var>, but which are within the
same target word as the <code>subreg</code> are undefined.  Bits which are
outside the target word are left unchanged.
</p>
<p>This class of patterns is special in several ways.  First of all, each
of these names up to and including full word size <em>must</em> be defined,
because there is no other way to copy a datum from one place to another.
If there are patterns accepting operands in larger modes,
&lsquo;<samp>mov<var>m</var></samp>&rsquo; must be defined for integer modes of those sizes.
</p>
<p>Second, these patterns are not used solely in the RTL generation pass.
Even the reload pass can generate move insns to copy values from stack
slots into temporary registers.  When it does so, one of the operands is
a hard register and the other is an operand that can need to be reloaded
into a register.
</p>
<a name="index-force_005freg"></a>
<p>Therefore, when given such a pair of operands, the pattern must generate
RTL which needs no reloading and needs no temporary registers&mdash;no
registers other than the operands.  For example, if you support the
pattern with a <code>define_expand</code>, then in such a case the
<code>define_expand</code> mustn&rsquo;t call <code>force_reg</code> or any other such
function which might generate new pseudo registers.
</p>
<p>This requirement exists even for subword modes on a RISC machine where
fetching those modes from memory normally requires several insns and
some temporary registers.
</p>
<a name="index-change_005faddress"></a>
<p>During reload a memory reference with an invalid address may be passed
as an operand.  Such an address will be replaced with a valid address
later in the reload pass.  In this case, nothing may be done with the
address except to use it as it stands.  If it is copied, it will not be
replaced with a valid address.  No attempt should be made to make such
an address into a valid address and no routine (such as
<code>change_address</code>) that will do so may be called.  Note that
<code>general_operand</code> will fail when applied to such an address.
</p>
<a name="index-reload_005fin_005fprogress"></a>
<p>The global variable <code>reload_in_progress</code> (which must be explicitly
declared if required) can be used to determine whether such special
handling is required.
</p>
<p>The variety of operands that have reloads depends on the rest of the
machine description, but typically on a RISC machine these can only be
pseudo registers that did not get hard registers, while on other
machines explicit memory references will get optional reloads.
</p>
<p>If a scratch register is required to move an object to or from memory,
it can be allocated using <code>gen_reg_rtx</code> prior to life analysis.
</p>
<p>If there are cases which need scratch registers during or after reload,
you must provide an appropriate secondary_reload target hook.
</p>
<a name="index-can_005fcreate_005fpseudo_005fp"></a>
<p>The macro <code>can_create_pseudo_p</code> can be used to determine if it
is unsafe to create new pseudo registers.  If this variable is nonzero, then
it is unsafe to call <code>gen_reg_rtx</code> to allocate a new pseudo.
</p>
<p>The constraints on a &lsquo;<samp>mov<var>m</var></samp>&rsquo; must permit moving any hard
register to any other hard register provided that
<code>HARD_REGNO_MODE_OK</code> permits mode <var>m</var> in both registers and
<code>TARGET_REGISTER_MOVE_COST</code> applied to their classes returns a value
of 2.
</p>
<p>It is obligatory to support floating point &lsquo;<samp>mov<var>m</var></samp>&rsquo;
instructions into and out of any registers that can hold fixed point
values, because unions and structures (which have modes <code>SImode</code> or
<code>DImode</code>) can be in those registers and they may have floating
point members.
</p>
<p>There may also be a need to support fixed point &lsquo;<samp>mov<var>m</var></samp>&rsquo;
instructions in and out of floating point registers.  Unfortunately, I
have forgotten why this was so, and I don&rsquo;t know whether it is still
true.  If <code>HARD_REGNO_MODE_OK</code> rejects fixed point values in
floating point registers, then the constraints of the fixed point
&lsquo;<samp>mov<var>m</var></samp>&rsquo; instructions must be designed to avoid ever trying to
reload into a floating point register.
</p>
<a name="index-reload_005fin-instruction-pattern"></a>
<a name="index-reload_005fout-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>reload_in<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>reload_out<var>m</var></samp>&rsquo;</dt>
<dd><p>These named patterns have been obsoleted by the target hook
<code>secondary_reload</code>.
</p>
<p>Like &lsquo;<samp>mov<var>m</var></samp>&rsquo;, but used when a scratch register is required to
move between operand 0 and operand 1.  Operand 2 describes the scratch
register.  See the discussion of the <code>SECONDARY_RELOAD_CLASS</code>
macro in see <a href="#Register-Classes">Register Classes</a>.
</p>
<p>There are special restrictions on the form of the <code>match_operand</code>s
used in these patterns.  First, only the predicate for the reload
operand is examined, i.e., <code>reload_in</code> examines operand 1, but not
the predicates for operand 0 or 2.  Second, there may be only one
alternative in the constraints.  Third, only a single register class
letter may be used for the constraint; subsequent constraint letters
are ignored.  As a special exception, an empty constraint string
matches the <code>ALL_REGS</code> register class.  This may relieve ports
of the burden of defining an <code>ALL_REGS</code> constraint letter just
for these patterns.
</p>
<a name="index-movstrictm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>movstrict<var>m</var></samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>mov<var>m</var></samp>&rsquo; except that if operand 0 is a <code>subreg</code>
with mode <var>m</var> of a register whose natural mode is wider,
the &lsquo;<samp>movstrict<var>m</var></samp>&rsquo; instruction is guaranteed not to alter
any of the register except the part which belongs to mode <var>m</var>.
</p>
<a name="index-movmisalignm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>movmisalign<var>m</var></samp>&rsquo;</dt>
<dd><p>This variant of a move pattern is designed to load or store a value
from a memory address that is not naturally aligned for its mode.
For a store, the memory will be in operand 0; for a load, the memory
will be in operand 1.  The other operand is guaranteed not to be a
memory, so that it&rsquo;s easy to tell whether this is a load or store.
</p>
<p>This pattern is used by the autovectorizer, and when expanding a
<code>MISALIGNED_INDIRECT_REF</code> expression.
</p>
<a name="index-load_005fmultiple-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>load_multiple</samp>&rsquo;</dt>
<dd><p>Load several consecutive memory locations into consecutive registers.
Operand 0 is the first of the consecutive registers, operand 1
is the first memory location, and operand 2 is a constant: the
number of consecutive registers.
</p>
<p>Define this only if the target machine really has such an instruction;
do not define this if the most efficient way of loading consecutive
registers from memory is to do them one at a time.
</p>
<p>On some machines, there are restrictions as to which consecutive
registers can be stored into memory, such as particular starting or
ending register numbers or only a range of valid counts.  For those
machines, use a <code>define_expand</code> (see <a href="#Expander-Definitions">Expander Definitions</a>)
and make the pattern fail if the restrictions are not met.
</p>
<p>Write the generated insn as a <code>parallel</code> with elements being a
<code>set</code> of one register from the appropriate memory location (you may
also need <code>use</code> or <code>clobber</code> elements).  Use a
<code>match_parallel</code> (see <a href="#RTL-Template">RTL Template</a>) to recognize the insn.  See
<samp>rs6000.md</samp> for examples of the use of this insn pattern.
</p>
<a name="index-store_005fmultiple-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>store_multiple</samp>&rsquo;</dt>
<dd><p>Similar to &lsquo;<samp>load_multiple</samp>&rsquo;, but store several consecutive registers
into consecutive memory locations.  Operand 0 is the first of the
consecutive memory locations, operand 1 is the first register, and
operand 2 is a constant: the number of consecutive registers.
</p>
<a name="index-vec_005fload_005flanesmn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_load_lanes<var>m</var><var>n</var></samp>&rsquo;</dt>
<dd><p>Perform an interleaved load of several vectors from memory operand 1
into register operand 0.  Both operands have mode <var>m</var>.  The register
operand is viewed as holding consecutive vectors of mode <var>n</var>,
while the memory operand is a flat array that contains the same number
of elements.  The operation is equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">int c = GET_MODE_SIZE (<var>m</var>) / GET_MODE_SIZE (<var>n</var>);
for (j = 0; j &lt; GET_MODE_NUNITS (<var>n</var>); j++)
  for (i = 0; i &lt; c; i++)
    operand0[i][j] = operand1[j * c + i];
</pre></div>

<p>For example, &lsquo;<samp>vec_load_lanestiv4hi</samp>&rsquo; loads 8 16-bit values
from memory into a register of mode &lsquo;<samp>TI</samp>&rsquo;.  The register
contains two consecutive vectors of mode &lsquo;<samp>V4HI</samp>&rsquo;.
</p>
<p>This pattern can only be used if:
</p><div class="smallexample">
<pre class="smallexample">TARGET_ARRAY_MODE_SUPPORTED_P (<var>n</var>, <var>c</var>)
</pre></div>
<p>is true.  GCC assumes that, if a target supports this kind of
instruction for some mode <var>n</var>, it also supports unaligned
loads for vectors of mode <var>n</var>.
</p>
<a name="index-vec_005fstore_005flanesmn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_store_lanes<var>m</var><var>n</var></samp>&rsquo;</dt>
<dd><p>Equivalent to &lsquo;<samp>vec_load_lanes<var>m</var><var>n</var></samp>&rsquo;, with the memory
and register operands reversed.  That is, the instruction is
equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">int c = GET_MODE_SIZE (<var>m</var>) / GET_MODE_SIZE (<var>n</var>);
for (j = 0; j &lt; GET_MODE_NUNITS (<var>n</var>); j++)
  for (i = 0; i &lt; c; i++)
    operand0[j * c + i] = operand1[i][j];
</pre></div>

<p>for a memory operand 0 and register operand 1.
</p>
<a name="index-vec_005fsetm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_set<var>m</var></samp>&rsquo;</dt>
<dd><p>Set given field in the vector value.  Operand 0 is the vector to modify,
operand 1 is new value of field and operand 2 specify the field index.
</p>
<a name="index-vec_005fextractm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_extract<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract given field from the vector value.  Operand 1 is the vector, operand 2
specify field index and operand 0 place to store value into.
</p>
<a name="index-vec_005finitm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_init<var>m</var></samp>&rsquo;</dt>
<dd><p>Initialize the vector to given values.  Operand 0 is the vector to initialize
and operand 1 is parallel containing values for individual fields.
</p>
<a name="index-vcondmn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vcond<var>m</var><var>n</var></samp>&rsquo;</dt>
<dd><p>Output a conditional vector move.  Operand 0 is the destination to
receive a combination of operand 1 and operand 2, which are of mode <var>m</var>,
dependent on the outcome of the predicate in operand 3 which is a
vector comparison with operands of mode <var>n</var> in operands 4 and 5.  The
modes <var>m</var> and <var>n</var> should have the same size.  Operand 0
will be set to the value <var>op1</var> &amp; <var>msk</var> | <var>op2</var> &amp; ~<var>msk</var>
where <var>msk</var> is computed by element-wise evaluation of the vector
comparison with a truth value of all-ones and a false value of all-zeros.
</p>
<a name="index-vec_005fpermm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_perm<var>m</var></samp>&rsquo;</dt>
<dd><p>Output a (variable) vector permutation.  Operand 0 is the destination
to receive elements from operand 1 and operand 2, which are of mode
<var>m</var>.  Operand 3 is the <em>selector</em>.  It is an integral mode
vector of the same width and number of elements as mode <var>m</var>.
</p>
<p>The input elements are numbered from 0 in operand 1 through
<em>2*<var>N</var>-1</em> in operand 2.  The elements of the selector must
be computed modulo <em>2*<var>N</var></em>.  Note that if
<code>rtx_equal_p(operand1, operand2)</code>, this can be implemented
with just operand 1 and selector elements modulo <var>N</var>.
</p>
<p>In order to make things easy for a number of targets, if there is no
&lsquo;<samp>vec_perm</samp>&rsquo; pattern for mode <var>m</var>, but there is for mode <var>q</var>
where <var>q</var> is a vector of <code>QImode</code> of the same width as <var>m</var>,
the middle-end will lower the mode <var>m</var> <code>VEC_PERM_EXPR</code> to
mode <var>q</var>.
</p>
<a name="index-vec_005fperm_005fconstm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_perm_const<var>m</var></samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>vec_perm</samp>&rsquo; except that the permutation is a compile-time
constant.  That is, operand 3, the <em>selector</em>, is a <code>CONST_VECTOR</code>.
</p>
<p>Some targets cannot perform a permutation with a variable selector,
but can efficiently perform a constant permutation.  Further, the
target hook <code>vec_perm_ok</code> is queried to determine if the 
specific constant permutation is available efficiently; the named
pattern is never expanded without <code>vec_perm_ok</code> returning true.
</p>
<p>There is no need for a target to supply both &lsquo;<samp>vec_perm<var>m</var></samp>&rsquo;
and &lsquo;<samp>vec_perm_const<var>m</var></samp>&rsquo; if the former can trivially implement
the operation with, say, the vector constant loaded into a register.
</p>
<a name="index-pushm1-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>push<var>m</var>1</samp>&rsquo;</dt>
<dd><p>Output a push instruction.  Operand 0 is value to push.  Used only when
<code>PUSH_ROUNDING</code> is defined.  For historical reason, this pattern may be
missing and in such case an <code>mov</code> expander is used instead, with a
<code>MEM</code> expression forming the push operation.  The <code>mov</code> expander
method is deprecated.
</p>
<a name="index-addm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>add<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Add operand 2 and operand 1, storing the result in operand 0.  All operands
must have mode <var>m</var>.  This can be used even on two-address machines, by
means of constraints requiring operands 1 and 0 to be the same location.
</p>
<a name="index-ssaddm3-instruction-pattern"></a>
<a name="index-usaddm3-instruction-pattern"></a>
<a name="index-subm3-instruction-pattern"></a>
<a name="index-sssubm3-instruction-pattern"></a>
<a name="index-ussubm3-instruction-pattern"></a>
<a name="index-mulm3-instruction-pattern"></a>
<a name="index-ssmulm3-instruction-pattern"></a>
<a name="index-usmulm3-instruction-pattern"></a>
<a name="index-divm3-instruction-pattern"></a>
<a name="index-ssdivm3-instruction-pattern"></a>
<a name="index-udivm3-instruction-pattern"></a>
<a name="index-usdivm3-instruction-pattern"></a>
<a name="index-modm3-instruction-pattern"></a>
<a name="index-umodm3-instruction-pattern"></a>
<a name="index-uminm3-instruction-pattern"></a>
<a name="index-umaxm3-instruction-pattern"></a>
<a name="index-andm3-instruction-pattern"></a>
<a name="index-iorm3-instruction-pattern"></a>
<a name="index-xorm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ssadd<var>m</var>3</samp>&rsquo;, &lsquo;<samp>usadd<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>sub<var>m</var>3</samp>&rsquo;, &lsquo;<samp>sssub<var>m</var>3</samp>&rsquo;, &lsquo;<samp>ussub<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>mul<var>m</var>3</samp>&rsquo;, &lsquo;<samp>ssmul<var>m</var>3</samp>&rsquo;, &lsquo;<samp>usmul<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>div<var>m</var>3</samp>&rsquo;, &lsquo;<samp>ssdiv<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>udiv<var>m</var>3</samp>&rsquo;, &lsquo;<samp>usdiv<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>mod<var>m</var>3</samp>&rsquo;, &lsquo;<samp>umod<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>umin<var>m</var>3</samp>&rsquo;, &lsquo;<samp>umax<var>m</var>3</samp>&rsquo;</dt>
<dt>&lsquo;<samp>and<var>m</var>3</samp>&rsquo;, &lsquo;<samp>ior<var>m</var>3</samp>&rsquo;, &lsquo;<samp>xor<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Similar, for other arithmetic operations.
</p>
<a name="index-fmam4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fma<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Multiply operand 2 and operand 1, then add operand 3, storing the
result in operand 0 without doing an intermediate rounding step.  All
operands must have mode <var>m</var>.  This pattern is used to implement
the <code>fma</code>, <code>fmaf</code>, and <code>fmal</code> builtin functions from
the ISO C99 standard.
</p>
<a name="index-fmsm4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fms<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>fma<var>m</var>4</code>, except operand 3 subtracted from the
product instead of added to the product.  This is represented
in the rtl as
</p>
<div class="smallexample">
<pre class="smallexample">(fma:<var>m</var> <var>op1</var> <var>op2</var> (neg:<var>m</var> <var>op3</var>))
</pre></div>

<a name="index-fnmam4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fnma<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>fma<var>m</var>4</code> except that the intermediate product
is negated before being added to operand 3.  This is represented
in the rtl as
</p>
<div class="smallexample">
<pre class="smallexample">(fma:<var>m</var> (neg:<var>m</var> <var>op1</var>) <var>op2</var> <var>op3</var>)
</pre></div>

<a name="index-fnmsm4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fnms<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>fms<var>m</var>4</code> except that the intermediate product
is negated before subtracting operand 3.  This is represented
in the rtl as
</p>
<div class="smallexample">
<pre class="smallexample">(fma:<var>m</var> (neg:<var>m</var> <var>op1</var>) <var>op2</var> (neg:<var>m</var> <var>op3</var>))
</pre></div>

<a name="index-minm3-instruction-pattern"></a>
<a name="index-maxm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>smin<var>m</var>3</samp>&rsquo;, &lsquo;<samp>smax<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Signed minimum and maximum operations.  When used with floating point,
if both operands are zeros, or if either operand is <code>NaN</code>, then
it is unspecified which of the two operands is returned as the result.
</p>
<a name="index-reduc_005fsmin_005fm-instruction-pattern"></a>
<a name="index-reduc_005fsmax_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>reduc_smin_<var>m</var></samp>&rsquo;, &lsquo;<samp>reduc_smax_<var>m</var></samp>&rsquo;</dt>
<dd><p>Find the signed minimum/maximum of the elements of a vector. The vector is
operand 1, and the scalar result is stored in the least significant bits of
operand 0 (also a vector). The output and input vector should have the same
modes.
</p>
<a name="index-reduc_005fumin_005fm-instruction-pattern"></a>
<a name="index-reduc_005fumax_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>reduc_umin_<var>m</var></samp>&rsquo;, &lsquo;<samp>reduc_umax_<var>m</var></samp>&rsquo;</dt>
<dd><p>Find the unsigned minimum/maximum of the elements of a vector. The vector is
operand 1, and the scalar result is stored in the least significant bits of
operand 0 (also a vector). The output and input vector should have the same
modes.
</p>
<a name="index-reduc_005fsplus_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>reduc_splus_<var>m</var></samp>&rsquo;</dt>
<dd><p>Compute the sum of the signed elements of a vector. The vector is operand 1,
and the scalar result is stored in the least significant bits of operand 0
(also a vector). The output and input vector should have the same modes.
</p>
<a name="index-reduc_005fuplus_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>reduc_uplus_<var>m</var></samp>&rsquo;</dt>
<dd><p>Compute the sum of the unsigned elements of a vector. The vector is operand 1,
and the scalar result is stored in the least significant bits of operand 0
(also a vector). The output and input vector should have the same modes.
</p>
<a name="index-sdot_005fprodm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sdot_prod<var>m</var></samp>&rsquo;</dt>
<dd><a name="index-udot_005fprodm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>udot_prod<var>m</var></samp>&rsquo;</dt>
<dd><p>Compute the sum of the products of two signed/unsigned elements.
Operand 1 and operand 2 are of the same mode. Their product, which is of a
wider mode, is computed and added to operand 3. Operand 3 is of a mode equal or
wider than the mode of the product. The result is placed in operand 0, which
is of the same mode as operand 3.
</p>
<a name="index-ssum_005fwidenm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ssum_widen<var>m3</var></samp>&rsquo;</dt>
<dd><a name="index-usum_005fwidenm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>usum_widen<var>m3</var></samp>&rsquo;</dt>
<dd><p>Operands 0 and 2 are of the same mode, which is wider than the mode of
operand 1. Add operand 1 to operand 2 and place the widened result in
operand 0. (This is used express accumulation of elements into an accumulator
of a wider mode.)
</p>
<a name="index-vec_005fshl_005fm-instruction-pattern"></a>
<a name="index-vec_005fshr_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_shl_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_shr_<var>m</var></samp>&rsquo;</dt>
<dd><p>Whole vector left/right shift in bits.
Operand 1 is a vector to be shifted.
Operand 2 is an integer shift amount in bits.
Operand 0 is where the resulting shifted vector is stored.
The output and input vectors should have the same modes.
</p>
<a name="index-vec_005fpack_005ftrunc_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_pack_trunc_<var>m</var></samp>&rsquo;</dt>
<dd><p>Narrow (demote) and merge the elements of two vectors. Operands 1 and 2
are vectors of the same mode having N integral or floating point elements
of size S.  Operand 0 is the resulting vector in which 2*N elements of
size N/2 are concatenated after narrowing them down using truncation.
</p>
<a name="index-vec_005fpack_005fssat_005fm-instruction-pattern"></a>
<a name="index-vec_005fpack_005fusat_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_pack_ssat_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_pack_usat_<var>m</var></samp>&rsquo;</dt>
<dd><p>Narrow (demote) and merge the elements of two vectors.  Operands 1 and 2
are vectors of the same mode having N integral elements of size S.
Operand 0 is the resulting vector in which the elements of the two input
vectors are concatenated after narrowing them down using signed/unsigned
saturating arithmetic.
</p>
<a name="index-vec_005fpack_005fsfix_005ftrunc_005fm-instruction-pattern"></a>
<a name="index-vec_005fpack_005fufix_005ftrunc_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_pack_sfix_trunc_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_pack_ufix_trunc_<var>m</var></samp>&rsquo;</dt>
<dd><p>Narrow, convert to signed/unsigned integral type and merge the elements
of two vectors.  Operands 1 and 2 are vectors of the same mode having N
floating point elements of size S.  Operand 0 is the resulting vector
in which 2*N elements of size N/2 are concatenated.
</p>
<a name="index-vec_005funpacks_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005funpacks_005flo_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_unpacks_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_unpacks_lo_<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract and widen (promote) the high/low part of a vector of signed
integral or floating point elements.  The input vector (operand 1) has N
elements of size S.  Widen (promote) the high/low elements of the vector
using signed or floating point extension and place the resulting N/2
values of size 2*S in the output vector (operand 0).
</p>
<a name="index-vec_005funpacku_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005funpacku_005flo_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_unpacku_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_unpacku_lo_<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract and widen (promote) the high/low part of a vector of unsigned
integral elements.  The input vector (operand 1) has N elements of size S.
Widen (promote) the high/low elements of the vector using zero extension and
place the resulting N/2 values of size 2*S in the output vector (operand 0).
</p>
<a name="index-vec_005funpacks_005ffloat_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005funpacks_005ffloat_005flo_005fm-instruction-pattern"></a>
<a name="index-vec_005funpacku_005ffloat_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005funpacku_005ffloat_005flo_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_unpacks_float_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_unpacks_float_lo_<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>vec_unpacku_float_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_unpacku_float_lo_<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract, convert to floating point type and widen the high/low part of a
vector of signed/unsigned integral elements.  The input vector (operand 1)
has N elements of size S.  Convert the high/low elements of the vector using
floating point conversion and place the resulting N/2 values of size 2*S in
the output vector (operand 0).
</p>
<a name="index-vec_005fwiden_005fumult_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fumult_005flo_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsmult_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsmult_005flo_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fumult_005feven_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fumult_005fodd_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsmult_005feven_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsmult_005fodd_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_widen_umult_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_umult_lo_<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>vec_widen_smult_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_smult_lo_<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>vec_widen_umult_even_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_umult_odd_<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>vec_widen_smult_even_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_smult_odd_<var>m</var></samp>&rsquo;</dt>
<dd><p>Signed/Unsigned widening multiplication.  The two inputs (operands 1 and 2)
are vectors with N signed/unsigned elements of size S.  Multiply the high/low
or even/odd elements of the two vectors, and put the N/2 products of size 2*S
in the output vector (operand 0).
</p>
<a name="index-vec_005fwiden_005fushiftl_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fushiftl_005flo_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsshiftl_005fhi_005fm-instruction-pattern"></a>
<a name="index-vec_005fwiden_005fsshiftl_005flo_005fm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vec_widen_ushiftl_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_ushiftl_lo_<var>m</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>vec_widen_sshiftl_hi_<var>m</var></samp>&rsquo;, &lsquo;<samp>vec_widen_sshiftl_lo_<var>m</var></samp>&rsquo;</dt>
<dd><p>Signed/Unsigned widening shift left.  The first input (operand 1) is a vector
with N signed/unsigned elements of size S.  Operand 2 is a constant.  Shift
the high/low elements of operand 1, and put the N/2 results of size 2*S in the
output vector (operand 0).
</p>
<a name="index-mulhisi3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mulhisi3</samp>&rsquo;</dt>
<dd><p>Multiply operands 1 and 2, which have mode <code>HImode</code>, and store
a <code>SImode</code> product in operand 0.
</p>
<a name="index-mulqihi3-instruction-pattern"></a>
<a name="index-mulsidi3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mulqihi3</samp>&rsquo;, &lsquo;<samp>mulsidi3</samp>&rsquo;</dt>
<dd><p>Similar widening-multiplication instructions of other widths.
</p>
<a name="index-umulqihi3-instruction-pattern"></a>
<a name="index-umulhisi3-instruction-pattern"></a>
<a name="index-umulsidi3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>umulqihi3</samp>&rsquo;, &lsquo;<samp>umulhisi3</samp>&rsquo;, &lsquo;<samp>umulsidi3</samp>&rsquo;</dt>
<dd><p>Similar widening-multiplication instructions that do unsigned
multiplication.
</p>
<a name="index-usmulqihi3-instruction-pattern"></a>
<a name="index-usmulhisi3-instruction-pattern"></a>
<a name="index-usmulsidi3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>usmulqihi3</samp>&rsquo;, &lsquo;<samp>usmulhisi3</samp>&rsquo;, &lsquo;<samp>usmulsidi3</samp>&rsquo;</dt>
<dd><p>Similar widening-multiplication instructions that interpret the first
operand as unsigned and the second operand as signed, then do a signed
multiplication.
</p>
<a name="index-smulm3_005fhighpart-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>smul<var>m</var>3_highpart</samp>&rsquo;</dt>
<dd><p>Perform a signed multiplication of operands 1 and 2, which have mode
<var>m</var>, and store the most significant half of the product in operand 0.
The least significant half of the product is discarded.
</p>
<a name="index-umulm3_005fhighpart-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>umul<var>m</var>3_highpart</samp>&rsquo;</dt>
<dd><p>Similar, but the multiplication is unsigned.
</p>
<a name="index-maddmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>madd<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Multiply operands 1 and 2, sign-extend them to mode <var>n</var>, add
operand 3, and store the result in operand 0.  Operands 1 and 2
have mode <var>m</var> and operands 0 and 3 have mode <var>n</var>.
Both modes must be integer or fixed-point modes and <var>n</var> must be twice
the size of <var>m</var>.
</p>
<p>In other words, <code>madd<var>m</var><var>n</var>4</code> is like
<code>mul<var>m</var><var>n</var>3</code> except that it also adds operand 3.
</p>
<p>These instructions are not allowed to <code>FAIL</code>.
</p>
<a name="index-umaddmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>umadd<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>madd<var>m</var><var>n</var>4</code>, but zero-extend the multiplication
operands instead of sign-extending them.
</p>
<a name="index-ssmaddmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ssmadd<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>madd<var>m</var><var>n</var>4</code>, but all involved operations must be
signed-saturating.
</p>
<a name="index-usmaddmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>usmadd<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>umadd<var>m</var><var>n</var>4</code>, but all involved operations must be
unsigned-saturating.
</p>
<a name="index-msubmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>msub<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Multiply operands 1 and 2, sign-extend them to mode <var>n</var>, subtract the
result from operand 3, and store the result in operand 0.  Operands 1 and 2
have mode <var>m</var> and operands 0 and 3 have mode <var>n</var>.
Both modes must be integer or fixed-point modes and <var>n</var> must be twice
the size of <var>m</var>.
</p>
<p>In other words, <code>msub<var>m</var><var>n</var>4</code> is like
<code>mul<var>m</var><var>n</var>3</code> except that it also subtracts the result
from operand 3.
</p>
<p>These instructions are not allowed to <code>FAIL</code>.
</p>
<a name="index-umsubmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>umsub<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>msub<var>m</var><var>n</var>4</code>, but zero-extend the multiplication
operands instead of sign-extending them.
</p>
<a name="index-ssmsubmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ssmsub<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>msub<var>m</var><var>n</var>4</code>, but all involved operations must be
signed-saturating.
</p>
<a name="index-usmsubmn4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>usmsub<var>m</var><var>n</var>4</samp>&rsquo;</dt>
<dd><p>Like <code>umsub<var>m</var><var>n</var>4</code>, but all involved operations must be
unsigned-saturating.
</p>
<a name="index-divmodm4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>divmod<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Signed division that produces both a quotient and a remainder.
Operand 1 is divided by operand 2 to produce a quotient stored
in operand 0 and a remainder stored in operand 3.
</p>
<p>For machines with an instruction that produces both a quotient and a
remainder, provide a pattern for &lsquo;<samp>divmod<var>m</var>4</samp>&rsquo; but do not
provide patterns for &lsquo;<samp>div<var>m</var>3</samp>&rsquo; and &lsquo;<samp>mod<var>m</var>3</samp>&rsquo;.  This
allows optimization in the relatively common case when both the quotient
and remainder are computed.
</p>
<p>If an instruction that just produces a quotient or just a remainder
exists and is more efficient than the instruction that produces both,
write the output routine of &lsquo;<samp>divmod<var>m</var>4</samp>&rsquo; to call
<code>find_reg_note</code> and look for a <code>REG_UNUSED</code> note on the
quotient or remainder and generate the appropriate instruction.
</p>
<a name="index-udivmodm4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>udivmod<var>m</var>4</samp>&rsquo;</dt>
<dd><p>Similar, but does unsigned division.
</p>
<a name="shift-patterns"></a><a name="index-ashlm3-instruction-pattern"></a>
<a name="index-ssashlm3-instruction-pattern"></a>
<a name="index-usashlm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ashl<var>m</var>3</samp>&rsquo;, &lsquo;<samp>ssashl<var>m</var>3</samp>&rsquo;, &lsquo;<samp>usashl<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Arithmetic-shift operand 1 left by a number of bits specified by operand
2, and store the result in operand 0.  Here <var>m</var> is the mode of
operand 0 and operand 1; operand 2&rsquo;s mode is specified by the
instruction pattern, and the compiler will convert the operand to that
mode before generating the instruction.  The meaning of out-of-range shift
counts can optionally be specified by <code>TARGET_SHIFT_TRUNCATION_MASK</code>.
See <a href="#TARGET_005fSHIFT_005fTRUNCATION_005fMASK">TARGET_SHIFT_TRUNCATION_MASK</a>.  Operand 2 is always a scalar type.
</p>
<a name="index-ashrm3-instruction-pattern"></a>
<a name="index-lshrm3-instruction-pattern"></a>
<a name="index-rotlm3-instruction-pattern"></a>
<a name="index-rotrm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ashr<var>m</var>3</samp>&rsquo;, &lsquo;<samp>lshr<var>m</var>3</samp>&rsquo;, &lsquo;<samp>rotl<var>m</var>3</samp>&rsquo;, &lsquo;<samp>rotr<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Other shift and rotate instructions, analogous to the
<code>ashl<var>m</var>3</code> instructions.  Operand 2 is always a scalar type.
</p>
<a name="index-vashlm3-instruction-pattern"></a>
<a name="index-vashrm3-instruction-pattern"></a>
<a name="index-vlshrm3-instruction-pattern"></a>
<a name="index-vrotlm3-instruction-pattern"></a>
<a name="index-vrotrm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>vashl<var>m</var>3</samp>&rsquo;, &lsquo;<samp>vashr<var>m</var>3</samp>&rsquo;, &lsquo;<samp>vlshr<var>m</var>3</samp>&rsquo;, &lsquo;<samp>vrotl<var>m</var>3</samp>&rsquo;, &lsquo;<samp>vrotr<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Vector shift and rotate instructions that take vectors as operand 2
instead of a scalar type.
</p>
<a name="index-bswapm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>bswap<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Reverse the order of bytes of operand 1 and store the result in operand 0.
</p>
<a name="index-negm2-instruction-pattern"></a>
<a name="index-ssnegm2-instruction-pattern"></a>
<a name="index-usnegm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>neg<var>m</var>2</samp>&rsquo;, &lsquo;<samp>ssneg<var>m</var>2</samp>&rsquo;, &lsquo;<samp>usneg<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Negate operand 1 and store the result in operand 0.
</p>
<a name="index-absm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>abs<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the absolute value of operand 1 into operand 0.
</p>
<a name="index-sqrtm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sqrt<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the square root of operand 1 into operand 0.
</p>
<p>The <code>sqrt</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>sqrtf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-fmodm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fmod<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store the remainder of dividing operand 1 by operand 2 into
operand 0, rounded towards zero to an integer.
</p>
<p>The <code>fmod</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>fmodf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-remainderm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>remainder<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store the remainder of dividing operand 1 by operand 2 into
operand 0, rounded to the nearest integer.
</p>
<p>The <code>remainder</code> built-in function of C always uses the mode
which corresponds to the C data type <code>double</code> and the
<code>remainderf</code> built-in function uses the mode which corresponds
to the C data type <code>float</code>.
</p>
<a name="index-cosm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cos<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the cosine of operand 1 into operand 0.
</p>
<p>The <code>cos</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>cosf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-sinm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sin<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the sine of operand 1 into operand 0.
</p>
<p>The <code>sin</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>sinf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-sincosm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sincos<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store the cosine of operand 2 into operand 0 and the sine of
operand 2 into operand 1.
</p>
<p>The <code>sin</code> and <code>cos</code> built-in functions of C always use the
mode which corresponds to the C data type <code>double</code> and the
<code>sinf</code> and <code>cosf</code> built-in function use the mode which
corresponds to the C data type <code>float</code>.
Targets that can calculate the sine and cosine simultaneously can
implement this pattern as opposed to implementing individual
<code>sin<var>m</var>2</code> and <code>cos<var>m</var>2</code> patterns.  The <code>sin</code>
and <code>cos</code> built-in functions will then be expanded to the
<code>sincos<var>m</var>3</code> pattern, with one of the output values
left unused.
</p>
<a name="index-expm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>exp<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the exponential of operand 1 into operand 0.
</p>
<p>The <code>exp</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>expf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-logm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>log<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the natural logarithm of operand 1 into operand 0.
</p>
<p>The <code>log</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>logf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-powm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>pow<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store the value of operand 1 raised to the exponent operand 2
into operand 0.
</p>
<p>The <code>pow</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>powf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-atan2m3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atan2<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store the arc tangent (inverse tangent) of operand 1 divided by
operand 2 into operand 0, using the signs of both arguments to
determine the quadrant of the result.
</p>
<p>The <code>atan2</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>atan2f</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-floorm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>floor<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the largest integral value not greater than argument.
</p>
<p>The <code>floor</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>floorf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-btruncm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>btrunc<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the argument rounded to integer towards zero.
</p>
<p>The <code>trunc</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>truncf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-roundm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>round<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the argument rounded to integer away from zero.
</p>
<p>The <code>round</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>roundf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-ceilm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ceil<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the argument rounded to integer away from zero.
</p>
<p>The <code>ceil</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>ceilf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-nearbyintm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>nearbyint<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the argument rounded according to the default rounding mode
</p>
<p>The <code>nearbyint</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>nearbyintf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-rintm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>rint<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the argument rounded according to the default rounding mode and
raise the inexact exception when the result differs in value from
the argument
</p>
<p>The <code>rint</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>rintf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-lrintmn2"></a>
</dd>
<dt>&lsquo;<samp>lrint<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as a signed number according to the current
rounding mode and store in operand 0 (which has mode <var>n</var>).
</p>
<a name="index-lroundmn2"></a>
</dd>
<dt>&lsquo;<samp>lround<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as a signed number rounding to nearest and away
from zero and store in operand 0 (which has mode <var>n</var>).
</p>
<a name="index-lfloormn2"></a>
</dd>
<dt>&lsquo;<samp>lfloor<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as a signed number rounding down and store in
operand 0 (which has mode <var>n</var>).
</p>
<a name="index-lceilmn2"></a>
</dd>
<dt>&lsquo;<samp>lceil<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as a signed number rounding up and store in
operand 0 (which has mode <var>n</var>).
</p>
<a name="index-copysignm3-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>copysign<var>m</var>3</samp>&rsquo;</dt>
<dd><p>Store a value with the magnitude of operand 1 and the sign of operand
2 into operand 0.
</p>
<p>The <code>copysign</code> built-in function of C always uses the mode which
corresponds to the C data type <code>double</code> and the <code>copysignf</code>
built-in function uses the mode which corresponds to the C data
type <code>float</code>.
</p>
<a name="index-ffsm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ffs<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store into operand 0 one plus the index of the least significant 1-bit
of operand 1.  If operand 1 is zero, store zero.  <var>m</var> is the mode
of operand 0; operand 1&rsquo;s mode is specified by the instruction
pattern, and the compiler will convert the operand to that mode before
generating the instruction.
</p>
<p>The <code>ffs</code> built-in function of C always uses the mode which
corresponds to the C data type <code>int</code>.
</p>
<a name="index-clzm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>clz<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store into operand 0 the number of leading 0-bits in <var>x</var>, starting
at the most significant bit position.  If <var>x</var> is 0, the
<code>CLZ_DEFINED_VALUE_AT_ZERO</code> (see <a href="#Misc">Misc</a>) macro defines if
the result is undefined or has a useful value.
<var>m</var> is the mode of operand 0; operand 1&rsquo;s mode is
specified by the instruction pattern, and the compiler will convert the
operand to that mode before generating the instruction.
</p>
<a name="index-ctzm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ctz<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store into operand 0 the number of trailing 0-bits in <var>x</var>, starting
at the least significant bit position.  If <var>x</var> is 0, the
<code>CTZ_DEFINED_VALUE_AT_ZERO</code> (see <a href="#Misc">Misc</a>) macro defines if
the result is undefined or has a useful value.
<var>m</var> is the mode of operand 0; operand 1&rsquo;s mode is
specified by the instruction pattern, and the compiler will convert the
operand to that mode before generating the instruction.
</p>
<a name="index-popcountm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>popcount<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store into operand 0 the number of 1-bits in <var>x</var>.  <var>m</var> is the
mode of operand 0; operand 1&rsquo;s mode is specified by the instruction
pattern, and the compiler will convert the operand to that mode before
generating the instruction.
</p>
<a name="index-paritym2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>parity<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store into operand 0 the parity of <var>x</var>, i.e. the number of 1-bits
in <var>x</var> modulo 2.  <var>m</var> is the mode of operand 0; operand 1&rsquo;s mode
is specified by the instruction pattern, and the compiler will convert
the operand to that mode before generating the instruction.
</p>
<a name="index-one_005fcmplm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>one_cmpl<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Store the bitwise-complement of operand 1 into operand 0.
</p>
<a name="index-movmemm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>movmem<var>m</var></samp>&rsquo;</dt>
<dd><p>Block move instruction.  The destination and source blocks of memory
are the first two operands, and both are <code>mem:BLK</code>s with an
address in mode <code>Pmode</code>.
</p>
<p>The number of bytes to move is the third operand, in mode <var>m</var>.
Usually, you specify <code>word_mode</code> for <var>m</var>.  However, if you can
generate better code knowing the range of valid lengths is smaller than
those representable in a full word, you should provide a pattern with a
mode corresponding to the range of values you can handle efficiently
(e.g., <code>QImode</code> for values in the range 0&ndash;127; note we avoid numbers
that appear negative) and also a pattern with <code>word_mode</code>.
</p>
<p>The fourth operand is the known shared alignment of the source and
destination, in the form of a <code>const_int</code> rtx.  Thus, if the
compiler knows that both source and destination are word-aligned,
it may provide the value 4 for this operand.
</p>
<p>Optional operands 5 and 6 specify expected alignment and size of block
respectively.  The expected alignment differs from alignment in operand 4
in a way that the blocks are not required to be aligned according to it in
all cases. This expected alignment is also in bytes, just like operand 4.
Expected size, when unknown, is set to <code>(const_int -1)</code>.
</p>
<p>Descriptions of multiple <code>movmem<var>m</var></code> patterns can only be
beneficial if the patterns for smaller modes have fewer restrictions
on their first, second and fourth operands.  Note that the mode <var>m</var>
in <code>movmem<var>m</var></code> does not impose any restriction on the mode of
individually moved data units in the block.
</p>
<p>These patterns need not give special consideration to the possibility
that the source and destination strings might overlap.
</p>
<a name="index-movstr-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>movstr</samp>&rsquo;</dt>
<dd><p>String copy instruction, with <code>stpcpy</code> semantics.  Operand 0 is
an output operand in mode <code>Pmode</code>.  The addresses of the
destination and source strings are operands 1 and 2, and both are
<code>mem:BLK</code>s with addresses in mode <code>Pmode</code>.  The execution of
the expansion of this pattern should store in operand 0 the address in
which the <code>NUL</code> terminator was stored in the destination string.
</p>
<a name="index-setmemm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>setmem<var>m</var></samp>&rsquo;</dt>
<dd><p>Block set instruction.  The destination string is the first operand,
given as a <code>mem:BLK</code> whose address is in mode <code>Pmode</code>.  The
number of bytes to set is the second operand, in mode <var>m</var>.  The value to
initialize the memory with is the third operand. Targets that only support the
clearing of memory should reject any value that is not the constant 0.  See
&lsquo;<samp>movmem<var>m</var></samp>&rsquo; for a discussion of the choice of mode.
</p>
<p>The fourth operand is the known alignment of the destination, in the form
of a <code>const_int</code> rtx.  Thus, if the compiler knows that the
destination is word-aligned, it may provide the value 4 for this
operand.
</p>
<p>Optional operands 5 and 6 specify expected alignment and size of block
respectively.  The expected alignment differs from alignment in operand 4
in a way that the blocks are not required to be aligned according to it in
all cases. This expected alignment is also in bytes, just like operand 4.
Expected size, when unknown, is set to <code>(const_int -1)</code>.
</p>
<p>The use for multiple <code>setmem<var>m</var></code> is as for <code>movmem<var>m</var></code>.
</p>
<a name="index-cmpstrnm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cmpstrn<var>m</var></samp>&rsquo;</dt>
<dd><p>String compare instruction, with five operands.  Operand 0 is the output;
it has mode <var>m</var>.  The remaining four operands are like the operands
of &lsquo;<samp>movmem<var>m</var></samp>&rsquo;.  The two memory blocks specified are compared
byte by byte in lexicographic order starting at the beginning of each
string.  The instruction is not allowed to prefetch more than one byte
at a time since either string may end in the first byte and reading past
that may access an invalid page or segment and cause a fault.  The
comparison terminates early if the fetched bytes are different or if
they are equal to zero.  The effect of the instruction is to store a
value in operand 0 whose sign indicates the result of the comparison.
</p>
<a name="index-cmpstrm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cmpstr<var>m</var></samp>&rsquo;</dt>
<dd><p>String compare instruction, without known maximum length.  Operand 0 is the
output; it has mode <var>m</var>.  The second and third operand are the blocks of
memory to be compared; both are <code>mem:BLK</code> with an address in mode
<code>Pmode</code>.
</p>
<p>The fourth operand is the known shared alignment of the source and
destination, in the form of a <code>const_int</code> rtx.  Thus, if the
compiler knows that both source and destination are word-aligned,
it may provide the value 4 for this operand.
</p>
<p>The two memory blocks specified are compared byte by byte in lexicographic
order starting at the beginning of each string.  The instruction is not allowed
to prefetch more than one byte at a time since either string may end in the
first byte and reading past that may access an invalid page or segment and
cause a fault.  The comparison will terminate when the fetched bytes
are different or if they are equal to zero.  The effect of the
instruction is to store a value in operand 0 whose sign indicates the
result of the comparison.
</p>
<a name="index-cmpmemm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cmpmem<var>m</var></samp>&rsquo;</dt>
<dd><p>Block compare instruction, with five operands like the operands
of &lsquo;<samp>cmpstr<var>m</var></samp>&rsquo;.  The two memory blocks specified are compared
byte by byte in lexicographic order starting at the beginning of each
block.  Unlike &lsquo;<samp>cmpstr<var>m</var></samp>&rsquo; the instruction can prefetch
any bytes in the two memory blocks.  Also unlike &lsquo;<samp>cmpstr<var>m</var></samp>&rsquo;
the comparison will not stop if both bytes are zero.  The effect of
the instruction is to store a value in operand 0 whose sign indicates
the result of the comparison.
</p>
<a name="index-strlenm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>strlen<var>m</var></samp>&rsquo;</dt>
<dd><p>Compute the length of a string, with three operands.
Operand 0 is the result (of mode <var>m</var>), operand 1 is
a <code>mem</code> referring to the first character of the string,
operand 2 is the character to search for (normally zero),
and operand 3 is a constant describing the known alignment
of the beginning of the string.
</p>
<a name="index-floatmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>float<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert signed integer operand 1 (valid for fixed point mode <var>m</var>) to
floating point mode <var>n</var> and store in operand 0 (which has mode
<var>n</var>).
</p>
<a name="index-floatunsmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>floatuns<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert unsigned integer operand 1 (valid for fixed point mode <var>m</var>)
to floating point mode <var>n</var> and store in operand 0 (which has mode
<var>n</var>).
</p>
<a name="index-fixmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fix<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as a signed number and store in operand 0 (which
has mode <var>n</var>).  This instruction&rsquo;s result is defined only when
the value of operand 1 is an integer.
</p>
<p>If the machine description defines this pattern, it also needs to
define the <code>ftrunc</code> pattern.
</p>
<a name="index-fixunsmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fixuns<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to fixed
point mode <var>n</var> as an unsigned number and store in operand 0 (which
has mode <var>n</var>).  This instruction&rsquo;s result is defined only when the
value of operand 1 is an integer.
</p>
<a name="index-ftruncm2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ftrunc<var>m</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 (valid for floating point mode <var>m</var>) to an
integer value, still represented in floating point mode <var>m</var>, and
store it in operand 0 (valid for floating point mode <var>m</var>).
</p>
<a name="index-fix_005ftruncmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fix_trunc<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>fix<var>m</var><var>n</var>2</samp>&rsquo; but works for any floating point value
of mode <var>m</var> by converting the value to an integer.
</p>
<a name="index-fixuns_005ftruncmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fixuns_trunc<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>fixuns<var>m</var><var>n</var>2</samp>&rsquo; but works for any floating point
value of mode <var>m</var> by converting the value to an integer.
</p>
<a name="index-truncmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>trunc<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Truncate operand 1 (valid for mode <var>m</var>) to mode <var>n</var> and
store in operand 0 (which has mode <var>n</var>).  Both modes must be fixed
point or both floating point.
</p>
<a name="index-extendmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extend<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Sign-extend operand 1 (valid for mode <var>m</var>) to mode <var>n</var> and
store in operand 0 (which has mode <var>n</var>).  Both modes must be fixed
point or both floating point.
</p>
<a name="index-zero_005fextendmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>zero_extend<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Zero-extend operand 1 (valid for mode <var>m</var>) to mode <var>n</var> and
store in operand 0 (which has mode <var>n</var>).  Both modes must be fixed
point.
</p>
<a name="index-fractmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fract<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 of mode <var>m</var> to mode <var>n</var> and store in
operand 0 (which has mode <var>n</var>).  Mode <var>m</var> and mode <var>n</var>
could be fixed-point to fixed-point, signed integer to fixed-point,
fixed-point to signed integer, floating-point to fixed-point,
or fixed-point to floating-point.
When overflows or underflows happen, the results are undefined.
</p>
<a name="index-satfractmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>satfract<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 of mode <var>m</var> to mode <var>n</var> and store in
operand 0 (which has mode <var>n</var>).  Mode <var>m</var> and mode <var>n</var>
could be fixed-point to fixed-point, signed integer to fixed-point,
or floating-point to fixed-point.
When overflows or underflows happen, the instruction saturates the
results to the maximum or the minimum.
</p>
<a name="index-fractunsmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>fractuns<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert operand 1 of mode <var>m</var> to mode <var>n</var> and store in
operand 0 (which has mode <var>n</var>).  Mode <var>m</var> and mode <var>n</var>
could be unsigned integer to fixed-point, or
fixed-point to unsigned integer.
When overflows or underflows happen, the results are undefined.
</p>
<a name="index-satfractunsmn2-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>satfractuns<var>m</var><var>n</var>2</samp>&rsquo;</dt>
<dd><p>Convert unsigned integer operand 1 of mode <var>m</var> to fixed-point mode
<var>n</var> and store in operand 0 (which has mode <var>n</var>).
When overflows or underflows happen, the instruction saturates the
results to the maximum or the minimum.
</p>
<a name="index-extvm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extv<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract a bit-field from register operand 1, sign-extend it, and store
it in operand 0.  Operand 2 specifies the width of the field in bits
and operand 3 the starting bit, which counts from the most significant
bit if &lsquo;<samp>BITS_BIG_ENDIAN</samp>&rsquo; is true and from the least significant bit
otherwise.
</p>
<p>Operands 0 and 1 both have mode <var>m</var>.  Operands 2 and 3 have a
target-specific mode.
</p>
<a name="index-extvmisalignm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extvmisalign<var>m</var></samp>&rsquo;</dt>
<dd><p>Extract a bit-field from memory operand 1, sign extend it, and store
it in operand 0.  Operand 2 specifies the width in bits and operand 3
the starting bit.  The starting bit is always somewhere in the first byte of
operand 1; it counts from the most significant bit if &lsquo;<samp>BITS_BIG_ENDIAN</samp>&rsquo;
is true and from the least significant bit otherwise.
</p>
<p>Operand 0 has mode <var>m</var> while operand 1 has <code>BLK</code> mode.
Operands 2 and 3 have a target-specific mode.
</p>
<p>The instruction must not read beyond the last byte of the bit-field.
</p>
<a name="index-extzvm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extzv<var>m</var></samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>extv<var>m</var></samp>&rsquo; except that the bit-field value is zero-extended.
</p>
<a name="index-extzvmisalignm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extzvmisalign<var>m</var></samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>extvmisalign<var>m</var></samp>&rsquo; except that the bit-field value is
zero-extended.
</p>
<a name="index-insvm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>insv<var>m</var></samp>&rsquo;</dt>
<dd><p>Insert operand 3 into a bit-field of register operand 0.  Operand 1
specifies the width of the field in bits and operand 2 the starting bit,
which counts from the most significant bit if &lsquo;<samp>BITS_BIG_ENDIAN</samp>&rsquo;
is true and from the least significant bit otherwise.
</p>
<p>Operands 0 and 3 both have mode <var>m</var>.  Operands 1 and 2 have a
target-specific mode.
</p>
<a name="index-insvmisalignm-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>insvmisalign<var>m</var></samp>&rsquo;</dt>
<dd><p>Insert operand 3 into a bit-field of memory operand 0.  Operand 1
specifies the width of the field in bits and operand 2 the starting bit.
The starting bit is always somewhere in the first byte of operand 0;
it counts from the most significant bit if &lsquo;<samp>BITS_BIG_ENDIAN</samp>&rsquo;
is true and from the least significant bit otherwise.
</p>
<p>Operand 3 has mode <var>m</var> while operand 0 has <code>BLK</code> mode.
Operands 1 and 2 have a target-specific mode.
</p>
<p>The instruction must not read or write beyond the last byte of the bit-field.
</p>
<a name="index-extv-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extv</samp>&rsquo;</dt>
<dd><p>Extract a bit-field from operand 1 (a register or memory operand), where
operand 2 specifies the width in bits and operand 3 the starting bit,
and store it in operand 0.  Operand 0 must have mode <code>word_mode</code>.
Operand 1 may have mode <code>byte_mode</code> or <code>word_mode</code>; often
<code>word_mode</code> is allowed only for registers.  Operands 2 and 3 must
be valid for <code>word_mode</code>.
</p>
<p>The RTL generation pass generates this instruction only with constants
for operands 2 and 3 and the constant is never zero for operand 2.
</p>
<p>The bit-field value is sign-extended to a full word integer
before it is stored in operand 0.
</p>
<p>This pattern is deprecated; please use &lsquo;<samp>extv<var>m</var></samp>&rsquo; and
<code>extvmisalign<var>m</var></code> instead.
</p>
<a name="index-extzv-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>extzv</samp>&rsquo;</dt>
<dd><p>Like &lsquo;<samp>extv</samp>&rsquo; except that the bit-field value is zero-extended.
</p>
<p>This pattern is deprecated; please use &lsquo;<samp>extzv<var>m</var></samp>&rsquo; and
<code>extzvmisalign<var>m</var></code> instead.
</p>
<a name="index-insv-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>insv</samp>&rsquo;</dt>
<dd><p>Store operand 3 (which must be valid for <code>word_mode</code>) into a
bit-field in operand 0, where operand 1 specifies the width in bits and
operand 2 the starting bit.  Operand 0 may have mode <code>byte_mode</code> or
<code>word_mode</code>; often <code>word_mode</code> is allowed only for registers.
Operands 1 and 2 must be valid for <code>word_mode</code>.
</p>
<p>The RTL generation pass generates this instruction only with constants
for operands 1 and 2 and the constant is never zero for operand 1.
</p>
<p>This pattern is deprecated; please use &lsquo;<samp>insv<var>m</var></samp>&rsquo; and
<code>insvmisalign<var>m</var></code> instead.
</p>
<a name="index-movmodecc-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mov<var>mode</var>cc</samp>&rsquo;</dt>
<dd><p>Conditionally move operand 2 or operand 3 into operand 0 according to the
comparison in operand 1.  If the comparison is true, operand 2 is moved
into operand 0, otherwise operand 3 is moved.
</p>
<p>The mode of the operands being compared need not be the same as the operands
being moved.  Some machines, sparc64 for example, have instructions that
conditionally move an integer value based on the floating point condition
codes and vice versa.
</p>
<p>If the machine does not have conditional move instructions, do not
define these patterns.
</p>
<a name="index-addmodecc-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>add<var>mode</var>cc</samp>&rsquo;</dt>
<dd><p>Similar to &lsquo;<samp>mov<var>mode</var>cc</samp>&rsquo; but for conditional addition.  Conditionally
move operand 2 or (operands 2 + operand 3) into operand 0 according to the
comparison in operand 1.  If the comparison is false, operand 2 is moved into
operand 0, otherwise (operand 2 + operand 3) is moved.
</p>
<a name="index-cstoremode4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo;</dt>
<dd><p>Store zero or nonzero in operand 0 according to whether a comparison
is true.  Operand 1 is a comparison operator.  Operand 2 and operand 3
are the first and second operand of the comparison, respectively.
You specify the mode that operand 0 must have when you write the
<code>match_operand</code> expression.  The compiler automatically sees which
mode you have used and supplies an operand of that mode.
</p>
<p>The value stored for a true condition must have 1 as its low bit, or
else must be negative.  Otherwise the instruction is not suitable and
you should omit it from the machine description.  You describe to the
compiler exactly which value is stored by defining the macro
<code>STORE_FLAG_VALUE</code> (see <a href="#Misc">Misc</a>).  If a description cannot be
found that can be used for all the possible comparison operators, you
should pick one and use a <code>define_expand</code> to map all results
onto the one you chose.
</p>
<p>These operations may <code>FAIL</code>, but should do so only in relatively
uncommon cases; if they would <code>FAIL</code> for common cases involving
integer comparisons, it is best to restrict the predicates to not
allow these operands.  Likewise if a given comparison operator will
always fail, independent of the operands (for floating-point modes, the
<code>ordered_comparison_operator</code> predicate is often useful in this case).
</p>
<p>If this pattern is omitted, the compiler will generate a conditional
branch&mdash;for example, it may copy a constant one to the target and branching
around an assignment of zero to the target&mdash;or a libcall.  If the predicate
for operand 1 only rejects some operators, it will also try reordering the
operands and/or inverting the result value (e.g. by an exclusive OR).
These possibilities could be cheaper or equivalent to the instructions
used for the &lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo; pattern followed by those required
to convert a positive result from <code>STORE_FLAG_VALUE</code> to 1; in this
case, you can and should make operand 1&rsquo;s predicate reject some operators
in the &lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo; pattern, or remove the pattern altogether
from the machine description.
</p>
<a name="index-cbranchmode4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>cbranch<var>mode</var>4</samp>&rsquo;</dt>
<dd><p>Conditional branch instruction combined with a compare instruction.
Operand 0 is a comparison operator.  Operand 1 and operand 2 are the
first and second operands of the comparison, respectively.  Operand 3
is a <code>label_ref</code> that refers to the label to jump to.
</p>
<a name="index-jump-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>jump</samp>&rsquo;</dt>
<dd><p>A jump inside a function; an unconditional branch.  Operand 0 is the
<code>label_ref</code> of the label to jump to.  This pattern name is mandatory
on all machines.
</p>
<a name="index-call-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>call</samp>&rsquo;</dt>
<dd><p>Subroutine call instruction returning no value.  Operand 0 is the
function to call; operand 1 is the number of bytes of arguments pushed
as a <code>const_int</code>; operand 2 is the number of registers used as
operands.
</p>
<p>On most machines, operand 2 is not actually stored into the RTL
pattern.  It is supplied for the sake of some RISC machines which need
to put this information into the assembler code; they can put it in
the RTL instead of operand 1.
</p>
<p>Operand 0 should be a <code>mem</code> RTX whose address is the address of the
function.  Note, however, that this address can be a <code>symbol_ref</code>
expression even if it would not be a legitimate memory address on the
target machine.  If it is also not a valid argument for a call
instruction, the pattern for this operation should be a
<code>define_expand</code> (see <a href="#Expander-Definitions">Expander Definitions</a>) that places the
address into a register and uses that register in the call instruction.
</p>
<a name="index-call_005fvalue-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>call_value</samp>&rsquo;</dt>
<dd><p>Subroutine call instruction returning a value.  Operand 0 is the hard
register in which the value is returned.  There are three more
operands, the same as the three operands of the &lsquo;<samp>call</samp>&rsquo;
instruction (but with numbers increased by one).
</p>
<p>Subroutines that return <code>BLKmode</code> objects use the &lsquo;<samp>call</samp>&rsquo;
insn.
</p>
<a name="index-call_005fpop-instruction-pattern"></a>
<a name="index-call_005fvalue_005fpop-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>call_pop</samp>&rsquo;, &lsquo;<samp>call_value_pop</samp>&rsquo;</dt>
<dd><p>Similar to &lsquo;<samp>call</samp>&rsquo; and &lsquo;<samp>call_value</samp>&rsquo;, except used if defined and
if <code>RETURN_POPS_ARGS</code> is nonzero.  They should emit a <code>parallel</code>
that contains both the function call and a <code>set</code> to indicate the
adjustment made to the frame pointer.
</p>
<p>For machines where <code>RETURN_POPS_ARGS</code> can be nonzero, the use of these
patterns increases the number of functions for which the frame pointer
can be eliminated, if desired.
</p>
<a name="index-untyped_005fcall-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>untyped_call</samp>&rsquo;</dt>
<dd><p>Subroutine call instruction returning a value of any type.  Operand 0 is
the function to call; operand 1 is a memory location where the result of
calling the function is to be stored; operand 2 is a <code>parallel</code>
expression where each element is a <code>set</code> expression that indicates
the saving of a function return value into the result block.
</p>
<p>This instruction pattern should be defined to support
<code>__builtin_apply</code> on machines where special instructions are needed
to call a subroutine with arbitrary arguments or to save the value
returned.  This instruction pattern is required on machines that have
multiple registers that can hold a return value
(i.e. <code>FUNCTION_VALUE_REGNO_P</code> is true for more than one register).
</p>
<a name="index-return-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>return</samp>&rsquo;</dt>
<dd><p>Subroutine return instruction.  This instruction pattern name should be
defined only if a single instruction can do all the work of returning
from a function.
</p>
<p>Like the &lsquo;<samp>mov<var>m</var></samp>&rsquo; patterns, this pattern is also used after the
RTL generation phase.  In this case it is to support machines where
multiple instructions are usually needed to return from a function, but
some class of functions only requires one instruction to implement a
return.  Normally, the applicable functions are those which do not need
to save any registers or allocate stack space.
</p>
<p>It is valid for this pattern to expand to an instruction using
<code>simple_return</code> if no epilogue is required.
</p>
<a name="index-simple_005freturn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>simple_return</samp>&rsquo;</dt>
<dd><p>Subroutine return instruction.  This instruction pattern name should be
defined only if a single instruction can do all the work of returning
from a function on a path where no epilogue is required.  This pattern
is very similar to the <code>return</code> instruction pattern, but it is emitted
only by the shrink-wrapping optimization on paths where the function
prologue has not been executed, and a function return should occur without
any of the effects of the epilogue.  Additional uses may be introduced on
paths where both the prologue and the epilogue have executed.
</p>
<a name="index-reload_005fcompleted"></a>
<a name="index-leaf_005ffunction_005fp"></a>
<p>For such machines, the condition specified in this pattern should only
be true when <code>reload_completed</code> is nonzero and the function&rsquo;s
epilogue would only be a single instruction.  For machines with register
windows, the routine <code>leaf_function_p</code> may be used to determine if
a register window push is required.
</p>
<p>Machines that have conditional return instructions should define patterns
such as
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (pc)
        (if_then_else (match_operator
                         0 &quot;comparison_operator&quot;
                         [(cc0) (const_int 0)])
                      (return)
                      (pc)))]
  &quot;<var>condition</var>&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>where <var>condition</var> would normally be the same condition specified on the
named &lsquo;<samp>return</samp>&rsquo; pattern.
</p>
<a name="index-untyped_005freturn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>untyped_return</samp>&rsquo;</dt>
<dd><p>Untyped subroutine return instruction.  This instruction pattern should
be defined to support <code>__builtin_return</code> on machines where special
instructions are needed to return a value of any type.
</p>
<p>Operand 0 is a memory location where the result of calling a function
with <code>__builtin_apply</code> is stored; operand 1 is a <code>parallel</code>
expression where each element is a <code>set</code> expression that indicates
the restoring of a function return value from the result block.
</p>
<a name="index-nop-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>nop</samp>&rsquo;</dt>
<dd><p>No-op instruction.  This instruction pattern name should always be defined
to output a no-op in assembler code.  <code>(const_int 0)</code> will do as an
RTL pattern.
</p>
<a name="index-indirect_005fjump-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>indirect_jump</samp>&rsquo;</dt>
<dd><p>An instruction to jump to an address which is operand zero.
This pattern name is mandatory on all machines.
</p>
<a name="index-casesi-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>casesi</samp>&rsquo;</dt>
<dd><p>Instruction to jump through a dispatch table, including bounds checking.
This instruction takes five operands:
</p>
<ol>
<li> The index to dispatch on, which has mode <code>SImode</code>.

</li><li> The lower bound for indices in the table, an integer constant.

</li><li> The total range of indices in the table&mdash;the largest index
minus the smallest one (both inclusive).

</li><li> A label that precedes the table itself.

</li><li> A label to jump to if the index has a value outside the bounds.
</li></ol>

<p>The table is an <code>addr_vec</code> or <code>addr_diff_vec</code> inside of a
<code>jump_insn</code>.  The number of elements in the table is one plus the
difference between the upper bound and the lower bound.
</p>
<a name="index-tablejump-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>tablejump</samp>&rsquo;</dt>
<dd><p>Instruction to jump to a variable address.  This is a low-level
capability which can be used to implement a dispatch table when there
is no &lsquo;<samp>casesi</samp>&rsquo; pattern.
</p>
<p>This pattern requires two operands: the address or offset, and a label
which should immediately precede the jump table.  If the macro
<code>CASE_VECTOR_PC_RELATIVE</code> evaluates to a nonzero value then the first
operand is an offset which counts from the address of the table; otherwise,
it is an absolute address to jump to.  In either case, the first operand has
mode <code>Pmode</code>.
</p>
<p>The &lsquo;<samp>tablejump</samp>&rsquo; insn is always the last insn before the jump
table it uses.  Its assembler code normally has no need to use the
second operand, but you should incorporate it in the RTL pattern so
that the jump optimizer will not delete the table as unreachable code.
</p>

<a name="index-decrement_005fand_005fbranch_005funtil_005fzero-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>decrement_and_branch_until_zero</samp>&rsquo;</dt>
<dd><p>Conditional branch instruction that decrements a register and
jumps if the register is nonzero.  Operand 0 is the register to
decrement and test; operand 1 is the label to jump to if the
register is nonzero.  See <a href="#Looping-Patterns">Looping Patterns</a>.
</p>
<p>This optional instruction pattern is only used by the combiner,
typically for loops reversed by the loop optimizer when strength
reduction is enabled.
</p>
<a name="index-doloop_005fend-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>doloop_end</samp>&rsquo;</dt>
<dd><p>Conditional branch instruction that decrements a register and jumps if
the register is nonzero.  This instruction takes five operands: Operand
0 is the register to decrement and test; operand 1 is the number of loop
iterations as a <code>const_int</code> or <code>const0_rtx</code> if this cannot be
determined until run-time; operand 2 is the actual or estimated maximum
number of iterations as a <code>const_int</code>; operand 3 is the number of
enclosed loops as a <code>const_int</code> (an innermost loop has a value of
1); operand 4 is the label to jump to if the register is nonzero;
operand 5 is const1_rtx if the loop in entered at its top, const0_rtx
otherwise.
See <a href="#Looping-Patterns">Looping Patterns</a>.
</p>
<p>This optional instruction pattern should be defined for machines with
low-overhead looping instructions as the loop optimizer will try to
modify suitable loops to utilize it.  If nested low-overhead looping is
not supported, use a <code>define_expand</code> (see <a href="#Expander-Definitions">Expander Definitions</a>)
and make the pattern fail if operand 3 is not <code>const1_rtx</code>.
Similarly, if the actual or estimated maximum number of iterations is
too large for this instruction, make it fail.
</p>
<a name="index-doloop_005fbegin-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>doloop_begin</samp>&rsquo;</dt>
<dd><p>Companion instruction to <code>doloop_end</code> required for machines that
need to perform some initialization, such as loading special registers
used by a low-overhead looping instruction.  If initialization insns do
not always need to be emitted, use a <code>define_expand</code>
(see <a href="#Expander-Definitions">Expander Definitions</a>) and make it fail.
</p>

<a name="index-canonicalize_005ffuncptr_005ffor_005fcompare-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>canonicalize_funcptr_for_compare</samp>&rsquo;</dt>
<dd><p>Canonicalize the function pointer in operand 1 and store the result
into operand 0.
</p>
<p>Operand 0 is always a <code>reg</code> and has mode <code>Pmode</code>; operand 1
may be a <code>reg</code>, <code>mem</code>, <code>symbol_ref</code>, <code>const_int</code>, etc
and also has mode <code>Pmode</code>.
</p>
<p>Canonicalization of a function pointer usually involves computing
the address of the function which would be called if the function
pointer were used in an indirect call.
</p>
<p>Only define this pattern if function pointers on the target machine
can have different values but still call the same function when
used in an indirect call.
</p>
<a name="index-save_005fstack_005fblock-instruction-pattern"></a>
<a name="index-save_005fstack_005ffunction-instruction-pattern"></a>
<a name="index-save_005fstack_005fnonlocal-instruction-pattern"></a>
<a name="index-restore_005fstack_005fblock-instruction-pattern"></a>
<a name="index-restore_005fstack_005ffunction-instruction-pattern"></a>
<a name="index-restore_005fstack_005fnonlocal-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>save_stack_block</samp>&rsquo;</dt>
<dt>&lsquo;<samp>save_stack_function</samp>&rsquo;</dt>
<dt>&lsquo;<samp>save_stack_nonlocal</samp>&rsquo;</dt>
<dt>&lsquo;<samp>restore_stack_block</samp>&rsquo;</dt>
<dt>&lsquo;<samp>restore_stack_function</samp>&rsquo;</dt>
<dt>&lsquo;<samp>restore_stack_nonlocal</samp>&rsquo;</dt>
<dd><p>Most machines save and restore the stack pointer by copying it to or
from an object of mode <code>Pmode</code>.  Do not define these patterns on
such machines.
</p>
<p>Some machines require special handling for stack pointer saves and
restores.  On those machines, define the patterns corresponding to the
non-standard cases by using a <code>define_expand</code> (see <a href="#Expander-Definitions">Expander Definitions</a>) that produces the required insns.  The three types of
saves and restores are:
</p>
<ol>
<li> &lsquo;<samp>save_stack_block</samp>&rsquo; saves the stack pointer at the start of a block
that allocates a variable-sized object, and &lsquo;<samp>restore_stack_block</samp>&rsquo;
restores the stack pointer when the block is exited.

</li><li> &lsquo;<samp>save_stack_function</samp>&rsquo; and &lsquo;<samp>restore_stack_function</samp>&rsquo; do a
similar job for the outermost block of a function and are used when the
function allocates variable-sized objects or calls <code>alloca</code>.  Only
the epilogue uses the restored stack pointer, allowing a simpler save or
restore sequence on some machines.

</li><li> &lsquo;<samp>save_stack_nonlocal</samp>&rsquo; is used in functions that contain labels
branched to by nested functions.  It saves the stack pointer in such a
way that the inner function can use &lsquo;<samp>restore_stack_nonlocal</samp>&rsquo; to
restore the stack pointer.  The compiler generates code to restore the
frame and argument pointer registers, but some machines require saving
and restoring additional data such as register window information or
stack backchains.  Place insns in these patterns to save and restore any
such required data.
</li></ol>

<p>When saving the stack pointer, operand 0 is the save area and operand 1
is the stack pointer.  The mode used to allocate the save area defaults
to <code>Pmode</code> but you can override that choice by defining the
<code>STACK_SAVEAREA_MODE</code> macro (see <a href="#Storage-Layout">Storage Layout</a>).  You must
specify an integral mode, or <code>VOIDmode</code> if no save area is needed
for a particular type of save (either because no save is needed or
because a machine-specific save area can be used).  Operand 0 is the
stack pointer and operand 1 is the save area for restore operations.  If
&lsquo;<samp>save_stack_block</samp>&rsquo; is defined, operand 0 must not be
<code>VOIDmode</code> since these saves can be arbitrarily nested.
</p>
<p>A save area is a <code>mem</code> that is at a constant offset from
<code>virtual_stack_vars_rtx</code> when the stack pointer is saved for use by
nonlocal gotos and a <code>reg</code> in the other two cases.
</p>
<a name="index-allocate_005fstack-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>allocate_stack</samp>&rsquo;</dt>
<dd><p>Subtract (or add if <code>STACK_GROWS_DOWNWARD</code> is undefined) operand 1 from
the stack pointer to create space for dynamically allocated data.
</p>
<p>Store the resultant pointer to this space into operand 0.  If you
are allocating space from the main stack, do this by emitting a
move insn to copy <code>virtual_stack_dynamic_rtx</code> to operand 0.
If you are allocating the space elsewhere, generate code to copy the
location of the space to operand 0.  In the latter case, you must
ensure this space gets freed when the corresponding space on the main
stack is free.
</p>
<p>Do not define this pattern if all that must be done is the subtraction.
Some machines require other operations such as stack probes or
maintaining the back chain.  Define this pattern to emit those
operations in addition to updating the stack pointer.
</p>
<a name="index-check_005fstack-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>check_stack</samp>&rsquo;</dt>
<dd><p>If stack checking (see <a href="#Stack-Checking">Stack Checking</a>) cannot be done on your system by
probing the stack, define this pattern to perform the needed check and signal
an error if the stack has overflowed.  The single operand is the address in
the stack farthest from the current stack pointer that you need to validate.
Normally, on platforms where this pattern is needed, you would obtain the
stack limit from a global or thread-specific variable or register.
</p>
<a name="index-probe_005fstack_005faddress-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>probe_stack_address</samp>&rsquo;</dt>
<dd><p>If stack checking (see <a href="#Stack-Checking">Stack Checking</a>) can be done on your system by
probing the stack but without the need to actually access it, define this
pattern and signal an error if the stack has overflowed.  The single operand
is the memory address in the stack that needs to be probed.
</p>
<a name="index-probe_005fstack-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>probe_stack</samp>&rsquo;</dt>
<dd><p>If stack checking (see <a href="#Stack-Checking">Stack Checking</a>) can be done on your system by
probing the stack but doing it with a &ldquo;store zero&rdquo; instruction is not valid
or optimal, define this pattern to do the probing differently and signal an
error if the stack has overflowed.  The single operand is the memory reference
in the stack that needs to be probed.
</p>
<a name="index-nonlocal_005fgoto-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>nonlocal_goto</samp>&rsquo;</dt>
<dd><p>Emit code to generate a non-local goto, e.g., a jump from one function
to a label in an outer function.  This pattern has four arguments,
each representing a value to be used in the jump.  The first
argument is to be loaded into the frame pointer, the second is
the address to branch to (code to dispatch to the actual label),
the third is the address of a location where the stack is saved,
and the last is the address of the label, to be placed in the
location for the incoming static chain.
</p>
<p>On most machines you need not define this pattern, since GCC will
already generate the correct code, which is to load the frame pointer
and static chain, restore the stack (using the
&lsquo;<samp>restore_stack_nonlocal</samp>&rsquo; pattern, if defined), and jump indirectly
to the dispatcher.  You need only define this pattern if this code will
not work on your machine.
</p>
<a name="index-nonlocal_005fgoto_005freceiver-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>nonlocal_goto_receiver</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, contains code needed at the target of a
nonlocal goto after the code already generated by GCC.  You will not
normally need to define this pattern.  A typical reason why you might
need this pattern is if some value, such as a pointer to a global table,
must be restored when the frame pointer is restored.  Note that a nonlocal
goto only occurs within a unit-of-translation, so a global table pointer
that is shared by all functions of a given module need not be restored.
There are no arguments.
</p>
<a name="index-exception_005freceiver-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>exception_receiver</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, contains code needed at the site of an
exception handler that isn&rsquo;t needed at the site of a nonlocal goto.  You
will not normally need to define this pattern.  A typical reason why you
might need this pattern is if some value, such as a pointer to a global
table, must be restored after control flow is branched to the handler of
an exception.  There are no arguments.
</p>
<a name="index-builtin_005fsetjmp_005fsetup-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>builtin_setjmp_setup</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, contains additional code needed to initialize
the <code>jmp_buf</code>.  You will not normally need to define this pattern.
A typical reason why you might need this pattern is if some value, such
as a pointer to a global table, must be restored.  Though it is
preferred that the pointer value be recalculated if possible (given the
address of a label for instance).  The single argument is a pointer to
the <code>jmp_buf</code>.  Note that the buffer is five words long and that
the first three are normally used by the generic mechanism.
</p>
<a name="index-builtin_005fsetjmp_005freceiver-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>builtin_setjmp_receiver</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, contains code needed at the site of a
built-in setjmp that isn&rsquo;t needed at the site of a nonlocal goto.  You
will not normally need to define this pattern.  A typical reason why you
might need this pattern is if some value, such as a pointer to a global
table, must be restored.  It takes one argument, which is the label
to which builtin_longjmp transferred control; this pattern may be emitted
at a small offset from that label.
</p>
<a name="index-builtin_005flongjmp-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>builtin_longjmp</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, performs the entire action of the longjmp.
You will not normally need to define this pattern unless you also define
<code>builtin_setjmp_setup</code>.  The single argument is a pointer to the
<code>jmp_buf</code>.
</p>
<a name="index-eh_005freturn-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>eh_return</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, affects the way <code>__builtin_eh_return</code>,
and thence the call frame exception handling library routines, are
built.  It is intended to handle non-trivial actions needed along
the abnormal return path.
</p>
<p>The address of the exception handler to which the function should return
is passed as operand to this pattern.  It will normally need to copied by
the pattern to some special register or memory location.
If the pattern needs to determine the location of the target call
frame in order to do so, it may use <code>EH_RETURN_STACKADJ_RTX</code>,
if defined; it will have already been assigned.
</p>
<p>If this pattern is not defined, the default action will be to simply
copy the return address to <code>EH_RETURN_HANDLER_RTX</code>.  Either
that macro or this pattern needs to be defined if call frame exception
handling is to be used.
</p>
<a name="index-prologue-instruction-pattern"></a>
<a name="prologue-instruction-pattern"></a></dd>
<dt>&lsquo;<samp>prologue</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, emits RTL for entry to a function.  The function
entry is responsible for setting up the stack frame, initializing the frame
pointer register, saving callee saved registers, etc.
</p>
<p>Using a prologue pattern is generally preferred over defining
<code>TARGET_ASM_FUNCTION_PROLOGUE</code> to emit assembly code for the prologue.
</p>
<p>The <code>prologue</code> pattern is particularly useful for targets which perform
instruction scheduling.
</p>
<a name="index-window_005fsave-instruction-pattern"></a>
<a name="window_005fsave-instruction-pattern"></a></dd>
<dt>&lsquo;<samp>window_save</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, emits RTL for a register window save.  It should
be defined if the target machine has register windows but the window events
are decoupled from calls to subroutines.  The canonical example is the SPARC
architecture.
</p>
<a name="index-epilogue-instruction-pattern"></a>
<a name="epilogue-instruction-pattern"></a></dd>
<dt>&lsquo;<samp>epilogue</samp>&rsquo;</dt>
<dd><p>This pattern emits RTL for exit from a function.  The function
exit is responsible for deallocating the stack frame, restoring callee saved
registers and emitting the return instruction.
</p>
<p>Using an epilogue pattern is generally preferred over defining
<code>TARGET_ASM_FUNCTION_EPILOGUE</code> to emit assembly code for the epilogue.
</p>
<p>The <code>epilogue</code> pattern is particularly useful for targets which perform
instruction scheduling or which have delay slots for their return instruction.
</p>
<a name="index-sibcall_005fepilogue-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sibcall_epilogue</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, emits RTL for exit from a function without the final
branch back to the calling function.  This pattern will be emitted before any
sibling call (aka tail call) sites.
</p>
<p>The <code>sibcall_epilogue</code> pattern must not clobber any arguments used for
parameter passing or any stack slots for arguments passed to the current
function.
</p>
<a name="index-trap-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>trap</samp>&rsquo;</dt>
<dd><p>This pattern, if defined, signals an error, typically by causing some
kind of signal to be raised.  Among other places, it is used by the Java
front end to signal &lsquo;invalid array index&rsquo; exceptions.
</p>
<a name="index-ctrapMM4-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>ctrap<var>MM</var>4</samp>&rsquo;</dt>
<dd><p>Conditional trap instruction.  Operand 0 is a piece of RTL which
performs a comparison, and operands 1 and 2 are the arms of the
comparison.  Operand 3 is the trap code, an integer.
</p>
<p>A typical <code>ctrap</code> pattern looks like
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;ctrapsi4&quot;
  [(trap_if (match_operator 0 &quot;trap_operator&quot;
             [(match_operand 1 &quot;register_operand&quot;)
              (match_operand 2 &quot;immediate_operand&quot;)])
            (match_operand 3 &quot;const_int_operand&quot; &quot;i&quot;))]
  &quot;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<a name="index-prefetch-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>prefetch</samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, emits code for a non-faulting data prefetch
instruction.  Operand 0 is the address of the memory to prefetch.  Operand 1
is a constant 1 if the prefetch is preparing for a write to the memory
address, or a constant 0 otherwise.  Operand 2 is the expected degree of
temporal locality of the data and is a value between 0 and 3, inclusive; 0
means that the data has no temporal locality, so it need not be left in the
cache after the access; 3 means that the data has a high degree of temporal
locality and should be left in all levels of cache possible;  1 and 2 mean,
respectively, a low or moderate degree of temporal locality.
</p>
<p>Targets that do not support write prefetches or locality hints can ignore
the values of operands 1 and 2.
</p>
<a name="index-blockage-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>blockage</samp>&rsquo;</dt>
<dd>
<p>This pattern defines a pseudo insn that prevents the instruction
scheduler and other passes from moving instructions and using register
equivalences across the boundary defined by the blockage insn.
This needs to be an UNSPEC_VOLATILE pattern or a volatile ASM.
</p>
<a name="index-memory_005fbarrier-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>memory_barrier</samp>&rsquo;</dt>
<dd>
<p>If the target memory model is not fully synchronous, then this pattern
should be defined to an instruction that orders both loads and stores
before the instruction with respect to loads and stores after the instruction.
This pattern has no operands.
</p>
<a name="index-sync_005fcompare_005fand_005fswapmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_compare_and_swap<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, emits code for an atomic compare-and-swap
operation.  Operand 1 is the memory on which the atomic operation is
performed.  Operand 2 is the &ldquo;old&rdquo; value to be compared against the
current contents of the memory location.  Operand 3 is the &ldquo;new&rdquo; value
to store in the memory if the compare succeeds.  Operand 0 is the result
of the operation; it should contain the contents of the memory
before the operation.  If the compare succeeds, this should obviously be
a copy of operand 2.
</p>
<p>This pattern must show that both operand 0 and operand 1 are modified.
</p>
<p>This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.
</p>
<p>For targets where the success or failure of the compare-and-swap
operation is available via the status flags, it is possible to
avoid a separate compare operation and issue the subsequent
branch or store-flag operation immediately after the compare-and-swap.
To this end, GCC will look for a <code>MODE_CC</code> set in the
output of <code>sync_compare_and_swap<var>mode</var></code>; if the machine
description includes such a set, the target should also define special
<code>cbranchcc4</code> and/or <code>cstorecc4</code> instructions.  GCC will then
be able to take the destination of the <code>MODE_CC</code> set and pass it
to the <code>cbranchcc4</code> or <code>cstorecc4</code> pattern as the first
operand of the comparison (the second will be <code>(const_int 0)</code>).
</p>
<p>For targets where the operating system may provide support for this
operation via library calls, the <code>sync_compare_and_swap_optab</code>
may be initialized to a function with the same interface as the
<code>__sync_val_compare_and_swap_<var>n</var></code> built-in.  If the entire
set of <var>__sync</var> builtins are supported via library calls, the
target can initialize all of the optabs at once with
<code>init_sync_libfuncs</code>.
For the purposes of C++11 <code>std::atomic::is_lock_free</code>, it is
assumed that these library calls do <em>not</em> use any kind of
interruptable locking.
</p>
<a name="index-sync_005faddmode-instruction-pattern"></a>
<a name="index-sync_005fsubmode-instruction-pattern"></a>
<a name="index-sync_005fiormode-instruction-pattern"></a>
<a name="index-sync_005fandmode-instruction-pattern"></a>
<a name="index-sync_005fxormode-instruction-pattern"></a>
<a name="index-sync_005fnandmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_add<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_sub<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_ior<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_and<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_xor<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_nand<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns emit code for an atomic operation on memory.
Operand 0 is the memory on which the atomic operation is performed.
Operand 1 is the second operand to the binary operator.
</p>
<p>This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.
</p>
<p>If these patterns are not defined, the operation will be constructed
from a compare-and-swap operation, if defined.
</p>
<a name="index-sync_005fold_005faddmode-instruction-pattern"></a>
<a name="index-sync_005fold_005fsubmode-instruction-pattern"></a>
<a name="index-sync_005fold_005fiormode-instruction-pattern"></a>
<a name="index-sync_005fold_005fandmode-instruction-pattern"></a>
<a name="index-sync_005fold_005fxormode-instruction-pattern"></a>
<a name="index-sync_005fold_005fnandmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_old_add<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_old_sub<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_old_ior<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_old_and<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_old_xor<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_old_nand<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns emit code for an atomic operation on memory,
and return the value that the memory contained before the operation.
Operand 0 is the result value, operand 1 is the memory on which the
atomic operation is performed, and operand 2 is the second operand
to the binary operator.
</p>
<p>This pattern must issue any memory barrier instructions such that all
memory operations before the atomic operation occur before the atomic
operation and all memory operations after the atomic operation occur
after the atomic operation.
</p>
<p>If these patterns are not defined, the operation will be constructed
from a compare-and-swap operation, if defined.
</p>
<a name="index-sync_005fnew_005faddmode-instruction-pattern"></a>
<a name="index-sync_005fnew_005fsubmode-instruction-pattern"></a>
<a name="index-sync_005fnew_005fiormode-instruction-pattern"></a>
<a name="index-sync_005fnew_005fandmode-instruction-pattern"></a>
<a name="index-sync_005fnew_005fxormode-instruction-pattern"></a>
<a name="index-sync_005fnew_005fnandmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_new_add<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_new_sub<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_new_ior<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_new_and<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>sync_new_xor<var>mode</var></samp>&rsquo;, &lsquo;<samp>sync_new_nand<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns are like their <code>sync_old_<var>op</var></code> counterparts,
except that they return the value that exists in the memory location
after the operation, rather than before the operation.
</p>
<a name="index-sync_005flock_005ftest_005fand_005fsetmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_lock_test_and_set<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>This pattern takes two forms, based on the capabilities of the target.
In either case, operand 0 is the result of the operand, operand 1 is
the memory on which the atomic operation is performed, and operand 2
is the value to set in the lock.
</p>
<p>In the ideal case, this operation is an atomic exchange operation, in
which the previous value in memory operand is copied into the result
operand, and the value operand is stored in the memory operand.
</p>
<p>For less capable targets, any value operand that is not the constant 1
should be rejected with <code>FAIL</code>.  In this case the target may use
an atomic test-and-set bit operation.  The result operand should contain
1 if the bit was previously set and 0 if the bit was previously clear.
The true contents of the memory operand are implementation defined.
</p>
<p>This pattern must issue any memory barrier instructions such that the
pattern as a whole acts as an acquire barrier, that is all memory
operations after the pattern do not occur until the lock is acquired.
</p>
<p>If this pattern is not defined, the operation will be constructed from
a compare-and-swap operation, if defined.
</p>
<a name="index-sync_005flock_005freleasemode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>sync_lock_release<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, releases a lock set by
<code>sync_lock_test_and_set<var>mode</var></code>.  Operand 0 is the memory
that contains the lock; operand 1 is the value to store in the lock.
</p>
<p>If the target doesn&rsquo;t implement full semantics for
<code>sync_lock_test_and_set<var>mode</var></code>, any value operand which is not
the constant 0 should be rejected with <code>FAIL</code>, and the true contents
of the memory operand are implementation defined.
</p>
<p>This pattern must issue any memory barrier instructions such that the
pattern as a whole acts as a release barrier, that is the lock is
released only after all previous memory operations have completed.
</p>
<p>If this pattern is not defined, then a <code>memory_barrier</code> pattern
will be emitted, followed by a store of the value to the memory operand.
</p>
<a name="index-atomic_005fcompare_005fand_005fswapmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_compare_and_swap<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern, if defined, emits code for an atomic compare-and-swap
operation with memory model semantics.  Operand 2 is the memory on which
the atomic operation is performed.  Operand 0 is an output operand which
is set to true or false based on whether the operation succeeded.  Operand
1 is an output operand which is set to the contents of the memory before
the operation was attempted.  Operand 3 is the value that is expected to
be in memory.  Operand 4 is the value to put in memory if the expected
value is found there.  Operand 5 is set to 1 if this compare and swap is to
be treated as a weak operation.  Operand 6 is the memory model to be used
if the operation is a success.  Operand 7 is the memory model to be used
if the operation fails.
</p>
<p>If memory referred to in operand 2 contains the value in operand 3, then
operand 4 is stored in memory pointed to by operand 2 and fencing based on
the memory model in operand 6 is issued.  
</p>
<p>If memory referred to in operand 2 does not contain the value in operand 3,
then fencing based on the memory model in operand 7 is issued.
</p>
<p>If a target does not support weak compare-and-swap operations, or the port
elects not to implement weak operations, the argument in operand 5 can be
ignored.  Note a strong implementation must be provided.
</p>
<p>If this pattern is not provided, the <code>__atomic_compare_exchange</code>
built-in functions will utilize the legacy <code>sync_compare_and_swap</code>
pattern with an <code>__ATOMIC_SEQ_CST</code> memory model.
</p>
<a name="index-atomic_005floadmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_load<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern implements an atomic load operation with memory model
semantics.  Operand 1 is the memory address being loaded from.  Operand 0
is the result of the load.  Operand 2 is the memory model to be used for
the load operation.
</p>
<p>If not present, the <code>__atomic_load</code> built-in function will either
resort to a normal load with memory barriers, or a compare-and-swap
operation if a normal load would not be atomic.
</p>
<a name="index-atomic_005fstoremode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_store<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern implements an atomic store operation with memory model
semantics.  Operand 0 is the memory address being stored to.  Operand 1
is the value to be written.  Operand 2 is the memory model to be used for
the operation.
</p>
<p>If not present, the <code>__atomic_store</code> built-in function will attempt to
perform a normal store and surround it with any required memory fences.  If
the store would not be atomic, then an <code>__atomic_exchange</code> is
attempted with the result being ignored.
</p>
<a name="index-atomic_005fexchangemode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_exchange<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern implements an atomic exchange operation with memory model
semantics.  Operand 1 is the memory location the operation is performed on.
Operand 0 is an output operand which is set to the original value contained
in the memory pointed to by operand 1.  Operand 2 is the value to be
stored.  Operand 3 is the memory model to be used.
</p>
<p>If this pattern is not present, the built-in function
<code>__atomic_exchange</code> will attempt to preform the operation with a
compare and swap loop.
</p>
<a name="index-atomic_005faddmode-instruction-pattern"></a>
<a name="index-atomic_005fsubmode-instruction-pattern"></a>
<a name="index-atomic_005formode-instruction-pattern"></a>
<a name="index-atomic_005fandmode-instruction-pattern"></a>
<a name="index-atomic_005fxormode-instruction-pattern"></a>
<a name="index-atomic_005fnandmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_add<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_sub<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_or<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_and<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_xor<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_nand<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns emit code for an atomic operation on memory with memory
model semantics. Operand 0 is the memory on which the atomic operation is
performed.  Operand 1 is the second operand to the binary operator.
Operand 2 is the memory model to be used by the operation.
</p>
<p>If these patterns are not defined, attempts will be made to use legacy
<code>sync</code> patterns, or equivalent patterns which return a result.  If
none of these are available a compare-and-swap loop will be used.
</p>
<a name="index-atomic_005ffetch_005faddmode-instruction-pattern"></a>
<a name="index-atomic_005ffetch_005fsubmode-instruction-pattern"></a>
<a name="index-atomic_005ffetch_005formode-instruction-pattern"></a>
<a name="index-atomic_005ffetch_005fandmode-instruction-pattern"></a>
<a name="index-atomic_005ffetch_005fxormode-instruction-pattern"></a>
<a name="index-atomic_005ffetch_005fnandmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_fetch_add<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_fetch_sub<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_fetch_or<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_fetch_and<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_fetch_xor<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_fetch_nand<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns emit code for an atomic operation on memory with memory
model semantics, and return the original value. Operand 0 is an output 
operand which contains the value of the memory location before the 
operation was performed.  Operand 1 is the memory on which the atomic 
operation is performed.  Operand 2 is the second operand to the binary
operator.  Operand 3 is the memory model to be used by the operation.
</p>
<p>If these patterns are not defined, attempts will be made to use legacy
<code>sync</code> patterns.  If none of these are available a compare-and-swap
loop will be used.
</p>
<a name="index-atomic_005fadd_005ffetchmode-instruction-pattern"></a>
<a name="index-atomic_005fsub_005ffetchmode-instruction-pattern"></a>
<a name="index-atomic_005for_005ffetchmode-instruction-pattern"></a>
<a name="index-atomic_005fand_005ffetchmode-instruction-pattern"></a>
<a name="index-atomic_005fxor_005ffetchmode-instruction-pattern"></a>
<a name="index-atomic_005fnand_005ffetchmode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_add_fetch<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_sub_fetch<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_or_fetch<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_and_fetch<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>atomic_xor_fetch<var>mode</var></samp>&rsquo;, &lsquo;<samp>atomic_nand_fetch<var>mode</var></samp>&rsquo;</dt>
<dd>
<p>These patterns emit code for an atomic operation on memory with memory
model semantics and return the result after the operation is performed.
Operand 0 is an output operand which contains the value after the
operation.  Operand 1 is the memory on which the atomic operation is
performed.  Operand 2 is the second operand to the binary operator.
Operand 3 is the memory model to be used by the operation.
</p>
<p>If these patterns are not defined, attempts will be made to use legacy
<code>sync</code> patterns, or equivalent patterns which return the result before
the operation followed by the arithmetic operation required to produce the
result.  If none of these are available a compare-and-swap loop will be
used.
</p>
<a name="index-atomic_005ftest_005fand_005fset-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>atomic_test_and_set</samp>&rsquo;</dt>
<dd>
<p>This pattern emits code for <code>__builtin_atomic_test_and_set</code>.
Operand 0 is an output operand which is set to true if the previous
previous contents of the byte was &quot;set&quot;, and false otherwise.  Operand 1
is the <code>QImode</code> memory to be modified.  Operand 2 is the memory
model to be used.
</p>
<p>The specific value that defines &quot;set&quot; is implementation defined, and
is normally based on what is performed by the native atomic test and set
instruction.
</p>
<a name="index-mem_005fthread_005ffencemode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mem_thread_fence<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern emits code required to implement a thread fence with
memory model semantics.  Operand 0 is the memory model to be used.
</p>
<p>If this pattern is not specified, all memory models except
<code>__ATOMIC_RELAXED</code> will result in issuing a <code>sync_synchronize</code>
barrier pattern.
</p>
<a name="index-mem_005fsignal_005ffencemode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>mem_signal_fence<var>mode</var></samp>&rsquo;</dt>
<dd><p>This pattern emits code required to implement a signal fence with
memory model semantics.  Operand 0 is the memory model to be used.
</p>
<p>This pattern should impact the compiler optimizers the same way that
mem_signal_fence does, but it does not need to issue any barrier
instructions.
</p>
<p>If this pattern is not specified, all memory models except
<code>__ATOMIC_RELAXED</code> will result in issuing a <code>sync_synchronize</code>
barrier pattern.
</p>
<a name="index-get_005fthread_005fpointermode-instruction-pattern"></a>
<a name="index-set_005fthread_005fpointermode-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>get_thread_pointer<var>mode</var></samp>&rsquo;</dt>
<dt>&lsquo;<samp>set_thread_pointer<var>mode</var></samp>&rsquo;</dt>
<dd><p>These patterns emit code that reads/sets the TLS thread pointer. Currently,
these are only needed if the target needs to support the
<code>__builtin_thread_pointer</code> and <code>__builtin_set_thread_pointer</code>
builtins.
</p>
<p>The get/set patterns have a single output/input operand respectively,
with <var>mode</var> intended to be <code>Pmode</code>.
</p>
<a name="index-stack_005fprotect_005fset-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>stack_protect_set</samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, moves a <code>ptr_mode</code> value from the memory
in operand 1 to the memory in operand 0 without leaving the value in
a register afterward.  This is to avoid leaking the value some place
that an attacker might use to rewrite the stack guard slot after
having clobbered it.
</p>
<p>If this pattern is not defined, then a plain move pattern is generated.
</p>
<a name="index-stack_005fprotect_005ftest-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>stack_protect_test</samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, compares a <code>ptr_mode</code> value from the
memory in operand 1 with the memory in operand 0 without leaving the
value in a register afterward and branches to operand 2 if the values
were equal.
</p>
<p>If this pattern is not defined, then a plain compare pattern and
conditional branch pattern is used.
</p>
<a name="index-clear_005fcache-instruction-pattern"></a>
</dd>
<dt>&lsquo;<samp>clear_cache</samp>&rsquo;</dt>
<dd>
<p>This pattern, if defined, flushes the instruction cache for a region of
memory.  The region is bounded to by the Pmode pointers in operand 0
inclusive and operand 1 exclusive.
</p>
<p>If this pattern is not defined, a call to the library function
<code>__clear_cache</code> is used.
</p>
</dd>
</dl>

<hr>
<a name="Pattern-Ordering"></a>
<div class="header">
<p>
Next: <a href="#Dependent-Patterns" accesskey="n" rel="next">Dependent Patterns</a>, Previous: <a href="#Standard-Names" accesskey="p" rel="prev">Standard Names</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="When-the-Order-of-Patterns-Matters"></a>
<h3 class="section">16.10 When the Order of Patterns Matters</h3>
<a name="index-Pattern-Ordering"></a>
<a name="index-Ordering-of-Patterns"></a>

<p>Sometimes an insn can match more than one instruction pattern.  Then the
pattern that appears first in the machine description is the one used.
Therefore, more specific patterns (patterns that will match fewer things)
and faster instructions (those that will produce better code when they
do match) should usually go first in the description.
</p>
<p>In some cases the effect of ordering the patterns can be used to hide
a pattern when it is not valid.  For example, the 68000 has an
instruction for converting a fullword to floating point and another
for converting a byte to floating point.  An instruction converting
an integer to floating point could match either one.  We put the
pattern to convert the fullword first to make sure that one will
be used rather than the other.  (Otherwise a large integer might
be generated as a single-byte immediate quantity, which would not work.)
Instead of using this pattern ordering it would be possible to make the
pattern for convert-a-byte smart enough to deal properly with any
constant value.
</p>
<hr>
<a name="Dependent-Patterns"></a>
<div class="header">
<p>
Next: <a href="#Jump-Patterns" accesskey="n" rel="next">Jump Patterns</a>, Previous: <a href="#Pattern-Ordering" accesskey="p" rel="prev">Pattern Ordering</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Interdependence-of-Patterns"></a>
<h3 class="section">16.11 Interdependence of Patterns</h3>
<a name="index-Dependent-Patterns"></a>
<a name="index-Interdependence-of-Patterns"></a>

<p>In some cases machines support instructions identical except for the
machine mode of one or more operands.  For example, there may be
&ldquo;sign-extend halfword&rdquo; and &ldquo;sign-extend byte&rdquo; instructions whose
patterns are
</p>
<div class="smallexample">
<pre class="smallexample">(set (match_operand:SI 0 &hellip;)
     (extend:SI (match_operand:HI 1 &hellip;)))

(set (match_operand:SI 0 &hellip;)
     (extend:SI (match_operand:QI 1 &hellip;)))
</pre></div>

<p>Constant integers do not specify a machine mode, so an instruction to
extend a constant value could match either pattern.  The pattern it
actually will match is the one that appears first in the file.  For correct
results, this must be the one for the widest possible mode (<code>HImode</code>,
here).  If the pattern matches the <code>QImode</code> instruction, the results
will be incorrect if the constant value does not actually fit that mode.
</p>
<p>Such instructions to extend constants are rarely generated because they are
optimized away, but they do occasionally happen in nonoptimized
compilations.
</p>
<p>If a constraint in a pattern allows a constant, the reload pass may
replace a register with a constant permitted by the constraint in some
cases.  Similarly for memory references.  Because of this substitution,
you should not provide separate patterns for increment and decrement
instructions.  Instead, they should be generated from the same pattern
that supports register-register add insns by examining the operands and
generating the appropriate machine instruction.
</p>
<hr>
<a name="Jump-Patterns"></a>
<div class="header">
<p>
Next: <a href="#Looping-Patterns" accesskey="n" rel="next">Looping Patterns</a>, Previous: <a href="#Dependent-Patterns" accesskey="p" rel="prev">Dependent Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Jump-Instruction-Patterns"></a>
<h3 class="section">16.12 Defining Jump Instruction Patterns</h3>
<a name="index-jump-instruction-patterns"></a>
<a name="index-defining-jump-instruction-patterns"></a>

<p>GCC does not assume anything about how the machine realizes jumps.
The machine description should define a single pattern, usually
a <code>define_expand</code>, which expands to all the required insns.
</p>
<p>Usually, this would be a comparison insn to set the condition code
and a separate branch insn testing the condition code and branching
or not according to its value.  For many machines, however,
separating compares and branches is limiting, which is why the
more flexible approach with one <code>define_expand</code> is used in GCC.
The machine description becomes clearer for architectures that
have compare-and-branch instructions but no condition code.  It also
works better when different sets of comparison operators are supported
by different kinds of conditional branches (e.g. integer vs. floating-point),
or by conditional branches with respect to conditional stores.
</p>
<p>Two separate insns are always used if the machine description represents
a condition code register using the legacy RTL expression <code>(cc0)</code>,
and on most machines that use a separate condition code register
(see <a href="#Condition-Code">Condition Code</a>).  For machines that use <code>(cc0)</code>, in
fact, the set and use of the condition code must be separate and
adjacent<a name="DOCF4" href="#FOOT4"><sup>4</sup></a>, thus
allowing flags in <code>cc_status</code> to be used (see <a href="#Condition-Code">Condition Code</a>) and
so that the comparison and branch insns could be located from each other
by using the functions <code>prev_cc0_setter</code> and <code>next_cc0_user</code>.
</p>
<p>Even in this case having a single entry point for conditional branches
is advantageous, because it handles equally well the case where a single
comparison instruction records the results of both signed and unsigned
comparison of the given operands (with the branch insns coming in distinct
signed and unsigned flavors) as in the x86 or SPARC, and the case where
there are distinct signed and unsigned compare instructions and only
one set of conditional branch instructions as in the PowerPC.
</p>
<hr>
<a name="Looping-Patterns"></a>
<div class="header">
<p>
Next: <a href="#Insn-Canonicalizations" accesskey="n" rel="next">Insn Canonicalizations</a>, Previous: <a href="#Jump-Patterns" accesskey="p" rel="prev">Jump Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Looping-Instruction-Patterns"></a>
<h3 class="section">16.13 Defining Looping Instruction Patterns</h3>
<a name="index-looping-instruction-patterns"></a>
<a name="index-defining-looping-instruction-patterns"></a>

<p>Some machines have special jump instructions that can be utilized to
make loops more efficient.  A common example is the 68000 &lsquo;<samp>dbra</samp>&rsquo;
instruction which performs a decrement of a register and a branch if the
result was greater than zero.  Other machines, in particular digital
signal processors (DSPs), have special block repeat instructions to
provide low-overhead loop support.  For example, the TI TMS320C3x/C4x
DSPs have a block repeat instruction that loads special registers to
mark the top and end of a loop and to count the number of loop
iterations.  This avoids the need for fetching and executing a
&lsquo;<samp>dbra</samp>&rsquo;-like instruction and avoids pipeline stalls associated with
the jump.
</p>
<p>GCC has three special named patterns to support low overhead looping.
They are &lsquo;<samp>decrement_and_branch_until_zero</samp>&rsquo;, &lsquo;<samp>doloop_begin</samp>&rsquo;,
and &lsquo;<samp>doloop_end</samp>&rsquo;.  The first pattern,
&lsquo;<samp>decrement_and_branch_until_zero</samp>&rsquo;, is not emitted during RTL
generation but may be emitted during the instruction combination phase.
This requires the assistance of the loop optimizer, using information
collected during strength reduction, to reverse a loop to count down to
zero.  Some targets also require the loop optimizer to add a
<code>REG_NONNEG</code> note to indicate that the iteration count is always
positive.  This is needed if the target performs a signed loop
termination test.  For example, the 68000 uses a pattern similar to the
following for its <code>dbra</code> instruction:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;decrement_and_branch_until_zero&quot;
  [(set (pc)
        (if_then_else
          (ge (plus:SI (match_operand:SI 0 &quot;general_operand&quot; &quot;+d*am&quot;)
                       (const_int -1))
              (const_int 0))
          (label_ref (match_operand 1 &quot;&quot; &quot;&quot;))
          (pc)))
   (set (match_dup 0)
        (plus:SI (match_dup 0)
                 (const_int -1)))]
  &quot;find_reg_note (insn, REG_NONNEG, 0)&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>Note that since the insn is both a jump insn and has an output, it must
deal with its own reloads, hence the &lsquo;m&rsquo; constraints.  Also note that
since this insn is generated by the instruction combination phase
combining two sequential insns together into an implicit parallel insn,
the iteration counter needs to be biased by the same amount as the
decrement operation, in this case -1.  Note that the following similar
pattern will not be matched by the combiner.
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;decrement_and_branch_until_zero&quot;
  [(set (pc)
        (if_then_else
          (ge (match_operand:SI 0 &quot;general_operand&quot; &quot;+d*am&quot;)
              (const_int 1))
          (label_ref (match_operand 1 &quot;&quot; &quot;&quot;))
          (pc)))
   (set (match_dup 0)
        (plus:SI (match_dup 0)
                 (const_int -1)))]
  &quot;find_reg_note (insn, REG_NONNEG, 0)&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>The other two special looping patterns, &lsquo;<samp>doloop_begin</samp>&rsquo; and
&lsquo;<samp>doloop_end</samp>&rsquo;, are emitted by the loop optimizer for certain
well-behaved loops with a finite number of loop iterations using
information collected during strength reduction.
</p>
<p>The &lsquo;<samp>doloop_end</samp>&rsquo; pattern describes the actual looping instruction
(or the implicit looping operation) and the &lsquo;<samp>doloop_begin</samp>&rsquo; pattern
is an optional companion pattern that can be used for initialization
needed for some low-overhead looping instructions.
</p>
<p>Note that some machines require the actual looping instruction to be
emitted at the top of the loop (e.g., the TMS320C3x/C4x DSPs).  Emitting
the true RTL for a looping instruction at the top of the loop can cause
problems with flow analysis.  So instead, a dummy <code>doloop</code> insn is
emitted at the end of the loop.  The machine dependent reorg pass checks
for the presence of this <code>doloop</code> insn and then searches back to
the top of the loop, where it inserts the true looping insn (provided
there are no instructions in the loop which would cause problems).  Any
additional labels can be emitted at this point.  In addition, if the
desired special iteration counter register was not allocated, this
machine dependent reorg pass could emit a traditional compare and jump
instruction pair.
</p>
<p>The essential difference between the
&lsquo;<samp>decrement_and_branch_until_zero</samp>&rsquo; and the &lsquo;<samp>doloop_end</samp>&rsquo;
patterns is that the loop optimizer allocates an additional pseudo
register for the latter as an iteration counter.  This pseudo register
cannot be used within the loop (i.e., general induction variables cannot
be derived from it), however, in many cases the loop induction variable
may become redundant and removed by the flow pass.
</p>

<hr>
<a name="Insn-Canonicalizations"></a>
<div class="header">
<p>
Next: <a href="#Expander-Definitions" accesskey="n" rel="next">Expander Definitions</a>, Previous: <a href="#Looping-Patterns" accesskey="p" rel="prev">Looping Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Canonicalization-of-Instructions"></a>
<h3 class="section">16.14 Canonicalization of Instructions</h3>
<a name="index-canonicalization-of-instructions"></a>
<a name="index-insn-canonicalization"></a>

<p>There are often cases where multiple RTL expressions could represent an
operation performed by a single machine instruction.  This situation is
most commonly encountered with logical, branch, and multiply-accumulate
instructions.  In such cases, the compiler attempts to convert these
multiple RTL expressions into a single canonical form to reduce the
number of insn patterns required.
</p>
<p>In addition to algebraic simplifications, following canonicalizations
are performed:
</p>
<ul>
<li> For commutative and comparison operators, a constant is always made the
second operand.  If a machine only supports a constant as the second
operand, only patterns that match a constant in the second operand need
be supplied.

</li><li> For associative operators, a sequence of operators will always chain
to the left; for instance, only the left operand of an integer <code>plus</code>
can itself be a <code>plus</code>.  <code>and</code>, <code>ior</code>, <code>xor</code>,
<code>plus</code>, <code>mult</code>, <code>smin</code>, <code>smax</code>, <code>umin</code>, and
<code>umax</code> are associative when applied to integers, and sometimes to
floating-point.

</li><li> <a name="index-neg_002c-canonicalization-of"></a>
<a name="index-not_002c-canonicalization-of"></a>
<a name="index-mult_002c-canonicalization-of"></a>
<a name="index-plus_002c-canonicalization-of"></a>
<a name="index-minus_002c-canonicalization-of"></a>
For these operators, if only one operand is a <code>neg</code>, <code>not</code>,
<code>mult</code>, <code>plus</code>, or <code>minus</code> expression, it will be the
first operand.

</li><li> In combinations of <code>neg</code>, <code>mult</code>, <code>plus</code>, and
<code>minus</code>, the <code>neg</code> operations (if any) will be moved inside
the operations as far as possible.  For instance,
<code>(neg (mult A B))</code> is canonicalized as <code>(mult (neg A) B)</code>, but
<code>(plus (mult (neg B) C) A)</code> is canonicalized as
<code>(minus A (mult B C))</code>.

</li><li> <a name="index-compare_002c-canonicalization-of"></a>
For the <code>compare</code> operator, a constant is always the second operand
if the first argument is a condition code register or <code>(cc0)</code>.

</li><li> An operand of <code>neg</code>, <code>not</code>, <code>mult</code>, <code>plus</code>, or
<code>minus</code> is made the first operand under the same conditions as
above.

</li><li> <code>(ltu (plus <var>a</var> <var>b</var>) <var>b</var>)</code> is converted to
<code>(ltu (plus <var>a</var> <var>b</var>) <var>a</var>)</code>. Likewise with <code>geu</code> instead
of <code>ltu</code>.

</li><li> <code>(minus <var>x</var> (const_int <var>n</var>))</code> is converted to
<code>(plus <var>x</var> (const_int <var>-n</var>))</code>.

</li><li> Within address computations (i.e., inside <code>mem</code>), a left shift is
converted into the appropriate multiplication by a power of two.

</li><li> <a name="index-ior_002c-canonicalization-of"></a>
<a name="index-and_002c-canonicalization-of"></a>
<a name="index-De-Morgan_0027s-law"></a>
De Morgan&rsquo;s Law is used to move bitwise negation inside a bitwise
logical-and or logical-or operation.  If this results in only one
operand being a <code>not</code> expression, it will be the first one.

<p>A machine that has an instruction that performs a bitwise logical-and of one
operand with the bitwise negation of the other should specify the pattern
for that instruction as
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:<var>m</var> 0 &hellip;)
        (and:<var>m</var> (not:<var>m</var> (match_operand:<var>m</var> 1 &hellip;))
                     (match_operand:<var>m</var> 2 &hellip;)))]
  &quot;&hellip;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>Similarly, a pattern for a &ldquo;NAND&rdquo; instruction should be written
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (match_operand:<var>m</var> 0 &hellip;)
        (ior:<var>m</var> (not:<var>m</var> (match_operand:<var>m</var> 1 &hellip;))
                     (not:<var>m</var> (match_operand:<var>m</var> 2 &hellip;))))]
  &quot;&hellip;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>In both cases, it is not necessary to include patterns for the many
logically equivalent RTL expressions.
</p>
</li><li> <a name="index-xor_002c-canonicalization-of"></a>
The only possible RTL expressions involving both bitwise exclusive-or
and bitwise negation are <code>(xor:<var>m</var> <var>x</var> <var>y</var>)</code>
and <code>(not:<var>m</var> (xor:<var>m</var> <var>x</var> <var>y</var>))</code>.

</li><li> The sum of three items, one of which is a constant, will only appear in
the form

<div class="smallexample">
<pre class="smallexample">(plus:<var>m</var> (plus:<var>m</var> <var>x</var> <var>y</var>) <var>constant</var>)
</pre></div>

</li><li> <a name="index-zero_005fextract_002c-canonicalization-of"></a>
<a name="index-sign_005fextract_002c-canonicalization-of"></a>
Equality comparisons of a group of bits (usually a single bit) with zero
will be written using <code>zero_extract</code> rather than the equivalent
<code>and</code> or <code>sign_extract</code> operations.

</li><li> <a name="index-mult_002c-canonicalization-of-1"></a>
<code>(sign_extend:<var>m1</var> (mult:<var>m2</var> (sign_extend:<var>m2</var> <var>x</var>)
(sign_extend:<var>m2</var> <var>y</var>)))</code> is converted to <code>(mult:<var>m1</var>
(sign_extend:<var>m1</var> <var>x</var>) (sign_extend:<var>m1</var> <var>y</var>))</code>, and likewise
for <code>zero_extend</code>.

</li><li> <code>(sign_extend:<var>m1</var> (mult:<var>m2</var> (ashiftrt:<var>m2</var>
<var>x</var> <var>s</var>) (sign_extend:<var>m2</var> <var>y</var>)))</code> is converted
to <code>(mult:<var>m1</var> (sign_extend:<var>m1</var> (ashiftrt:<var>m2</var>
<var>x</var> <var>s</var>)) (sign_extend:<var>m1</var> <var>y</var>))</code>, and likewise for
patterns using <code>zero_extend</code> and <code>lshiftrt</code>.  If the second
operand of <code>mult</code> is also a shift, then that is extended also.
This transformation is only applied when it can be proven that the
original operation had sufficient precision to prevent overflow.

</li></ul>

<p>Further canonicalization rules are defined in the function
<code>commutative_operand_precedence</code> in <samp>gcc/rtlanal.c</samp>.
</p>
<hr>
<a name="Expander-Definitions"></a>
<div class="header">
<p>
Next: <a href="#Insn-Splitting" accesskey="n" rel="next">Insn Splitting</a>, Previous: <a href="#Insn-Canonicalizations" accesskey="p" rel="prev">Insn Canonicalizations</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-RTL-Sequences-for-Code-Generation"></a>
<h3 class="section">16.15 Defining RTL Sequences for Code Generation</h3>
<a name="index-expander-definitions"></a>
<a name="index-code-generation-RTL-sequences"></a>
<a name="index-defining-RTL-sequences-for-code-generation"></a>

<p>On some target machines, some standard pattern names for RTL generation
cannot be handled with single insn, but a sequence of RTL insns can
represent them.  For these target machines, you can write a
<code>define_expand</code> to specify how to generate the sequence of RTL.
</p>
<a name="index-define_005fexpand"></a>
<p>A <code>define_expand</code> is an RTL expression that looks almost like a
<code>define_insn</code>; but, unlike the latter, a <code>define_expand</code> is used
only for RTL generation and it can produce more than one RTL insn.
</p>
<p>A <code>define_expand</code> RTX has four operands:
</p>
<ul>
<li> The name.  Each <code>define_expand</code> must have a name, since the only
use for it is to refer to it by name.

</li><li> The RTL template.  This is a vector of RTL expressions representing
a sequence of separate instructions.  Unlike <code>define_insn</code>, there
is no implicit surrounding <code>PARALLEL</code>.

</li><li> The condition, a string containing a C expression.  This expression is
used to express how the availability of this pattern depends on
subclasses of target machine, selected by command-line options when GCC
is run.  This is just like the condition of a <code>define_insn</code> that
has a standard name.  Therefore, the condition (if present) may not
depend on the data in the insn being matched, but only the
target-machine-type flags.  The compiler needs to test these conditions
during initialization in order to learn exactly which named instructions
are available in a particular run.

</li><li> The preparation statements, a string containing zero or more C
statements which are to be executed before RTL code is generated from
the RTL template.

<p>Usually these statements prepare temporary registers for use as
internal operands in the RTL template, but they can also generate RTL
insns directly by calling routines such as <code>emit_insn</code>, etc.
Any such insns precede the ones that come from the RTL template.
</p>
</li><li> Optionally, a vector containing the values of attributes. See <a href="#Insn-Attributes">Insn Attributes</a>.
</li></ul>

<p>Every RTL insn emitted by a <code>define_expand</code> must match some
<code>define_insn</code> in the machine description.  Otherwise, the compiler
will crash when trying to generate code for the insn or trying to optimize
it.
</p>
<p>The RTL template, in addition to controlling generation of RTL insns,
also describes the operands that need to be specified when this pattern
is used.  In particular, it gives a predicate for each operand.
</p>
<p>A true operand, which needs to be specified in order to generate RTL from
the pattern, should be described with a <code>match_operand</code> in its first
occurrence in the RTL template.  This enters information on the operand&rsquo;s
predicate into the tables that record such things.  GCC uses the
information to preload the operand into a register if that is required for
valid RTL code.  If the operand is referred to more than once, subsequent
references should use <code>match_dup</code>.
</p>
<p>The RTL template may also refer to internal &ldquo;operands&rdquo; which are
temporary registers or labels used only within the sequence made by the
<code>define_expand</code>.  Internal operands are substituted into the RTL
template with <code>match_dup</code>, never with <code>match_operand</code>.  The
values of the internal operands are not passed in as arguments by the
compiler when it requests use of this pattern.  Instead, they are computed
within the pattern, in the preparation statements.  These statements
compute the values and store them into the appropriate elements of
<code>operands</code> so that <code>match_dup</code> can find them.
</p>
<p>There are two special macros defined for use in the preparation statements:
<code>DONE</code> and <code>FAIL</code>.  Use them with a following semicolon,
as a statement.
</p>
<dl compact="compact">
<dd>
<a name="index-DONE"></a>
</dd>
<dt><code>DONE</code></dt>
<dd><p>Use the <code>DONE</code> macro to end RTL generation for the pattern.  The
only RTL insns resulting from the pattern on this occasion will be
those already emitted by explicit calls to <code>emit_insn</code> within the
preparation statements; the RTL template will not be generated.
</p>
<a name="index-FAIL"></a>
</dd>
<dt><code>FAIL</code></dt>
<dd><p>Make the pattern fail on this occasion.  When a pattern fails, it means
that the pattern was not truly available.  The calling routines in the
compiler will try other strategies for code generation using other patterns.
</p>
<p>Failure is currently supported only for binary (addition, multiplication,
shifting, etc.) and bit-field (<code>extv</code>, <code>extzv</code>, and <code>insv</code>)
operations.
</p></dd>
</dl>

<p>If the preparation falls through (invokes neither <code>DONE</code> nor
<code>FAIL</code>), then the <code>define_expand</code> acts like a
<code>define_insn</code> in that the RTL template is used to generate the
insn.
</p>
<p>The RTL template is not used for matching, only for generating the
initial insn list.  If the preparation statement always invokes
<code>DONE</code> or <code>FAIL</code>, the RTL template may be reduced to a simple
list of operands, such as this example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_expand &quot;addsi3&quot;
  [(match_operand:SI 0 &quot;register_operand&quot; &quot;&quot;)
   (match_operand:SI 1 &quot;register_operand&quot; &quot;&quot;)
   (match_operand:SI 2 &quot;register_operand&quot; &quot;&quot;)]
</pre><pre class="smallexample">  &quot;&quot;
  &quot;
{
  handle_add (operands[0], operands[1], operands[2]);
  DONE;
}&quot;)
</pre></div>

<p>Here is an example, the definition of left-shift for the SPUR chip:
</p>
<div class="smallexample">
<pre class="smallexample">(define_expand &quot;ashlsi3&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;&quot;)
        (ashift:SI
</pre><pre class="smallexample">          (match_operand:SI 1 &quot;register_operand&quot; &quot;&quot;)
          (match_operand:SI 2 &quot;nonmemory_operand&quot; &quot;&quot;)))]
  &quot;&quot;
  &quot;
</pre></div>

<div class="smallexample">
<pre class="smallexample">{
  if (GET_CODE (operands[2]) != CONST_INT
      || (unsigned) INTVAL (operands[2]) &gt; 3)
    FAIL;
}&quot;)
</pre></div>

<p>This example uses <code>define_expand</code> so that it can generate an RTL insn
for shifting when the shift-count is in the supported range of 0 to 3 but
fail in other cases where machine insns aren&rsquo;t available.  When it fails,
the compiler tries another strategy using different patterns (such as, a
library call).
</p>
<p>If the compiler were able to handle nontrivial condition-strings in
patterns with names, then it would be possible to use a
<code>define_insn</code> in that case.  Here is another case (zero-extension
on the 68000) which makes more use of the power of <code>define_expand</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(define_expand &quot;zero_extendhisi2&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;&quot;)
        (const_int 0))
   (set (strict_low_part
          (subreg:HI
            (match_dup 0)
            0))
        (match_operand:HI 1 &quot;general_operand&quot; &quot;&quot;))]
  &quot;&quot;
  &quot;operands[1] = make_safe_from (operands[1], operands[0]);&quot;)
</pre></div>

<a name="index-make_005fsafe_005ffrom"></a>
<p>Here two RTL insns are generated, one to clear the entire output operand
and the other to copy the input operand into its low half.  This sequence
is incorrect if the input operand refers to [the old value of] the output
operand, so the preparation statement makes sure this isn&rsquo;t so.  The
function <code>make_safe_from</code> copies the <code>operands[1]</code> into a
temporary register if it refers to <code>operands[0]</code>.  It does this
by emitting another RTL insn.
</p>
<p>Finally, a third example shows the use of an internal operand.
Zero-extension on the SPUR chip is done by <code>and</code>-ing the result
against a halfword mask.  But this mask cannot be represented by a
<code>const_int</code> because the constant value is too large to be legitimate
on this machine.  So it must be copied into a register with
<code>force_reg</code> and then the register used in the <code>and</code>.
</p>
<div class="smallexample">
<pre class="smallexample">(define_expand &quot;zero_extendhisi2&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;&quot;)
        (and:SI (subreg:SI
                  (match_operand:HI 1 &quot;register_operand&quot; &quot;&quot;)
                  0)
                (match_dup 2)))]
  &quot;&quot;
  &quot;operands[2]
     = force_reg (SImode, GEN_INT (65535)); &quot;)
</pre></div>

<p><em>Note:</em> If the <code>define_expand</code> is used to serve a
standard binary or unary arithmetic operation or a bit-field operation,
then the last insn it generates must not be a <code>code_label</code>,
<code>barrier</code> or <code>note</code>.  It must be an <code>insn</code>,
<code>jump_insn</code> or <code>call_insn</code>.  If you don&rsquo;t need a real insn
at the end, emit an insn to copy the result of the operation into
itself.  Such an insn will generate no code, but it can avoid problems
in the compiler.
</p>
<hr>
<a name="Insn-Splitting"></a>
<div class="header">
<p>
Next: <a href="#Including-Patterns" accesskey="n" rel="next">Including Patterns</a>, Previous: <a href="#Expander-Definitions" accesskey="p" rel="prev">Expander Definitions</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-How-to-Split-Instructions"></a>
<h3 class="section">16.16 Defining How to Split Instructions</h3>
<a name="index-insn-splitting"></a>
<a name="index-instruction-splitting"></a>
<a name="index-splitting-instructions"></a>

<p>There are two cases where you should specify how to split a pattern
into multiple insns.  On machines that have instructions requiring
delay slots (see <a href="#Delay-Slots">Delay Slots</a>) or that have instructions whose
output is not available for multiple cycles (see <a href="#Processor-pipeline-description">Processor pipeline description</a>), the compiler phases that optimize these cases need to
be able to move insns into one-instruction delay slots.  However, some
insns may generate more than one machine instruction.  These insns
cannot be placed into a delay slot.
</p>
<p>Often you can rewrite the single insn as a list of individual insns,
each corresponding to one machine instruction.  The disadvantage of
doing so is that it will cause the compilation to be slower and require
more space.  If the resulting insns are too complex, it may also
suppress some optimizations.  The compiler splits the insn if there is a
reason to believe that it might improve instruction or delay slot
scheduling.
</p>
<p>The insn combiner phase also splits putative insns.  If three insns are
merged into one insn with a complex expression that cannot be matched by
some <code>define_insn</code> pattern, the combiner phase attempts to split
the complex pattern into two insns that are recognized.  Usually it can
break the complex pattern into two patterns by splitting out some
subexpression.  However, in some other cases, such as performing an
addition of a large constant in two insns on a RISC machine, the way to
split the addition into two insns is machine-dependent.
</p>
<a name="index-define_005fsplit"></a>
<p>The <code>define_split</code> definition tells the compiler how to split a
complex insn into several simpler insns.  It looks like this:
</p>
<div class="smallexample">
<pre class="smallexample">(define_split
  [<var>insn-pattern</var>]
  &quot;<var>condition</var>&quot;
  [<var>new-insn-pattern-1</var>
   <var>new-insn-pattern-2</var>
   &hellip;]
  &quot;<var>preparation-statements</var>&quot;)
</pre></div>

<p><var>insn-pattern</var> is a pattern that needs to be split and
<var>condition</var> is the final condition to be tested, as in a
<code>define_insn</code>.  When an insn matching <var>insn-pattern</var> and
satisfying <var>condition</var> is found, it is replaced in the insn list
with the insns given by <var>new-insn-pattern-1</var>,
<var>new-insn-pattern-2</var>, etc.
</p>
<p>The <var>preparation-statements</var> are similar to those statements that
are specified for <code>define_expand</code> (see <a href="#Expander-Definitions">Expander Definitions</a>)
and are executed before the new RTL is generated to prepare for the
generated code or emit some insns whose pattern is not fixed.  Unlike
those in <code>define_expand</code>, however, these statements must not
generate any new pseudo-registers.  Once reload has completed, they also
must not allocate any space in the stack frame.
</p>
<p>Patterns are matched against <var>insn-pattern</var> in two different
circumstances.  If an insn needs to be split for delay slot scheduling
or insn scheduling, the insn is already known to be valid, which means
that it must have been matched by some <code>define_insn</code> and, if
<code>reload_completed</code> is nonzero, is known to satisfy the constraints
of that <code>define_insn</code>.  In that case, the new insn patterns must
also be insns that are matched by some <code>define_insn</code> and, if
<code>reload_completed</code> is nonzero, must also satisfy the constraints
of those definitions.
</p>
<p>As an example of this usage of <code>define_split</code>, consider the following
example from <samp>a29k.md</samp>, which splits a <code>sign_extend</code> from
<code>HImode</code> to <code>SImode</code> into a pair of shift insns:
</p>
<div class="smallexample">
<pre class="smallexample">(define_split
  [(set (match_operand:SI 0 &quot;gen_reg_operand&quot; &quot;&quot;)
        (sign_extend:SI (match_operand:HI 1 &quot;gen_reg_operand&quot; &quot;&quot;)))]
  &quot;&quot;
  [(set (match_dup 0)
        (ashift:SI (match_dup 1)
                   (const_int 16)))
   (set (match_dup 0)
        (ashiftrt:SI (match_dup 0)
                     (const_int 16)))]
  &quot;
{ operands[1] = gen_lowpart (SImode, operands[1]); }&quot;)
</pre></div>

<p>When the combiner phase tries to split an insn pattern, it is always the
case that the pattern is <em>not</em> matched by any <code>define_insn</code>.
The combiner pass first tries to split a single <code>set</code> expression
and then the same <code>set</code> expression inside a <code>parallel</code>, but
followed by a <code>clobber</code> of a pseudo-reg to use as a scratch
register.  In these cases, the combiner expects exactly two new insn
patterns to be generated.  It will verify that these patterns match some
<code>define_insn</code> definitions, so you need not do this test in the
<code>define_split</code> (of course, there is no point in writing a
<code>define_split</code> that will never produce insns that match).
</p>
<p>Here is an example of this use of <code>define_split</code>, taken from
<samp>rs6000.md</samp>:
</p>
<div class="smallexample">
<pre class="smallexample">(define_split
  [(set (match_operand:SI 0 &quot;gen_reg_operand&quot; &quot;&quot;)
        (plus:SI (match_operand:SI 1 &quot;gen_reg_operand&quot; &quot;&quot;)
                 (match_operand:SI 2 &quot;non_add_cint_operand&quot; &quot;&quot;)))]
  &quot;&quot;
  [(set (match_dup 0) (plus:SI (match_dup 1) (match_dup 3)))
   (set (match_dup 0) (plus:SI (match_dup 0) (match_dup 4)))]
&quot;
{
  int low = INTVAL (operands[2]) &amp; 0xffff;
  int high = (unsigned) INTVAL (operands[2]) &gt;&gt; 16;

  if (low &amp; 0x8000)
    high++, low |= 0xffff0000;

  operands[3] = GEN_INT (high &lt;&lt; 16);
  operands[4] = GEN_INT (low);
}&quot;)
</pre></div>

<p>Here the predicate <code>non_add_cint_operand</code> matches any
<code>const_int</code> that is <em>not</em> a valid operand of a single add
insn.  The add with the smaller displacement is written so that it
can be substituted into the address of a subsequent operation.
</p>
<p>An example that uses a scratch register, from the same file, generates
an equality comparison of a register and a large constant:
</p>
<div class="smallexample">
<pre class="smallexample">(define_split
  [(set (match_operand:CC 0 &quot;cc_reg_operand&quot; &quot;&quot;)
        (compare:CC (match_operand:SI 1 &quot;gen_reg_operand&quot; &quot;&quot;)
                    (match_operand:SI 2 &quot;non_short_cint_operand&quot; &quot;&quot;)))
   (clobber (match_operand:SI 3 &quot;gen_reg_operand&quot; &quot;&quot;))]
  &quot;find_single_use (operands[0], insn, 0)
   &amp;&amp; (GET_CODE (*find_single_use (operands[0], insn, 0)) == EQ
       || GET_CODE (*find_single_use (operands[0], insn, 0)) == NE)&quot;
  [(set (match_dup 3) (xor:SI (match_dup 1) (match_dup 4)))
   (set (match_dup 0) (compare:CC (match_dup 3) (match_dup 5)))]
  &quot;
{
  /* <span class="roman">Get the constant we are comparing against, C, and see what it
     looks like sign-extended to 16 bits.  Then see what constant
     could be XOR&rsquo;ed with C to get the sign-extended value.</span>  */

  int c = INTVAL (operands[2]);
  int sextc = (c &lt;&lt; 16) &gt;&gt; 16;
  int xorv = c ^ sextc;

  operands[4] = GEN_INT (xorv);
  operands[5] = GEN_INT (sextc);
}&quot;)
</pre></div>

<p>To avoid confusion, don&rsquo;t write a single <code>define_split</code> that
accepts some insns that match some <code>define_insn</code> as well as some
insns that don&rsquo;t.  Instead, write two separate <code>define_split</code>
definitions, one for the insns that are valid and one for the insns that
are not valid.
</p>
<p>The splitter is allowed to split jump instructions into sequence of
jumps or create new jumps in while splitting non-jump instructions.  As
the central flowgraph and branch prediction information needs to be updated,
several restriction apply.
</p>
<p>Splitting of jump instruction into sequence that over by another jump
instruction is always valid, as compiler expect identical behavior of new
jump.  When new sequence contains multiple jump instructions or new labels,
more assistance is needed.  Splitter is required to create only unconditional
jumps, or simple conditional jump instructions.  Additionally it must attach a
<code>REG_BR_PROB</code> note to each conditional jump.  A global variable
<code>split_branch_probability</code> holds the probability of the original branch in case
it was a simple conditional jump, -1 otherwise.  To simplify
recomputing of edge frequencies, the new sequence is required to have only
forward jumps to the newly created labels.
</p>
<a name="index-define_005finsn_005fand_005fsplit"></a>
<p>For the common case where the pattern of a define_split exactly matches the
pattern of a define_insn, use <code>define_insn_and_split</code>.  It looks like
this:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn_and_split
  [<var>insn-pattern</var>]
  &quot;<var>condition</var>&quot;
  &quot;<var>output-template</var>&quot;
  &quot;<var>split-condition</var>&quot;
  [<var>new-insn-pattern-1</var>
   <var>new-insn-pattern-2</var>
   &hellip;]
  &quot;<var>preparation-statements</var>&quot;
  [<var>insn-attributes</var>])

</pre></div>

<p><var>insn-pattern</var>, <var>condition</var>, <var>output-template</var>, and
<var>insn-attributes</var> are used as in <code>define_insn</code>.  The
<var>new-insn-pattern</var> vector and the <var>preparation-statements</var> are used as
in a <code>define_split</code>.  The <var>split-condition</var> is also used as in
<code>define_split</code>, with the additional behavior that if the condition starts
with &lsquo;<samp>&amp;&amp;</samp>&rsquo;, the condition used for the split will be the constructed as a
logical &ldquo;and&rdquo; of the split condition with the insn condition.  For example,
from i386.md:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn_and_split &quot;zero_extendhisi2_and&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=r&quot;)
     (zero_extend:SI (match_operand:HI 1 &quot;register_operand&quot; &quot;0&quot;)))
   (clobber (reg:CC 17))]
  &quot;TARGET_ZERO_EXTEND_WITH_AND &amp;&amp; !optimize_size&quot;
  &quot;#&quot;
  &quot;&amp;&amp; reload_completed&quot;
  [(parallel [(set (match_dup 0)
                   (and:SI (match_dup 0) (const_int 65535)))
              (clobber (reg:CC 17))])]
  &quot;&quot;
  [(set_attr &quot;type&quot; &quot;alu1&quot;)])

</pre></div>

<p>In this case, the actual split condition will be
&lsquo;<samp>TARGET_ZERO_EXTEND_WITH_AND &amp;&amp; !optimize_size &amp;&amp; reload_completed</samp>&rsquo;.
</p>
<p>The <code>define_insn_and_split</code> construction provides exactly the same
functionality as two separate <code>define_insn</code> and <code>define_split</code>
patterns.  It exists for compactness, and as a maintenance tool to prevent
having to ensure the two patterns&rsquo; templates match.
</p>
<hr>
<a name="Including-Patterns"></a>
<div class="header">
<p>
Next: <a href="#Peephole-Definitions" accesskey="n" rel="next">Peephole Definitions</a>, Previous: <a href="#Insn-Splitting" accesskey="p" rel="prev">Insn Splitting</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Including-Patterns-in-Machine-Descriptions_002e"></a>
<h3 class="section">16.17 Including Patterns in Machine Descriptions.</h3>
<a name="index-insn-includes"></a>

<a name="index-include"></a>
<p>The <code>include</code> pattern tells the compiler tools where to
look for patterns that are in files other than in the file
<samp>.md</samp>.  This is used only at build time and there is no preprocessing allowed.
</p>
<p>It looks like:
</p>
<div class="smallexample">
<pre class="smallexample">
(include
  <var>pathname</var>)
</pre></div>

<p>For example:
</p>
<div class="smallexample">
<pre class="smallexample">
(include &quot;filestuff&quot;)

</pre></div>

<p>Where <var>pathname</var> is a string that specifies the location of the file,
specifies the include file to be in <samp>gcc/config/target/filestuff</samp>.  The
directory <samp>gcc/config/target</samp> is regarded as the default directory.
</p>

<p>Machine descriptions may be split up into smaller more manageable subsections
and placed into subdirectories.
</p>
<p>By specifying:
</p>
<div class="smallexample">
<pre class="smallexample">
(include &quot;BOGUS/filestuff&quot;)

</pre></div>

<p>the include file is specified to be in <samp>gcc/config/<var>target</var>/BOGUS/filestuff</samp>.
</p>
<p>Specifying an absolute path for the include file such as;
</p><div class="smallexample">
<pre class="smallexample">
(include &quot;/u2/BOGUS/filestuff&quot;)

</pre></div>
<p>is permitted but is not encouraged.
</p>
<a name="RTL-Generation-Tool-Options-for-Directory-Search"></a>
<h4 class="subsection">16.17.1 RTL Generation Tool Options for Directory Search</h4>
<a name="index-directory-options-_002emd"></a>
<a name="index-options_002c-directory-search"></a>
<a name="index-search-options"></a>

<p>The <samp>-I<var>dir</var></samp> option specifies directories to search for machine descriptions.
For example:
</p>
<div class="smallexample">
<pre class="smallexample">
genrecog -I/p1/abc/proc1 -I/p2/abcd/pro2 target.md

</pre></div>


<p>Add the directory <var>dir</var> to the head of the list of directories to be
searched for header files.  This can be used to override a system machine definition
file, substituting your own version, since these directories are
searched before the default machine description file directories.  If you use more than
one <samp>-I</samp> option, the directories are scanned in left-to-right
order; the standard default directory come after.
</p>

<hr>
<a name="Peephole-Definitions"></a>
<div class="header">
<p>
Next: <a href="#Insn-Attributes" accesskey="n" rel="next">Insn Attributes</a>, Previous: <a href="#Including-Patterns" accesskey="p" rel="prev">Including Patterns</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Machine_002dSpecific-Peephole-Optimizers"></a>
<h3 class="section">16.18 Machine-Specific Peephole Optimizers</h3>
<a name="index-peephole-optimizer-definitions"></a>
<a name="index-defining-peephole-optimizers"></a>

<p>In addition to instruction patterns the <samp>md</samp> file may contain
definitions of machine-specific peephole optimizations.
</p>
<p>The combiner does not notice certain peephole optimizations when the data
flow in the program does not suggest that it should try them.  For example,
sometimes two consecutive insns related in purpose can be combined even
though the second one does not appear to use a register computed in the
first one.  A machine-specific peephole optimizer can detect such
opportunities.
</p>
<p>There are two forms of peephole definitions that may be used.  The
original <code>define_peephole</code> is run at assembly output time to
match insns and substitute assembly text.  Use of <code>define_peephole</code>
is deprecated.
</p>
<p>A newer <code>define_peephole2</code> matches insns and substitutes new
insns.  The <code>peephole2</code> pass is run after register allocation
but before scheduling, which may result in much better code for
targets that do scheduling.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#define_005fpeephole" accesskey="1">define_peephole</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">RTL to Text Peephole Optimizers
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#define_005fpeephole2" accesskey="2">define_peephole2</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">RTL to RTL Peephole Optimizers
</td></tr>
</table>

<hr>
<a name="define_005fpeephole"></a>
<div class="header">
<p>
Next: <a href="#define_005fpeephole2" accesskey="n" rel="next">define_peephole2</a>, Up: <a href="#Peephole-Definitions" accesskey="u" rel="up">Peephole Definitions</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-to-Text-Peephole-Optimizers"></a>
<h4 class="subsection">16.18.1 RTL to Text Peephole Optimizers</h4>
<a name="index-define_005fpeephole"></a>

<p>A definition looks like this:
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole
  [<var>insn-pattern-1</var>
   <var>insn-pattern-2</var>
   &hellip;]
  &quot;<var>condition</var>&quot;
  &quot;<var>template</var>&quot;
  &quot;<var>optional-insn-attributes</var>&quot;)
</pre></div>

<p>The last string operand may be omitted if you are not using any
machine-specific information in this machine description.  If present,
it must obey the same rules as in a <code>define_insn</code>.
</p>
<p>In this skeleton, <var>insn-pattern-1</var> and so on are patterns to match
consecutive insns.  The optimization applies to a sequence of insns when
<var>insn-pattern-1</var> matches the first one, <var>insn-pattern-2</var> matches
the next, and so on.
</p>
<p>Each of the insns matched by a peephole must also match a
<code>define_insn</code>.  Peepholes are checked only at the last stage just
before code generation, and only optionally.  Therefore, any insn which
would match a peephole but no <code>define_insn</code> will cause a crash in code
generation in an unoptimized compilation, or at various optimization
stages.
</p>
<p>The operands of the insns are matched with <code>match_operands</code>,
<code>match_operator</code>, and <code>match_dup</code>, as usual.  What is not
usual is that the operand numbers apply to all the insn patterns in the
definition.  So, you can check for identical operands in two insns by
using <code>match_operand</code> in one insn and <code>match_dup</code> in the
other.
</p>
<p>The operand constraints used in <code>match_operand</code> patterns do not have
any direct effect on the applicability of the peephole, but they will
be validated afterward, so make sure your constraints are general enough
to apply whenever the peephole matches.  If the peephole matches
but the constraints are not satisfied, the compiler will crash.
</p>
<p>It is safe to omit constraints in all the operands of the peephole; or
you can write constraints which serve as a double-check on the criteria
previously tested.
</p>
<p>Once a sequence of insns matches the patterns, the <var>condition</var> is
checked.  This is a C expression which makes the final decision whether to
perform the optimization (we do so if the expression is nonzero).  If
<var>condition</var> is omitted (in other words, the string is empty) then the
optimization is applied to every sequence of insns that matches the
patterns.
</p>
<p>The defined peephole optimizations are applied after register allocation
is complete.  Therefore, the peephole definition can check which
operands have ended up in which kinds of registers, just by looking at
the operands.
</p>
<a name="index-prev_005factive_005finsn"></a>
<p>The way to refer to the operands in <var>condition</var> is to write
<code>operands[<var>i</var>]</code> for operand number <var>i</var> (as matched by
<code>(match_operand <var>i</var> &hellip;)</code>).  Use the variable <code>insn</code>
to refer to the last of the insns being matched; use
<code>prev_active_insn</code> to find the preceding insns.
</p>
<a name="index-dead_005for_005fset_005fp"></a>
<p>When optimizing computations with intermediate results, you can use
<var>condition</var> to match only when the intermediate results are not used
elsewhere.  Use the C expression <code>dead_or_set_p (<var>insn</var>,
<var>op</var>)</code>, where <var>insn</var> is the insn in which you expect the value
to be used for the last time (from the value of <code>insn</code>, together
with use of <code>prev_nonnote_insn</code>), and <var>op</var> is the intermediate
value (from <code>operands[<var>i</var>]</code>).
</p>
<p>Applying the optimization means replacing the sequence of insns with one
new insn.  The <var>template</var> controls ultimate output of assembler code
for this combined insn.  It works exactly like the template of a
<code>define_insn</code>.  Operand numbers in this template are the same ones
used in matching the original sequence of insns.
</p>
<p>The result of a defined peephole optimizer does not need to match any of
the insn patterns in the machine description; it does not even have an
opportunity to match them.  The peephole optimizer definition itself serves
as the insn pattern to control how the insn is output.
</p>
<p>Defined peephole optimizers are run as assembler code is being output,
so the insns they produce are never combined or rearranged in any way.
</p>
<p>Here is an example, taken from the 68000 machine description:
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole
  [(set (reg:SI 15) (plus:SI (reg:SI 15) (const_int 4)))
   (set (match_operand:DF 0 &quot;register_operand&quot; &quot;=f&quot;)
        (match_operand:DF 1 &quot;register_operand&quot; &quot;ad&quot;))]
  &quot;FP_REG_P (operands[0]) &amp;&amp; ! FP_REG_P (operands[1])&quot;
{
  rtx xoperands[2];
  xoperands[1] = gen_rtx_REG (SImode, REGNO (operands[1]) + 1);
#ifdef MOTOROLA
  output_asm_insn (&quot;move.l %1,(sp)&quot;, xoperands);
  output_asm_insn (&quot;move.l %1,-(sp)&quot;, operands);
  return &quot;fmove.d (sp)+,%0&quot;;
#else
  output_asm_insn (&quot;movel %1,sp@&quot;, xoperands);
  output_asm_insn (&quot;movel %1,sp@-&quot;, operands);
  return &quot;fmoved sp@+,%0&quot;;
#endif
})
</pre></div>

<p>The effect of this optimization is to change
</p>
<div class="smallexample">
<pre class="smallexample">jbsr _foobar
addql #4,sp
movel d1,sp@-
movel d0,sp@-
fmoved sp@+,fp0
</pre></div>

<p>into
</p>
<div class="smallexample">
<pre class="smallexample">jbsr _foobar
movel d1,sp@
movel d0,sp@-
fmoved sp@+,fp0
</pre></div>


<p><var>insn-pattern-1</var> and so on look <em>almost</em> like the second
operand of <code>define_insn</code>.  There is one important difference: the
second operand of <code>define_insn</code> consists of one or more RTX&rsquo;s
enclosed in square brackets.  Usually, there is only one: then the same
action can be written as an element of a <code>define_peephole</code>.  But
when there are multiple actions in a <code>define_insn</code>, they are
implicitly enclosed in a <code>parallel</code>.  Then you must explicitly
write the <code>parallel</code>, and the square brackets within it, in the
<code>define_peephole</code>.  Thus, if an insn pattern looks like this,
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;divmodsi4&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=d&quot;)
        (div:SI (match_operand:SI 1 &quot;general_operand&quot; &quot;0&quot;)
                (match_operand:SI 2 &quot;general_operand&quot; &quot;dmsK&quot;)))
   (set (match_operand:SI 3 &quot;general_operand&quot; &quot;=d&quot;)
        (mod:SI (match_dup 1) (match_dup 2)))]
  &quot;TARGET_68020&quot;
  &quot;divsl%.l %2,%3:%0&quot;)
</pre></div>

<p>then the way to mention this insn in a peephole is as follows:
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole
  [&hellip;
   (parallel
    [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=d&quot;)
          (div:SI (match_operand:SI 1 &quot;general_operand&quot; &quot;0&quot;)
                  (match_operand:SI 2 &quot;general_operand&quot; &quot;dmsK&quot;)))
     (set (match_operand:SI 3 &quot;general_operand&quot; &quot;=d&quot;)
          (mod:SI (match_dup 1) (match_dup 2)))])
   &hellip;]
  &hellip;)
</pre></div>

<hr>
<a name="define_005fpeephole2"></a>
<div class="header">
<p>
Previous: <a href="#define_005fpeephole" accesskey="p" rel="prev">define_peephole</a>, Up: <a href="#Peephole-Definitions" accesskey="u" rel="up">Peephole Definitions</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-to-RTL-Peephole-Optimizers"></a>
<h4 class="subsection">16.18.2 RTL to RTL Peephole Optimizers</h4>
<a name="index-define_005fpeephole2"></a>

<p>The <code>define_peephole2</code> definition tells the compiler how to
substitute one sequence of instructions for another sequence,
what additional scratch registers may be needed and what their
lifetimes must be.
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole2
  [<var>insn-pattern-1</var>
   <var>insn-pattern-2</var>
   &hellip;]
  &quot;<var>condition</var>&quot;
  [<var>new-insn-pattern-1</var>
   <var>new-insn-pattern-2</var>
   &hellip;]
  &quot;<var>preparation-statements</var>&quot;)
</pre></div>

<p>The definition is almost identical to <code>define_split</code>
(see <a href="#Insn-Splitting">Insn Splitting</a>) except that the pattern to match is not a
single instruction, but a sequence of instructions.
</p>
<p>It is possible to request additional scratch registers for use in the
output template.  If appropriate registers are not free, the pattern
will simply not match.
</p>
<a name="index-match_005fscratch-1"></a>
<a name="index-match_005fdup-1"></a>
<p>Scratch registers are requested with a <code>match_scratch</code> pattern at
the top level of the input pattern.  The allocated register (initially) will
be dead at the point requested within the original sequence.  If the scratch
is used at more than a single point, a <code>match_dup</code> pattern at the
top level of the input pattern marks the last position in the input sequence
at which the register must be available.
</p>
<p>Here is an example from the IA-32 machine description:
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole2
  [(match_scratch:SI 2 &quot;r&quot;)
   (parallel [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;&quot;)
                   (match_operator:SI 3 &quot;arith_or_logical_operator&quot;
                     [(match_dup 0)
                      (match_operand:SI 1 &quot;memory_operand&quot; &quot;&quot;)]))
              (clobber (reg:CC 17))])]
  &quot;! optimize_size &amp;&amp; ! TARGET_READ_MODIFY&quot;
  [(set (match_dup 2) (match_dup 1))
   (parallel [(set (match_dup 0)
                   (match_op_dup 3 [(match_dup 0) (match_dup 2)]))
              (clobber (reg:CC 17))])]
  &quot;&quot;)
</pre></div>

<p>This pattern tries to split a load from its use in the hopes that we&rsquo;ll be
able to schedule around the memory load latency.  It allocates a single
<code>SImode</code> register of class <code>GENERAL_REGS</code> (<code>&quot;r&quot;</code>) that needs
to be live only at the point just before the arithmetic.
</p>
<p>A real example requiring extended scratch lifetimes is harder to come by,
so here&rsquo;s a silly made-up example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_peephole2
  [(match_scratch:SI 4 &quot;r&quot;)
   (set (match_operand:SI 0 &quot;&quot; &quot;&quot;) (match_operand:SI 1 &quot;&quot; &quot;&quot;))
   (set (match_operand:SI 2 &quot;&quot; &quot;&quot;) (match_dup 1))
   (match_dup 4)
   (set (match_operand:SI 3 &quot;&quot; &quot;&quot;) (match_dup 1))]
  &quot;/* <span class="roman">determine 1 does not overlap 0 and 2</span> */&quot;
  [(set (match_dup 4) (match_dup 1))
   (set (match_dup 0) (match_dup 4))
   (set (match_dup 2) (match_dup 4))]
   (set (match_dup 3) (match_dup 4))]
  &quot;&quot;)
</pre></div>

<p>If we had not added the <code>(match_dup 4)</code> in the middle of the input
sequence, it might have been the case that the register we chose at the
beginning of the sequence is killed by the first or second <code>set</code>.
</p>
<hr>
<a name="Insn-Attributes"></a>
<div class="header">
<p>
Next: <a href="#Conditional-Execution" accesskey="n" rel="next">Conditional Execution</a>, Previous: <a href="#Peephole-Definitions" accesskey="p" rel="prev">Peephole Definitions</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Instruction-Attributes"></a>
<h3 class="section">16.19 Instruction Attributes</h3>
<a name="index-insn-attributes"></a>
<a name="index-instruction-attributes"></a>

<p>In addition to describing the instruction supported by the target machine,
the <samp>md</samp> file also defines a group of <em>attributes</em> and a set of
values for each.  Every generated insn is assigned a value for each attribute.
One possible attribute would be the effect that the insn has on the machine&rsquo;s
condition code.  This attribute can then be used by <code>NOTICE_UPDATE_CC</code>
to track the condition codes.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Defining-Attributes" accesskey="1">Defining Attributes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Specifying attributes and their values.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Expressions" accesskey="2">Expressions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Valid expressions for attribute values.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tagging-Insns" accesskey="3">Tagging Insns</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Assigning attribute values to insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Attr-Example" accesskey="4">Attr Example</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">An example of assigning attributes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Insn-Lengths" accesskey="5">Insn Lengths</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Computing the length of insns.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Constant-Attributes" accesskey="6">Constant Attributes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining attributes that are constant.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Delay-Slots" accesskey="7">Delay Slots</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining delay slots required for a machine.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Processor-pipeline-description" accesskey="8">Processor pipeline description</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Specifying information for insn scheduling.
</td></tr>
</table>

<hr>
<a name="Defining-Attributes"></a>
<div class="header">
<p>
Next: <a href="#Expressions" accesskey="n" rel="next">Expressions</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Attributes-and-their-Values"></a>
<h4 class="subsection">16.19.1 Defining Attributes and their Values</h4>
<a name="index-defining-attributes-and-their-values"></a>
<a name="index-attributes_002c-defining"></a>

<a name="index-define_005fattr"></a>
<p>The <code>define_attr</code> expression is used to define each attribute required
by the target machine.  It looks like:
</p>
<div class="smallexample">
<pre class="smallexample">(define_attr <var>name</var> <var>list-of-values</var> <var>default</var>)
</pre></div>

<p><var>name</var> is a string specifying the name of the attribute being defined.
Some attributes are used in a special way by the rest of the compiler. The
<code>enabled</code> attribute can be used to conditionally enable or disable
insn alternatives (see <a href="#Disable-Insn-Alternatives">Disable Insn Alternatives</a>). The <code>predicable</code>
attribute, together with a suitable <code>define_cond_exec</code>
(see <a href="#Conditional-Execution">Conditional Execution</a>), can be used to automatically generate
conditional variants of instruction patterns. The compiler internally uses
the names <code>ce_enabled</code> and <code>nonce_enabled</code>, so they should not be
used elsewhere as alternative names.
</p>
<p><var>list-of-values</var> is either a string that specifies a comma-separated
list of values that can be assigned to the attribute, or a null string to
indicate that the attribute takes numeric values.
</p>
<p><var>default</var> is an attribute expression that gives the value of this
attribute for insns that match patterns whose definition does not include
an explicit value for this attribute.  See <a href="#Attr-Example">Attr Example</a>, for more
information on the handling of defaults.  See <a href="#Constant-Attributes">Constant Attributes</a>,
for information on attributes that do not depend on any particular insn.
</p>
<a name="index-insn_002dattr_002eh"></a>
<p>For each defined attribute, a number of definitions are written to the
<samp>insn-attr.h</samp> file.  For cases where an explicit set of values is
specified for an attribute, the following are defined:
</p>
<ul>
<li> A &lsquo;<samp>#define</samp>&rsquo; is written for the symbol &lsquo;<samp>HAVE_ATTR_<var>name</var></samp>&rsquo;.

</li><li> An enumerated class is defined for &lsquo;<samp>attr_<var>name</var></samp>&rsquo; with
elements of the form &lsquo;<samp><var>upper-name</var>_<var>upper-value</var></samp>&rsquo; where
the attribute name and value are first converted to uppercase.

</li><li> A function &lsquo;<samp>get_attr_<var>name</var></samp>&rsquo; is defined that is passed an insn and
returns the attribute value for that insn.
</li></ul>

<p>For example, if the following is present in the <samp>md</samp> file:
</p>
<div class="smallexample">
<pre class="smallexample">(define_attr &quot;type&quot; &quot;branch,fp,load,store,arith&quot; &hellip;)
</pre></div>

<p>the following lines will be written to the file <samp>insn-attr.h</samp>.
</p>
<div class="smallexample">
<pre class="smallexample">#define HAVE_ATTR_type 1
enum attr_type {TYPE_BRANCH, TYPE_FP, TYPE_LOAD,
                 TYPE_STORE, TYPE_ARITH};
extern enum attr_type get_attr_type ();
</pre></div>

<p>If the attribute takes numeric values, no <code>enum</code> type will be
defined and the function to obtain the attribute&rsquo;s value will return
<code>int</code>.
</p>
<p>There are attributes which are tied to a specific meaning.  These
attributes are not free to use for other purposes:
</p>
<dl compact="compact">
<dt><code>length</code></dt>
<dd><p>The <code>length</code> attribute is used to calculate the length of emitted
code chunks.  This is especially important when verifying branch
distances. See <a href="#Insn-Lengths">Insn Lengths</a>.
</p>
</dd>
<dt><code>enabled</code></dt>
<dd><p>The <code>enabled</code> attribute can be defined to prevent certain
alternatives of an insn definition from being used during code
generation. See <a href="#Disable-Insn-Alternatives">Disable Insn Alternatives</a>.
</p></dd>
</dl>

<p>For each of these special attributes, the corresponding
&lsquo;<samp>HAVE_ATTR_<var>name</var></samp>&rsquo; &lsquo;<samp>#define</samp>&rsquo; is also written when the
attribute is not defined; in that case, it is defined as &lsquo;<samp>0</samp>&rsquo;.
</p>
<a name="index-define_005fenum_005fattr"></a>
<a name="define_005fenum_005fattr"></a><p>Another way of defining an attribute is to use:
</p>
<div class="smallexample">
<pre class="smallexample">(define_enum_attr &quot;<var>attr</var>&quot; &quot;<var>enum</var>&quot; <var>default</var>)
</pre></div>

<p>This works in just the same way as <code>define_attr</code>, except that
the list of values is taken from a separate enumeration called
<var>enum</var> (see <a href="#define_005fenum">define_enum</a>).  This form allows you to use
the same list of values for several attributes without having to
repeat the list each time.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_enum &quot;processor&quot; [
  model_a
  model_b
  &hellip;
])
(define_enum_attr &quot;arch&quot; &quot;processor&quot;
  (const (symbol_ref &quot;target_arch&quot;)))
(define_enum_attr &quot;tune&quot; &quot;processor&quot;
  (const (symbol_ref &quot;target_tune&quot;)))
</pre></div>

<p>defines the same attributes as:
</p>
<div class="smallexample">
<pre class="smallexample">(define_attr &quot;arch&quot; &quot;model_a,model_b,&hellip;&quot;
  (const (symbol_ref &quot;target_arch&quot;)))
(define_attr &quot;tune&quot; &quot;model_a,model_b,&hellip;&quot;
  (const (symbol_ref &quot;target_tune&quot;)))
</pre></div>

<p>but without duplicating the processor list.  The second example defines two
separate C enums (<code>attr_arch</code> and <code>attr_tune</code>) whereas the first
defines a single C enum (<code>processor</code>).
</p><hr>
<a name="Expressions"></a>
<div class="header">
<p>
Next: <a href="#Tagging-Insns" accesskey="n" rel="next">Tagging Insns</a>, Previous: <a href="#Defining-Attributes" accesskey="p" rel="prev">Defining Attributes</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Attribute-Expressions"></a>
<h4 class="subsection">16.19.2 Attribute Expressions</h4>
<a name="index-attribute-expressions"></a>

<p>RTL expressions used to define attributes use the codes described above
plus a few specific to attribute definitions, to be discussed below.
Attribute value expressions must have one of the following forms:
</p>
<dl compact="compact">
<dd><a name="index-const_005fint-and-attributes"></a>
</dd>
<dt><code>(const_int <var>i</var>)</code></dt>
<dd><p>The integer <var>i</var> specifies the value of a numeric attribute.  <var>i</var>
must be non-negative.
</p>
<p>The value of a numeric attribute can be specified either with a
<code>const_int</code>, or as an integer represented as a string in
<code>const_string</code>, <code>eq_attr</code> (see below), <code>attr</code>,
<code>symbol_ref</code>, simple arithmetic expressions, and <code>set_attr</code>
overrides on specific instructions (see <a href="#Tagging-Insns">Tagging Insns</a>).
</p>
<a name="index-const_005fstring-and-attributes"></a>
</dd>
<dt><code>(const_string <var>value</var>)</code></dt>
<dd><p>The string <var>value</var> specifies a constant attribute value.
If <var>value</var> is specified as &lsquo;<samp>&quot;*&quot;</samp>&rsquo;, it means that the default value of
the attribute is to be used for the insn containing this expression.
&lsquo;<samp>&quot;*&quot;</samp>&rsquo; obviously cannot be used in the <var>default</var> expression
of a <code>define_attr</code>.
</p>
<p>If the attribute whose value is being specified is numeric, <var>value</var>
must be a string containing a non-negative integer (normally
<code>const_int</code> would be used in this case).  Otherwise, it must
contain one of the valid values for the attribute.
</p>
<a name="index-if_005fthen_005felse-and-attributes"></a>
</dd>
<dt><code>(if_then_else <var>test</var> <var>true-value</var> <var>false-value</var>)</code></dt>
<dd><p><var>test</var> specifies an attribute test, whose format is defined below.
The value of this expression is <var>true-value</var> if <var>test</var> is true,
otherwise it is <var>false-value</var>.
</p>
<a name="index-cond-and-attributes"></a>
</dd>
<dt><code>(cond [<var>test1</var> <var>value1</var> &hellip;] <var>default</var>)</code></dt>
<dd><p>The first operand of this expression is a vector containing an even
number of expressions and consisting of pairs of <var>test</var> and <var>value</var>
expressions.  The value of the <code>cond</code> expression is that of the
<var>value</var> corresponding to the first true <var>test</var> expression.  If
none of the <var>test</var> expressions are true, the value of the <code>cond</code>
expression is that of the <var>default</var> expression.
</p></dd>
</dl>

<p><var>test</var> expressions can have one of the following forms:
</p>
<dl compact="compact">
<dd><a name="index-const_005fint-and-attribute-tests"></a>
</dd>
<dt><code>(const_int <var>i</var>)</code></dt>
<dd><p>This test is true if <var>i</var> is nonzero and false otherwise.
</p>
<a name="index-not-and-attributes"></a>
<a name="index-ior-and-attributes"></a>
<a name="index-and-and-attributes"></a>
</dd>
<dt><code>(not <var>test</var>)</code></dt>
<dt><code>(ior <var>test1</var> <var>test2</var>)</code></dt>
<dt><code>(and <var>test1</var> <var>test2</var>)</code></dt>
<dd><p>These tests are true if the indicated logical function is true.
</p>
<a name="index-match_005foperand-and-attributes"></a>
</dd>
<dt><code>(match_operand:<var>m</var> <var>n</var> <var>pred</var> <var>constraints</var>)</code></dt>
<dd><p>This test is true if operand <var>n</var> of the insn whose attribute value
is being determined has mode <var>m</var> (this part of the test is ignored
if <var>m</var> is <code>VOIDmode</code>) and the function specified by the string
<var>pred</var> returns a nonzero value when passed operand <var>n</var> and mode
<var>m</var> (this part of the test is ignored if <var>pred</var> is the null
string).
</p>
<p>The <var>constraints</var> operand is ignored and should be the null string.
</p>
<a name="index-match_005ftest-and-attributes"></a>
</dd>
<dt><code>(match_test <var>c-expr</var>)</code></dt>
<dd><p>The test is true if C expression <var>c-expr</var> is true.  In non-constant
attributes, <var>c-expr</var> has access to the following variables:
</p>
<dl compact="compact">
<dt><var>insn</var></dt>
<dd><p>The rtl instruction under test.
</p></dd>
<dt><var>which_alternative</var></dt>
<dd><p>The <code>define_insn</code> alternative that <var>insn</var> matches.
See <a href="#Output-Statement">Output Statement</a>.
</p></dd>
<dt><var>operands</var></dt>
<dd><p>An array of <var>insn</var>&rsquo;s rtl operands.
</p></dd>
</dl>

<p><var>c-expr</var> behaves like the condition in a C <code>if</code> statement,
so there is no need to explicitly convert the expression into a boolean
0 or 1 value.  For example, the following two tests are equivalent:
</p>
<div class="smallexample">
<pre class="smallexample">(match_test &quot;x &amp; 2&quot;)
(match_test &quot;(x &amp; 2) != 0&quot;)
</pre></div>

<a name="index-le-and-attributes"></a>
<a name="index-leu-and-attributes"></a>
<a name="index-lt-and-attributes"></a>
<a name="index-gt-and-attributes"></a>
<a name="index-gtu-and-attributes"></a>
<a name="index-ge-and-attributes"></a>
<a name="index-geu-and-attributes"></a>
<a name="index-ne-and-attributes"></a>
<a name="index-eq-and-attributes"></a>
<a name="index-plus-and-attributes"></a>
<a name="index-minus-and-attributes"></a>
<a name="index-mult-and-attributes"></a>
<a name="index-div-and-attributes"></a>
<a name="index-mod-and-attributes"></a>
<a name="index-abs-and-attributes"></a>
<a name="index-neg-and-attributes"></a>
<a name="index-ashift-and-attributes"></a>
<a name="index-lshiftrt-and-attributes"></a>
<a name="index-ashiftrt-and-attributes"></a>
</dd>
<dt><code>(le <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(leu <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(lt <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(ltu <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(gt <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(gtu <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(ge <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(geu <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(ne <var>arith1</var> <var>arith2</var>)</code></dt>
<dt><code>(eq <var>arith1</var> <var>arith2</var>)</code></dt>
<dd><p>These tests are true if the indicated comparison of the two arithmetic
expressions is true.  Arithmetic expressions are formed with
<code>plus</code>, <code>minus</code>, <code>mult</code>, <code>div</code>, <code>mod</code>,
<code>abs</code>, <code>neg</code>, <code>and</code>, <code>ior</code>, <code>xor</code>, <code>not</code>,
<code>ashift</code>, <code>lshiftrt</code>, and <code>ashiftrt</code> expressions.
</p>
<a name="index-get_005fattr"></a>
<p><code>const_int</code> and <code>symbol_ref</code> are always valid terms (see <a href="#Insn-Lengths">Insn Lengths</a>,for additional forms).  <code>symbol_ref</code> is a string
denoting a C expression that yields an <code>int</code> when evaluated by the
&lsquo;<samp>get_attr_&hellip;</samp>&rsquo; routine.  It should normally be a global
variable.
</p>
<a name="index-eq_005fattr"></a>
</dd>
<dt><code>(eq_attr <var>name</var> <var>value</var>)</code></dt>
<dd><p><var>name</var> is a string specifying the name of an attribute.
</p>
<p><var>value</var> is a string that is either a valid value for attribute
<var>name</var>, a comma-separated list of values, or &lsquo;<samp>!</samp>&rsquo; followed by a
value or list.  If <var>value</var> does not begin with a &lsquo;<samp>!</samp>&rsquo;, this
test is true if the value of the <var>name</var> attribute of the current
insn is in the list specified by <var>value</var>.  If <var>value</var> begins
with a &lsquo;<samp>!</samp>&rsquo;, this test is true if the attribute&rsquo;s value is
<em>not</em> in the specified list.
</p>
<p>For example,
</p>
<div class="smallexample">
<pre class="smallexample">(eq_attr &quot;type&quot; &quot;load,store&quot;)
</pre></div>

<p>is equivalent to
</p>
<div class="smallexample">
<pre class="smallexample">(ior (eq_attr &quot;type&quot; &quot;load&quot;) (eq_attr &quot;type&quot; &quot;store&quot;))
</pre></div>

<p>If <var>name</var> specifies an attribute of &lsquo;<samp>alternative</samp>&rsquo;, it refers to the
value of the compiler variable <code>which_alternative</code>
(see <a href="#Output-Statement">Output Statement</a>) and the values must be small integers.  For
example,
</p>
<div class="smallexample">
<pre class="smallexample">(eq_attr &quot;alternative&quot; &quot;2,3&quot;)
</pre></div>

<p>is equivalent to
</p>
<div class="smallexample">
<pre class="smallexample">(ior (eq (symbol_ref &quot;which_alternative&quot;) (const_int 2))
     (eq (symbol_ref &quot;which_alternative&quot;) (const_int 3)))
</pre></div>

<p>Note that, for most attributes, an <code>eq_attr</code> test is simplified in cases
where the value of the attribute being tested is known for all insns matching
a particular pattern.  This is by far the most common case.
</p>
<a name="index-attr_005fflag"></a>
</dd>
<dt><code>(attr_flag <var>name</var>)</code></dt>
<dd><p>The value of an <code>attr_flag</code> expression is true if the flag
specified by <var>name</var> is true for the <code>insn</code> currently being
scheduled.
</p>
<p><var>name</var> is a string specifying one of a fixed set of flags to test.
Test the flags <code>forward</code> and <code>backward</code> to determine the
direction of a conditional branch.
</p>
<p>This example describes a conditional branch delay slot which
can be nullified for forward branches that are taken (annul-true) or
for backward branches which are not taken (annul-false).
</p>
<div class="smallexample">
<pre class="smallexample">(define_delay (eq_attr &quot;type&quot; &quot;cbranch&quot;)
  [(eq_attr &quot;in_branch_delay&quot; &quot;true&quot;)
   (and (eq_attr &quot;in_branch_delay&quot; &quot;true&quot;)
        (attr_flag &quot;forward&quot;))
   (and (eq_attr &quot;in_branch_delay&quot; &quot;true&quot;)
        (attr_flag &quot;backward&quot;))])
</pre></div>

<p>The <code>forward</code> and <code>backward</code> flags are false if the current
<code>insn</code> being scheduled is not a conditional branch.
</p>
<p><code>attr_flag</code> is only used during delay slot scheduling and has no
meaning to other passes of the compiler.
</p>
<a name="index-attr"></a>
</dd>
<dt><code>(attr <var>name</var>)</code></dt>
<dd><p>The value of another attribute is returned.  This is most useful
for numeric attributes, as <code>eq_attr</code> and <code>attr_flag</code>
produce more efficient code for non-numeric attributes.
</p></dd>
</dl>

<hr>
<a name="Tagging-Insns"></a>
<div class="header">
<p>
Next: <a href="#Attr-Example" accesskey="n" rel="next">Attr Example</a>, Previous: <a href="#Expressions" accesskey="p" rel="prev">Expressions</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Assigning-Attribute-Values-to-Insns"></a>
<h4 class="subsection">16.19.3 Assigning Attribute Values to Insns</h4>
<a name="index-tagging-insns"></a>
<a name="index-assigning-attribute-values-to-insns"></a>

<p>The value assigned to an attribute of an insn is primarily determined by
which pattern is matched by that insn (or which <code>define_peephole</code>
generated it).  Every <code>define_insn</code> and <code>define_peephole</code> can
have an optional last argument to specify the values of attributes for
matching insns.  The value of any attribute not specified in a particular
insn is set to the default value for that attribute, as specified in its
<code>define_attr</code>.  Extensive use of default values for attributes
permits the specification of the values for only one or two attributes
in the definition of most insn patterns, as seen in the example in the
next section.
</p>
<p>The optional last argument of <code>define_insn</code> and
<code>define_peephole</code> is a vector of expressions, each of which defines
the value for a single attribute.  The most general way of assigning an
attribute&rsquo;s value is to use a <code>set</code> expression whose first operand is an
<code>attr</code> expression giving the name of the attribute being set.  The
second operand of the <code>set</code> is an attribute expression
(see <a href="#Expressions">Expressions</a>) giving the value of the attribute.
</p>
<p>When the attribute value depends on the &lsquo;<samp>alternative</samp>&rsquo; attribute
(i.e., which is the applicable alternative in the constraint of the
insn), the <code>set_attr_alternative</code> expression can be used.  It
allows the specification of a vector of attribute expressions, one for
each alternative.
</p>
<a name="index-set_005fattr"></a>
<p>When the generality of arbitrary attribute expressions is not required,
the simpler <code>set_attr</code> expression can be used, which allows
specifying a string giving either a single attribute value or a list
of attribute values, one for each alternative.
</p>
<p>The form of each of the above specifications is shown below.  In each case,
<var>name</var> is a string specifying the attribute to be set.
</p>
<dl compact="compact">
<dt><code>(set_attr <var>name</var> <var>value-string</var>)</code></dt>
<dd><p><var>value-string</var> is either a string giving the desired attribute value,
or a string containing a comma-separated list giving the values for
succeeding alternatives.  The number of elements must match the number
of alternatives in the constraint of the insn pattern.
</p>
<p>Note that it may be useful to specify &lsquo;<samp>*</samp>&rsquo; for some alternative, in
which case the attribute will assume its default value for insns matching
that alternative.
</p>
<a name="index-set_005fattr_005falternative"></a>
</dd>
<dt><code>(set_attr_alternative <var>name</var> [<var>value1</var> <var>value2</var> &hellip;])</code></dt>
<dd><p>Depending on the alternative of the insn, the value will be one of the
specified values.  This is a shorthand for using a <code>cond</code> with
tests on the &lsquo;<samp>alternative</samp>&rsquo; attribute.
</p>
<a name="index-attr-1"></a>
</dd>
<dt><code>(set (attr <var>name</var>) <var>value</var>)</code></dt>
<dd><p>The first operand of this <code>set</code> must be the special RTL expression
<code>attr</code>, whose sole operand is a string giving the name of the
attribute being set.  <var>value</var> is the value of the attribute.
</p></dd>
</dl>

<p>The following shows three different ways of representing the same
attribute value specification:
</p>
<div class="smallexample">
<pre class="smallexample">(set_attr &quot;type&quot; &quot;load,store,arith&quot;)

(set_attr_alternative &quot;type&quot;
                      [(const_string &quot;load&quot;) (const_string &quot;store&quot;)
                       (const_string &quot;arith&quot;)])

(set (attr &quot;type&quot;)
     (cond [(eq_attr &quot;alternative&quot; &quot;1&quot;) (const_string &quot;load&quot;)
            (eq_attr &quot;alternative&quot; &quot;2&quot;) (const_string &quot;store&quot;)]
           (const_string &quot;arith&quot;)))
</pre></div>

<a name="index-define_005fasm_005fattributes"></a>
<p>The <code>define_asm_attributes</code> expression provides a mechanism to
specify the attributes assigned to insns produced from an <code>asm</code>
statement.  It has the form:
</p>
<div class="smallexample">
<pre class="smallexample">(define_asm_attributes [<var>attr-sets</var>])
</pre></div>

<p>where <var>attr-sets</var> is specified the same as for both the
<code>define_insn</code> and the <code>define_peephole</code> expressions.
</p>
<p>These values will typically be the &ldquo;worst case&rdquo; attribute values.  For
example, they might indicate that the condition code will be clobbered.
</p>
<p>A specification for a <code>length</code> attribute is handled specially.  The
way to compute the length of an <code>asm</code> insn is to multiply the
length specified in the expression <code>define_asm_attributes</code> by the
number of machine instructions specified in the <code>asm</code> statement,
determined by counting the number of semicolons and newlines in the
string.  Therefore, the value of the <code>length</code> attribute specified
in a <code>define_asm_attributes</code> should be the maximum possible length
of a single machine instruction.
</p>
<hr>
<a name="Attr-Example"></a>
<div class="header">
<p>
Next: <a href="#Insn-Lengths" accesskey="n" rel="next">Insn Lengths</a>, Previous: <a href="#Tagging-Insns" accesskey="p" rel="prev">Tagging Insns</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Example-of-Attribute-Specifications"></a>
<h4 class="subsection">16.19.4 Example of Attribute Specifications</h4>
<a name="index-attribute-specifications-example"></a>
<a name="index-attribute-specifications"></a>

<p>The judicious use of defaulting is important in the efficient use of
insn attributes.  Typically, insns are divided into <em>types</em> and an
attribute, customarily called <code>type</code>, is used to represent this
value.  This attribute is normally used only to define the default value
for other attributes.  An example will clarify this usage.
</p>
<p>Assume we have a RISC machine with a condition code and in which only
full-word operations are performed in registers.  Let us assume that we
can divide all insns into loads, stores, (integer) arithmetic
operations, floating point operations, and branches.
</p>
<p>Here we will concern ourselves with determining the effect of an insn on
the condition code and will limit ourselves to the following possible
effects:  The condition code can be set unpredictably (clobbered), not
be changed, be set to agree with the results of the operation, or only
changed if the item previously set into the condition code has been
modified.
</p>
<p>Here is part of a sample <samp>md</samp> file for such a machine:
</p>
<div class="smallexample">
<pre class="smallexample">(define_attr &quot;type&quot; &quot;load,store,arith,fp,branch&quot; (const_string &quot;arith&quot;))

(define_attr &quot;cc&quot; &quot;clobber,unchanged,set,change0&quot;
             (cond [(eq_attr &quot;type&quot; &quot;load&quot;)
                        (const_string &quot;change0&quot;)
                    (eq_attr &quot;type&quot; &quot;store,branch&quot;)
                        (const_string &quot;unchanged&quot;)
                    (eq_attr &quot;type&quot; &quot;arith&quot;)
                        (if_then_else (match_operand:SI 0 &quot;&quot; &quot;&quot;)
                                      (const_string &quot;set&quot;)
                                      (const_string &quot;clobber&quot;))]
                   (const_string &quot;clobber&quot;)))

(define_insn &quot;&quot;
  [(set (match_operand:SI 0 &quot;general_operand&quot; &quot;=r,r,m&quot;)
        (match_operand:SI 1 &quot;general_operand&quot; &quot;r,m,r&quot;))]
  &quot;&quot;
  &quot;@
   move %0,%1
   load %0,%1
   store %0,%1&quot;
  [(set_attr &quot;type&quot; &quot;arith,load,store&quot;)])
</pre></div>

<p>Note that we assume in the above example that arithmetic operations
performed on quantities smaller than a machine word clobber the condition
code since they will set the condition code to a value corresponding to the
full-word result.
</p>
<hr>
<a name="Insn-Lengths"></a>
<div class="header">
<p>
Next: <a href="#Constant-Attributes" accesskey="n" rel="next">Constant Attributes</a>, Previous: <a href="#Attr-Example" accesskey="p" rel="prev">Attr Example</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Computing-the-Length-of-an-Insn"></a>
<h4 class="subsection">16.19.5 Computing the Length of an Insn</h4>
<a name="index-insn-lengths_002c-computing"></a>
<a name="index-computing-the-length-of-an-insn"></a>

<p>For many machines, multiple types of branch instructions are provided, each
for different length branch displacements.  In most cases, the assembler
will choose the correct instruction to use.  However, when the assembler
cannot do so, GCC can when a special attribute, the <code>length</code>
attribute, is defined.  This attribute must be defined to have numeric
values by specifying a null string in its <code>define_attr</code>.
</p>
<p>In the case of the <code>length</code> attribute, two additional forms of
arithmetic terms are allowed in test expressions:
</p>
<dl compact="compact">
<dd><a name="index-match_005fdup-and-attributes"></a>
</dd>
<dt><code>(match_dup <var>n</var>)</code></dt>
<dd><p>This refers to the address of operand <var>n</var> of the current insn, which
must be a <code>label_ref</code>.
</p>
<a name="index-pc-and-attributes"></a>
</dd>
<dt><code>(pc)</code></dt>
<dd><p>For non-branch instructions and backward branch instructions, this refers
to the address of the current insn.  But for forward branch instructions,
this refers to the address of the next insn, because the length of the
current insn is to be computed.
</p></dd>
</dl>

<a name="index-addr_005fvec_002c-length-of"></a>
<a name="index-addr_005fdiff_005fvec_002c-length-of"></a>
<p>For normal insns, the length will be determined by value of the
<code>length</code> attribute.  In the case of <code>addr_vec</code> and
<code>addr_diff_vec</code> insn patterns, the length is computed as
the number of vectors multiplied by the size of each vector.
</p>
<p>Lengths are measured in addressable storage units (bytes).
</p>
<p>The following macros can be used to refine the length computation:
</p>
<dl compact="compact">
<dd><a name="index-ADJUST_005fINSN_005fLENGTH"></a>
</dd>
<dt><code>ADJUST_INSN_LENGTH (<var>insn</var>, <var>length</var>)</code></dt>
<dd><p>If defined, modifies the length assigned to instruction <var>insn</var> as a
function of the context in which it is used.  <var>length</var> is an lvalue
that contains the initially computed length of the insn and should be
updated with the correct length of the insn.
</p>
<p>This macro will normally not be required.  A case in which it is
required is the ROMP.  On this machine, the size of an <code>addr_vec</code>
insn must be increased by two to compensate for the fact that alignment
may be required.
</p></dd>
</dl>

<a name="index-get_005fattr_005flength"></a>
<p>The routine that returns <code>get_attr_length</code> (the value of the
<code>length</code> attribute) can be used by the output routine to
determine the form of the branch instruction to be written, as the
example below illustrates.
</p>
<p>As an example of the specification of variable-length branches, consider
the IBM 360.  If we adopt the convention that a register will be set to
the starting address of a function, we can jump to labels within 4k of
the start using a four-byte instruction.  Otherwise, we need a six-byte
sequence to load the address from memory and then branch to it.
</p>
<p>On such a machine, a pattern for a branch instruction might be specified
as follows:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;jump&quot;
  [(set (pc)
        (label_ref (match_operand 0 &quot;&quot; &quot;&quot;)))]
  &quot;&quot;
{
   return (get_attr_length (insn) == 4
           ? &quot;b %l0&quot; : &quot;l r15,=a(%l0); br r15&quot;);
}
  [(set (attr &quot;length&quot;)
        (if_then_else (lt (match_dup 0) (const_int 4096))
                      (const_int 4)
                      (const_int 6)))])
</pre></div>

<hr>
<a name="Constant-Attributes"></a>
<div class="header">
<p>
Next: <a href="#Delay-Slots" accesskey="n" rel="next">Delay Slots</a>, Previous: <a href="#Insn-Lengths" accesskey="p" rel="prev">Insn Lengths</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constant-Attributes-1"></a>
<h4 class="subsection">16.19.6 Constant Attributes</h4>
<a name="index-constant-attributes"></a>

<p>A special form of <code>define_attr</code>, where the expression for the
default value is a <code>const</code> expression, indicates an attribute that
is constant for a given run of the compiler.  Constant attributes may be
used to specify which variety of processor is used.  For example,
</p>
<div class="smallexample">
<pre class="smallexample">(define_attr &quot;cpu&quot; &quot;m88100,m88110,m88000&quot;
 (const
  (cond [(symbol_ref &quot;TARGET_88100&quot;) (const_string &quot;m88100&quot;)
         (symbol_ref &quot;TARGET_88110&quot;) (const_string &quot;m88110&quot;)]
        (const_string &quot;m88000&quot;))))

(define_attr &quot;memory&quot; &quot;fast,slow&quot;
 (const
  (if_then_else (symbol_ref &quot;TARGET_FAST_MEM&quot;)
                (const_string &quot;fast&quot;)
                (const_string &quot;slow&quot;))))
</pre></div>

<p>The routine generated for constant attributes has no parameters as it
does not depend on any particular insn.  RTL expressions used to define
the value of a constant attribute may use the <code>symbol_ref</code> form,
but may not use either the <code>match_operand</code> form or <code>eq_attr</code>
forms involving insn attributes.
</p>
<hr>
<a name="Delay-Slots"></a>
<div class="header">
<p>
Next: <a href="#Processor-pipeline-description" accesskey="n" rel="next">Processor pipeline description</a>, Previous: <a href="#Constant-Attributes" accesskey="p" rel="prev">Constant Attributes</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Delay-Slot-Scheduling"></a>
<h4 class="subsection">16.19.7 Delay Slot Scheduling</h4>
<a name="index-delay-slots_002c-defining"></a>

<p>The insn attribute mechanism can be used to specify the requirements for
delay slots, if any, on a target machine.  An instruction is said to
require a <em>delay slot</em> if some instructions that are physically
after the instruction are executed as if they were located before it.
Classic examples are branch and call instructions, which often execute
the following instruction before the branch or call is performed.
</p>
<p>On some machines, conditional branch instructions can optionally
<em>annul</em> instructions in the delay slot.  This means that the
instruction will not be executed for certain branch outcomes.  Both
instructions that annul if the branch is true and instructions that
annul if the branch is false are supported.
</p>
<p>Delay slot scheduling differs from instruction scheduling in that
determining whether an instruction needs a delay slot is dependent only
on the type of instruction being generated, not on data flow between the
instructions.  See the next section for a discussion of data-dependent
instruction scheduling.
</p>
<a name="index-define_005fdelay"></a>
<p>The requirement of an insn needing one or more delay slots is indicated
via the <code>define_delay</code> expression.  It has the following form:
</p>
<div class="smallexample">
<pre class="smallexample">(define_delay <var>test</var>
              [<var>delay-1</var> <var>annul-true-1</var> <var>annul-false-1</var>
               <var>delay-2</var> <var>annul-true-2</var> <var>annul-false-2</var>
               &hellip;])
</pre></div>

<p><var>test</var> is an attribute test that indicates whether this
<code>define_delay</code> applies to a particular insn.  If so, the number of
required delay slots is determined by the length of the vector specified
as the second argument.  An insn placed in delay slot <var>n</var> must
satisfy attribute test <var>delay-n</var>.  <var>annul-true-n</var> is an
attribute test that specifies which insns may be annulled if the branch
is true.  Similarly, <var>annul-false-n</var> specifies which insns in the
delay slot may be annulled if the branch is false.  If annulling is not
supported for that delay slot, <code>(nil)</code> should be coded.
</p>
<p>For example, in the common case where branch and call insns require
a single delay slot, which may contain any insn other than a branch or
call, the following would be placed in the <samp>md</samp> file:
</p>
<div class="smallexample">
<pre class="smallexample">(define_delay (eq_attr &quot;type&quot; &quot;branch,call&quot;)
              [(eq_attr &quot;type&quot; &quot;!branch,call&quot;) (nil) (nil)])
</pre></div>

<p>Multiple <code>define_delay</code> expressions may be specified.  In this
case, each such expression specifies different delay slot requirements
and there must be no insn for which tests in two <code>define_delay</code>
expressions are both true.
</p>
<p>For example, if we have a machine that requires one delay slot for branches
but two for calls,  no delay slot can contain a branch or call insn,
and any valid insn in the delay slot for the branch can be annulled if the
branch is true, we might represent this as follows:
</p>
<div class="smallexample">
<pre class="smallexample">(define_delay (eq_attr &quot;type&quot; &quot;branch&quot;)
   [(eq_attr &quot;type&quot; &quot;!branch,call&quot;)
    (eq_attr &quot;type&quot; &quot;!branch,call&quot;)
    (nil)])

(define_delay (eq_attr &quot;type&quot; &quot;call&quot;)
              [(eq_attr &quot;type&quot; &quot;!branch,call&quot;) (nil) (nil)
               (eq_attr &quot;type&quot; &quot;!branch,call&quot;) (nil) (nil)])
</pre></div>

<hr>
<a name="Processor-pipeline-description"></a>
<div class="header">
<p>
Previous: <a href="#Delay-Slots" accesskey="p" rel="prev">Delay Slots</a>, Up: <a href="#Insn-Attributes" accesskey="u" rel="up">Insn Attributes</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Specifying-processor-pipeline-description"></a>
<h4 class="subsection">16.19.8 Specifying processor pipeline description</h4>
<a name="index-processor-pipeline-description"></a>
<a name="index-processor-functional-units"></a>
<a name="index-instruction-latency-time"></a>
<a name="index-interlock-delays"></a>
<a name="index-data-dependence-delays"></a>
<a name="index-reservation-delays"></a>
<a name="index-pipeline-hazard-recognizer"></a>
<a name="index-automaton-based-pipeline-description"></a>
<a name="index-regular-expressions"></a>
<a name="index-deterministic-finite-state-automaton"></a>
<a name="index-automaton-based-scheduler"></a>
<a name="index-RISC"></a>
<a name="index-VLIW"></a>

<p>To achieve better performance, most modern processors
(super-pipelined, superscalar <acronym>RISC</acronym>, and <acronym>VLIW</acronym>
processors) have many <em>functional units</em> on which several
instructions can be executed simultaneously.  An instruction starts
execution if its issue conditions are satisfied.  If not, the
instruction is stalled until its conditions are satisfied.  Such
<em>interlock (pipeline) delay</em> causes interruption of the fetching
of successor instructions (or demands nop instructions, e.g. for some
MIPS processors).
</p>
<p>There are two major kinds of interlock delays in modern processors.
The first one is a data dependence delay determining <em>instruction
latency time</em>.  The instruction execution is not started until all
source data have been evaluated by prior instructions (there are more
complex cases when the instruction execution starts even when the data
are not available but will be ready in given time after the
instruction execution start).  Taking the data dependence delays into
account is simple.  The data dependence (true, output, and
anti-dependence) delay between two instructions is given by a
constant.  In most cases this approach is adequate.  The second kind
of interlock delays is a reservation delay.  The reservation delay
means that two instructions under execution will be in need of shared
processors resources, i.e. buses, internal registers, and/or
functional units, which are reserved for some time.  Taking this kind
of delay into account is complex especially for modern <acronym>RISC</acronym>
processors.
</p>
<p>The task of exploiting more processor parallelism is solved by an
instruction scheduler.  For a better solution to this problem, the
instruction scheduler has to have an adequate description of the
processor parallelism (or <em>pipeline description</em>).  GCC
machine descriptions describe processor parallelism and functional
unit reservations for groups of instructions with the aid of
<em>regular expressions</em>.
</p>
<p>The GCC instruction scheduler uses a <em>pipeline hazard recognizer</em> to
figure out the possibility of the instruction issue by the processor
on a given simulated processor cycle.  The pipeline hazard recognizer is
automatically generated from the processor pipeline description.  The
pipeline hazard recognizer generated from the machine description
is based on a deterministic finite state automaton (<acronym>DFA</acronym>):
the instruction issue is possible if there is a transition from one
automaton state to another one.  This algorithm is very fast, and
furthermore, its speed is not dependent on processor
complexity<a name="DOCF5" href="#FOOT5"><sup>5</sup></a>.
</p>
<a name="index-automaton-based-pipeline-description-1"></a>
<p>The rest of this section describes the directives that constitute
an automaton-based processor pipeline description.  The order of
these constructions within the machine description file is not
important.
</p>
<a name="index-define_005fautomaton"></a>
<a name="index-pipeline-hazard-recognizer-1"></a>
<p>The following optional construction describes names of automata
generated and used for the pipeline hazards recognition.  Sometimes
the generated finite state automaton used by the pipeline hazard
recognizer is large.  If we use more than one automaton and bind functional
units to the automata, the total size of the automata is usually
less than the size of the single automaton.  If there is no one such
construction, only one finite state automaton is generated.
</p>
<div class="smallexample">
<pre class="smallexample">(define_automaton <var>automata-names</var>)
</pre></div>

<p><var>automata-names</var> is a string giving names of the automata.  The
names are separated by commas.  All the automata should have unique names.
The automaton name is used in the constructions <code>define_cpu_unit</code> and
<code>define_query_cpu_unit</code>.
</p>
<a name="index-define_005fcpu_005funit"></a>
<a name="index-processor-functional-units-1"></a>
<p>Each processor functional unit used in the description of instruction
reservations should be described by the following construction.
</p>
<div class="smallexample">
<pre class="smallexample">(define_cpu_unit <var>unit-names</var> [<var>automaton-name</var>])
</pre></div>

<p><var>unit-names</var> is a string giving the names of the functional units
separated by commas.  Don&rsquo;t use name &lsquo;<samp>nothing</samp>&rsquo;, it is reserved
for other goals.
</p>
<p><var>automaton-name</var> is a string giving the name of the automaton with
which the unit is bound.  The automaton should be described in
construction <code>define_automaton</code>.  You should give
<em>automaton-name</em>, if there is a defined automaton.
</p>
<p>The assignment of units to automata are constrained by the uses of the
units in insn reservations.  The most important constraint is: if a
unit reservation is present on a particular cycle of an alternative
for an insn reservation, then some unit from the same automaton must
be present on the same cycle for the other alternatives of the insn
reservation.  The rest of the constraints are mentioned in the
description of the subsequent constructions.
</p>
<a name="index-define_005fquery_005fcpu_005funit"></a>
<a name="index-querying-function-unit-reservations"></a>
<p>The following construction describes CPU functional units analogously
to <code>define_cpu_unit</code>.  The reservation of such units can be
queried for an automaton state.  The instruction scheduler never
queries reservation of functional units for given automaton state.  So
as a rule, you don&rsquo;t need this construction.  This construction could
be used for future code generation goals (e.g. to generate
<acronym>VLIW</acronym> insn templates).
</p>
<div class="smallexample">
<pre class="smallexample">(define_query_cpu_unit <var>unit-names</var> [<var>automaton-name</var>])
</pre></div>

<p><var>unit-names</var> is a string giving names of the functional units
separated by commas.
</p>
<p><var>automaton-name</var> is a string giving the name of the automaton with
which the unit is bound.
</p>
<a name="index-define_005finsn_005freservation"></a>
<a name="index-instruction-latency-time-1"></a>
<a name="index-regular-expressions-1"></a>
<a name="index-data-bypass"></a>
<p>The following construction is the major one to describe pipeline
characteristics of an instruction.
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn_reservation <var>insn-name</var> <var>default_latency</var>
                         <var>condition</var> <var>regexp</var>)
</pre></div>

<p><var>default_latency</var> is a number giving latency time of the
instruction.  There is an important difference between the old
description and the automaton based pipeline description.  The latency
time is used for all dependencies when we use the old description.  In
the automaton based pipeline description, the given latency time is only
used for true dependencies.  The cost of anti-dependencies is always
zero and the cost of output dependencies is the difference between
latency times of the producing and consuming insns (if the difference
is negative, the cost is considered to be zero).  You can always
change the default costs for any description by using the target hook
<code>TARGET_SCHED_ADJUST_COST</code> (see <a href="#Scheduling">Scheduling</a>).
</p>
<p><var>insn-name</var> is a string giving the internal name of the insn.  The
internal names are used in constructions <code>define_bypass</code> and in
the automaton description file generated for debugging.  The internal
name has nothing in common with the names in <code>define_insn</code>.  It is a
good practice to use insn classes described in the processor manual.
</p>
<p><var>condition</var> defines what RTL insns are described by this
construction.  You should remember that you will be in trouble if
<var>condition</var> for two or more different
<code>define_insn_reservation</code> constructions is TRUE for an insn.  In
this case what reservation will be used for the insn is not defined.
Such cases are not checked during generation of the pipeline hazards
recognizer because in general recognizing that two conditions may have
the same value is quite difficult (especially if the conditions
contain <code>symbol_ref</code>).  It is also not checked during the
pipeline hazard recognizer work because it would slow down the
recognizer considerably.
</p>
<p><var>regexp</var> is a string describing the reservation of the cpu&rsquo;s functional
units by the instruction.  The reservations are described by a regular
expression according to the following syntax:
</p>
<div class="smallexample">
<pre class="smallexample">       regexp = regexp &quot;,&quot; oneof
              | oneof

       oneof = oneof &quot;|&quot; allof
             | allof

       allof = allof &quot;+&quot; repeat
             | repeat

       repeat = element &quot;*&quot; number
              | element

       element = cpu_function_unit_name
               | reservation_name
               | result_name
               | &quot;nothing&quot;
               | &quot;(&quot; regexp &quot;)&quot;
</pre></div>

<ul>
<li> &lsquo;<samp>,</samp>&rsquo; is used for describing the start of the next cycle in
the reservation.

</li><li> &lsquo;<samp>|</samp>&rsquo; is used for describing a reservation described by the first
regular expression <strong>or</strong> a reservation described by the second
regular expression <strong>or</strong> etc.

</li><li> &lsquo;<samp>+</samp>&rsquo; is used for describing a reservation described by the first
regular expression <strong>and</strong> a reservation described by the
second regular expression <strong>and</strong> etc.

</li><li> &lsquo;<samp>*</samp>&rsquo; is used for convenience and simply means a sequence in which
the regular expression are repeated <var>number</var> times with cycle
advancing (see &lsquo;<samp>,</samp>&rsquo;).

</li><li> &lsquo;<samp>cpu_function_unit_name</samp>&rsquo; denotes reservation of the named
functional unit.

</li><li> &lsquo;<samp>reservation_name</samp>&rsquo; &mdash; see description of construction
&lsquo;<samp>define_reservation</samp>&rsquo;.

</li><li> &lsquo;<samp>nothing</samp>&rsquo; denotes no unit reservations.
</li></ul>

<a name="index-define_005freservation"></a>
<p>Sometimes unit reservations for different insns contain common parts.
In such case, you can simplify the pipeline description by describing
the common part by the following construction
</p>
<div class="smallexample">
<pre class="smallexample">(define_reservation <var>reservation-name</var> <var>regexp</var>)
</pre></div>

<p><var>reservation-name</var> is a string giving name of <var>regexp</var>.
Functional unit names and reservation names are in the same name
space.  So the reservation names should be different from the
functional unit names and can not be the reserved name &lsquo;<samp>nothing</samp>&rsquo;.
</p>
<a name="index-define_005fbypass"></a>
<a name="index-instruction-latency-time-2"></a>
<a name="index-data-bypass-1"></a>
<p>The following construction is used to describe exceptions in the
latency time for given instruction pair.  This is so called bypasses.
</p>
<div class="smallexample">
<pre class="smallexample">(define_bypass <var>number</var> <var>out_insn_names</var> <var>in_insn_names</var>
               [<var>guard</var>])
</pre></div>

<p><var>number</var> defines when the result generated by the instructions
given in string <var>out_insn_names</var> will be ready for the
instructions given in string <var>in_insn_names</var>.  Each of these
strings is a comma-separated list of filename-style globs and
they refer to the names of <code>define_insn_reservation</code>s.
For example:
</p><div class="smallexample">
<pre class="smallexample">(define_bypass 1 &quot;cpu1_load_*, cpu1_store_*&quot; &quot;cpu1_load_*&quot;)
</pre></div>
<p>defines a bypass between instructions that start with
&lsquo;<samp>cpu1_load_</samp>&rsquo; or &lsquo;<samp>cpu1_store_</samp>&rsquo; and those that start with
&lsquo;<samp>cpu1_load_</samp>&rsquo;.
</p>
<p><var>guard</var> is an optional string giving the name of a C function which
defines an additional guard for the bypass.  The function will get the
two insns as parameters.  If the function returns zero the bypass will
be ignored for this case.  The additional guard is necessary to
recognize complicated bypasses, e.g. when the consumer is only an address
of insn &lsquo;<samp>store</samp>&rsquo; (not a stored value).
</p>
<p>If there are more one bypass with the same output and input insns, the
chosen bypass is the first bypass with a guard in description whose
guard function returns nonzero.  If there is no such bypass, then
bypass without the guard function is chosen.
</p>
<a name="index-exclusion_005fset"></a>
<a name="index-presence_005fset"></a>
<a name="index-final_005fpresence_005fset"></a>
<a name="index-absence_005fset"></a>
<a name="index-final_005fabsence_005fset"></a>
<a name="index-VLIW-1"></a>
<a name="index-RISC-1"></a>
<p>The following five constructions are usually used to describe
<acronym>VLIW</acronym> processors, or more precisely, to describe a placement
of small instructions into <acronym>VLIW</acronym> instruction slots.  They
can be used for <acronym>RISC</acronym> processors, too.
</p>
<div class="smallexample">
<pre class="smallexample">(exclusion_set <var>unit-names</var> <var>unit-names</var>)
(presence_set <var>unit-names</var> <var>patterns</var>)
(final_presence_set <var>unit-names</var> <var>patterns</var>)
(absence_set <var>unit-names</var> <var>patterns</var>)
(final_absence_set <var>unit-names</var> <var>patterns</var>)
</pre></div>

<p><var>unit-names</var> is a string giving names of functional units
separated by commas.
</p>
<p><var>patterns</var> is a string giving patterns of functional units
separated by comma.  Currently pattern is one unit or units
separated by white-spaces.
</p>
<p>The first construction (&lsquo;<samp>exclusion_set</samp>&rsquo;) means that each
functional unit in the first string can not be reserved simultaneously
with a unit whose name is in the second string and vice versa.  For
example, the construction is useful for describing processors
(e.g. some SPARC processors) with a fully pipelined floating point
functional unit which can execute simultaneously only single floating
point insns or only double floating point insns.
</p>
<p>The second construction (&lsquo;<samp>presence_set</samp>&rsquo;) means that each
functional unit in the first string can not be reserved unless at
least one of pattern of units whose names are in the second string is
reserved.  This is an asymmetric relation.  For example, it is useful
for description that <acronym>VLIW</acronym> &lsquo;<samp>slot1</samp>&rsquo; is reserved after
&lsquo;<samp>slot0</samp>&rsquo; reservation.  We could describe it by the following
construction
</p>
<div class="smallexample">
<pre class="smallexample">(presence_set &quot;slot1&quot; &quot;slot0&quot;)
</pre></div>

<p>Or &lsquo;<samp>slot1</samp>&rsquo; is reserved only after &lsquo;<samp>slot0</samp>&rsquo; and unit &lsquo;<samp>b0</samp>&rsquo;
reservation.  In this case we could write
</p>
<div class="smallexample">
<pre class="smallexample">(presence_set &quot;slot1&quot; &quot;slot0 b0&quot;)
</pre></div>

<p>The third construction (&lsquo;<samp>final_presence_set</samp>&rsquo;) is analogous to
&lsquo;<samp>presence_set</samp>&rsquo;.  The difference between them is when checking is
done.  When an instruction is issued in given automaton state
reflecting all current and planned unit reservations, the automaton
state is changed.  The first state is a source state, the second one
is a result state.  Checking for &lsquo;<samp>presence_set</samp>&rsquo; is done on the
source state reservation, checking for &lsquo;<samp>final_presence_set</samp>&rsquo; is
done on the result reservation.  This construction is useful to
describe a reservation which is actually two subsequent reservations.
For example, if we use
</p>
<div class="smallexample">
<pre class="smallexample">(presence_set &quot;slot1&quot; &quot;slot0&quot;)
</pre></div>

<p>the following insn will be never issued (because &lsquo;<samp>slot1</samp>&rsquo; requires
&lsquo;<samp>slot0</samp>&rsquo; which is absent in the source state).
</p>
<div class="smallexample">
<pre class="smallexample">(define_reservation &quot;insn_and_nop&quot; &quot;slot0 + slot1&quot;)
</pre></div>

<p>but it can be issued if we use analogous &lsquo;<samp>final_presence_set</samp>&rsquo;.
</p>
<p>The forth construction (&lsquo;<samp>absence_set</samp>&rsquo;) means that each functional
unit in the first string can be reserved only if each pattern of units
whose names are in the second string is not reserved.  This is an
asymmetric relation (actually &lsquo;<samp>exclusion_set</samp>&rsquo; is analogous to
this one but it is symmetric).  For example it might be useful in a
<acronym>VLIW</acronym> description to say that &lsquo;<samp>slot0</samp>&rsquo; cannot be reserved
after either &lsquo;<samp>slot1</samp>&rsquo; or &lsquo;<samp>slot2</samp>&rsquo; have been reserved.  This
can be described as:
</p>
<div class="smallexample">
<pre class="smallexample">(absence_set &quot;slot0&quot; &quot;slot1, slot2&quot;)
</pre></div>

<p>Or &lsquo;<samp>slot2</samp>&rsquo; can not be reserved if &lsquo;<samp>slot0</samp>&rsquo; and unit &lsquo;<samp>b0</samp>&rsquo;
are reserved or &lsquo;<samp>slot1</samp>&rsquo; and unit &lsquo;<samp>b1</samp>&rsquo; are reserved.  In
this case we could write
</p>
<div class="smallexample">
<pre class="smallexample">(absence_set &quot;slot2&quot; &quot;slot0 b0, slot1 b1&quot;)
</pre></div>

<p>All functional units mentioned in a set should belong to the same
automaton.
</p>
<p>The last construction (&lsquo;<samp>final_absence_set</samp>&rsquo;) is analogous to
&lsquo;<samp>absence_set</samp>&rsquo; but checking is done on the result (state)
reservation.  See comments for &lsquo;<samp>final_presence_set</samp>&rsquo;.
</p>
<a name="index-automata_005foption"></a>
<a name="index-deterministic-finite-state-automaton-1"></a>
<a name="index-nondeterministic-finite-state-automaton"></a>
<a name="index-finite-state-automaton-minimization"></a>
<p>You can control the generator of the pipeline hazard recognizer with
the following construction.
</p>
<div class="smallexample">
<pre class="smallexample">(automata_option <var>options</var>)
</pre></div>

<p><var>options</var> is a string giving options which affect the generated
code.  Currently there are the following options:
</p>
<ul>
<li> <em>no-minimization</em> makes no minimization of the automaton.  This is
only worth to do when we are debugging the description and need to
look more accurately at reservations of states.

</li><li> <em>time</em> means printing time statistics about the generation of
automata.

</li><li> <em>stats</em> means printing statistics about the generated automata
such as the number of DFA states, NDFA states and arcs.

</li><li> <em>v</em> means a generation of the file describing the result automata.
The file has suffix &lsquo;<samp>.dfa</samp>&rsquo; and can be used for the description
verification and debugging.

</li><li> <em>w</em> means a generation of warning instead of error for
non-critical errors.

</li><li> <em>no-comb-vect</em> prevents the automaton generator from generating
two data structures and comparing them for space efficiency.  Using
a comb vector to represent transitions may be better, but it can be
very expensive to construct.  This option is useful if the build
process spends an unacceptably long time in genautomata.

</li><li> <em>ndfa</em> makes nondeterministic finite state automata.  This affects
the treatment of operator &lsquo;<samp>|</samp>&rsquo; in the regular expressions.  The
usual treatment of the operator is to try the first alternative and,
if the reservation is not possible, the second alternative.  The
nondeterministic treatment means trying all alternatives, some of them
may be rejected by reservations in the subsequent insns.

</li><li> <em>collapse-ndfa</em> modifies the behaviour of the generator when
producing an automaton.  An additional state transition to collapse a
nondeterministic <acronym>NDFA</acronym> state to a deterministic <acronym>DFA</acronym>
state is generated.  It can be triggered by passing <code>const0_rtx</code> to
state_transition.  In such an automaton, cycle advance transitions are
available only for these collapsed states.  This option is useful for
ports that want to use the <code>ndfa</code> option, but also want to use
<code>define_query_cpu_unit</code> to assign units to insns issued in a cycle.

</li><li> <em>progress</em> means output of a progress bar showing how many states
were generated so far for automaton being processed.  This is useful
during debugging a <acronym>DFA</acronym> description.  If you see too many
generated states, you could interrupt the generator of the pipeline
hazard recognizer and try to figure out a reason for generation of the
huge automaton.
</li></ul>

<p>As an example, consider a superscalar <acronym>RISC</acronym> machine which can
issue three insns (two integer insns and one floating point insn) on
the cycle but can finish only two insns.  To describe this, we define
the following functional units.
</p>
<div class="smallexample">
<pre class="smallexample">(define_cpu_unit &quot;i0_pipeline, i1_pipeline, f_pipeline&quot;)
(define_cpu_unit &quot;port0, port1&quot;)
</pre></div>

<p>All simple integer insns can be executed in any integer pipeline and
their result is ready in two cycles.  The simple integer insns are
issued into the first pipeline unless it is reserved, otherwise they
are issued into the second pipeline.  Integer division and
multiplication insns can be executed only in the second integer
pipeline and their results are ready correspondingly in 8 and 4
cycles.  The integer division is not pipelined, i.e. the subsequent
integer division insn can not be issued until the current division
insn finished.  Floating point insns are fully pipelined and their
results are ready in 3 cycles.  Where the result of a floating point
insn is used by an integer insn, an additional delay of one cycle is
incurred.  To describe all of this we could specify
</p>
<div class="smallexample">
<pre class="smallexample">(define_cpu_unit &quot;div&quot;)

(define_insn_reservation &quot;simple&quot; 2 (eq_attr &quot;type&quot; &quot;int&quot;)
                         &quot;(i0_pipeline | i1_pipeline), (port0 | port1)&quot;)

(define_insn_reservation &quot;mult&quot; 4 (eq_attr &quot;type&quot; &quot;mult&quot;)
                         &quot;i1_pipeline, nothing*2, (port0 | port1)&quot;)

(define_insn_reservation &quot;div&quot; 8 (eq_attr &quot;type&quot; &quot;div&quot;)
                         &quot;i1_pipeline, div*7, div + (port0 | port1)&quot;)

(define_insn_reservation &quot;float&quot; 3 (eq_attr &quot;type&quot; &quot;float&quot;)
                         &quot;f_pipeline, nothing, (port0 | port1))

(define_bypass 4 &quot;float&quot; &quot;simple,mult,div&quot;)
</pre></div>

<p>To simplify the description we could describe the following reservation
</p>
<div class="smallexample">
<pre class="smallexample">(define_reservation &quot;finish&quot; &quot;port0|port1&quot;)
</pre></div>

<p>and use it in all <code>define_insn_reservation</code> as in the following
construction
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn_reservation &quot;simple&quot; 2 (eq_attr &quot;type&quot; &quot;int&quot;)
                         &quot;(i0_pipeline | i1_pipeline), finish&quot;)
</pre></div>


<hr>
<a name="Conditional-Execution"></a>
<div class="header">
<p>
Next: <a href="#Define-Subst" accesskey="n" rel="next">Define Subst</a>, Previous: <a href="#Insn-Attributes" accesskey="p" rel="prev">Insn Attributes</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Conditional-Execution-1"></a>
<h3 class="section">16.20 Conditional Execution</h3>
<a name="index-conditional-execution"></a>
<a name="index-predication"></a>

<p>A number of architectures provide for some form of conditional
execution, or predication.  The hallmark of this feature is the
ability to nullify most of the instructions in the instruction set.
When the instruction set is large and not entirely symmetric, it
can be quite tedious to describe these forms directly in the
<samp>.md</samp> file.  An alternative is the <code>define_cond_exec</code> template.
</p>
<a name="index-define_005fcond_005fexec"></a>
<div class="smallexample">
<pre class="smallexample">(define_cond_exec
  [<var>predicate-pattern</var>]
  &quot;<var>condition</var>&quot;
  &quot;<var>output-template</var>&quot;)
</pre></div>

<p><var>predicate-pattern</var> is the condition that must be true for the
insn to be executed at runtime and should match a relational operator.
One can use <code>match_operator</code> to match several relational operators
at once.  Any <code>match_operand</code> operands must have no more than one
alternative.
</p>
<p><var>condition</var> is a C expression that must be true for the generated
pattern to match.
</p>
<a name="index-current_005finsn_005fpredicate"></a>
<p><var>output-template</var> is a string similar to the <code>define_insn</code>
output template (see <a href="#Output-Template">Output Template</a>), except that the &lsquo;<samp>*</samp>&rsquo;
and &lsquo;<samp>@</samp>&rsquo; special cases do not apply.  This is only useful if the
assembly text for the predicate is a simple prefix to the main insn.
In order to handle the general case, there is a global variable
<code>current_insn_predicate</code> that will contain the entire predicate
if the current insn is predicated, and will otherwise be <code>NULL</code>.
</p>
<p>When <code>define_cond_exec</code> is used, an implicit reference to
the <code>predicable</code> instruction attribute is made.
See <a href="#Insn-Attributes">Insn Attributes</a>.  This attribute must be a boolean (i.e. have
exactly two elements in its <var>list-of-values</var>), with the possible
values being <code>no</code> and <code>yes</code>.  The default and all uses in
the insns must be a simple constant, not a complex expressions.  It
may, however, depend on the alternative, by using a comma-separated
list of values.  If that is the case, the port should also define an
<code>enabled</code> attribute (see <a href="#Disable-Insn-Alternatives">Disable Insn Alternatives</a>), which
should also allow only <code>no</code> and <code>yes</code> as its values.
</p>
<p>For each <code>define_insn</code> for which the <code>predicable</code>
attribute is true, a new <code>define_insn</code> pattern will be
generated that matches a predicated version of the instruction.
For example,
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;addsi&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;r&quot;)
        (plus:SI (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
                 (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;)))]
  &quot;<var>test1</var>&quot;
  &quot;add %2,%1,%0&quot;)

(define_cond_exec
  [(ne (match_operand:CC 0 &quot;register_operand&quot; &quot;c&quot;)
       (const_int 0))]
  &quot;<var>test2</var>&quot;
  &quot;(%0)&quot;)
</pre></div>

<p>generates a new pattern
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(cond_exec
     (ne (match_operand:CC 3 &quot;register_operand&quot; &quot;c&quot;) (const_int 0))
     (set (match_operand:SI 0 &quot;register_operand&quot; &quot;r&quot;)
          (plus:SI (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
                   (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;))))]
  &quot;(<var>test2</var>) &amp;&amp; (<var>test1</var>)&quot;
  &quot;(%3) add %2,%1,%0&quot;)
</pre></div>

<hr>
<a name="Define-Subst"></a>
<div class="header">
<p>
Next: <a href="#Constant-Definitions" accesskey="n" rel="next">Constant Definitions</a>, Previous: <a href="#Conditional-Execution" accesskey="p" rel="prev">Conditional Execution</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="RTL-Templates-Transformations"></a>
<h3 class="section">16.21 RTL Templates Transformations</h3>
<a name="index-define_005fsubst"></a>

<p>For some hardware architectures there are common cases when the RTL
templates for the instructions can be derived from the other RTL
templates using simple transformations.  E.g., <samp>i386.md</samp> contains
an RTL template for the ordinary <code>sub</code> instruction&mdash;
<code>*subsi_1</code>, and for the <code>sub</code> instruction with subsequent
zero-extension&mdash;<code>*subsi_1_zext</code>.  Such cases can be easily
implemented by a single meta-template capable of generating a modified
case based on the initial one:
</p>
<a name="index-define_005fsubst-4"></a>
<div class="smallexample">
<pre class="smallexample">(define_subst &quot;<var>name</var>&quot;
  [<var>input-template</var>]
  &quot;<var>condition</var>&quot;
  [<var>output-template</var>])
</pre></div>
<p><var>input-template</var> is a pattern describing the source RTL template,
which will be transformed.
</p>
<p><var>condition</var> is a C expression that is conjunct with the condition
from the input-template to generate a condition to be used in the
output-template.
</p>
<p><var>output-template</var> is a pattern that will be used in the resulting
template.
</p>
<p><code>define_subst</code> mechanism is tightly coupled with the notion of the
subst attribute (see <a href="#Subst-Iterators">Subst Iterators</a>).  The use of
<code>define_subst</code> is triggered by a reference to a subst attribute in
the transforming RTL template.  This reference initiates duplication of
the source RTL template and substitution of the attributes with their
values.  The source RTL template is left unchanged, while the copy is
transformed by <code>define_subst</code>.  This transformation can fail in the
case when the source RTL template is not matched against the
input-template of the <code>define_subst</code>.  In such case the copy is
deleted.
</p>
<p><code>define_subst</code> can be used only in <code>define_insn</code> and
<code>define_expand</code>, it cannot be used in other expressions (e.g. in
<code>define_insn_and_split</code>).
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Define-Subst-Example" accesskey="1">Define Subst Example</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Example of <code>define_subst</code> work.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Define-Subst-Pattern-Matching" accesskey="2">Define Subst Pattern Matching</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Process of template comparison.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Define-Subst-Output-Template" accesskey="3">Define Subst Output Template</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generation of output template.
</td></tr>
</table>

<hr>
<a name="Define-Subst-Example"></a>
<div class="header">
<p>
Next: <a href="#Define-Subst-Pattern-Matching" accesskey="n" rel="next">Define Subst Pattern Matching</a>, Up: <a href="#Define-Subst" accesskey="u" rel="up">Define Subst</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="define_005fsubst-Example"></a>
<h4 class="subsection">16.21.1 <code>define_subst</code> Example</h4>
<a name="index-define_005fsubst-1"></a>

<p>To illustrate how <code>define_subst</code> works, let us examine a simple
template transformation.
</p>
<p>Suppose there are two kinds of instructions: one that touches flags and
the other that does not.  The instructions of the second type could be
generated with the following <code>define_subst</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(define_subst &quot;add_clobber_subst&quot;
  [(set (match_operand:SI 0 &quot;&quot; &quot;&quot;)
        (match_operand:SI 1 &quot;&quot; &quot;&quot;))]
  &quot;&quot;
  [(set (match_dup 0)
        (match_dup 1))
   (clobber (reg:CC FLAGS_REG))]
</pre></div>

<p>This <code>define_subst</code> can be applied to any RTL pattern containing
<code>set</code> of mode SI and generates a copy with clobber when it is
applied.
</p>
<p>Assume there is an RTL template for a <code>max</code> instruction to be used
in <code>define_subst</code> mentioned above:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;maxsi&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=r&quot;)
        (max:SI
          (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
          (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;)))]
  &quot;&quot;
  &quot;max\t{%2, %1, %0|%0, %1, %2}&quot;
 [&hellip;])
</pre></div>

<p>To mark the RTL template for <code>define_subst</code> application,
subst-attributes are used.  They should be declared in advance:
</p>
<div class="smallexample">
<pre class="smallexample">(define_subst_attr &quot;add_clobber_name&quot; &quot;add_clobber_subst&quot; &quot;_noclobber&quot; &quot;_clobber&quot;)
</pre></div>

<p>Here &lsquo;<samp>add_clobber_name</samp>&rsquo; is the attribute name,
&lsquo;<samp>add_clobber_subst</samp>&rsquo; is the name of the corresponding
<code>define_subst</code>, the third argument (&lsquo;<samp>_noclobber</samp>&rsquo;) is the
attribute value that would be substituted into the unchanged version of
the source RTL template, and the last argument (&lsquo;<samp>_clobber</samp>&rsquo;) is the
value that would be substituted into the second, transformed,
version of the RTL template.
</p>
<p>Once the subst-attribute has been defined, it should be used in RTL
templates which need to be processed by the <code>define_subst</code>.  So,
the original RTL template should be changed:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;maxsi&lt;add_clobber_name&gt;&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=r&quot;)
        (max:SI
          (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
          (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;)))]
  &quot;&quot;
  &quot;max\t{%2, %1, %0|%0, %1, %2}&quot;
 [&hellip;])
</pre></div>

<p>The result of the <code>define_subst</code> usage would look like the following:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;maxsi_noclobber&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=r&quot;)
        (max:SI
          (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
          (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;)))]
  &quot;&quot;
  &quot;max\t{%2, %1, %0|%0, %1, %2}&quot;
 [&hellip;])
(define_insn &quot;maxsi_clobber&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=r&quot;)
        (max:SI
          (match_operand:SI 1 &quot;register_operand&quot; &quot;r&quot;)
          (match_operand:SI 2 &quot;register_operand&quot; &quot;r&quot;)))
   (clobber (reg:CC FLAGS_REG))]
  &quot;&quot;
  &quot;max\t{%2, %1, %0|%0, %1, %2}&quot;
 [&hellip;])
</pre></div>

<hr>
<a name="Define-Subst-Pattern-Matching"></a>
<div class="header">
<p>
Next: <a href="#Define-Subst-Output-Template" accesskey="n" rel="next">Define Subst Output Template</a>, Previous: <a href="#Define-Subst-Example" accesskey="p" rel="prev">Define Subst Example</a>, Up: <a href="#Define-Subst" accesskey="u" rel="up">Define Subst</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Pattern-Matching-in-define_005fsubst"></a>
<h4 class="subsection">16.21.2 Pattern Matching in <code>define_subst</code></h4>
<a name="index-define_005fsubst-2"></a>

<p>All expressions, allowed in <code>define_insn</code> or <code>define_expand</code>,
are allowed in the input-template of <code>define_subst</code>, except
<code>match_par_dup</code>, <code>match_scratch</code>, <code>match_parallel</code>. The
meanings of expressions in the input-template were changed:
</p>
<p><code>match_operand</code> matches any expression (possibly, a subtree in
RTL-template), if modes of the <code>match_operand</code> and this expression
are the same, or mode of the <code>match_operand</code> is <code>VOIDmode</code>, or
this expression is <code>match_dup</code>, <code>match_op_dup</code>.  If the
expression is <code>match_operand</code> too, and predicate of
<code>match_operand</code> from the input pattern is not empty, then the
predicates are compared.  That can be used for more accurate filtering
of accepted RTL-templates.
</p>
<p><code>match_operator</code> matches common operators (like <code>plus</code>,
<code>minus</code>), <code>unspec</code>, <code>unspec_volatile</code> operators and
<code>match_operator</code>s from the original pattern if the modes match and
<code>match_operator</code> from the input pattern has the same number of
operands as the operator from the original pattern.
</p>
<hr>
<a name="Define-Subst-Output-Template"></a>
<div class="header">
<p>
Previous: <a href="#Define-Subst-Pattern-Matching" accesskey="p" rel="prev">Define Subst Pattern Matching</a>, Up: <a href="#Define-Subst" accesskey="u" rel="up">Define Subst</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Generation-of-output-template-in-define_005fsubst"></a>
<h4 class="subsection">16.21.3 Generation of output template in <code>define_subst</code></h4>
<a name="index-define_005fsubst-3"></a>

<p>If all necessary checks for <code>define_subst</code> application pass, a new
RTL-pattern, based on the output-template, is created to replace the old
template.  Like in input-patterns, meanings of some RTL expressions are
changed when they are used in output-patterns of a <code>define_subst</code>.
Thus, <code>match_dup</code> is used for copying the whole expression from the
original pattern, which matched corresponding <code>match_operand</code> from
the input pattern.
</p>
<p><code>match_dup N</code> is used in the output template to be replaced with
the expression from the original pattern, which matched
<code>match_operand N</code> from the input pattern.  As a consequence,
<code>match_dup</code> cannot be used to point to <code>match_operand</code>s from
the output pattern, it should always refer to a <code>match_operand</code>
from the input pattern.
</p>
<p>In the output template one can refer to the expressions from the
original pattern and create new ones.  For instance, some operands could
be added by means of standard <code>match_operand</code>.
</p>
<p>After replacing <code>match_dup</code> with some RTL-subtree from the original
pattern, it could happen that several <code>match_operand</code>s in the
output pattern have the same indexes.  It is unknown, how many and what
indexes would be used in the expression which would replace
<code>match_dup</code>, so such conflicts in indexes are inevitable.  To
overcome this issue, <code>match_operands</code> and <code>match_operators</code>,
which were introduced into the output pattern, are renumerated when all
<code>match_dup</code>s are replaced.
</p>
<p>Number of alternatives in <code>match_operand</code>s introduced into the
output template <code>M</code> could differ from the number of alternatives in
the original pattern <code>N</code>, so in the resultant pattern there would
be <code>N*M</code> alternatives.  Thus, constraints from the original pattern
would be duplicated <code>N</code> times, constraints from the output pattern
would be duplicated <code>M</code> times, producing all possible combinations.
</p>
<hr>
<a name="Constant-Definitions"></a>
<div class="header">
<p>
Next: <a href="#Iterators" accesskey="n" rel="next">Iterators</a>, Previous: <a href="#Define-Subst" accesskey="p" rel="prev">Define Subst</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Constant-Definitions-1"></a>
<h3 class="section">16.22 Constant Definitions</h3>
<a name="index-constant-definitions"></a>
<a name="index-define_005fconstants"></a>

<p>Using literal constants inside instruction patterns reduces legibility and
can be a maintenance problem.
</p>
<p>To overcome this problem, you may use the <code>define_constants</code>
expression.  It contains a vector of name-value pairs.  From that
point on, wherever any of the names appears in the MD file, it is as
if the corresponding value had been written instead.  You may use
<code>define_constants</code> multiple times; each appearance adds more
constants to the table.  It is an error to redefine a constant with
a different value.
</p>
<p>To come back to the a29k load multiple example, instead of
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(match_parallel 0 &quot;load_multiple_operation&quot;
     [(set (match_operand:SI 1 &quot;gpc_reg_operand&quot; &quot;=r&quot;)
           (match_operand:SI 2 &quot;memory_operand&quot; &quot;m&quot;))
      (use (reg:SI 179))
      (clobber (reg:SI 179))])]
  &quot;&quot;
  &quot;loadm 0,0,%1,%2&quot;)
</pre></div>

<p>You could write:
</p>
<div class="smallexample">
<pre class="smallexample">(define_constants [
    (R_BP 177)
    (R_FC 178)
    (R_CR 179)
    (R_Q  180)
])

(define_insn &quot;&quot;
  [(match_parallel 0 &quot;load_multiple_operation&quot;
     [(set (match_operand:SI 1 &quot;gpc_reg_operand&quot; &quot;=r&quot;)
           (match_operand:SI 2 &quot;memory_operand&quot; &quot;m&quot;))
      (use (reg:SI R_CR))
      (clobber (reg:SI R_CR))])]
  &quot;&quot;
  &quot;loadm 0,0,%1,%2&quot;)
</pre></div>

<p>The constants that are defined with a define_constant are also output
in the insn-codes.h header file as #defines.
</p>
<a name="index-enumerations"></a>
<a name="index-define_005fc_005fenum"></a>
<p>You can also use the machine description file to define enumerations.
Like the constants defined by <code>define_constant</code>, these enumerations
are visible to both the machine description file and the main C code.
</p>
<p>The syntax is as follows:
</p>
<div class="smallexample">
<pre class="smallexample">(define_c_enum &quot;<var>name</var>&quot; [
  <var>value0</var>
  <var>value1</var>
  &hellip;
  <var>valuen</var>
])
</pre></div>

<p>This definition causes the equivalent of the following C code to appear
in <samp>insn-constants.h</samp>:
</p>
<div class="smallexample">
<pre class="smallexample">enum <var>name</var> {
  <var>value0</var> = 0,
  <var>value1</var> = 1,
  &hellip;
  <var>valuen</var> = <var>n</var>
};
#define NUM_<var>cname</var>_VALUES (<var>n</var> + 1)
</pre></div>

<p>where <var>cname</var> is the capitalized form of <var>name</var>.
It also makes each <var>valuei</var> available in the machine description
file, just as if it had been declared with:
</p>
<div class="smallexample">
<pre class="smallexample">(define_constants [(<var>valuei</var> <var>i</var>)])
</pre></div>

<p>Each <var>valuei</var> is usually an upper-case identifier and usually
begins with <var>cname</var>.
</p>
<p>You can split the enumeration definition into as many statements as
you like.  The above example is directly equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">(define_c_enum &quot;<var>name</var>&quot; [<var>value0</var>])
(define_c_enum &quot;<var>name</var>&quot; [<var>value1</var>])
&hellip;
(define_c_enum &quot;<var>name</var>&quot; [<var>valuen</var>])
</pre></div>

<p>Splitting the enumeration helps to improve the modularity of each
individual <code>.md</code> file.  For example, if a port defines its
synchronization instructions in a separate <samp>sync.md</samp> file,
it is convenient to define all synchronization-specific enumeration
values in <samp>sync.md</samp> rather than in the main <samp>.md</samp> file.
</p>
<p>Some enumeration names have special significance to GCC:
</p>
<dl compact="compact">
<dt><code>unspecv</code></dt>
<dd><a name="index-unspec_005fvolatile-1"></a>
<p>If an enumeration called <code>unspecv</code> is defined, GCC will use it
when printing out <code>unspec_volatile</code> expressions.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_c_enum &quot;unspecv&quot; [
  UNSPECV_BLOCKAGE
])
</pre></div>

<p>causes GCC to print &lsquo;<samp>(unspec_volatile &hellip; 0)</samp>&rsquo; as:
</p>
<div class="smallexample">
<pre class="smallexample">(unspec_volatile ... UNSPECV_BLOCKAGE)
</pre></div>

</dd>
<dt><code>unspec</code></dt>
<dd><a name="index-unspec-1"></a>
<p>If an enumeration called <code>unspec</code> is defined, GCC will use
it when printing out <code>unspec</code> expressions.  GCC will also use
it when printing out <code>unspec_volatile</code> expressions unless an
<code>unspecv</code> enumeration is also defined.  You can therefore
decide whether to keep separate enumerations for volatile and
non-volatile expressions or whether to use the same enumeration
for both.
</p></dd>
</dl>

<a name="index-define_005fenum"></a>
<a name="define_005fenum"></a><p>Another way of defining an enumeration is to use <code>define_enum</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(define_enum &quot;<var>name</var>&quot; [
  <var>value0</var>
  <var>value1</var>
  &hellip;
  <var>valuen</var>
])
</pre></div>

<p>This directive implies:
</p>
<div class="smallexample">
<pre class="smallexample">(define_c_enum &quot;<var>name</var>&quot; [
  <var>cname</var>_<var>cvalue0</var>
  <var>cname</var>_<var>cvalue1</var>
  &hellip;
  <var>cname</var>_<var>cvaluen</var>
])
</pre></div>

<a name="index-define_005fenum_005fattr-1"></a>
<p>where <var>cvaluei</var> is the capitalized form of <var>valuei</var>.
However, unlike <code>define_c_enum</code>, the enumerations defined
by <code>define_enum</code> can be used in attribute specifications
(see <a href="#define_005fenum_005fattr">define_enum_attr</a>).
</p><hr>
<a name="Iterators"></a>
<div class="header">
<p>
Previous: <a href="#Constant-Definitions" accesskey="p" rel="prev">Constant Definitions</a>, Up: <a href="#Machine-Desc" accesskey="u" rel="up">Machine Desc</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Iterators-1"></a>
<h3 class="section">16.23 Iterators</h3>
<a name="index-iterators-in-_002emd-files"></a>

<p>Ports often need to define similar patterns for more than one machine
mode or for more than one rtx code.  GCC provides some simple iterator
facilities to make this process easier.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Mode-Iterators" accesskey="1">Mode Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating variations of patterns for different modes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Code-Iterators" accesskey="2">Code Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Doing the same for codes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Int-Iterators" accesskey="3">Int Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Doing the same for integers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Subst-Iterators" accesskey="4">Subst Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Generating variations of patterns for define_subst.
</td></tr>
</table>

<hr>
<a name="Mode-Iterators"></a>
<div class="header">
<p>
Next: <a href="#Code-Iterators" accesskey="n" rel="next">Code Iterators</a>, Up: <a href="#Iterators" accesskey="u" rel="up">Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Mode-Iterators-1"></a>
<h4 class="subsection">16.23.1 Mode Iterators</h4>
<a name="index-mode-iterators-in-_002emd-files"></a>

<p>Ports often need to define similar patterns for two or more different modes.
For example:
</p>
<ul>
<li> If a processor has hardware support for both single and double
floating-point arithmetic, the <code>SFmode</code> patterns tend to be
very similar to the <code>DFmode</code> ones.

</li><li> If a port uses <code>SImode</code> pointers in one configuration and
<code>DImode</code> pointers in another, it will usually have very similar
<code>SImode</code> and <code>DImode</code> patterns for manipulating pointers.
</li></ul>

<p>Mode iterators allow several patterns to be instantiated from one
<samp>.md</samp> file template.  They can be used with any type of
rtx-based construct, such as a <code>define_insn</code>,
<code>define_split</code>, or <code>define_peephole2</code>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Defining-Mode-Iterators" accesskey="1">Defining Mode Iterators</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining a new mode iterator.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Substitutions" accesskey="2">Substitutions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Combining mode iterators with substitutions
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Examples" accesskey="3">Examples</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Examples
</td></tr>
</table>

<hr>
<a name="Defining-Mode-Iterators"></a>
<div class="header">
<p>
Next: <a href="#Substitutions" accesskey="n" rel="next">Substitutions</a>, Up: <a href="#Mode-Iterators" accesskey="u" rel="up">Mode Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-Mode-Iterators-1"></a>
<h4 class="subsubsection">16.23.1.1 Defining Mode Iterators</h4>
<a name="index-define_005fmode_005fiterator"></a>

<p>The syntax for defining a mode iterator is:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator <var>name</var> [(<var>mode1</var> &quot;<var>cond1</var>&quot;) &hellip; (<var>moden</var> &quot;<var>condn</var>&quot;)])
</pre></div>

<p>This allows subsequent <samp>.md</samp> file constructs to use the mode suffix
<code>:<var>name</var></code>.  Every construct that does so will be expanded
<var>n</var> times, once with every use of <code>:<var>name</var></code> replaced by
<code>:<var>mode1</var></code>, once with every use replaced by <code>:<var>mode2</var></code>,
and so on.  In the expansion for a particular <var>modei</var>, every
C condition will also require that <var>condi</var> be true.
</p>
<p>For example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator P [(SI &quot;Pmode == SImode&quot;) (DI &quot;Pmode == DImode&quot;)])
</pre></div>

<p>defines a new mode suffix <code>:P</code>.  Every construct that uses
<code>:P</code> will be expanded twice, once with every <code>:P</code> replaced
by <code>:SI</code> and once with every <code>:P</code> replaced by <code>:DI</code>.
The <code>:SI</code> version will only apply if <code>Pmode == SImode</code> and
the <code>:DI</code> version will only apply if <code>Pmode == DImode</code>.
</p>
<p>As with other <samp>.md</samp> conditions, an empty string is treated
as &ldquo;always true&rdquo;.  <code>(<var>mode</var> &quot;&quot;)</code> can also be abbreviated
to <code><var>mode</var></code>.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator GPR [SI (DI &quot;TARGET_64BIT&quot;)])
</pre></div>

<p>means that the <code>:DI</code> expansion only applies if <code>TARGET_64BIT</code>
but that the <code>:SI</code> expansion has no such constraint.
</p>
<p>Iterators are applied in the order they are defined.  This can be
significant if two iterators are used in a construct that requires
substitutions.  See <a href="#Substitutions">Substitutions</a>.
</p>
<hr>
<a name="Substitutions"></a>
<div class="header">
<p>
Next: <a href="#Examples" accesskey="n" rel="next">Examples</a>, Previous: <a href="#Defining-Mode-Iterators" accesskey="p" rel="prev">Defining Mode Iterators</a>, Up: <a href="#Mode-Iterators" accesskey="u" rel="up">Mode Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Substitution-in-Mode-Iterators"></a>
<h4 class="subsubsection">16.23.1.2 Substitution in Mode Iterators</h4>
<a name="index-define_005fmode_005fattr"></a>

<p>If an <samp>.md</samp> file construct uses mode iterators, each version of the
construct will often need slightly different strings or modes.  For
example:
</p>
<ul>
<li> When a <code>define_expand</code> defines several <code>add<var>m</var>3</code> patterns
(see <a href="#Standard-Names">Standard Names</a>), each expander will need to use the
appropriate mode name for <var>m</var>.

</li><li> When a <code>define_insn</code> defines several instruction patterns,
each instruction will often use a different assembler mnemonic.

</li><li> When a <code>define_insn</code> requires operands with different modes,
using an iterator for one of the operand modes usually requires a specific
mode for the other operand(s).
</li></ul>

<p>GCC supports such variations through a system of &ldquo;mode attributes&rdquo;.
There are two standard attributes: <code>mode</code>, which is the name of
the mode in lower case, and <code>MODE</code>, which is the same thing in
upper case.  You can define other attributes using:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_attr <var>name</var> [(<var>mode1</var> &quot;<var>value1</var>&quot;) &hellip; (<var>moden</var> &quot;<var>valuen</var>&quot;)])
</pre></div>

<p>where <var>name</var> is the name of the attribute and <var>valuei</var>
is the value associated with <var>modei</var>.
</p>
<p>When GCC replaces some <var>:iterator</var> with <var>:mode</var>, it will scan
each string and mode in the pattern for sequences of the form
<code>&lt;<var>iterator</var>:<var>attr</var>&gt;</code>, where <var>attr</var> is the name of a
mode attribute.  If the attribute is defined for <var>mode</var>, the whole
<code>&lt;&hellip;&gt;</code> sequence will be replaced by the appropriate attribute
value.
</p>
<p>For example, suppose an <samp>.md</samp> file has:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator P [(SI &quot;Pmode == SImode&quot;) (DI &quot;Pmode == DImode&quot;)])
(define_mode_attr load [(SI &quot;lw&quot;) (DI &quot;ld&quot;)])
</pre></div>

<p>If one of the patterns that uses <code>:P</code> contains the string
<code>&quot;&lt;P:load&gt;\t%0,%1&quot;</code>, the <code>SI</code> version of that pattern
will use <code>&quot;lw\t%0,%1&quot;</code> and the <code>DI</code> version will use
<code>&quot;ld\t%0,%1&quot;</code>.
</p>
<p>Here is an example of using an attribute for a mode:
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator LONG [SI DI])
(define_mode_attr SHORT [(SI &quot;HI&quot;) (DI &quot;SI&quot;)])
(define_insn &hellip;
  (sign_extend:LONG (match_operand:&lt;LONG:SHORT&gt; &hellip;)) &hellip;)
</pre></div>

<p>The <code><var>iterator</var>:</code> prefix may be omitted, in which case the
substitution will be attempted for every iterator expansion.
</p>
<hr>
<a name="Examples"></a>
<div class="header">
<p>
Previous: <a href="#Substitutions" accesskey="p" rel="prev">Substitutions</a>, Up: <a href="#Mode-Iterators" accesskey="u" rel="up">Mode Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Mode-Iterator-Examples"></a>
<h4 class="subsubsection">16.23.1.3 Mode Iterator Examples</h4>

<p>Here is an example from the MIPS port.  It defines the following
modes and attributes (among others):
</p>
<div class="smallexample">
<pre class="smallexample">(define_mode_iterator GPR [SI (DI &quot;TARGET_64BIT&quot;)])
(define_mode_attr d [(SI &quot;&quot;) (DI &quot;d&quot;)])
</pre></div>

<p>and uses the following template to define both <code>subsi3</code>
and <code>subdi3</code>:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;sub&lt;mode&gt;3&quot;
  [(set (match_operand:GPR 0 &quot;register_operand&quot; &quot;=d&quot;)
        (minus:GPR (match_operand:GPR 1 &quot;register_operand&quot; &quot;d&quot;)
                   (match_operand:GPR 2 &quot;register_operand&quot; &quot;d&quot;)))]
  &quot;&quot;
  &quot;&lt;d&gt;subu\t%0,%1,%2&quot;
  [(set_attr &quot;type&quot; &quot;arith&quot;)
   (set_attr &quot;mode&quot; &quot;&lt;MODE&gt;&quot;)])
</pre></div>

<p>This is exactly equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;subsi3&quot;
  [(set (match_operand:SI 0 &quot;register_operand&quot; &quot;=d&quot;)
        (minus:SI (match_operand:SI 1 &quot;register_operand&quot; &quot;d&quot;)
                  (match_operand:SI 2 &quot;register_operand&quot; &quot;d&quot;)))]
  &quot;&quot;
  &quot;subu\t%0,%1,%2&quot;
  [(set_attr &quot;type&quot; &quot;arith&quot;)
   (set_attr &quot;mode&quot; &quot;SI&quot;)])

(define_insn &quot;subdi3&quot;
  [(set (match_operand:DI 0 &quot;register_operand&quot; &quot;=d&quot;)
        (minus:DI (match_operand:DI 1 &quot;register_operand&quot; &quot;d&quot;)
                  (match_operand:DI 2 &quot;register_operand&quot; &quot;d&quot;)))]
  &quot;&quot;
  &quot;dsubu\t%0,%1,%2&quot;
  [(set_attr &quot;type&quot; &quot;arith&quot;)
   (set_attr &quot;mode&quot; &quot;DI&quot;)])
</pre></div>

<hr>
<a name="Code-Iterators"></a>
<div class="header">
<p>
Next: <a href="#Int-Iterators" accesskey="n" rel="next">Int Iterators</a>, Previous: <a href="#Mode-Iterators" accesskey="p" rel="prev">Mode Iterators</a>, Up: <a href="#Iterators" accesskey="u" rel="up">Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Code-Iterators-1"></a>
<h4 class="subsection">16.23.2 Code Iterators</h4>
<a name="index-code-iterators-in-_002emd-files"></a>
<a name="index-define_005fcode_005fiterator"></a>
<a name="index-define_005fcode_005fattr"></a>

<p>Code iterators operate in a similar way to mode iterators.  See <a href="#Mode-Iterators">Mode Iterators</a>.
</p>
<p>The construct:
</p>
<div class="smallexample">
<pre class="smallexample">(define_code_iterator <var>name</var> [(<var>code1</var> &quot;<var>cond1</var>&quot;) &hellip; (<var>coden</var> &quot;<var>condn</var>&quot;)])
</pre></div>

<p>defines a pseudo rtx code <var>name</var> that can be instantiated as
<var>codei</var> if condition <var>condi</var> is true.  Each <var>codei</var>
must have the same rtx format.  See <a href="#RTL-Classes">RTL Classes</a>.
</p>
<p>As with mode iterators, each pattern that uses <var>name</var> will be
expanded <var>n</var> times, once with all uses of <var>name</var> replaced by
<var>code1</var>, once with all uses replaced by <var>code2</var>, and so on.
See <a href="#Defining-Mode-Iterators">Defining Mode Iterators</a>.
</p>
<p>It is possible to define attributes for codes as well as for modes.
There are two standard code attributes: <code>code</code>, the name of the
code in lower case, and <code>CODE</code>, the name of the code in upper case.
Other attributes are defined using:
</p>
<div class="smallexample">
<pre class="smallexample">(define_code_attr <var>name</var> [(<var>code1</var> &quot;<var>value1</var>&quot;) &hellip; (<var>coden</var> &quot;<var>valuen</var>&quot;)])
</pre></div>

<p>Here&rsquo;s an example of code iterators in action, taken from the MIPS port:
</p>
<div class="smallexample">
<pre class="smallexample">(define_code_iterator any_cond [unordered ordered unlt unge uneq ltgt unle ungt
                                eq ne gt ge lt le gtu geu ltu leu])

(define_expand &quot;b&lt;code&gt;&quot;
  [(set (pc)
        (if_then_else (any_cond:CC (cc0)
                                   (const_int 0))
                      (label_ref (match_operand 0 &quot;&quot;))
                      (pc)))]
  &quot;&quot;
{
  gen_conditional_branch (operands, &lt;CODE&gt;);
  DONE;
})
</pre></div>

<p>This is equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">(define_expand &quot;bunordered&quot;
  [(set (pc)
        (if_then_else (unordered:CC (cc0)
                                    (const_int 0))
                      (label_ref (match_operand 0 &quot;&quot;))
                      (pc)))]
  &quot;&quot;
{
  gen_conditional_branch (operands, UNORDERED);
  DONE;
})

(define_expand &quot;bordered&quot;
  [(set (pc)
        (if_then_else (ordered:CC (cc0)
                                  (const_int 0))
                      (label_ref (match_operand 0 &quot;&quot;))
                      (pc)))]
  &quot;&quot;
{
  gen_conditional_branch (operands, ORDERED);
  DONE;
})

&hellip;
</pre></div>

<hr>
<a name="Int-Iterators"></a>
<div class="header">
<p>
Next: <a href="#Subst-Iterators" accesskey="n" rel="next">Subst Iterators</a>, Previous: <a href="#Code-Iterators" accesskey="p" rel="prev">Code Iterators</a>, Up: <a href="#Iterators" accesskey="u" rel="up">Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Int-Iterators-1"></a>
<h4 class="subsection">16.23.3 Int Iterators</h4>
<a name="index-int-iterators-in-_002emd-files"></a>
<a name="index-define_005fint_005fiterator"></a>
<a name="index-define_005fint_005fattr"></a>

<p>Int iterators operate in a similar way to code iterators.  See <a href="#Code-Iterators">Code Iterators</a>.
</p>
<p>The construct:
</p>
<div class="smallexample">
<pre class="smallexample">(define_int_iterator <var>name</var> [(<var>int1</var> &quot;<var>cond1</var>&quot;) &hellip; (<var>intn</var> &quot;<var>condn</var>&quot;)])
</pre></div>

<p>defines a pseudo integer constant <var>name</var> that can be instantiated as
<var>inti</var> if condition <var>condi</var> is true.  Each <var>int</var>
must have the same rtx format.  See <a href="#RTL-Classes">RTL Classes</a>. Int iterators can appear
in only those rtx fields that have &rsquo;i&rsquo; as the specifier. This means that
each <var>int</var> has to be a constant defined using define_constant or
define_c_enum.
</p>
<p>As with mode and code iterators, each pattern that uses <var>name</var> will be
expanded <var>n</var> times, once with all uses of <var>name</var> replaced by
<var>int1</var>, once with all uses replaced by <var>int2</var>, and so on.
See <a href="#Defining-Mode-Iterators">Defining Mode Iterators</a>.
</p>
<p>It is possible to define attributes for ints as well as for codes and modes.
Attributes are defined using:
</p>
<div class="smallexample">
<pre class="smallexample">(define_int_attr <var>name</var> [(<var>int1</var> &quot;<var>value1</var>&quot;) &hellip; (<var>intn</var> &quot;<var>valuen</var>&quot;)])
</pre></div>

<p>Here&rsquo;s an example of int iterators in action, taken from the ARM port:
</p>
<div class="smallexample">
<pre class="smallexample">(define_int_iterator QABSNEG [UNSPEC_VQABS UNSPEC_VQNEG])

(define_int_attr absneg [(UNSPEC_VQABS &quot;abs&quot;) (UNSPEC_VQNEG &quot;neg&quot;)])

(define_insn &quot;neon_vq&lt;absneg&gt;&lt;mode&gt;&quot;
  [(set (match_operand:VDQIW 0 &quot;s_register_operand&quot; &quot;=w&quot;)
	(unspec:VDQIW [(match_operand:VDQIW 1 &quot;s_register_operand&quot; &quot;w&quot;)
		       (match_operand:SI 2 &quot;immediate_operand&quot; &quot;i&quot;)]
		      QABSNEG))]
  &quot;TARGET_NEON&quot;
  &quot;vq&lt;absneg&gt;.&lt;V_s_elem&gt;\t%&lt;V_reg&gt;0, %&lt;V_reg&gt;1&quot;
  [(set_attr &quot;neon_type&quot; &quot;neon_vqneg_vqabs&quot;)]
)

</pre></div>

<p>This is equivalent to:
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;neon_vqabs&lt;mode&gt;&quot;
  [(set (match_operand:VDQIW 0 &quot;s_register_operand&quot; &quot;=w&quot;)
	(unspec:VDQIW [(match_operand:VDQIW 1 &quot;s_register_operand&quot; &quot;w&quot;)
		       (match_operand:SI 2 &quot;immediate_operand&quot; &quot;i&quot;)]
		      UNSPEC_VQABS))]
  &quot;TARGET_NEON&quot;
  &quot;vqabs.&lt;V_s_elem&gt;\t%&lt;V_reg&gt;0, %&lt;V_reg&gt;1&quot;
  [(set_attr &quot;neon_type&quot; &quot;neon_vqneg_vqabs&quot;)]
)

(define_insn &quot;neon_vqneg&lt;mode&gt;&quot;
  [(set (match_operand:VDQIW 0 &quot;s_register_operand&quot; &quot;=w&quot;)
	(unspec:VDQIW [(match_operand:VDQIW 1 &quot;s_register_operand&quot; &quot;w&quot;)
		       (match_operand:SI 2 &quot;immediate_operand&quot; &quot;i&quot;)]
		      UNSPEC_VQNEG))]
  &quot;TARGET_NEON&quot;
  &quot;vqneg.&lt;V_s_elem&gt;\t%&lt;V_reg&gt;0, %&lt;V_reg&gt;1&quot;
  [(set_attr &quot;neon_type&quot; &quot;neon_vqneg_vqabs&quot;)]
)

</pre></div>

<hr>
<a name="Subst-Iterators"></a>
<div class="header">
<p>
Previous: <a href="#Int-Iterators" accesskey="p" rel="prev">Int Iterators</a>, Up: <a href="#Iterators" accesskey="u" rel="up">Iterators</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Subst-Iterators-1"></a>
<h4 class="subsection">16.23.4 Subst Iterators</h4>
<a name="index-subst-iterators-in-_002emd-files"></a>
<a name="index-define_005fsubst-5"></a>
<a name="index-define_005fsubst_005fattr"></a>

<p>Subst iterators are special type of iterators with the following
restrictions: they could not be declared explicitly, they always have
only two values, and they do not have explicit dedicated name.
Subst-iterators are triggered only when corresponding subst-attribute is
used in RTL-pattern.
</p>
<p>Subst iterators transform templates in the following way: the templates
are duplicated, the subst-attributes in these templates are replaced
with the corresponding values, and a new attribute is implicitly added
to the given <code>define_insn</code>/<code>define_expand</code>.  The name of the
added attribute matches the name of <code>define_subst</code>.  Such
attributes are declared implicitly, and it is not allowed to have a
<code>define_attr</code> named as a <code>define_subst</code>.
</p>
<p>Each subst iterator is linked to a <code>define_subst</code>.  It is declared
implicitly by the first appearance of the corresponding
<code>define_subst_attr</code>, and it is not allowed to define it explicitly.
</p>
<p>Declarations of subst-attributes have the following syntax:
</p>
<a name="index-define_005fsubst_005fattr-1"></a>
<div class="smallexample">
<pre class="smallexample">(define_subst_attr &quot;<var>name</var>&quot;
  &quot;<var>subst-name</var>&quot;
  &quot;<var>no-subst-value</var>&quot;
  &quot;<var>subst-applied-value</var>&quot;)
</pre></div>

<p><var>name</var> is a string with which the given subst-attribute could be
referred to.
</p>
<p><var>subst-name</var> shows which <code>define_subst</code> should be applied to an
RTL-template if the given subst-attribute is present in the
RTL-template.
</p>
<p><var>no-subst-value</var> is a value with which subst-attribute would be
replaced in the first copy of the original RTL-template.
</p>
<p><var>subst-applied-value</var> is a value with which subst-attribute would be
replaced in the second copy of the original RTL-template.
</p>

<hr>
<a name="Target-Macros"></a>
<div class="header">
<p>
Next: <a href="#Host-Config" accesskey="n" rel="next">Host Config</a>, Previous: <a href="#Machine-Desc" accesskey="p" rel="prev">Machine Desc</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Target-Description-Macros-and-Functions"></a>
<h2 class="chapter">17 Target Description Macros and Functions</h2>
<a name="index-machine-description-macros"></a>
<a name="index-target-description-macros"></a>
<a name="index-macros_002c-target-description"></a>
<a name="index-tm_002eh-macros"></a>

<p>In addition to the file <samp><var>machine</var>.md</samp>, a machine description
includes a C header file conventionally given the name
<samp><var>machine</var>.h</samp> and a C source file named <samp><var>machine</var>.c</samp>.
The header file defines numerous macros that convey the information
about the target machine that does not fit into the scheme of the
<samp>.md</samp> file.  The file <samp>tm.h</samp> should be a link to
<samp><var>machine</var>.h</samp>.  The header file <samp>config.h</samp> includes
<samp>tm.h</samp> and most compiler source files include <samp>config.h</samp>.  The
source file defines a variable <code>targetm</code>, which is a structure
containing pointers to functions and data relating to the target
machine.  <samp><var>machine</var>.c</samp> should also contain their definitions,
if they are not defined elsewhere in GCC, and other functions called
through the macros defined in the <samp>.h</samp> file.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Target-Structure" accesskey="1">Target Structure</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The <code>targetm</code> variable.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Driver" accesskey="2">Driver</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Controlling how the driver runs the compilation passes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Run_002dtime-Target" accesskey="3">Run-time Target</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining &lsquo;<samp>-m</samp>&rsquo; options like <samp>-m68000</samp> and <samp>-m68020</samp>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Per_002dFunction-Data" accesskey="4">Per-Function Data</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining data structures for per-function information.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Storage-Layout" accesskey="5">Storage Layout</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining sizes and alignments of data.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Type-Layout" accesskey="6">Type Layout</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining sizes and properties of basic user data types.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Registers" accesskey="7">Registers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Naming and describing the hardware registers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Register-Classes" accesskey="8">Register Classes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining the classes of hardware registers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Old-Constraints" accesskey="9">Old Constraints</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The old way to define machine-specific constraints.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Stack-and-Calling">Stack and Calling</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining which way the stack grows and by how much.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Varargs">Varargs</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining the varargs macros.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Trampolines">Trampolines</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Code set up at run time to enter a nested function.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Library-Calls">Library Calls</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Controlling how library routines are implicitly called.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Addressing-Modes">Addressing Modes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining addressing modes valid for memory operands.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Anchored-Addresses">Anchored Addresses</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining how <samp>-fsection-anchors</samp> should work.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Condition-Code">Condition Code</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining how insns update the condition code.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Costs">Costs</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining relative costs of different operations.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Scheduling">Scheduling</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Adjusting the behavior of the instruction scheduler.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Sections">Sections</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Dividing storage into text, data, and other sections.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#PIC">PIC</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros for position independent code.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Assembler-Format">Assembler Format</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining how to write insns and pseudo-ops to output.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Debugging-Info">Debugging Info</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining the format of debugging output.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Floating-Point">Floating Point</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Handling floating point for cross-compilers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Mode-Switching">Mode Switching</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Insertion of mode-switching instructions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Target-Attributes">Target Attributes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Defining target-specific uses of <code>__attribute__</code>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Emulated-TLS">Emulated TLS</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Emulated TLS support.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#MIPS-Coprocessors">MIPS Coprocessors</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">MIPS coprocessor support and how to customize it.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#PCH-Target">PCH Target</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Validity checking for precompiled headers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#C_002b_002b-ABI">C++ ABI</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Controlling C++ ABI changes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Named-Address-Spaces">Named Address Spaces</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Adding support for named address spaces
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Misc">Misc</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Everything else.
</td></tr>
</table>

<hr>
<a name="Target-Structure"></a>
<div class="header">
<p>
Next: <a href="#Driver" accesskey="n" rel="next">Driver</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Global-targetm-Variable"></a>
<h3 class="section">17.1 The Global <code>targetm</code> Variable</h3>
<a name="index-target-hooks"></a>
<a name="index-target-functions"></a>

<dl>
<dt><a name="index-targetm"></a>Variable: <em>struct gcc_target</em> <strong>targetm</strong></dt>
<dd><p>The target <samp>.c</samp> file must define the global <code>targetm</code> variable
which contains pointers to functions and data relating to the target
machine.  The variable is declared in <samp>target.h</samp>;
<samp>target-def.h</samp> defines the macro <code>TARGET_INITIALIZER</code> which is
used to initialize the variable, and macros for the default initializers
for elements of the structure.  The <samp>.c</samp> file should override those
macros for which the default definition is inappropriate.  For example:
</p><div class="smallexample">
<pre class="smallexample">#include &quot;target.h&quot;
#include &quot;target-def.h&quot;

/* <span class="roman">Initialize the GCC target structure.</span>  */

#undef TARGET_COMP_TYPE_ATTRIBUTES
#define TARGET_COMP_TYPE_ATTRIBUTES <var>machine</var>_comp_type_attributes

struct gcc_target targetm = TARGET_INITIALIZER;
</pre></div>
</dd></dl>

<p>Where a macro should be defined in the <samp>.c</samp> file in this manner to
form part of the <code>targetm</code> structure, it is documented below as a
&ldquo;Target Hook&rdquo; with a prototype.  Many macros will change in future
from being defined in the <samp>.h</samp> file to being part of the
<code>targetm</code> structure.
</p>
<p>Similarly, there is a <code>targetcm</code> variable for hooks that are
specific to front ends for C-family languages, documented as &ldquo;C
Target Hook&rdquo;.  This is declared in <samp>c-family/c-target.h</samp>, the
initializer <code>TARGETCM_INITIALIZER</code> in
<samp>c-family/c-target-def.h</samp>.  If targets initialize <code>targetcm</code>
themselves, they should set <code>target_has_targetcm=yes</code> in
<samp>config.gcc</samp>; otherwise a default definition is used.
</p>
<p>Similarly, there is a <code>targetm_common</code> variable for hooks that
are shared between the compiler driver and the compilers proper,
documented as &ldquo;Common Target Hook&rdquo;.  This is declared in
<samp>common/common-target.h</samp>, the initializer
<code>TARGETM_COMMON_INITIALIZER</code> in
<samp>common/common-target-def.h</samp>.  If targets initialize
<code>targetm_common</code> themselves, they should set
<code>target_has_targetm_common=yes</code> in <samp>config.gcc</samp>; otherwise a
default definition is used.
</p>
<hr>
<a name="Driver"></a>
<div class="header">
<p>
Next: <a href="#Run_002dtime-Target" accesskey="n" rel="next">Run-time Target</a>, Previous: <a href="#Target-Structure" accesskey="p" rel="prev">Target Structure</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Controlling-the-Compilation-Driver_002c-gcc"></a>
<h3 class="section">17.2 Controlling the Compilation Driver, <samp>gcc</samp></h3>
<a name="index-driver"></a>
<a name="index-controlling-the-compilation-driver"></a>

<p>You can control the compilation driver.
</p>
<dl>
<dt><a name="index-DRIVER_005fSELF_005fSPECS"></a>Macro: <strong>DRIVER_SELF_SPECS</strong></dt>
<dd><p>A list of specs for the driver itself.  It should be a suitable
initializer for an array of strings, with no surrounding braces.
</p>
<p>The driver applies these specs to its own command line between loading
default <samp>specs</samp> files (but not command-line specified ones) and
choosing the multilib directory or running any subcommands.  It
applies them in the order given, so each spec can depend on the
options added by earlier ones.  It is also possible to remove options
using &lsquo;<samp>%&lt;<var>option</var></samp>&rsquo; in the usual way.
</p>
<p>This macro can be useful when a port has several interdependent target
options.  It provides a way of standardizing the command line so
that the other specs are easier to write.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-OPTION_005fDEFAULT_005fSPECS"></a>Macro: <strong>OPTION_DEFAULT_SPECS</strong></dt>
<dd><p>A list of specs used to support configure-time default options (i.e.
<samp>--with</samp> options) in the driver.  It should be a suitable initializer
for an array of structures, each containing two strings, without the
outermost pair of surrounding braces.
</p>
<p>The first item in the pair is the name of the default.  This must match
the code in <samp>config.gcc</samp> for the target.  The second item is a spec
to apply if a default with this name was specified.  The string
&lsquo;<samp>%(VALUE)</samp>&rsquo; in the spec will be replaced by the value of the default
everywhere it occurs.
</p>
<p>The driver will apply these specs to its own command line between loading
default <samp>specs</samp> files and processing <code>DRIVER_SELF_SPECS</code>, using
the same mechanism as <code>DRIVER_SELF_SPECS</code>.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-CPP_005fSPEC"></a>Macro: <strong>CPP_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program options to
pass to CPP.  It can also specify how to translate options you
give to GCC into options for GCC to pass to the CPP.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-CPLUSPLUS_005fCPP_005fSPEC"></a>Macro: <strong>CPLUSPLUS_CPP_SPEC</strong></dt>
<dd><p>This macro is just like <code>CPP_SPEC</code>, but is used for C++, rather
than C.  If you do not define this macro, then the value of
<code>CPP_SPEC</code> (if any) will be used instead.
</p></dd></dl>

<dl>
<dt><a name="index-CC1_005fSPEC"></a>Macro: <strong>CC1_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program options to
pass to <code>cc1</code>, <code>cc1plus</code>, <code>f771</code>, and the other language
front ends.
It can also specify how to translate options you give to GCC into options
for GCC to pass to front ends.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-CC1PLUS_005fSPEC"></a>Macro: <strong>CC1PLUS_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program options to
pass to <code>cc1plus</code>.  It can also specify how to translate options you
give to GCC into options for GCC to pass to the <code>cc1plus</code>.
</p>
<p>Do not define this macro if it does not need to do anything.
Note that everything defined in CC1_SPEC is already passed to
<code>cc1plus</code> so there is no need to duplicate the contents of
CC1_SPEC in CC1PLUS_SPEC.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fSPEC"></a>Macro: <strong>ASM_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program options to
pass to the assembler.  It can also specify how to translate options
you give to GCC into options for GCC to pass to the assembler.
See the file <samp>sun3.h</samp> for an example of this.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fFINAL_005fSPEC"></a>Macro: <strong>ASM_FINAL_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program how to
run any programs which cleanup after the normal assembler.
Normally, this is not needed.  See the file <samp>mips.h</samp> for
an example of this.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-AS_005fNEEDS_005fDASH_005fFOR_005fPIPED_005fINPUT"></a>Macro: <strong>AS_NEEDS_DASH_FOR_PIPED_INPUT</strong></dt>
<dd><p>Define this macro, with no value, if the driver should give the assembler
an argument consisting of a single dash, <samp>-</samp>, to instruct it to
read from its standard input (which will be a pipe connected to the
output of the compiler proper).  This argument is given after any
<samp>-o</samp> option specifying the name of the output file.
</p>
<p>If you do not define this macro, the assembler is assumed to read its
standard input if given no non-option arguments.  If your assembler
cannot read standard input at all, use a &lsquo;<samp>%{pipe:%e}</samp>&rsquo; construct;
see <samp>mips.h</samp> for instance.
</p></dd></dl>

<dl>
<dt><a name="index-LINK_005fSPEC"></a>Macro: <strong>LINK_SPEC</strong></dt>
<dd><p>A C string constant that tells the GCC driver program options to
pass to the linker.  It can also specify how to translate options you
give to GCC into options for GCC to pass to the linker.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-LIB_005fSPEC"></a>Macro: <strong>LIB_SPEC</strong></dt>
<dd><p>Another C string constant used much like <code>LINK_SPEC</code>.  The difference
between the two is that <code>LIB_SPEC</code> is used at the end of the
command given to the linker.
</p>
<p>If this macro is not defined, a default is provided that
loads the standard C library from the usual place.  See <samp>gcc.c</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC_005fSPEC"></a>Macro: <strong>LIBGCC_SPEC</strong></dt>
<dd><p>Another C string constant that tells the GCC driver program
how and when to place a reference to <samp>libgcc.a</samp> into the
linker command line.  This constant is placed both before and after
the value of <code>LIB_SPEC</code>.
</p>
<p>If this macro is not defined, the GCC driver provides a default that
passes the string <samp>-lgcc</samp> to the linker.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fLIBGCC_005fSPEC"></a>Macro: <strong>REAL_LIBGCC_SPEC</strong></dt>
<dd><p>By default, if <code>ENABLE_SHARED_LIBGCC</code> is defined, the
<code>LIBGCC_SPEC</code> is not directly used by the driver program but is
instead modified to refer to different versions of <samp>libgcc.a</samp>
depending on the values of the command line flags <samp>-static</samp>,
<samp>-shared</samp>, <samp>-static-libgcc</samp>, and <samp>-shared-libgcc</samp>.  On
targets where these modifications are inappropriate, define
<code>REAL_LIBGCC_SPEC</code> instead.  <code>REAL_LIBGCC_SPEC</code> tells the
driver how to place a reference to <samp>libgcc</samp> on the link command
line, but, unlike <code>LIBGCC_SPEC</code>, it is used unmodified.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fLD_005fAS_005fNEEDED"></a>Macro: <strong>USE_LD_AS_NEEDED</strong></dt>
<dd><p>A macro that controls the modifications to <code>LIBGCC_SPEC</code>
mentioned in <code>REAL_LIBGCC_SPEC</code>.  If nonzero, a spec will be
generated that uses &ndash;as-needed and the shared libgcc in place of the
static exception handler library, when linking without any of
<code>-static</code>, <code>-static-libgcc</code>, or <code>-shared-libgcc</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LINK_005fEH_005fSPEC"></a>Macro: <strong>LINK_EH_SPEC</strong></dt>
<dd><p>If defined, this C string constant is added to <code>LINK_SPEC</code>.
When <code>USE_LD_AS_NEEDED</code> is zero or undefined, it also affects
the modifications to <code>LIBGCC_SPEC</code> mentioned in
<code>REAL_LIBGCC_SPEC</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STARTFILE_005fSPEC"></a>Macro: <strong>STARTFILE_SPEC</strong></dt>
<dd><p>Another C string constant used much like <code>LINK_SPEC</code>.  The
difference between the two is that <code>STARTFILE_SPEC</code> is used at
the very beginning of the command given to the linker.
</p>
<p>If this macro is not defined, a default is provided that loads the
standard C startup file from the usual place.  See <samp>gcc.c</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-ENDFILE_005fSPEC"></a>Macro: <strong>ENDFILE_SPEC</strong></dt>
<dd><p>Another C string constant used much like <code>LINK_SPEC</code>.  The
difference between the two is that <code>ENDFILE_SPEC</code> is used at
the very end of the command given to the linker.
</p>
<p>Do not define this macro if it does not need to do anything.
</p></dd></dl>

<dl>
<dt><a name="index-THREAD_005fMODEL_005fSPEC"></a>Macro: <strong>THREAD_MODEL_SPEC</strong></dt>
<dd><p>GCC <code>-v</code> will print the thread model GCC was configured to use.
However, this doesn&rsquo;t work on platforms that are multilibbed on thread
models, such as AIX 4.3.  On such platforms, define
<code>THREAD_MODEL_SPEC</code> such that it evaluates to a string without
blanks that names one of the recognized thread models.  <code>%*</code>, the
default value of this macro, will expand to the value of
<code>thread_file</code> set in <samp>config.gcc</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-SYSROOT_005fSUFFIX_005fSPEC"></a>Macro: <strong>SYSROOT_SUFFIX_SPEC</strong></dt>
<dd><p>Define this macro to add a suffix to the target sysroot when GCC is
configured with a sysroot.  This will cause GCC to search for usr/lib,
et al, within sysroot+suffix.
</p></dd></dl>

<dl>
<dt><a name="index-SYSROOT_005fHEADERS_005fSUFFIX_005fSPEC"></a>Macro: <strong>SYSROOT_HEADERS_SUFFIX_SPEC</strong></dt>
<dd><p>Define this macro to add a headers_suffix to the target sysroot when
GCC is configured with a sysroot.  This will cause GCC to pass the
updated sysroot+headers_suffix to CPP, causing it to search for
usr/include, et al, within sysroot+headers_suffix.
</p></dd></dl>

<dl>
<dt><a name="index-EXTRA_005fSPECS"></a>Macro: <strong>EXTRA_SPECS</strong></dt>
<dd><p>Define this macro to provide additional specifications to put in the
<samp>specs</samp> file that can be used in various specifications like
<code>CC1_SPEC</code>.
</p>
<p>The definition should be an initializer for an array of structures,
containing a string constant, that defines the specification name, and a
string constant that provides the specification.
</p>
<p>Do not define this macro if it does not need to do anything.
</p>
<p><code>EXTRA_SPECS</code> is useful when an architecture contains several
related targets, which have various <code>&hellip;_SPECS</code> which are similar
to each other, and the maintainer would like one central place to keep
these definitions.
</p>
<p>For example, the PowerPC System V.4 targets use <code>EXTRA_SPECS</code> to
define either <code>_CALL_SYSV</code> when the System V calling sequence is
used or <code>_CALL_AIX</code> when the older AIX-based calling sequence is
used.
</p>
<p>The <samp>config/rs6000/rs6000.h</samp> target file defines:
</p>
<div class="smallexample">
<pre class="smallexample">#define EXTRA_SPECS \
  { &quot;cpp_sysv_default&quot;, CPP_SYSV_DEFAULT },

#define CPP_SYS_DEFAULT &quot;&quot;
</pre></div>

<p>The <samp>config/rs6000/sysv.h</samp> target file defines:
</p><div class="smallexample">
<pre class="smallexample">#undef CPP_SPEC
#define CPP_SPEC \
&quot;%{posix: -D_POSIX_SOURCE } \
%{mcall-sysv: -D_CALL_SYSV } \
%{!mcall-sysv: %(cpp_sysv_default) } \
%{msoft-float: -D_SOFT_FLOAT} %{mcpu=403: -D_SOFT_FLOAT}&quot;

#undef CPP_SYSV_DEFAULT
#define CPP_SYSV_DEFAULT &quot;-D_CALL_SYSV&quot;
</pre></div>

<p>while the <samp>config/rs6000/eabiaix.h</samp> target file defines
<code>CPP_SYSV_DEFAULT</code> as:
</p>
<div class="smallexample">
<pre class="smallexample">#undef CPP_SYSV_DEFAULT
#define CPP_SYSV_DEFAULT &quot;-D_CALL_AIX&quot;
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-LINK_005fLIBGCC_005fSPECIAL_005f1"></a>Macro: <strong>LINK_LIBGCC_SPECIAL_1</strong></dt>
<dd><p>Define this macro if the driver program should find the library
<samp>libgcc.a</samp>.  If you do not define this macro, the driver program will pass
the argument <samp>-lgcc</samp> to tell the linker to do the search.
</p></dd></dl>

<dl>
<dt><a name="index-LINK_005fGCC_005fC_005fSEQUENCE_005fSPEC"></a>Macro: <strong>LINK_GCC_C_SEQUENCE_SPEC</strong></dt>
<dd><p>The sequence in which libgcc and libc are specified to the linker.
By default this is <code>%G %L %G</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LINK_005fCOMMAND_005fSPEC"></a>Macro: <strong>LINK_COMMAND_SPEC</strong></dt>
<dd><p>A C string constant giving the complete command line need to execute the
linker.  When you do this, you will need to update your port each time a
change is made to the link command line within <samp>gcc.c</samp>.  Therefore,
define this macro only if you need to completely redefine the command
line for invoking the linker and there is no other way to accomplish
the effect you need.  Overriding this macro may be avoidable by overriding
<code>LINK_GCC_C_SEQUENCE_SPEC</code> instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fALWAYS_005fSTRIP_005fDOTDOT"></a>Common Target Hook: <em>bool</em> <strong>TARGET_ALWAYS_STRIP_DOTDOT</strong></dt>
<dd><p>True if <samp>..</samp> components should always be removed from directory names computed relative to GCC&rsquo;s internal directories, false (default) if such components should be preserved and directory names containing them passed to other tools such as the linker.
</p></dd></dl>

<dl>
<dt><a name="index-MULTILIB_005fDEFAULTS"></a>Macro: <strong>MULTILIB_DEFAULTS</strong></dt>
<dd><p>Define this macro as a C expression for the initializer of an array of
string to tell the driver program which options are defaults for this
target and thus do not need to be handled specially when using
<code>MULTILIB_OPTIONS</code>.
</p>
<p>Do not define this macro if <code>MULTILIB_OPTIONS</code> is not defined in
the target makefile fragment or if none of the options listed in
<code>MULTILIB_OPTIONS</code> are set by default.
See <a href="#Target-Fragment">Target Fragment</a>.
</p></dd></dl>

<dl>
<dt><a name="index-RELATIVE_005fPREFIX_005fNOT_005fLINKDIR"></a>Macro: <strong>RELATIVE_PREFIX_NOT_LINKDIR</strong></dt>
<dd><p>Define this macro to tell <code>gcc</code> that it should only translate
a <samp>-B</samp> prefix into a <samp>-L</samp> linker option if the prefix
indicates an absolute file name.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fEXEC_005fPREFIX"></a>Macro: <strong>MD_EXEC_PREFIX</strong></dt>
<dd><p>If defined, this macro is an additional prefix to try after
<code>STANDARD_EXEC_PREFIX</code>.  <code>MD_EXEC_PREFIX</code> is not searched
when the compiler is built as a cross
compiler.  If you define <code>MD_EXEC_PREFIX</code>, then be sure to add it
to the list of directories used to find the assembler in <samp>configure.in</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-STANDARD_005fSTARTFILE_005fPREFIX"></a>Macro: <strong>STANDARD_STARTFILE_PREFIX</strong></dt>
<dd><p>Define this macro as a C string constant if you wish to override the
standard choice of <code>libdir</code> as the default prefix to
try when searching for startup files such as <samp>crt0.o</samp>.
<code>STANDARD_STARTFILE_PREFIX</code> is not searched when the compiler
is built as a cross compiler.
</p></dd></dl>

<dl>
<dt><a name="index-STANDARD_005fSTARTFILE_005fPREFIX_005f1"></a>Macro: <strong>STANDARD_STARTFILE_PREFIX_1</strong></dt>
<dd><p>Define this macro as a C string constant if you wish to override the
standard choice of <code>/lib</code> as a prefix to try after the default prefix
when searching for startup files such as <samp>crt0.o</samp>.
<code>STANDARD_STARTFILE_PREFIX_1</code> is not searched when the compiler
is built as a cross compiler.
</p></dd></dl>

<dl>
<dt><a name="index-STANDARD_005fSTARTFILE_005fPREFIX_005f2"></a>Macro: <strong>STANDARD_STARTFILE_PREFIX_2</strong></dt>
<dd><p>Define this macro as a C string constant if you wish to override the
standard choice of <code>/lib</code> as yet another prefix to try after the
default prefix when searching for startup files such as <samp>crt0.o</samp>.
<code>STANDARD_STARTFILE_PREFIX_2</code> is not searched when the compiler
is built as a cross compiler.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fSTARTFILE_005fPREFIX"></a>Macro: <strong>MD_STARTFILE_PREFIX</strong></dt>
<dd><p>If defined, this macro supplies an additional prefix to try after the
standard prefixes.  <code>MD_EXEC_PREFIX</code> is not searched when the
compiler is built as a cross compiler.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fSTARTFILE_005fPREFIX_005f1"></a>Macro: <strong>MD_STARTFILE_PREFIX_1</strong></dt>
<dd><p>If defined, this macro supplies yet another prefix to try after the
standard prefixes.  It is not searched when the compiler is built as a
cross compiler.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fENVIRONMENT"></a>Macro: <strong>INIT_ENVIRONMENT</strong></dt>
<dd><p>Define this macro as a C string constant if you wish to set environment
variables for programs called by the driver, such as the assembler and
loader.  The driver passes the value of this macro to <code>putenv</code> to
initialize the necessary environment variables.
</p></dd></dl>

<dl>
<dt><a name="index-LOCAL_005fINCLUDE_005fDIR"></a>Macro: <strong>LOCAL_INCLUDE_DIR</strong></dt>
<dd><p>Define this macro as a C string constant if you wish to override the
standard choice of <samp>/usr/local/include</samp> as the default prefix to
try when searching for local header files.  <code>LOCAL_INCLUDE_DIR</code>
comes before <code>NATIVE_SYSTEM_HEADER_DIR</code> (set in
<samp>config.gcc</samp>, normally <samp>/usr/include</samp>) in the search order.
</p>
<p>Cross compilers do not search either <samp>/usr/local/include</samp> or its
replacement.
</p></dd></dl>

<dl>
<dt><a name="index-NATIVE_005fSYSTEM_005fHEADER_005fCOMPONENT"></a>Macro: <strong>NATIVE_SYSTEM_HEADER_COMPONENT</strong></dt>
<dd><p>The &ldquo;component&rdquo; corresponding to <code>NATIVE_SYSTEM_HEADER_DIR</code>.
See <code>INCLUDE_DEFAULTS</code>, below, for the description of components.
If you do not define this macro, no component is used.
</p></dd></dl>

<dl>
<dt><a name="index-INCLUDE_005fDEFAULTS"></a>Macro: <strong>INCLUDE_DEFAULTS</strong></dt>
<dd><p>Define this macro if you wish to override the entire default search path
for include files.  For a native compiler, the default search path
usually consists of <code>GCC_INCLUDE_DIR</code>, <code>LOCAL_INCLUDE_DIR</code>,
<code>GPLUSPLUS_INCLUDE_DIR</code>, and
<code>NATIVE_SYSTEM_HEADER_DIR</code>.  In addition, <code>GPLUSPLUS_INCLUDE_DIR</code>
and <code>GCC_INCLUDE_DIR</code> are defined automatically by <samp>Makefile</samp>,
and specify private search areas for GCC.  The directory
<code>GPLUSPLUS_INCLUDE_DIR</code> is used only for C++ programs.
</p>
<p>The definition should be an initializer for an array of structures.
Each array element should have four elements: the directory name (a
string constant), the component name (also a string constant), a flag
for C++-only directories,
and a flag showing that the includes in the directory don&rsquo;t need to be
wrapped in <code>extern &lsquo;<samp>C</samp>&rsquo;</code> when compiling C++.  Mark the end of
the array with a null element.
</p>
<p>The component name denotes what GNU package the include file is part of,
if any, in all uppercase letters.  For example, it might be &lsquo;<samp>GCC</samp>&rsquo;
or &lsquo;<samp>BINUTILS</samp>&rsquo;.  If the package is part of a vendor-supplied
operating system, code the component name as &lsquo;<samp>0</samp>&rsquo;.
</p>
<p>For example, here is the definition used for VAX/VMS:
</p>
<div class="smallexample">
<pre class="smallexample">#define INCLUDE_DEFAULTS \
{                                       \
  { &quot;GNU_GXX_INCLUDE:&quot;, &quot;G++&quot;, 1, 1},   \
  { &quot;GNU_CC_INCLUDE:&quot;, &quot;GCC&quot;, 0, 0},    \
  { &quot;SYS$SYSROOT:[SYSLIB.]&quot;, 0, 0, 0},  \
  { &quot;.&quot;, 0, 0, 0},                      \
  { 0, 0, 0, 0}                         \
}
</pre></div>
</dd></dl>

<p>Here is the order of prefixes tried for exec files:
</p>
<ol>
<li> Any prefixes specified by the user with <samp>-B</samp>.

</li><li> The environment variable <code>GCC_EXEC_PREFIX</code> or, if <code>GCC_EXEC_PREFIX</code>
is not set and the compiler has not been installed in the configure-time
<var>prefix</var>, the location in which the compiler has actually been installed.

</li><li> The directories specified by the environment variable <code>COMPILER_PATH</code>.

</li><li> The macro <code>STANDARD_EXEC_PREFIX</code>, if the compiler has been installed
in the configured-time <var>prefix</var>.

</li><li> The location <samp>/usr/libexec/gcc/</samp>, but only if this is a native compiler.

</li><li> The location <samp>/usr/lib/gcc/</samp>, but only if this is a native compiler.

</li><li> The macro <code>MD_EXEC_PREFIX</code>, if defined, but only if this is a native
compiler.
</li></ol>

<p>Here is the order of prefixes tried for startfiles:
</p>
<ol>
<li> Any prefixes specified by the user with <samp>-B</samp>.

</li><li> The environment variable <code>GCC_EXEC_PREFIX</code> or its automatically determined
value based on the installed toolchain location.

</li><li> The directories specified by the environment variable <code>LIBRARY_PATH</code>
(or port-specific name; native only, cross compilers do not use this).

</li><li> The macro <code>STANDARD_EXEC_PREFIX</code>, but only if the toolchain is installed
in the configured <var>prefix</var> or this is a native compiler.

</li><li> The location <samp>/usr/lib/gcc/</samp>, but only if this is a native compiler.

</li><li> The macro <code>MD_EXEC_PREFIX</code>, if defined, but only if this is a native
compiler.

</li><li> The macro <code>MD_STARTFILE_PREFIX</code>, if defined, but only if this is a
native compiler, or we have a target system root.

</li><li> The macro <code>MD_STARTFILE_PREFIX_1</code>, if defined, but only if this is a
native compiler, or we have a target system root.

</li><li> The macro <code>STANDARD_STARTFILE_PREFIX</code>, with any sysroot modifications.
If this path is relative it will be prefixed by <code>GCC_EXEC_PREFIX</code> and
the machine suffix or <code>STANDARD_EXEC_PREFIX</code> and the machine suffix.

</li><li> The macro <code>STANDARD_STARTFILE_PREFIX_1</code>, but only if this is a native
compiler, or we have a target system root. The default for this macro is
<samp>/lib/</samp>.

</li><li> The macro <code>STANDARD_STARTFILE_PREFIX_2</code>, but only if this is a native
compiler, or we have a target system root. The default for this macro is
<samp>/usr/lib/</samp>.
</li></ol>

<hr>
<a name="Run_002dtime-Target"></a>
<div class="header">
<p>
Next: <a href="#Per_002dFunction-Data" accesskey="n" rel="next">Per-Function Data</a>, Previous: <a href="#Driver" accesskey="p" rel="prev">Driver</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Run_002dtime-Target-Specification"></a>
<h3 class="section">17.3 Run-time Target Specification</h3>
<a name="index-run_002dtime-target-specification"></a>
<a name="index-predefined-macros"></a>
<a name="index-target-specifications"></a>

<p>Here are run-time target specifications.
</p>
<dl>
<dt><a name="index-TARGET_005fCPU_005fCPP_005fBUILTINS"></a>Macro: <strong>TARGET_CPU_CPP_BUILTINS</strong> <em>()</em></dt>
<dd><p>This function-like macro expands to a block of code that defines
built-in preprocessor macros and assertions for the target CPU, using
the functions <code>builtin_define</code>, <code>builtin_define_std</code> and
<code>builtin_assert</code>.  When the front end
calls this macro it provides a trailing semicolon, and since it has
finished command line option processing your code can use those
results freely.
</p>
<p><code>builtin_assert</code> takes a string in the form you pass to the
command-line option <samp>-A</samp>, such as <code>cpu=mips</code>, and creates
the assertion.  <code>builtin_define</code> takes a string in the form
accepted by option <samp>-D</samp> and unconditionally defines the macro.
</p>
<p><code>builtin_define_std</code> takes a string representing the name of an
object-like macro.  If it doesn&rsquo;t lie in the user&rsquo;s namespace,
<code>builtin_define_std</code> defines it unconditionally.  Otherwise, it
defines a version with two leading underscores, and another version
with two leading and trailing underscores, and defines the original
only if an ISO standard was not requested on the command line.  For
example, passing <code>unix</code> defines <code>__unix</code>, <code>__unix__</code>
and possibly <code>unix</code>; passing <code>_mips</code> defines <code>__mips</code>,
<code>__mips__</code> and possibly <code>_mips</code>, and passing <code>_ABI64</code>
defines only <code>_ABI64</code>.
</p>
<p>You can also test for the C dialect being compiled.  The variable
<code>c_language</code> is set to one of <code>clk_c</code>, <code>clk_cplusplus</code>
or <code>clk_objective_c</code>.  Note that if we are preprocessing
assembler, this variable will be <code>clk_c</code> but the function-like
macro <code>preprocessing_asm_p()</code> will return true, so you might want
to check for that first.  If you need to check for strict ANSI, the
variable <code>flag_iso</code> can be used.  The function-like macro
<code>preprocessing_trad_p()</code> can be used to check for traditional
preprocessing.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOS_005fCPP_005fBUILTINS"></a>Macro: <strong>TARGET_OS_CPP_BUILTINS</strong> <em>()</em></dt>
<dd><p>Similarly to <code>TARGET_CPU_CPP_BUILTINS</code> but this macro is optional
and is used for the target operating system instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOBJFMT_005fCPP_005fBUILTINS"></a>Macro: <strong>TARGET_OBJFMT_CPP_BUILTINS</strong> <em>()</em></dt>
<dd><p>Similarly to <code>TARGET_CPU_CPP_BUILTINS</code> but this macro is optional
and is used for the target object format.  <samp>elfos.h</samp> uses this
macro to define <code>__ELF__</code>, so you probably do not need to define
it yourself.
</p></dd></dl>

<dl>
<dt><a name="index-target_005fflags"></a>Variable: <em>extern int</em> <strong>target_flags</strong></dt>
<dd><p>This variable is declared in <samp>options.h</samp>, which is included before
any target-specific headers.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDEFAULT_005fTARGET_005fFLAGS"></a>Common Target Hook: <em>int</em> <strong>TARGET_DEFAULT_TARGET_FLAGS</strong></dt>
<dd><p>This variable specifies the initial value of <code>target_flags</code>.
Its default setting is 0.
</p></dd></dl>

<a name="index-optional-hardware-or-system-features"></a>
<a name="index-features_002c-optional_002c-in-system-conventions"></a>

<dl>
<dt><a name="index-TARGET_005fHANDLE_005fOPTION"></a>Common Target Hook: <em>bool</em> <strong>TARGET_HANDLE_OPTION</strong> <em>(struct gcc_options *<var>opts</var>, struct gcc_options *<var>opts_set</var>, const struct cl_decoded_option *<var>decoded</var>, location_t <var>loc</var>)</em></dt>
<dd><p>This hook is called whenever the user specifies one of the
target-specific options described by the <samp>.opt</samp> definition files
(see <a href="#Options">Options</a>).  It has the opportunity to do some option-specific
processing and should return true if the option is valid.  The default
definition does nothing but return true.
</p>
<p><var>decoded</var> specifies the option and its arguments.  <var>opts</var> and
<var>opts_set</var> are the <code>gcc_options</code> structures to be used for
storing option state, and <var>loc</var> is the location at which the
option was passed (<code>UNKNOWN_LOCATION</code> except for options passed
via attributes).
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHANDLE_005fC_005fOPTION"></a>C Target Hook: <em>bool</em> <strong>TARGET_HANDLE_C_OPTION</strong> <em>(size_t <var>code</var>, const char *<var>arg</var>, int <var>value</var>)</em></dt>
<dd><p>This target hook is called whenever the user specifies one of the
target-specific C language family options described by the <samp>.opt</samp>
definition files(see <a href="#Options">Options</a>).  It has the opportunity to do some
option-specific processing and should return true if the option is
valid.  The arguments are like for <code>TARGET_HANDLE_OPTION</code>.  The
default definition does nothing but return false.
</p>
<p>In general, you should use <code>TARGET_HANDLE_OPTION</code> to handle
options.  However, if processing an option requires routines that are
only available in the C (and related language) front ends, then you
should use <code>TARGET_HANDLE_C_OPTION</code> instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOBJC_005fCONSTRUCT_005fSTRING_005fOBJECT"></a>C Target Hook: <em>tree</em> <strong>TARGET_OBJC_CONSTRUCT_STRING_OBJECT</strong> <em>(tree <var>string</var>)</em></dt>
<dd><p>Targets may provide a string object type that can be used within and between C, C++ and their respective Objective-C dialects. A string object might, for example, embed encoding and length information. These objects are considered opaque to the compiler and handled as references. An ideal implementation makes the composition of the string object match that of the Objective-C <code>NSString</code> (<code>NXString</code> for GNUStep), allowing efficient interworking between C-only and Objective-C code. If a target implements string objects then this hook should return a reference to such an object constructed from the normal &lsquo;C&rsquo; string representation provided in <var>string</var>. At present, the hook is used by Objective-C only, to obtain a common-format string object when the target provides one.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOBJC_005fDECLARE_005fUNRESOLVED_005fCLASS_005fREFERENCE"></a>C Target Hook: <em>void</em> <strong>TARGET_OBJC_DECLARE_UNRESOLVED_CLASS_REFERENCE</strong> <em>(const char *<var>classname</var>)</em></dt>
<dd><p>Declare that Objective C class <var>classname</var> is referenced  by the current TU.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOBJC_005fDECLARE_005fCLASS_005fDEFINITION"></a>C Target Hook: <em>void</em> <strong>TARGET_OBJC_DECLARE_CLASS_DEFINITION</strong> <em>(const char *<var>classname</var>)</em></dt>
<dd><p>Declare that Objective C class <var>classname</var> is defined  by the current TU.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTRING_005fOBJECT_005fREF_005fTYPE_005fP"></a>C Target Hook: <em>bool</em> <strong>TARGET_STRING_OBJECT_REF_TYPE_P</strong> <em>(const_tree <var>stringref</var>)</em></dt>
<dd><p>If a target implements string objects then this hook should return <code>true</code> if <var>stringref</var> is a valid reference to such an object.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCHECK_005fSTRING_005fOBJECT_005fFORMAT_005fARG"></a>C Target Hook: <em>void</em> <strong>TARGET_CHECK_STRING_OBJECT_FORMAT_ARG</strong> <em>(tree <var>format_arg</var>, tree <var>args_list</var>)</em></dt>
<dd><p>If a target implements string objects then this hook should should  provide a facility to check the function arguments in <var>args_list</var>  against the format specifiers in <var>format_arg</var> where the type of  <var>format_arg</var> is one recognized as a valid string reference type.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOVERRIDE_005fOPTIONS_005fAFTER_005fCHANGE"></a>Target Hook: <em>void</em> <strong>TARGET_OVERRIDE_OPTIONS_AFTER_CHANGE</strong> <em>(void)</em></dt>
<dd><p>This target function is similar to the hook <code>TARGET_OPTION_OVERRIDE</code>
but is called when the optimize level is changed via an attribute or
pragma or when it is reset at the end of the code affected by the
attribute or pragma.  It is not called at the beginning of compilation
when <code>TARGET_OPTION_OVERRIDE</code> is called so if you want to perform these
actions then, you should have <code>TARGET_OPTION_OVERRIDE</code> call
<code>TARGET_OVERRIDE_OPTIONS_AFTER_CHANGE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-C_005fCOMMON_005fOVERRIDE_005fOPTIONS"></a>Macro: <strong>C_COMMON_OVERRIDE_OPTIONS</strong></dt>
<dd><p>This is similar to the <code>TARGET_OPTION_OVERRIDE</code> hook
but is only used in the C
language frontends (C, Objective-C, C++, Objective-C++) and so can be
used to alter option flag variables which only exist in those
frontends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fOPTIMIZATION_005fTABLE"></a>Common Target Hook: <em>const struct default_options *</em> <strong>TARGET_OPTION_OPTIMIZATION_TABLE</strong></dt>
<dd><p>Some machines may desire to change what optimizations are performed for
various optimization levels.   This variable, if defined, describes
options to enable at particular sets of optimization levels.  These
options are processed once
just after the optimization level is determined and before the remainder
of the command options have been parsed, so may be overridden by other
options passed explicitly.
</p>
<p>This processing is run once at program startup and when the optimization
options are changed via <code>#pragma GCC optimize</code> or by using the
<code>optimize</code> attribute.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fINIT_005fSTRUCT"></a>Common Target Hook: <em>void</em> <strong>TARGET_OPTION_INIT_STRUCT</strong> <em>(struct gcc_options *<var>opts</var>)</em></dt>
<dd><p>Set target-dependent initial values of fields in <var>opts</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fDEFAULT_005fPARAMS"></a>Common Target Hook: <em>void</em> <strong>TARGET_OPTION_DEFAULT_PARAMS</strong> <em>(void)</em></dt>
<dd><p>Set target-dependent default values for <samp>--param</samp> settings, using calls to <code>set_default_param_value</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SWITCHABLE_005fTARGET"></a>Macro: <strong>SWITCHABLE_TARGET</strong></dt>
<dd><p>Some targets need to switch between substantially different subtargets
during compilation.  For example, the MIPS target has one subtarget for
the traditional MIPS architecture and another for MIPS16.  Source code
can switch between these two subarchitectures using the <code>mips16</code>
and <code>nomips16</code> attributes.
</p>
<p>Such subtargets can differ in things like the set of available
registers, the set of available instructions, the costs of various
operations, and so on.  GCC caches a lot of this type of information
in global variables, and recomputing them for each subtarget takes a
significant amount of time.  The compiler therefore provides a facility
for maintaining several versions of the global variables and quickly
switching between them; see <samp>target-globals.h</samp> for details.
</p>
<p>Define this macro to 1 if your target needs this facility.  The default
is 0.
</p></dd></dl>

<hr>
<a name="Per_002dFunction-Data"></a>
<div class="header">
<p>
Next: <a href="#Storage-Layout" accesskey="n" rel="next">Storage Layout</a>, Previous: <a href="#Run_002dtime-Target" accesskey="p" rel="prev">Run-time Target</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-data-structures-for-per_002dfunction-information_002e"></a>
<h3 class="section">17.4 Defining data structures for per-function information.</h3>
<a name="index-per_002dfunction-data"></a>
<a name="index-data-structures"></a>

<p>If the target needs to store information on a per-function basis, GCC
provides a macro and a couple of variables to allow this.  Note, just
using statics to store the information is a bad idea, since GCC supports
nested functions, so you can be halfway through encoding one function
when another one comes along.
</p>
<p>GCC defines a data structure called <code>struct function</code> which
contains all of the data specific to an individual function.  This
structure contains a field called <code>machine</code> whose type is
<code>struct machine_function *</code>, which can be used by targets to point
to their own specific data.
</p>
<p>If a target needs per-function specific data it should define the type
<code>struct machine_function</code> and also the macro <code>INIT_EXPANDERS</code>.
This macro should be used to initialize the function pointer
<code>init_machine_status</code>.  This pointer is explained below.
</p>
<p>One typical use of per-function, target specific data is to create an
RTX to hold the register containing the function&rsquo;s return address.  This
RTX can then be used to implement the <code>__builtin_return_address</code>
function, for level 0.
</p>
<p>Note&mdash;earlier implementations of GCC used a single data area to hold
all of the per-function information.  Thus when processing of a nested
function began the old per-function data had to be pushed onto a
stack, and when the processing was finished, it had to be popped off the
stack.  GCC used to provide function pointers called
<code>save_machine_status</code> and <code>restore_machine_status</code> to handle
the saving and restoring of the target specific information.  Since the
single data area approach is no longer used, these pointers are no
longer supported.
</p>
<dl>
<dt><a name="index-INIT_005fEXPANDERS"></a>Macro: <strong>INIT_EXPANDERS</strong></dt>
<dd><p>Macro called to initialize any target specific information.  This macro
is called once per function, before generation of any RTL has begun.
The intention of this macro is to allow the initialization of the
function pointer <code>init_machine_status</code>.
</p></dd></dl>

<dl>
<dt><a name="index-init_005fmachine_005fstatus"></a>Variable: <em>void (*)(struct function *)</em> <strong>init_machine_status</strong></dt>
<dd><p>If this function pointer is non-<code>NULL</code> it will be called once per
function, before function compilation starts, in order to allow the
target to perform any target specific initialization of the
<code>struct function</code> structure.  It is intended that this would be
used to initialize the <code>machine</code> of that structure.
</p>
<p><code>struct machine_function</code> structures are expected to be freed by GC.
Generally, any memory that they reference must be allocated by using
GC allocation, including the structure itself.
</p></dd></dl>

<hr>
<a name="Storage-Layout"></a>
<div class="header">
<p>
Next: <a href="#Type-Layout" accesskey="n" rel="next">Type Layout</a>, Previous: <a href="#Per_002dFunction-Data" accesskey="p" rel="prev">Per-Function Data</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Storage-Layout-1"></a>
<h3 class="section">17.5 Storage Layout</h3>
<a name="index-storage-layout"></a>

<p>Note that the definitions of the macros in this table which are sizes or
alignments measured in bits do not need to be constant.  They can be C
expressions that refer to static variables, such as the <code>target_flags</code>.
See <a href="#Run_002dtime-Target">Run-time Target</a>.
</p>
<dl>
<dt><a name="index-BITS_005fBIG_005fENDIAN"></a>Macro: <strong>BITS_BIG_ENDIAN</strong></dt>
<dd><p>Define this macro to have the value 1 if the most significant bit in a
byte has the lowest number; otherwise define it to have the value zero.
This means that bit-field instructions count from the most significant
bit.  If the machine has no bit-field instructions, then this must still
be defined, but it doesn&rsquo;t matter which value it is defined to.  This
macro need not be a constant.
</p>
<p>This macro does not affect the way structure fields are packed into
bytes or words; that is controlled by <code>BYTES_BIG_ENDIAN</code>.
</p></dd></dl>

<dl>
<dt><a name="index-BYTES_005fBIG_005fENDIAN"></a>Macro: <strong>BYTES_BIG_ENDIAN</strong></dt>
<dd><p>Define this macro to have the value 1 if the most significant byte in a
word has the lowest number.  This macro need not be a constant.
</p></dd></dl>

<dl>
<dt><a name="index-WORDS_005fBIG_005fENDIAN"></a>Macro: <strong>WORDS_BIG_ENDIAN</strong></dt>
<dd><p>Define this macro to have the value 1 if, in a multiword object, the
most significant word has the lowest number.  This applies to both
memory locations and registers; see <code>REG_WORDS_BIG_ENDIAN</code> if the
order of words in memory is not the same as the order in registers.  This
macro need not be a constant.
</p></dd></dl>

<dl>
<dt><a name="index-REG_005fWORDS_005fBIG_005fENDIAN"></a>Macro: <strong>REG_WORDS_BIG_ENDIAN</strong></dt>
<dd><p>On some machines, the order of words in a multiword object differs between
registers in memory.  In such a situation, define this macro to describe
the order of words in a register.  The macro <code>WORDS_BIG_ENDIAN</code> controls
the order of words in memory.
</p></dd></dl>

<dl>
<dt><a name="index-FLOAT_005fWORDS_005fBIG_005fENDIAN"></a>Macro: <strong>FLOAT_WORDS_BIG_ENDIAN</strong></dt>
<dd><p>Define this macro to have the value 1 if <code>DFmode</code>, <code>XFmode</code> or
<code>TFmode</code> floating point numbers are stored in memory with the word
containing the sign bit at the lowest address; otherwise define it to
have the value 0.  This macro need not be a constant.
</p>
<p>You need not define this macro if the ordering is the same as for
multi-word integers.
</p></dd></dl>

<dl>
<dt><a name="index-BITS_005fPER_005fUNIT"></a>Macro: <strong>BITS_PER_UNIT</strong></dt>
<dd><p>Define this macro to be the number of bits in an addressable storage
unit (byte).  If you do not define this macro the default is 8.
</p></dd></dl>

<dl>
<dt><a name="index-BITS_005fPER_005fWORD"></a>Macro: <strong>BITS_PER_WORD</strong></dt>
<dd><p>Number of bits in a word.  If you do not define this macro, the default
is <code>BITS_PER_UNIT * UNITS_PER_WORD</code>.
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fBITS_005fPER_005fWORD"></a>Macro: <strong>MAX_BITS_PER_WORD</strong></dt>
<dd><p>Maximum number of bits in a word.  If this is undefined, the default is
<code>BITS_PER_WORD</code>.  Otherwise, it is the constant value that is the
largest value that <code>BITS_PER_WORD</code> can have at run-time.
</p></dd></dl>

<dl>
<dt><a name="index-UNITS_005fPER_005fWORD"></a>Macro: <strong>UNITS_PER_WORD</strong></dt>
<dd><p>Number of storage units in a word; normally the size of a general-purpose
register, a power of two from 1 or 8.
</p></dd></dl>

<dl>
<dt><a name="index-MIN_005fUNITS_005fPER_005fWORD"></a>Macro: <strong>MIN_UNITS_PER_WORD</strong></dt>
<dd><p>Minimum number of units in a word.  If this is undefined, the default is
<code>UNITS_PER_WORD</code>.  Otherwise, it is the constant value that is the
smallest value that <code>UNITS_PER_WORD</code> can have at run-time.
</p></dd></dl>

<dl>
<dt><a name="index-POINTER_005fSIZE"></a>Macro: <strong>POINTER_SIZE</strong></dt>
<dd><p>Width of a pointer, in bits.  You must specify a value no wider than the
width of <code>Pmode</code>.  If it is not equal to the width of <code>Pmode</code>,
you must define <code>POINTERS_EXTEND_UNSIGNED</code>.  If you do not specify
a value the default is <code>BITS_PER_WORD</code>.
</p></dd></dl>

<dl>
<dt><a name="index-POINTERS_005fEXTEND_005fUNSIGNED"></a>Macro: <strong>POINTERS_EXTEND_UNSIGNED</strong></dt>
<dd><p>A C expression that determines how pointers should be extended from
<code>ptr_mode</code> to either <code>Pmode</code> or <code>word_mode</code>.  It is
greater than zero if pointers should be zero-extended, zero if they
should be sign-extended, and negative if some other sort of conversion
is needed.  In the last case, the extension is done by the target&rsquo;s
<code>ptr_extend</code> instruction.
</p>
<p>You need not define this macro if the <code>ptr_mode</code>, <code>Pmode</code>
and <code>word_mode</code> are all the same width.
</p></dd></dl>

<dl>
<dt><a name="index-PROMOTE_005fMODE"></a>Macro: <strong>PROMOTE_MODE</strong> <em>(<var>m</var>, <var>unsignedp</var>, <var>type</var>)</em></dt>
<dd><p>A macro to update <var>m</var> and <var>unsignedp</var> when an object whose type
is <var>type</var> and which has the specified mode and signedness is to be
stored in a register.  This macro is only called when <var>type</var> is a
scalar type.
</p>
<p>On most RISC machines, which only have operations that operate on a full
register, define this macro to set <var>m</var> to <code>word_mode</code> if
<var>m</var> is an integer mode narrower than <code>BITS_PER_WORD</code>.  In most
cases, only integer modes should be widened because wider-precision
floating-point operations are usually more expensive than their narrower
counterparts.
</p>
<p>For most machines, the macro definition does not change <var>unsignedp</var>.
However, some machines, have instructions that preferentially handle
either signed or unsigned quantities of certain modes.  For example, on
the DEC Alpha, 32-bit loads from memory and 32-bit add instructions
sign-extend the result to 64 bits.  On such machines, set
<var>unsignedp</var> according to which kind of extension is more efficient.
</p>
<p>Do not define this macro if it would never modify <var>m</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPROMOTE_005fFUNCTION_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_PROMOTE_FUNCTION_MODE</strong> <em>(const_tree <var>type</var>, enum machine_mode <var>mode</var>, int *<var>punsignedp</var>, const_tree <var>funtype</var>, int <var>for_return</var>)</em></dt>
<dd><p>Like <code>PROMOTE_MODE</code>, but it is applied to outgoing function arguments or
function return values.  The target hook should return the new mode
and possibly change <code>*<var>punsignedp</var></code> if the promotion should
change signedness.  This function is called only for scalar <em>or
pointer</em> types.
</p>
<p><var>for_return</var> allows to distinguish the promotion of arguments and
return values.  If it is <code>1</code>, a return value is being promoted and
<code>TARGET_FUNCTION_VALUE</code> must perform the same promotions done here.
If it is <code>2</code>, the returned mode should be that of the register in
which an incoming parameter is copied, or the outgoing result is computed;
then the hook should return the same mode as <code>promote_mode</code>, though
the signedness may be different.
</p>
<p><var>type</var> can be NULL when promoting function arguments of libcalls.
</p>
<p>The default is to not promote arguments and return values.  You can
also define the hook to <code>default_promote_function_mode_always_promote</code>
if you would like to apply the same rules given by <code>PROMOTE_MODE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PARM_005fBOUNDARY"></a>Macro: <strong>PARM_BOUNDARY</strong></dt>
<dd><p>Normal alignment required for function parameters on the stack, in
bits.  All stack parameters receive at least this much alignment
regardless of data type.  On most machines, this is the same as the
size of an integer.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fBOUNDARY"></a>Macro: <strong>STACK_BOUNDARY</strong></dt>
<dd><p>Define this macro to the minimum alignment enforced by hardware for the
stack pointer on this machine.  The definition is a C expression for the
desired alignment (measured in bits).  This value is used as a default
if <code>PREFERRED_STACK_BOUNDARY</code> is not defined.  On most machines,
this should be the same as <code>PARM_BOUNDARY</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PREFERRED_005fSTACK_005fBOUNDARY"></a>Macro: <strong>PREFERRED_STACK_BOUNDARY</strong></dt>
<dd><p>Define this macro if you wish to preserve a certain alignment for the
stack pointer, greater than what the hardware enforces.  The definition
is a C expression for the desired alignment (measured in bits).  This
macro must evaluate to a value equal to or larger than
<code>STACK_BOUNDARY</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INCOMING_005fSTACK_005fBOUNDARY"></a>Macro: <strong>INCOMING_STACK_BOUNDARY</strong></dt>
<dd><p>Define this macro if the incoming stack boundary may be different
from <code>PREFERRED_STACK_BOUNDARY</code>.  This macro must evaluate
to a value equal to or larger than <code>STACK_BOUNDARY</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fBOUNDARY"></a>Macro: <strong>FUNCTION_BOUNDARY</strong></dt>
<dd><p>Alignment required for a function entry point, in bits.
</p></dd></dl>

<dl>
<dt><a name="index-BIGGEST_005fALIGNMENT"></a>Macro: <strong>BIGGEST_ALIGNMENT</strong></dt>
<dd><p>Biggest alignment that any data type can require on this machine, in
bits.  Note that this is not the biggest alignment that is supported,
just the biggest alignment that, when violated, may cause a fault.
</p></dd></dl>

<dl>
<dt><a name="index-MALLOC_005fABI_005fALIGNMENT"></a>Macro: <strong>MALLOC_ABI_ALIGNMENT</strong></dt>
<dd><p>Alignment, in bits, a C conformant malloc implementation has to
provide.  If not defined, the default value is <code>BITS_PER_WORD</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ATTRIBUTE_005fALIGNED_005fVALUE"></a>Macro: <strong>ATTRIBUTE_ALIGNED_VALUE</strong></dt>
<dd><p>Alignment used by the <code>__attribute__ ((aligned))</code> construct.  If
not defined, the default value is <code>BIGGEST_ALIGNMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-MINIMUM_005fATOMIC_005fALIGNMENT"></a>Macro: <strong>MINIMUM_ATOMIC_ALIGNMENT</strong></dt>
<dd><p>If defined, the smallest alignment, in bits, that can be given to an
object that can be referenced in one operation, without disturbing any
nearby object.  Normally, this is <code>BITS_PER_UNIT</code>, but may be larger
on machines that don&rsquo;t have byte or half-word store operations.
</p></dd></dl>

<dl>
<dt><a name="index-BIGGEST_005fFIELD_005fALIGNMENT"></a>Macro: <strong>BIGGEST_FIELD_ALIGNMENT</strong></dt>
<dd><p>Biggest alignment that any structure or union field can require on this
machine, in bits.  If defined, this overrides <code>BIGGEST_ALIGNMENT</code> for
structure and union fields only, unless the field alignment has been set
by the <code>__attribute__ ((aligned (<var>n</var>)))</code> construct.
</p></dd></dl>

<dl>
<dt><a name="index-ADJUST_005fFIELD_005fALIGN"></a>Macro: <strong>ADJUST_FIELD_ALIGN</strong> <em>(<var>field</var>, <var>computed</var>)</em></dt>
<dd><p>An expression for the alignment of a structure field <var>field</var> if the
alignment computed in the usual way (including applying of
<code>BIGGEST_ALIGNMENT</code> and <code>BIGGEST_FIELD_ALIGNMENT</code> to the
alignment) is <var>computed</var>.  It overrides alignment only if the
field alignment has not been set by the
<code>__attribute__ ((aligned (<var>n</var>)))</code> construct.
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fSTACK_005fALIGNMENT"></a>Macro: <strong>MAX_STACK_ALIGNMENT</strong></dt>
<dd><p>Biggest stack alignment guaranteed by the backend.  Use this macro
to specify the maximum alignment of a variable on stack.
</p>
<p>If not defined, the default value is <code>STACK_BOUNDARY</code>.
</p>
</dd></dl>

<dl>
<dt><a name="index-MAX_005fOFILE_005fALIGNMENT"></a>Macro: <strong>MAX_OFILE_ALIGNMENT</strong></dt>
<dd><p>Biggest alignment supported by the object file format of this machine.
Use this macro to limit the alignment which can be specified using the
<code>__attribute__ ((aligned (<var>n</var>)))</code> construct.  If not defined,
the default value is <code>BIGGEST_ALIGNMENT</code>.
</p>
<p>On systems that use ELF, the default (in <samp>config/elfos.h</samp>) is
the largest supported 32-bit ELF section alignment representable on
a 32-bit host e.g. &lsquo;<samp>(((unsigned HOST_WIDEST_INT) 1 &lt;&lt; 28) * 8)</samp>&rsquo;.
On 32-bit ELF the largest supported section alignment in bits is
&lsquo;<samp>(0x80000000 * 8)</samp>&rsquo;, but this is not representable on 32-bit hosts.
</p></dd></dl>

<dl>
<dt><a name="index-DATA_005fALIGNMENT"></a>Macro: <strong>DATA_ALIGNMENT</strong> <em>(<var>type</var>, <var>basic-align</var>)</em></dt>
<dd><p>If defined, a C expression to compute the alignment for a variable in
the static store.  <var>type</var> is the data type, and <var>basic-align</var> is
the alignment that the object would ordinarily have.  The value of this
macro is used instead of that alignment to align the object.
</p>
<p>If this macro is not defined, then <var>basic-align</var> is used.
</p>
<a name="index-strcpy"></a>
<p>One use of this macro is to increase alignment of medium-size data to
make it all fit in fewer cache lines.  Another is to cause character
arrays to be word-aligned so that <code>strcpy</code> calls that copy
constants to character arrays can be done inline.
</p></dd></dl>

<dl>
<dt><a name="index-CONSTANT_005fALIGNMENT"></a>Macro: <strong>CONSTANT_ALIGNMENT</strong> <em>(<var>constant</var>, <var>basic-align</var>)</em></dt>
<dd><p>If defined, a C expression to compute the alignment given to a constant
that is being placed in memory.  <var>constant</var> is the constant and
<var>basic-align</var> is the alignment that the object would ordinarily
have.  The value of this macro is used instead of that alignment to
align the object.
</p>
<p>If this macro is not defined, then <var>basic-align</var> is used.
</p>
<p>The typical use of this macro is to increase alignment for string
constants to be word aligned so that <code>strcpy</code> calls that copy
constants can be done inline.
</p></dd></dl>

<dl>
<dt><a name="index-LOCAL_005fALIGNMENT"></a>Macro: <strong>LOCAL_ALIGNMENT</strong> <em>(<var>type</var>, <var>basic-align</var>)</em></dt>
<dd><p>If defined, a C expression to compute the alignment for a variable in
the local store.  <var>type</var> is the data type, and <var>basic-align</var> is
the alignment that the object would ordinarily have.  The value of this
macro is used instead of that alignment to align the object.
</p>
<p>If this macro is not defined, then <var>basic-align</var> is used.
</p>
<p>One use of this macro is to increase alignment of medium-size data to
make it all fit in fewer cache lines.
</p>
<p>If the value of this macro has a type, it should be an unsigned type.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTOR_005fALIGNMENT"></a>Target Hook: <em>HOST_WIDE_INT</em> <strong>TARGET_VECTOR_ALIGNMENT</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>This hook can be used to define the alignment for a vector of type
<var>type</var>, in order to comply with a platform ABI.  The default is to
require natural alignment for vector types.  The alignment returned by
this hook must be a power-of-two multiple of the default alignment of
the vector element type.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fSLOT_005fALIGNMENT"></a>Macro: <strong>STACK_SLOT_ALIGNMENT</strong> <em>(<var>type</var>, <var>mode</var>, <var>basic-align</var>)</em></dt>
<dd><p>If defined, a C expression to compute the alignment for stack slot.
<var>type</var> is the data type, <var>mode</var> is the widest mode available,
and <var>basic-align</var> is the alignment that the slot would ordinarily
have.  The value of this macro is used instead of that alignment to
align the slot.
</p>
<p>If this macro is not defined, then <var>basic-align</var> is used when
<var>type</var> is <code>NULL</code>.  Otherwise, <code>LOCAL_ALIGNMENT</code> will
be used.
</p>
<p>This macro is to set alignment of stack slot to the maximum alignment
of all possible modes which the slot may have.
</p>
<p>If the value of this macro has a type, it should be an unsigned type.
</p></dd></dl>

<dl>
<dt><a name="index-LOCAL_005fDECL_005fALIGNMENT"></a>Macro: <strong>LOCAL_DECL_ALIGNMENT</strong> <em>(<var>decl</var>)</em></dt>
<dd><p>If defined, a C expression to compute the alignment for a local
variable <var>decl</var>.
</p>
<p>If this macro is not defined, then
<code>LOCAL_ALIGNMENT (TREE_TYPE (<var>decl</var>), DECL_ALIGN (<var>decl</var>))</code>
is used.
</p>
<p>One use of this macro is to increase alignment of medium-size data to
make it all fit in fewer cache lines.
</p>
<p>If the value of this macro has a type, it should be an unsigned type.
</p></dd></dl>

<dl>
<dt><a name="index-MINIMUM_005fALIGNMENT"></a>Macro: <strong>MINIMUM_ALIGNMENT</strong> <em>(<var>exp</var>, <var>mode</var>, <var>align</var>)</em></dt>
<dd><p>If defined, a C expression to compute the minimum required alignment
for dynamic stack realignment purposes for <var>exp</var> (a type or decl),
<var>mode</var>, assuming normal alignment <var>align</var>.
</p>
<p>If this macro is not defined, then <var>align</var> will be used.
</p></dd></dl>

<dl>
<dt><a name="index-EMPTY_005fFIELD_005fBOUNDARY"></a>Macro: <strong>EMPTY_FIELD_BOUNDARY</strong></dt>
<dd><p>Alignment in bits to be given to a structure bit-field that follows an
empty field such as <code>int : 0;</code>.
</p>
<p>If <code>PCC_BITFIELD_TYPE_MATTERS</code> is true, it overrides this macro.
</p></dd></dl>

<dl>
<dt><a name="index-STRUCTURE_005fSIZE_005fBOUNDARY"></a>Macro: <strong>STRUCTURE_SIZE_BOUNDARY</strong></dt>
<dd><p>Number of bits which any structure or union&rsquo;s size must be a multiple of.
Each structure or union&rsquo;s size is rounded up to a multiple of this.
</p>
<p>If you do not define this macro, the default is the same as
<code>BITS_PER_UNIT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STRICT_005fALIGNMENT"></a>Macro: <strong>STRICT_ALIGNMENT</strong></dt>
<dd><p>Define this macro to be the value 1 if instructions will fail to work
if given data not on the nominal alignment.  If instructions will merely
go slower in that case, define this macro as 0.
</p></dd></dl>

<dl>
<dt><a name="index-PCC_005fBITFIELD_005fTYPE_005fMATTERS"></a>Macro: <strong>PCC_BITFIELD_TYPE_MATTERS</strong></dt>
<dd><p>Define this if you wish to imitate the way many other C compilers handle
alignment of bit-fields and the structures that contain them.
</p>
<p>The behavior is that the type written for a named bit-field (<code>int</code>,
<code>short</code>, or other integer type) imposes an alignment for the entire
structure, as if the structure really did contain an ordinary field of
that type.  In addition, the bit-field is placed within the structure so
that it would fit within such a field, not crossing a boundary for it.
</p>
<p>Thus, on most machines, a named bit-field whose type is written as
<code>int</code> would not cross a four-byte boundary, and would force
four-byte alignment for the whole structure.  (The alignment used may
not be four bytes; it is controlled by the other alignment parameters.)
</p>
<p>An unnamed bit-field will not affect the alignment of the containing
structure.
</p>
<p>If the macro is defined, its definition should be a C expression;
a nonzero value for the expression enables this behavior.
</p>
<p>Note that if this macro is not defined, or its value is zero, some
bit-fields may cross more than one alignment boundary.  The compiler can
support such references if there are &lsquo;<samp>insv</samp>&rsquo;, &lsquo;<samp>extv</samp>&rsquo;, and
&lsquo;<samp>extzv</samp>&rsquo; insns that can directly reference memory.
</p>
<p>The other known way of making bit-fields work is to define
<code>STRUCTURE_SIZE_BOUNDARY</code> as large as <code>BIGGEST_ALIGNMENT</code>.
Then every structure can be accessed with fullwords.
</p>
<p>Unless the machine has bit-field instructions or you define
<code>STRUCTURE_SIZE_BOUNDARY</code> that way, you must define
<code>PCC_BITFIELD_TYPE_MATTERS</code> to have a nonzero value.
</p>
<p>If your aim is to make GCC use the same conventions for laying out
bit-fields as are used by another compiler, here is how to investigate
what the other compiler does.  Compile and run this program:
</p>
<div class="smallexample">
<pre class="smallexample">struct foo1
{
  char x;
  char :0;
  char y;
};

struct foo2
{
  char x;
  int :0;
  char y;
};

main ()
{
  printf (&quot;Size of foo1 is %d\n&quot;,
          sizeof (struct foo1));
  printf (&quot;Size of foo2 is %d\n&quot;,
          sizeof (struct foo2));
  exit (0);
}
</pre></div>

<p>If this prints 2 and 5, then the compiler&rsquo;s behavior is what you would
get from <code>PCC_BITFIELD_TYPE_MATTERS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-BITFIELD_005fNBYTES_005fLIMITED"></a>Macro: <strong>BITFIELD_NBYTES_LIMITED</strong></dt>
<dd><p>Like <code>PCC_BITFIELD_TYPE_MATTERS</code> except that its effect is limited
to aligning a bit-field within the structure.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fALIGN_005fANON_005fBITFIELD"></a>Target Hook: <em>bool</em> <strong>TARGET_ALIGN_ANON_BITFIELD</strong> <em>(void)</em></dt>
<dd><p>When <code>PCC_BITFIELD_TYPE_MATTERS</code> is true this hook will determine
whether unnamed bitfields affect the alignment of the containing
structure.  The hook should return true if the structure should inherit
the alignment requirements of an unnamed bitfield&rsquo;s type.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fNARROW_005fVOLATILE_005fBITFIELD"></a>Target Hook: <em>bool</em> <strong>TARGET_NARROW_VOLATILE_BITFIELD</strong> <em>(void)</em></dt>
<dd><p>This target hook should return <code>true</code> if accesses to volatile bitfields
should use the narrowest mode possible.  It should return <code>false</code> if
these accesses should use the bitfield container type.
</p>
<p>The default is <code>!TARGET_STRICT_ALIGN</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMEMBER_005fTYPE_005fFORCES_005fBLK"></a>Target Hook: <em>bool</em> <strong>TARGET_MEMBER_TYPE_FORCES_BLK</strong> <em>(const_tree <var>field</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Return true if a structure, union or array containing <var>field</var> should
be accessed using <code>BLKMODE</code>.
</p>
<p>If <var>field</var> is the only field in the structure, <var>mode</var> is its
mode, otherwise <var>mode</var> is VOIDmode.  <var>mode</var> is provided in the
case where structures of one field would require the structure&rsquo;s mode to
retain the field&rsquo;s mode.
</p>
<p>Normally, this is not needed.
</p></dd></dl>

<dl>
<dt><a name="index-ROUND_005fTYPE_005fALIGN"></a>Macro: <strong>ROUND_TYPE_ALIGN</strong> <em>(<var>type</var>, <var>computed</var>, <var>specified</var>)</em></dt>
<dd><p>Define this macro as an expression for the alignment of a type (given
by <var>type</var> as a tree node) if the alignment computed in the usual
way is <var>computed</var> and the alignment explicitly specified was
<var>specified</var>.
</p>
<p>The default is to use <var>specified</var> if it is larger; otherwise, use
the smaller of <var>computed</var> and <code>BIGGEST_ALIGNMENT</code>
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fFIXED_005fMODE_005fSIZE"></a>Macro: <strong>MAX_FIXED_MODE_SIZE</strong></dt>
<dd><p>An integer expression for the size in bits of the largest integer
machine mode that should actually be used.  All integer machine modes of
this size or smaller can be used for structures and unions with the
appropriate sizes.  If this macro is undefined, <code>GET_MODE_BITSIZE
(DImode)</code> is assumed.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fSAVEAREA_005fMODE"></a>Macro: <strong>STACK_SAVEAREA_MODE</strong> <em>(<var>save_level</var>)</em></dt>
<dd><p>If defined, an expression of type <code>enum machine_mode</code> that
specifies the mode of the save area operand of a
<code>save_stack_<var>level</var></code> named pattern (see <a href="#Standard-Names">Standard Names</a>).
<var>save_level</var> is one of <code>SAVE_BLOCK</code>, <code>SAVE_FUNCTION</code>, or
<code>SAVE_NONLOCAL</code> and selects which of the three named patterns is
having its mode specified.
</p>
<p>You need not define this macro if it always returns <code>Pmode</code>.  You
would most commonly define this macro if the
<code>save_stack_<var>level</var></code> patterns need to support both a 32- and a
64-bit mode.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fSIZE_005fMODE"></a>Macro: <strong>STACK_SIZE_MODE</strong></dt>
<dd><p>If defined, an expression of type <code>enum machine_mode</code> that
specifies the mode of the size increment operand of an
<code>allocate_stack</code> named pattern (see <a href="#Standard-Names">Standard Names</a>).
</p>
<p>You need not define this macro if it always returns <code>word_mode</code>.
You would most commonly define this macro if the <code>allocate_stack</code>
pattern needs to support both a 32- and a 64-bit mode.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIBGCC_005fCMP_005fRETURN_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_LIBGCC_CMP_RETURN_MODE</strong> <em>(void)</em></dt>
<dd><p>This target hook should return the mode to be used for the return value
of compare instructions expanded to libgcc calls.  If not defined
<code>word_mode</code> is returned which is the right choice for a majority of
targets.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIBGCC_005fSHIFT_005fCOUNT_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_LIBGCC_SHIFT_COUNT_MODE</strong> <em>(void)</em></dt>
<dd><p>This target hook should return the mode to be used for the shift count operand
of shift instructions expanded to libgcc calls.  If not defined
<code>word_mode</code> is returned which is the right choice for a majority of
targets.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUNWIND_005fWORD_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_UNWIND_WORD_MODE</strong> <em>(void)</em></dt>
<dd><p>Return machine mode to be used for <code>_Unwind_Word</code> type.
The default is to use <code>word_mode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ROUND_005fTOWARDS_005fZERO"></a>Macro: <strong>ROUND_TOWARDS_ZERO</strong></dt>
<dd><p>If defined, this macro should be true if the prevailing rounding
mode is towards zero.
</p>
<p>Defining this macro only affects the way <samp>libgcc.a</samp> emulates
floating-point arithmetic.
</p>
<p>Not defining this macro is equivalent to returning zero.
</p></dd></dl>

<dl>
<dt><a name="index-LARGEST_005fEXPONENT_005fIS_005fNORMAL"></a>Macro: <strong>LARGEST_EXPONENT_IS_NORMAL</strong> <em>(<var>size</var>)</em></dt>
<dd><p>This macro should return true if floats with <var>size</var>
bits do not have a NaN or infinity representation, but use the largest
exponent for normal numbers instead.
</p>
<p>Defining this macro only affects the way <samp>libgcc.a</samp> emulates
floating-point arithmetic.
</p>
<p>The default definition of this macro returns false for all sizes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMS_005fBITFIELD_005fLAYOUT_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_MS_BITFIELD_LAYOUT_P</strong> <em>(const_tree <var>record_type</var>)</em></dt>
<dd><p>This target hook returns <code>true</code> if bit-fields in the given
<var>record_type</var> are to be laid out following the rules of Microsoft
Visual C/C++, namely: (i) a bit-field won&rsquo;t share the same storage
unit with the previous bit-field if their underlying types have
different sizes, and the bit-field will be aligned to the highest
alignment of the underlying types of itself and of the previous
bit-field; (ii) a zero-sized bit-field will affect the alignment of
the whole enclosing structure, even if it is unnamed; except that
(iii) a zero-sized bit-field will be disregarded unless it follows
another bit-field of nonzero size.  If this hook returns <code>true</code>,
other macros that control bit-field layout are ignored.
</p>
<p>When a bit-field is inserted into a packed record, the whole size
of the underlying type is used by one or more same-size adjacent
bit-fields (that is, if its long:3, 32 bits is used in the record,
and any additional adjacent long bit-fields are packed into the same
chunk of 32 bits.  However, if the size changes, a new field of that
size is allocated).  In an unpacked record, this is the same as using
alignment, but not equivalent when packing.
</p>
<p>If both MS bit-fields and &lsquo;<samp>__attribute__((packed))</samp>&rsquo; are used,
the latter will take precedence.  If &lsquo;<samp>__attribute__((packed))</samp>&rsquo; is
used on a single field when MS bit-fields are in use, it will take
precedence for that field, but the alignment of the rest of the structure
may affect its placement.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDECIMAL_005fFLOAT_005fSUPPORTED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_DECIMAL_FLOAT_SUPPORTED_P</strong> <em>(void)</em></dt>
<dd><p>Returns true if the target supports decimal floating point.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFIXED_005fPOINT_005fSUPPORTED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_FIXED_POINT_SUPPORTED_P</strong> <em>(void)</em></dt>
<dd><p>Returns true if the target supports fixed-point arithmetic.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXPAND_005fTO_005fRTL_005fHOOK"></a>Target Hook: <em>void</em> <strong>TARGET_EXPAND_TO_RTL_HOOK</strong> <em>(void)</em></dt>
<dd><p>This hook is called just before expansion into rtl, allowing the target
to perform additional initializations or analysis before the expansion.
For example, the rs6000 port uses it to allocate a scratch stack slot
for use in copying SDmode values between memory and floating point
registers whenever the function being expanded has any SDmode
usage.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINSTANTIATE_005fDECLS"></a>Target Hook: <em>void</em> <strong>TARGET_INSTANTIATE_DECLS</strong> <em>(void)</em></dt>
<dd><p>This hook allows the backend to perform additional instantiations on rtl
that are not actually in any insns yet, but will be later.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMANGLE_005fTYPE"></a>Target Hook: <em>const char *</em> <strong>TARGET_MANGLE_TYPE</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>If your target defines any fundamental types, or any types your target
uses should be mangled differently from the default, define this hook
to return the appropriate encoding for these types as part of a C++
mangled name.  The <var>type</var> argument is the tree structure representing
the type to be mangled.  The hook may be applied to trees which are
not target-specific fundamental types; it should return <code>NULL</code>
for all such types, as well as arguments it does not recognize.  If the
return value is not <code>NULL</code>, it must point to a statically-allocated
string constant.
</p>
<p>Target-specific fundamental types might be new fundamental types or
qualified versions of ordinary fundamental types.  Encode new
fundamental types as &lsquo;<samp>u&nbsp;<var>n</var>&nbsp;<var>name</var><!-- /@w --></samp>&rsquo;, where <var>name</var>
is the name used for the type in source code, and <var>n</var> is the
length of <var>name</var> in decimal.  Encode qualified versions of
ordinary types as &lsquo;<samp>U&nbsp;<var>n</var>&nbsp;<var>name</var>&nbsp;<var>code</var><!-- /@w --></samp>&rsquo;, where
<var>name</var> is the name used for the type qualifier in source code,
<var>n</var> is the length of <var>name</var> as above, and <var>code</var> is the
code used to represent the unqualified version of this type.  (See
<code>write_builtin_type</code> in <samp>cp/mangle.c</samp> for the list of
codes.)  In both cases the spaces are for clarity; do not include any
spaces in your string.
</p>
<p>This hook is applied to types prior to typedef resolution.  If the mangled
name for a particular type depends only on that type&rsquo;s main variant, you
can perform typedef resolution yourself using <code>TYPE_MAIN_VARIANT</code>
before mangling.
</p>
<p>The default version of this hook always returns <code>NULL</code>, which is
appropriate for a target that does not define any new fundamental
types.
</p></dd></dl>

<hr>
<a name="Type-Layout"></a>
<div class="header">
<p>
Next: <a href="#Registers" accesskey="n" rel="next">Registers</a>, Previous: <a href="#Storage-Layout" accesskey="p" rel="prev">Storage Layout</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Layout-of-Source-Language-Data-Types"></a>
<h3 class="section">17.6 Layout of Source Language Data Types</h3>

<p>These macros define the sizes and other characteristics of the standard
basic data types used in programs being compiled.  Unlike the macros in
the previous section, these apply to specific features of C and related
languages, rather than to fundamental aspects of storage layout.
</p>
<dl>
<dt><a name="index-INT_005fTYPE_005fSIZE"></a>Macro: <strong>INT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>int</code> on the
target machine.  If you don&rsquo;t define this, the default is one word.
</p></dd></dl>

<dl>
<dt><a name="index-SHORT_005fTYPE_005fSIZE"></a>Macro: <strong>SHORT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>short</code> on the
target machine.  If you don&rsquo;t define this, the default is half a word.
(If this would be less than one storage unit, it is rounded up to one
unit.)
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long</code> on the
target machine.  If you don&rsquo;t define this, the default is one word.
</p></dd></dl>

<dl>
<dt><a name="index-ADA_005fLONG_005fTYPE_005fSIZE"></a>Macro: <strong>ADA_LONG_TYPE_SIZE</strong></dt>
<dd><p>On some machines, the size used for the Ada equivalent of the type
<code>long</code> by a native Ada compiler differs from that used by C.  In
that situation, define this macro to be a C expression to be used for
the size of that type.  If you don&rsquo;t define this, the default is the
value of <code>LONG_TYPE_SIZE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fLONG_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_LONG_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long long</code> on the
target machine.  If you don&rsquo;t define this, the default is two
words.  If you want to support GNU Ada on your machine, the value of this
macro must be at least 64.
</p></dd></dl>

<dl>
<dt><a name="index-CHAR_005fTYPE_005fSIZE"></a>Macro: <strong>CHAR_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>char</code> on the
target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-BOOL_005fTYPE_005fSIZE"></a>Macro: <strong>BOOL_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the C++ type <code>bool</code> and
C99 type <code>_Bool</code> on the target machine.  If you don&rsquo;t define
this, and you probably shouldn&rsquo;t, the default is <code>CHAR_TYPE_SIZE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FLOAT_005fTYPE_005fSIZE"></a>Macro: <strong>FLOAT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>float</code> on the
target machine.  If you don&rsquo;t define this, the default is one word.
</p></dd></dl>

<dl>
<dt><a name="index-DOUBLE_005fTYPE_005fSIZE"></a>Macro: <strong>DOUBLE_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>double</code> on the
target machine.  If you don&rsquo;t define this, the default is two
words.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fDOUBLE_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_DOUBLE_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long double</code> on
the target machine.  If you don&rsquo;t define this, the default is two
words.
</p></dd></dl>

<dl>
<dt><a name="index-SHORT_005fFRACT_005fTYPE_005fSIZE"></a>Macro: <strong>SHORT_FRACT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>short _Fract</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FRACT_005fTYPE_005fSIZE"></a>Macro: <strong>FRACT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>_Fract</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 2</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fFRACT_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_FRACT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long _Fract</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 4</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fLONG_005fFRACT_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_LONG_FRACT_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long long _Fract</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 8</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SHORT_005fACCUM_005fTYPE_005fSIZE"></a>Macro: <strong>SHORT_ACCUM_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>short _Accum</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 2</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ACCUM_005fTYPE_005fSIZE"></a>Macro: <strong>ACCUM_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>_Accum</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 4</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fACCUM_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_ACCUM_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long _Accum</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 8</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LONG_005fLONG_005fACCUM_005fTYPE_005fSIZE"></a>Macro: <strong>LONG_LONG_ACCUM_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the type <code>long long _Accum</code> on
the target machine.  If you don&rsquo;t define this, the default is
<code>BITS_PER_UNIT * 16</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fLONG_005fDOUBLE_005fTYPE_005fSIZE"></a>Macro: <strong>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</strong></dt>
<dd><p>Define this macro if <code>LONG_DOUBLE_TYPE_SIZE</code> is not constant or
if you want routines in <samp>libgcc2.a</samp> for a size other than
<code>LONG_DOUBLE_TYPE_SIZE</code>.  If you don&rsquo;t define this, the
default is <code>LONG_DOUBLE_TYPE_SIZE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fHAS_005fDF_005fMODE"></a>Macro: <strong>LIBGCC2_HAS_DF_MODE</strong></dt>
<dd><p>Define this macro if neither <code>DOUBLE_TYPE_SIZE</code> nor
<code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code> is
<code>DFmode</code> but you want <code>DFmode</code> routines in <samp>libgcc2.a</samp>
anyway.  If you don&rsquo;t define this and either <code>DOUBLE_TYPE_SIZE</code>
or <code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code> is 64 then the default is 1,
otherwise it is 0.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fHAS_005fXF_005fMODE"></a>Macro: <strong>LIBGCC2_HAS_XF_MODE</strong></dt>
<dd><p>Define this macro if <code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code> is not
<code>XFmode</code> but you want <code>XFmode</code> routines in <samp>libgcc2.a</samp>
anyway.  If you don&rsquo;t define this and <code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code>
is 80 then the default is 1, otherwise it is 0.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fHAS_005fTF_005fMODE"></a>Macro: <strong>LIBGCC2_HAS_TF_MODE</strong></dt>
<dd><p>Define this macro if <code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code> is not
<code>TFmode</code> but you want <code>TFmode</code> routines in <samp>libgcc2.a</samp>
anyway.  If you don&rsquo;t define this and <code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code>
is 128 then the default is 1, otherwise it is 0.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fGNU_005fPREFIX"></a>Macro: <strong>LIBGCC2_GNU_PREFIX</strong></dt>
<dd><p>This macro corresponds to the <code>TARGET_LIBFUNC_GNU_PREFIX</code> target
hook and should be defined if that hook is overriden to be true.  It
causes function names in libgcc to be changed to use a <code>__gnu_</code>
prefix for their name rather than the default <code>__</code>.  A port which
uses this macro should also arrange to use <samp>t-gnu-prefix</samp> in
the libgcc <samp>config.host</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-SF_005fSIZE"></a>Macro: <strong>SF_SIZE</strong></dt>
<dt><a name="index-DF_005fSIZE"></a>Macro: <strong>DF_SIZE</strong></dt>
<dt><a name="index-XF_005fSIZE"></a>Macro: <strong>XF_SIZE</strong></dt>
<dt><a name="index-TF_005fSIZE"></a>Macro: <strong>TF_SIZE</strong></dt>
<dd><p>Define these macros to be the size in bits of the mantissa of
<code>SFmode</code>, <code>DFmode</code>, <code>XFmode</code> and <code>TFmode</code> values,
if the defaults in <samp>libgcc2.h</samp> are inappropriate.  By default,
<code>FLT_MANT_DIG</code> is used for <code>SF_SIZE</code>, <code>LDBL_MANT_DIG</code>
for <code>XF_SIZE</code> and <code>TF_SIZE</code>, and <code>DBL_MANT_DIG</code> or
<code>LDBL_MANT_DIG</code> for <code>DF_SIZE</code> according to whether
<code>DOUBLE_TYPE_SIZE</code> or
<code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code> is 64.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFLT_005fEVAL_005fMETHOD"></a>Macro: <strong>TARGET_FLT_EVAL_METHOD</strong></dt>
<dd><p>A C expression for the value for <code>FLT_EVAL_METHOD</code> in <samp>float.h</samp>,
assuming, if applicable, that the floating-point control word is in its
default state.  If you do not define this macro the value of
<code>FLT_EVAL_METHOD</code> will be zero.
</p></dd></dl>

<dl>
<dt><a name="index-WIDEST_005fHARDWARE_005fFP_005fSIZE"></a>Macro: <strong>WIDEST_HARDWARE_FP_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the widest floating-point format
supported by the hardware.  If you define this macro, you must specify a
value less than or equal to the value of <code>LONG_DOUBLE_TYPE_SIZE</code>.
If you do not define this macro, the value of <code>LONG_DOUBLE_TYPE_SIZE</code>
is the default.
</p></dd></dl>

<dl>
<dt><a name="index-DEFAULT_005fSIGNED_005fCHAR"></a>Macro: <strong>DEFAULT_SIGNED_CHAR</strong></dt>
<dd><p>An expression whose value is 1 or 0, according to whether the type
<code>char</code> should be signed or unsigned by default.  The user can
always override this default with the options <samp>-fsigned-char</samp>
and <samp>-funsigned-char</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDEFAULT_005fSHORT_005fENUMS"></a>Target Hook: <em>bool</em> <strong>TARGET_DEFAULT_SHORT_ENUMS</strong> <em>(void)</em></dt>
<dd><p>This target hook should return true if the compiler should give an
<code>enum</code> type only as many bytes as it takes to represent the range
of possible values of that type.  It should return false if all
<code>enum</code> types should be allocated like <code>int</code>.
</p>
<p>The default is to return false.
</p></dd></dl>

<dl>
<dt><a name="index-SIZE_005fTYPE"></a>Macro: <strong>SIZE_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type to use
for size values.  The typedef name <code>size_t</code> is defined using the
contents of the string.
</p>
<p>The string can contain more than one keyword.  If so, separate them with
spaces, and write first any length keyword, then <code>unsigned</code> if
appropriate, and finally <code>int</code>.  The string must exactly match one
of the data type names defined in the function
<code>c_common_nodes_and_builtins</code> in the file <samp>c-family/c-common.c</samp>.
You may not omit <code>int</code> or change the order&mdash;that would cause the
compiler to crash on startup.
</p>
<p>If you don&rsquo;t define this macro, the default is <code>&quot;long unsigned
int&quot;</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SIZETYPE"></a>Macro: <strong>SIZETYPE</strong></dt>
<dd><p>GCC defines internal types (<code>sizetype</code>, <code>ssizetype</code>,
<code>bitsizetype</code> and <code>sbitsizetype</code>) for expressions
dealing with size.  This macro is a C expression for a string describing
the name of the data type from which the precision of <code>sizetype</code>
is extracted.
</p>
<p>The string has the same restrictions as <code>SIZE_TYPE</code> string.
</p>
<p>If you don&rsquo;t define this macro, the default is <code>SIZE_TYPE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PTRDIFF_005fTYPE"></a>Macro: <strong>PTRDIFF_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type to use
for the result of subtracting two pointers.  The typedef name
<code>ptrdiff_t</code> is defined using the contents of the string.  See
<code>SIZE_TYPE</code> above for more information.
</p>
<p>If you don&rsquo;t define this macro, the default is <code>&quot;long int&quot;</code>.
</p></dd></dl>

<dl>
<dt><a name="index-WCHAR_005fTYPE"></a>Macro: <strong>WCHAR_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type to use
for wide characters.  The typedef name <code>wchar_t</code> is defined using
the contents of the string.  See <code>SIZE_TYPE</code> above for more
information.
</p>
<p>If you don&rsquo;t define this macro, the default is <code>&quot;int&quot;</code>.
</p></dd></dl>

<dl>
<dt><a name="index-WCHAR_005fTYPE_005fSIZE"></a>Macro: <strong>WCHAR_TYPE_SIZE</strong></dt>
<dd><p>A C expression for the size in bits of the data type for wide
characters.  This is used in <code>cpp</code>, which cannot make use of
<code>WCHAR_TYPE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-WINT_005fTYPE"></a>Macro: <strong>WINT_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type to
use for wide characters passed to <code>printf</code> and returned from
<code>getwc</code>.  The typedef name <code>wint_t</code> is defined using the
contents of the string.  See <code>SIZE_TYPE</code> above for more
information.
</p>
<p>If you don&rsquo;t define this macro, the default is <code>&quot;unsigned int&quot;</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INTMAX_005fTYPE"></a>Macro: <strong>INTMAX_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type that
can represent any value of any standard or extended signed integer type.
The typedef name <code>intmax_t</code> is defined using the contents of the
string.  See <code>SIZE_TYPE</code> above for more information.
</p>
<p>If you don&rsquo;t define this macro, the default is the first of
<code>&quot;int&quot;</code>, <code>&quot;long int&quot;</code>, or <code>&quot;long long int&quot;</code> that has as
much precision as <code>long long int</code>.
</p></dd></dl>

<dl>
<dt><a name="index-UINTMAX_005fTYPE"></a>Macro: <strong>UINTMAX_TYPE</strong></dt>
<dd><p>A C expression for a string describing the name of the data type that
can represent any value of any standard or extended unsigned integer
type.  The typedef name <code>uintmax_t</code> is defined using the contents
of the string.  See <code>SIZE_TYPE</code> above for more information.
</p>
<p>If you don&rsquo;t define this macro, the default is the first of
<code>&quot;unsigned int&quot;</code>, <code>&quot;long unsigned int&quot;</code>, or <code>&quot;long long
unsigned int&quot;</code> that has as much precision as <code>long long unsigned
int</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SIG_005fATOMIC_005fTYPE"></a>Macro: <strong>SIG_ATOMIC_TYPE</strong></dt>
<dt><a name="index-INT8_005fTYPE"></a>Macro: <strong>INT8_TYPE</strong></dt>
<dt><a name="index-INT16_005fTYPE"></a>Macro: <strong>INT16_TYPE</strong></dt>
<dt><a name="index-INT32_005fTYPE"></a>Macro: <strong>INT32_TYPE</strong></dt>
<dt><a name="index-INT64_005fTYPE"></a>Macro: <strong>INT64_TYPE</strong></dt>
<dt><a name="index-UINT8_005fTYPE"></a>Macro: <strong>UINT8_TYPE</strong></dt>
<dt><a name="index-UINT16_005fTYPE"></a>Macro: <strong>UINT16_TYPE</strong></dt>
<dt><a name="index-UINT32_005fTYPE"></a>Macro: <strong>UINT32_TYPE</strong></dt>
<dt><a name="index-UINT64_005fTYPE"></a>Macro: <strong>UINT64_TYPE</strong></dt>
<dt><a name="index-INT_005fLEAST8_005fTYPE"></a>Macro: <strong>INT_LEAST8_TYPE</strong></dt>
<dt><a name="index-INT_005fLEAST16_005fTYPE"></a>Macro: <strong>INT_LEAST16_TYPE</strong></dt>
<dt><a name="index-INT_005fLEAST32_005fTYPE"></a>Macro: <strong>INT_LEAST32_TYPE</strong></dt>
<dt><a name="index-INT_005fLEAST64_005fTYPE"></a>Macro: <strong>INT_LEAST64_TYPE</strong></dt>
<dt><a name="index-UINT_005fLEAST8_005fTYPE"></a>Macro: <strong>UINT_LEAST8_TYPE</strong></dt>
<dt><a name="index-UINT_005fLEAST16_005fTYPE"></a>Macro: <strong>UINT_LEAST16_TYPE</strong></dt>
<dt><a name="index-UINT_005fLEAST32_005fTYPE"></a>Macro: <strong>UINT_LEAST32_TYPE</strong></dt>
<dt><a name="index-UINT_005fLEAST64_005fTYPE"></a>Macro: <strong>UINT_LEAST64_TYPE</strong></dt>
<dt><a name="index-INT_005fFAST8_005fTYPE"></a>Macro: <strong>INT_FAST8_TYPE</strong></dt>
<dt><a name="index-INT_005fFAST16_005fTYPE"></a>Macro: <strong>INT_FAST16_TYPE</strong></dt>
<dt><a name="index-INT_005fFAST32_005fTYPE"></a>Macro: <strong>INT_FAST32_TYPE</strong></dt>
<dt><a name="index-INT_005fFAST64_005fTYPE"></a>Macro: <strong>INT_FAST64_TYPE</strong></dt>
<dt><a name="index-UINT_005fFAST8_005fTYPE"></a>Macro: <strong>UINT_FAST8_TYPE</strong></dt>
<dt><a name="index-UINT_005fFAST16_005fTYPE"></a>Macro: <strong>UINT_FAST16_TYPE</strong></dt>
<dt><a name="index-UINT_005fFAST32_005fTYPE"></a>Macro: <strong>UINT_FAST32_TYPE</strong></dt>
<dt><a name="index-UINT_005fFAST64_005fTYPE"></a>Macro: <strong>UINT_FAST64_TYPE</strong></dt>
<dt><a name="index-INTPTR_005fTYPE"></a>Macro: <strong>INTPTR_TYPE</strong></dt>
<dt><a name="index-UINTPTR_005fTYPE"></a>Macro: <strong>UINTPTR_TYPE</strong></dt>
<dd><p>C expressions for the standard types <code>sig_atomic_t</code>,
<code>int8_t</code>, <code>int16_t</code>, <code>int32_t</code>, <code>int64_t</code>,
<code>uint8_t</code>, <code>uint16_t</code>, <code>uint32_t</code>, <code>uint64_t</code>,
<code>int_least8_t</code>, <code>int_least16_t</code>, <code>int_least32_t</code>,
<code>int_least64_t</code>, <code>uint_least8_t</code>, <code>uint_least16_t</code>,
<code>uint_least32_t</code>, <code>uint_least64_t</code>, <code>int_fast8_t</code>,
<code>int_fast16_t</code>, <code>int_fast32_t</code>, <code>int_fast64_t</code>,
<code>uint_fast8_t</code>, <code>uint_fast16_t</code>, <code>uint_fast32_t</code>,
<code>uint_fast64_t</code>, <code>intptr_t</code>, and <code>uintptr_t</code>.  See
<code>SIZE_TYPE</code> above for more information.
</p>
<p>If any of these macros evaluates to a null pointer, the corresponding
type is not supported; if GCC is configured to provide
<code>&lt;stdint.h&gt;</code> in such a case, the header provided may not conform
to C99, depending on the type in question.  The defaults for all of
these macros are null pointers.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPTRMEMFUNC_005fVBIT_005fLOCATION"></a>Macro: <strong>TARGET_PTRMEMFUNC_VBIT_LOCATION</strong></dt>
<dd><p>The C++ compiler represents a pointer-to-member-function with a struct
that looks like:
</p>
<div class="smallexample">
<pre class="smallexample">  struct {
    union {
      void (*fn)();
      ptrdiff_t vtable_index;
    };
    ptrdiff_t delta;
  };
</pre></div>

<p>The C++ compiler must use one bit to indicate whether the function that
will be called through a pointer-to-member-function is virtual.
Normally, we assume that the low-order bit of a function pointer must
always be zero.  Then, by ensuring that the vtable_index is odd, we can
distinguish which variant of the union is in use.  But, on some
platforms function pointers can be odd, and so this doesn&rsquo;t work.  In
that case, we use the low-order bit of the <code>delta</code> field, and shift
the remainder of the <code>delta</code> field to the left.
</p>
<p>GCC will automatically make the right selection about where to store
this bit using the <code>FUNCTION_BOUNDARY</code> setting for your platform.
However, some platforms such as ARM/Thumb have <code>FUNCTION_BOUNDARY</code>
set such that functions always start at even addresses, but the lowest
bit of pointers to functions indicate whether the function at that
address is in ARM or Thumb mode.  If this is the case of your
architecture, you should define this macro to
<code>ptrmemfunc_vbit_in_delta</code>.
</p>
<p>In general, you should not have to define this macro.  On architectures
in which function addresses are always even, according to
<code>FUNCTION_BOUNDARY</code>, GCC will automatically define this macro to
<code>ptrmemfunc_vbit_in_pfn</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVTABLE_005fUSES_005fDESCRIPTORS"></a>Macro: <strong>TARGET_VTABLE_USES_DESCRIPTORS</strong></dt>
<dd><p>Normally, the C++ compiler uses function pointers in vtables.  This
macro allows the target to change to use &ldquo;function descriptors&rdquo;
instead.  Function descriptors are found on targets for whom a
function pointer is actually a small data structure.  Normally the
data structure consists of the actual code address plus a data
pointer to which the function&rsquo;s data is relative.
</p>
<p>If vtables are used, the value of this macro should be the number
of words that the function descriptor occupies.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVTABLE_005fENTRY_005fALIGN"></a>Macro: <strong>TARGET_VTABLE_ENTRY_ALIGN</strong></dt>
<dd><p>By default, the vtable entries are void pointers, the so the alignment
is the same as pointer alignment.  The value of this macro specifies
the alignment of the vtable entry in bits.  It should be defined only
when special alignment is necessary. */
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVTABLE_005fDATA_005fENTRY_005fDISTANCE"></a>Macro: <strong>TARGET_VTABLE_DATA_ENTRY_DISTANCE</strong></dt>
<dd><p>There are a few non-descriptor entries in the vtable at offsets below
zero.  If these entries must be padded (say, to preserve the alignment
specified by <code>TARGET_VTABLE_ENTRY_ALIGN</code>), set this to the number
of words in each data entry.
</p></dd></dl>

<hr>
<a name="Registers"></a>
<div class="header">
<p>
Next: <a href="#Register-Classes" accesskey="n" rel="next">Register Classes</a>, Previous: <a href="#Type-Layout" accesskey="p" rel="prev">Type Layout</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Register-Usage"></a>
<h3 class="section">17.7 Register Usage</h3>
<a name="index-register-usage"></a>

<p>This section explains how to describe what registers the target machine
has, and how (in general) they can be used.
</p>
<p>The description of which registers a specific instruction can use is
done with register classes; see <a href="#Register-Classes">Register Classes</a>.  For information
on using registers to access a stack frame, see <a href="#Frame-Registers">Frame Registers</a>.
For passing values in registers, see <a href="#Register-Arguments">Register Arguments</a>.
For returning values in registers, see <a href="#Scalar-Return">Scalar Return</a>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Register-Basics" accesskey="1">Register Basics</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Number and kinds of registers.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Allocation-Order" accesskey="2">Allocation Order</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Order in which registers are allocated.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Values-in-Registers" accesskey="3">Values in Registers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">What kinds of values each reg can hold.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Leaf-Functions" accesskey="4">Leaf Functions</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Renumbering registers for leaf functions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Stack-Registers" accesskey="5">Stack Registers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Handling a register stack such as 80387.
</td></tr>
</table>

<hr>
<a name="Register-Basics"></a>
<div class="header">
<p>
Next: <a href="#Allocation-Order" accesskey="n" rel="next">Allocation Order</a>, Up: <a href="#Registers" accesskey="u" rel="up">Registers</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Basic-Characteristics-of-Registers"></a>
<h4 class="subsection">17.7.1 Basic Characteristics of Registers</h4>

<p>Registers have various characteristics.
</p>
<dl>
<dt><a name="index-FIRST_005fPSEUDO_005fREGISTER"></a>Macro: <strong>FIRST_PSEUDO_REGISTER</strong></dt>
<dd><p>Number of hardware registers known to the compiler.  They receive
numbers 0 through <code>FIRST_PSEUDO_REGISTER-1</code>; thus, the first
pseudo register&rsquo;s number really is assigned the number
<code>FIRST_PSEUDO_REGISTER</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FIXED_005fREGISTERS"></a>Macro: <strong>FIXED_REGISTERS</strong></dt>
<dd><a name="index-fixed-register"></a>
<p>An initializer that says which registers are used for fixed purposes
all throughout the compiled code and are therefore not available for
general allocation.  These would include the stack pointer, the frame
pointer (except on machines where that can be used as a general
register when no frame pointer is needed), the program counter on
machines where that is considered one of the addressable registers,
and any other numbered register with a standard use.
</p>
<p>This information is expressed as a sequence of numbers, separated by
commas and surrounded by braces.  The <var>n</var>th number is 1 if
register <var>n</var> is fixed, 0 otherwise.
</p>
<p>The table initialized from this macro, and the table initialized by
the following one, may be overridden at run time either automatically,
by the actions of the macro <code>CONDITIONAL_REGISTER_USAGE</code>, or by
the user with the command options <samp>-ffixed-<var>reg</var></samp>,
<samp>-fcall-used-<var>reg</var></samp> and <samp>-fcall-saved-<var>reg</var></samp>.
</p></dd></dl>

<dl>
<dt><a name="index-CALL_005fUSED_005fREGISTERS"></a>Macro: <strong>CALL_USED_REGISTERS</strong></dt>
<dd><a name="index-call_002dused-register"></a>
<a name="index-call_002dclobbered-register"></a>
<a name="index-call_002dsaved-register"></a>
<p>Like <code>FIXED_REGISTERS</code> but has 1 for each register that is
clobbered (in general) by function calls as well as for fixed
registers.  This macro therefore identifies the registers that are not
available for general allocation of values that must live across
function calls.
</p>
<p>If a register has 0 in <code>CALL_USED_REGISTERS</code>, the compiler
automatically saves it on function entry and restores it on function
exit, if the register is used within the function.
</p></dd></dl>

<dl>
<dt><a name="index-CALL_005fREALLY_005fUSED_005fREGISTERS"></a>Macro: <strong>CALL_REALLY_USED_REGISTERS</strong></dt>
<dd><a name="index-call_002dused-register-1"></a>
<a name="index-call_002dclobbered-register-1"></a>
<a name="index-call_002dsaved-register-1"></a>
<p>Like <code>CALL_USED_REGISTERS</code> except this macro doesn&rsquo;t require
that the entire set of <code>FIXED_REGISTERS</code> be included.
(<code>CALL_USED_REGISTERS</code> must be a superset of <code>FIXED_REGISTERS</code>).
This macro is optional.  If not specified, it defaults to the value
of <code>CALL_USED_REGISTERS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fCALL_005fPART_005fCLOBBERED"></a>Macro: <strong>HARD_REGNO_CALL_PART_CLOBBERED</strong> <em>(<var>regno</var>, <var>mode</var>)</em></dt>
<dd><a name="index-call_002dused-register-2"></a>
<a name="index-call_002dclobbered-register-2"></a>
<a name="index-call_002dsaved-register-2"></a>
<p>A C expression that is nonzero if it is not permissible to store a
value of mode <var>mode</var> in hard register number <var>regno</var> across a
call without some part of it being clobbered.  For most machines this
macro need not be defined.  It is only required for machines that do not
preserve the entire contents of a register across a call.
</p></dd></dl>

<a name="index-fixed_005fregs"></a>
<a name="index-call_005fused_005fregs"></a>
<a name="index-global_005fregs"></a>
<a name="index-reg_005fnames"></a>
<a name="index-reg_005fclass_005fcontents"></a>
<dl>
<dt><a name="index-TARGET_005fCONDITIONAL_005fREGISTER_005fUSAGE"></a>Target Hook: <em>void</em> <strong>TARGET_CONDITIONAL_REGISTER_USAGE</strong> <em>(void)</em></dt>
<dd><p>This hook may conditionally modify five variables
<code>fixed_regs</code>, <code>call_used_regs</code>, <code>global_regs</code>,
<code>reg_names</code>, and <code>reg_class_contents</code>, to take into account
any dependence of these register sets on target flags.  The first three
of these are of type <code>char []</code> (interpreted as Boolean vectors).
<code>global_regs</code> is a <code>const char *[]</code>, and
<code>reg_class_contents</code> is a <code>HARD_REG_SET</code>.  Before the macro is
called, <code>fixed_regs</code>, <code>call_used_regs</code>,
<code>reg_class_contents</code>, and <code>reg_names</code> have been initialized
from <code>FIXED_REGISTERS</code>, <code>CALL_USED_REGISTERS</code>,
<code>REG_CLASS_CONTENTS</code>, and <code>REGISTER_NAMES</code>, respectively.
<code>global_regs</code> has been cleared, and any <samp>-ffixed-<var>reg</var></samp>,
<samp>-fcall-used-<var>reg</var></samp> and <samp>-fcall-saved-<var>reg</var></samp>
command options have been applied.
</p>
<a name="index-disabling-certain-registers"></a>
<a name="index-controlling-register-usage"></a>
<p>If the usage of an entire class of registers depends on the target
flags, you may indicate this to GCC by using this macro to modify
<code>fixed_regs</code> and <code>call_used_regs</code> to 1 for each of the
registers in the classes which should not be used by GCC.  Also define
the macro <code>REG_CLASS_FROM_LETTER</code> / <code>REG_CLASS_FROM_CONSTRAINT</code>
to return <code>NO_REGS</code> if it
is called with a letter for a class that shouldn&rsquo;t be used.
</p>
<p>(However, if this class is not included in <code>GENERAL_REGS</code> and all
of the insn patterns whose constraints permit this class are
controlled by target switches, then GCC will automatically avoid using
these registers when the target switches are opposed to them.)
</p></dd></dl>

<dl>
<dt><a name="index-INCOMING_005fREGNO"></a>Macro: <strong>INCOMING_REGNO</strong> <em>(<var>out</var>)</em></dt>
<dd><p>Define this macro if the target machine has register windows.  This C
expression returns the register number as seen by the called function
corresponding to the register number <var>out</var> as seen by the calling
function.  Return <var>out</var> if register number <var>out</var> is not an
outbound register.
</p></dd></dl>

<dl>
<dt><a name="index-OUTGOING_005fREGNO"></a>Macro: <strong>OUTGOING_REGNO</strong> <em>(<var>in</var>)</em></dt>
<dd><p>Define this macro if the target machine has register windows.  This C
expression returns the register number as seen by the calling function
corresponding to the register number <var>in</var> as seen by the called
function.  Return <var>in</var> if register number <var>in</var> is not an inbound
register.
</p></dd></dl>

<dl>
<dt><a name="index-LOCAL_005fREGNO"></a>Macro: <strong>LOCAL_REGNO</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>Define this macro if the target machine has register windows.  This C
expression returns true if the register is call-saved but is in the
register window.  Unlike most call-saved registers, such registers
need not be explicitly restored on function exit or during non-local
gotos.
</p></dd></dl>

<dl>
<dt><a name="index-PC_005fREGNUM"></a>Macro: <strong>PC_REGNUM</strong></dt>
<dd><p>If the program counter has a register number, define this as that
register number.  Otherwise, do not define it.
</p></dd></dl>

<hr>
<a name="Allocation-Order"></a>
<div class="header">
<p>
Next: <a href="#Values-in-Registers" accesskey="n" rel="next">Values in Registers</a>, Previous: <a href="#Register-Basics" accesskey="p" rel="prev">Register Basics</a>, Up: <a href="#Registers" accesskey="u" rel="up">Registers</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Order-of-Allocation-of-Registers"></a>
<h4 class="subsection">17.7.2 Order of Allocation of Registers</h4>
<a name="index-order-of-register-allocation"></a>
<a name="index-register-allocation-order"></a>

<p>Registers are allocated in order.
</p>
<dl>
<dt><a name="index-REG_005fALLOC_005fORDER"></a>Macro: <strong>REG_ALLOC_ORDER</strong></dt>
<dd><p>If defined, an initializer for a vector of integers, containing the
numbers of hard registers in the order in which GCC should prefer
to use them (from most preferred to least).
</p>
<p>If this macro is not defined, registers are used lowest numbered first
(all else being equal).
</p>
<p>One use of this macro is on machines where the highest numbered
registers must always be saved and the save-multiple-registers
instruction supports only sequences of consecutive registers.  On such
machines, define <code>REG_ALLOC_ORDER</code> to be an initializer that lists
the highest numbered allocable register first.
</p></dd></dl>

<dl>
<dt><a name="index-ADJUST_005fREG_005fALLOC_005fORDER"></a>Macro: <strong>ADJUST_REG_ALLOC_ORDER</strong></dt>
<dd><p>A C statement (sans semicolon) to choose the order in which to allocate
hard registers for pseudo-registers local to a basic block.
</p>
<p>Store the desired register order in the array <code>reg_alloc_order</code>.
Element 0 should be the register to allocate first; element 1, the next
register; and so on.
</p>
<p>The macro body should not assume anything about the contents of
<code>reg_alloc_order</code> before execution of the macro.
</p>
<p>On most machines, it is not necessary to define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-HONOR_005fREG_005fALLOC_005fORDER"></a>Macro: <strong>HONOR_REG_ALLOC_ORDER</strong></dt>
<dd><p>Normally, IRA tries to estimate the costs for saving a register in the
prologue and restoring it in the epilogue.  This discourages it from
using call-saved registers.  If a machine wants to ensure that IRA
allocates registers in the order given by REG_ALLOC_ORDER even if some
call-saved registers appear earlier than call-used ones, this macro
should be defined.
</p></dd></dl>

<dl>
<dt><a name="index-IRA_005fHARD_005fREGNO_005fADD_005fCOST_005fMULTIPLIER"></a>Macro: <strong>IRA_HARD_REGNO_ADD_COST_MULTIPLIER</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>In some case register allocation order is not enough for the
Integrated Register Allocator (<acronym>IRA</acronym>) to generate a good code.
If this macro is defined, it should return a floating point value
based on <var>regno</var>.  The cost of using <var>regno</var> for a pseudo will
be increased by approximately the pseudo&rsquo;s usage frequency times the
value returned by this macro.  Not defining this macro is equivalent
to having it always return <code>0.0</code>.
</p>
<p>On most machines, it is not necessary to define this macro.
</p></dd></dl>

<hr>
<a name="Values-in-Registers"></a>
<div class="header">
<p>
Next: <a href="#Leaf-Functions" accesskey="n" rel="next">Leaf Functions</a>, Previous: <a href="#Allocation-Order" accesskey="p" rel="prev">Allocation Order</a>, Up: <a href="#Registers" accesskey="u" rel="up">Registers</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="How-Values-Fit-in-Registers"></a>
<h4 class="subsection">17.7.3 How Values Fit in Registers</h4>

<p>This section discusses the macros that describe which kinds of values
(specifically, which machine modes) each register can hold, and how many
consecutive registers are needed for a given mode.
</p>
<dl>
<dt><a name="index-HARD_005fREGNO_005fNREGS"></a>Macro: <strong>HARD_REGNO_NREGS</strong> <em>(<var>regno</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression for the number of consecutive hard registers, starting
at register number <var>regno</var>, required to hold a value of mode
<var>mode</var>.  This macro must never return zero, even if a register
cannot hold the requested mode - indicate that with HARD_REGNO_MODE_OK
and/or CANNOT_CHANGE_MODE_CLASS instead.
</p>
<p>On a machine where all registers are exactly one word, a suitable
definition of this macro is
</p>
<div class="smallexample">
<pre class="smallexample">#define HARD_REGNO_NREGS(REGNO, MODE)            \
   ((GET_MODE_SIZE (MODE) + UNITS_PER_WORD - 1)  \
    / UNITS_PER_WORD)
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fNREGS_005fHAS_005fPADDING"></a>Macro: <strong>HARD_REGNO_NREGS_HAS_PADDING</strong> <em>(<var>regno</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression that is nonzero if a value of mode <var>mode</var>, stored
in memory, ends with padding that causes it to take up more space than
in registers starting at register number <var>regno</var> (as determined by
multiplying GCC&rsquo;s notion of the size of the register when containing
this mode by the number of registers returned by
<code>HARD_REGNO_NREGS</code>).  By default this is zero.
</p>
<p>For example, if a floating-point value is stored in three 32-bit
registers but takes up 128 bits in memory, then this would be
nonzero.
</p>
<p>This macros only needs to be defined if there are cases where
<code>subreg_get_info</code>
would otherwise wrongly determine that a <code>subreg</code> can be
represented by an offset to the register number, when in fact such a
<code>subreg</code> would contain some of the padding not stored in
registers and so not be representable.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fNREGS_005fWITH_005fPADDING"></a>Macro: <strong>HARD_REGNO_NREGS_WITH_PADDING</strong> <em>(<var>regno</var>, <var>mode</var>)</em></dt>
<dd><p>For values of <var>regno</var> and <var>mode</var> for which
<code>HARD_REGNO_NREGS_HAS_PADDING</code> returns nonzero, a C expression
returning the greater number of registers required to hold the value
including any padding.  In the example above, the value would be four.
</p></dd></dl>

<dl>
<dt><a name="index-REGMODE_005fNATURAL_005fSIZE"></a>Macro: <strong>REGMODE_NATURAL_SIZE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>Define this macro if the natural size of registers that hold values
of mode <var>mode</var> is not the word size.  It is a C expression that
should give the natural size in bytes for the specified mode.  It is
used by the register allocator to try to optimize its results.  This
happens for example on SPARC 64-bit where the natural size of
floating-point registers is still 32-bit.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fMODE_005fOK"></a>Macro: <strong>HARD_REGNO_MODE_OK</strong> <em>(<var>regno</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression that is nonzero if it is permissible to store a value
of mode <var>mode</var> in hard register number <var>regno</var> (or in several
registers starting with that one).  For a machine where all registers
are equivalent, a suitable definition is
</p>
<div class="smallexample">
<pre class="smallexample">#define HARD_REGNO_MODE_OK(REGNO, MODE) 1
</pre></div>

<p>You need not include code to check for the numbers of fixed registers,
because the allocation mechanism considers them to be always occupied.
</p>
<a name="index-register-pairs"></a>
<p>On some machines, double-precision values must be kept in even/odd
register pairs.  You can implement that by defining this macro to reject
odd register numbers for such modes.
</p>
<p>The minimum requirement for a mode to be OK in a register is that the
&lsquo;<samp>mov<var>mode</var></samp>&rsquo; instruction pattern support moves between the
register and other hard register in the same class and that moving a
value into the register and back out not alter it.
</p>
<p>Since the same instruction used to move <code>word_mode</code> will work for
all narrower integer modes, it is not necessary on any machine for
<code>HARD_REGNO_MODE_OK</code> to distinguish between these modes, provided
you define patterns &lsquo;<samp>movhi</samp>&rsquo;, etc., to take advantage of this.  This
is useful because of the interaction between <code>HARD_REGNO_MODE_OK</code>
and <code>MODES_TIEABLE_P</code>; it is very desirable for all integer modes
to be tieable.
</p>
<p>Many machines have special registers for floating point arithmetic.
Often people assume that floating point machine modes are allowed only
in floating point registers.  This is not true.  Any registers that
can hold integers can safely <em>hold</em> a floating point machine
mode, whether or not floating arithmetic can be done on it in those
registers.  Integer move instructions can be used to move the values.
</p>
<p>On some machines, though, the converse is true: fixed-point machine
modes may not go in floating registers.  This is true if the floating
registers normalize any value stored in them, because storing a
non-floating value there would garble it.  In this case,
<code>HARD_REGNO_MODE_OK</code> should reject fixed-point machine modes in
floating registers.  But if the floating registers do not automatically
normalize, if you can store any bit pattern in one and retrieve it
unchanged without a trap, then any machine mode may go in a floating
register, so you can define this macro to say so.
</p>
<p>The primary significance of special floating registers is rather that
they are the registers acceptable in floating point arithmetic
instructions.  However, this is of no concern to
<code>HARD_REGNO_MODE_OK</code>.  You handle it by writing the proper
constraints for those instructions.
</p>
<p>On some machines, the floating registers are especially slow to access,
so that it is better to store a value in a stack frame than in such a
register if floating point arithmetic is not being done.  As long as the
floating registers are not in class <code>GENERAL_REGS</code>, they will not
be used unless some pattern&rsquo;s constraint asks for one.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fRENAME_005fOK"></a>Macro: <strong>HARD_REGNO_RENAME_OK</strong> <em>(<var>from</var>, <var>to</var>)</em></dt>
<dd><p>A C expression that is nonzero if it is OK to rename a hard register
<var>from</var> to another hard register <var>to</var>.
</p>
<p>One common use of this macro is to prevent renaming of a register to
another register that is not saved by a prologue in an interrupt
handler.
</p>
<p>The default is always nonzero.
</p></dd></dl>

<dl>
<dt><a name="index-MODES_005fTIEABLE_005fP"></a>Macro: <strong>MODES_TIEABLE_P</strong> <em>(<var>mode1</var>, <var>mode2</var>)</em></dt>
<dd><p>A C expression that is nonzero if a value of mode
<var>mode1</var> is accessible in mode <var>mode2</var> without copying.
</p>
<p>If <code>HARD_REGNO_MODE_OK (<var>r</var>, <var>mode1</var>)</code> and
<code>HARD_REGNO_MODE_OK (<var>r</var>, <var>mode2</var>)</code> are always the same for
any <var>r</var>, then <code>MODES_TIEABLE_P (<var>mode1</var>, <var>mode2</var>)</code>
should be nonzero.  If they differ for any <var>r</var>, you should define
this macro to return zero unless some other mechanism ensures the
accessibility of the value in a narrower mode.
</p>
<p>You should define this macro to return nonzero in as many cases as
possible since doing so will allow GCC to perform better register
allocation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHARD_005fREGNO_005fSCRATCH_005fOK"></a>Target Hook: <em>bool</em> <strong>TARGET_HARD_REGNO_SCRATCH_OK</strong> <em>(unsigned int <var>regno</var>)</em></dt>
<dd><p>This target hook should return <code>true</code> if it is OK to use a hard register
<var>regno</var> as scratch reg in peephole2.
</p>
<p>One common use of this macro is to prevent using of a register that
is not saved by a prologue in an interrupt handler.
</p>
<p>The default version of this hook always returns <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-AVOID_005fCCMODE_005fCOPIES"></a>Macro: <strong>AVOID_CCMODE_COPIES</strong></dt>
<dd><p>Define this macro if the compiler should avoid copies to/from <code>CCmode</code>
registers.  You should only define this macro if support for copying to/from
<code>CCmode</code> is incomplete.
</p></dd></dl>

<hr>
<a name="Leaf-Functions"></a>
<div class="header">
<p>
Next: <a href="#Stack-Registers" accesskey="n" rel="next">Stack Registers</a>, Previous: <a href="#Values-in-Registers" accesskey="p" rel="prev">Values in Registers</a>, Up: <a href="#Registers" accesskey="u" rel="up">Registers</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Handling-Leaf-Functions"></a>
<h4 class="subsection">17.7.4 Handling Leaf Functions</h4>

<a name="index-leaf-functions"></a>
<a name="index-functions_002c-leaf"></a>
<p>On some machines, a leaf function (i.e., one which makes no calls) can run
more efficiently if it does not make its own register window.  Often this
means it is required to receive its arguments in the registers where they
are passed by the caller, instead of the registers where they would
normally arrive.
</p>
<p>The special treatment for leaf functions generally applies only when
other conditions are met; for example, often they may use only those
registers for its own variables and temporaries.  We use the term &ldquo;leaf
function&rdquo; to mean a function that is suitable for this special
handling, so that functions with no calls are not necessarily &ldquo;leaf
functions&rdquo;.
</p>
<p>GCC assigns register numbers before it knows whether the function is
suitable for leaf function treatment.  So it needs to renumber the
registers in order to output a leaf function.  The following macros
accomplish this.
</p>
<dl>
<dt><a name="index-LEAF_005fREGISTERS"></a>Macro: <strong>LEAF_REGISTERS</strong></dt>
<dd><p>Name of a char vector, indexed by hard register number, which
contains 1 for a register that is allowable in a candidate for leaf
function treatment.
</p>
<p>If leaf function treatment involves renumbering the registers, then the
registers marked here should be the ones before renumbering&mdash;those that
GCC would ordinarily allocate.  The registers which will actually be
used in the assembler code, after renumbering, should not be marked with 1
in this vector.
</p>
<p>Define this macro only if the target machine offers a way to optimize
the treatment of leaf functions.
</p></dd></dl>

<dl>
<dt><a name="index-LEAF_005fREG_005fREMAP"></a>Macro: <strong>LEAF_REG_REMAP</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>A C expression whose value is the register number to which <var>regno</var>
should be renumbered, when a function is treated as a leaf function.
</p>
<p>If <var>regno</var> is a register number which should not appear in a leaf
function before renumbering, then the expression should yield -1, which
will cause the compiler to abort.
</p>
<p>Define this macro only if the target machine offers a way to optimize the
treatment of leaf functions, and registers need to be renumbered to do
this.
</p></dd></dl>

<a name="index-current_005ffunction_005fis_005fleaf"></a>
<a name="index-current_005ffunction_005fuses_005fonly_005fleaf_005fregs"></a>
<p><code>TARGET_ASM_FUNCTION_PROLOGUE</code> and
<code>TARGET_ASM_FUNCTION_EPILOGUE</code> must usually treat leaf functions
specially.  They can test the C variable <code>current_function_is_leaf</code>
which is nonzero for leaf functions.  <code>current_function_is_leaf</code> is
set prior to local register allocation and is valid for the remaining
compiler passes.  They can also test the C variable
<code>current_function_uses_only_leaf_regs</code> which is nonzero for leaf
functions which only use leaf registers.
<code>current_function_uses_only_leaf_regs</code> is valid after all passes
that modify the instructions have been run and is only useful if
<code>LEAF_REGISTERS</code> is defined.
</p>
<hr>
<a name="Stack-Registers"></a>
<div class="header">
<p>
Previous: <a href="#Leaf-Functions" accesskey="p" rel="prev">Leaf Functions</a>, Up: <a href="#Registers" accesskey="u" rel="up">Registers</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Registers-That-Form-a-Stack"></a>
<h4 class="subsection">17.7.5 Registers That Form a Stack</h4>

<p>There are special features to handle computers where some of the
&ldquo;registers&rdquo; form a stack.  Stack registers are normally written by
pushing onto the stack, and are numbered relative to the top of the
stack.
</p>
<p>Currently, GCC can only handle one group of stack-like registers, and
they must be consecutively numbered.  Furthermore, the existing
support for stack-like registers is specific to the 80387 floating
point coprocessor.  If you have a new architecture that uses
stack-like registers, you will need to do substantial work on
<samp>reg-stack.c</samp> and write your machine description to cooperate
with it, as well as defining these macros.
</p>
<dl>
<dt><a name="index-STACK_005fREGS"></a>Macro: <strong>STACK_REGS</strong></dt>
<dd><p>Define this if the machine has any stack-like registers.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fREG_005fCOVER_005fCLASS"></a>Macro: <strong>STACK_REG_COVER_CLASS</strong></dt>
<dd><p>This is a cover class containing the stack registers.  Define this if
the machine has any stack-like registers.
</p></dd></dl>

<dl>
<dt><a name="index-FIRST_005fSTACK_005fREG"></a>Macro: <strong>FIRST_STACK_REG</strong></dt>
<dd><p>The number of the first stack-like register.  This one is the top
of the stack.
</p></dd></dl>

<dl>
<dt><a name="index-LAST_005fSTACK_005fREG"></a>Macro: <strong>LAST_STACK_REG</strong></dt>
<dd><p>The number of the last stack-like register.  This one is the bottom of
the stack.
</p></dd></dl>

<hr>
<a name="Register-Classes"></a>
<div class="header">
<p>
Next: <a href="#Old-Constraints" accesskey="n" rel="next">Old Constraints</a>, Previous: <a href="#Registers" accesskey="p" rel="prev">Registers</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Register-Classes-1"></a>
<h3 class="section">17.8 Register Classes</h3>
<a name="index-register-class-definitions"></a>
<a name="index-class-definitions_002c-register"></a>

<p>On many machines, the numbered registers are not all equivalent.
For example, certain registers may not be allowed for indexed addressing;
certain registers may not be allowed in some instructions.  These machine
restrictions are described to the compiler using <em>register classes</em>.
</p>
<p>You define a number of register classes, giving each one a name and saying
which of the registers belong to it.  Then you can specify register classes
that are allowed as operands to particular instruction patterns.
</p>
<a name="index-ALL_005fREGS"></a>
<a name="index-NO_005fREGS"></a>
<p>In general, each register will belong to several classes.  In fact, one
class must be named <code>ALL_REGS</code> and contain all the registers.  Another
class must be named <code>NO_REGS</code> and contain no registers.  Often the
union of two classes will be another class; however, this is not required.
</p>
<a name="index-GENERAL_005fREGS"></a>
<p>One of the classes must be named <code>GENERAL_REGS</code>.  There is nothing
terribly special about the name, but the operand constraint letters
&lsquo;<samp>r</samp>&rsquo; and &lsquo;<samp>g</samp>&rsquo; specify this class.  If <code>GENERAL_REGS</code> is
the same as <code>ALL_REGS</code>, just define it as a macro which expands
to <code>ALL_REGS</code>.
</p>
<p>Order the classes so that if class <var>x</var> is contained in class <var>y</var>
then <var>x</var> has a lower class number than <var>y</var>.
</p>
<p>The way classes other than <code>GENERAL_REGS</code> are specified in operand
constraints is through machine-dependent operand constraint letters.
You can define such letters to correspond to various classes, then use
them in operand constraints.
</p>
<p>You must define the narrowest register classes for allocatable
registers, so that each class either has no subclasses, or that for
some mode, the move cost between registers within the class is
cheaper than moving a register in the class to or from memory
(see <a href="#Costs">Costs</a>).
</p>
<p>You should define a class for the union of two classes whenever some
instruction allows both classes.  For example, if an instruction allows
either a floating point (coprocessor) register or a general register for a
certain operand, you should define a class <code>FLOAT_OR_GENERAL_REGS</code>
which includes both of them.  Otherwise you will get suboptimal code,
or even internal compiler errors when reload cannot find a register in the
class computed via <code>reg_class_subunion</code>.
</p>
<p>You must also specify certain redundant information about the register
classes: for each class, which classes contain it and which ones are
contained in it; for each pair of classes, the largest class contained
in their union.
</p>
<p>When a value occupying several consecutive registers is expected in a
certain class, all the registers used must belong to that class.
Therefore, register classes cannot be used to enforce a requirement for
a register pair to start with an even-numbered register.  The way to
specify this requirement is with <code>HARD_REGNO_MODE_OK</code>.
</p>
<p>Register classes used for input-operands of bitwise-and or shift
instructions have a special requirement: each such class must have, for
each fixed-point machine mode, a subclass whose registers can transfer that
mode to or from memory.  For example, on some machines, the operations for
single-byte values (<code>QImode</code>) are limited to certain registers.  When
this is so, each register class that is used in a bitwise-and or shift
instruction must have a subclass consisting of registers from which
single-byte values can be loaded or stored.  This is so that
<code>PREFERRED_RELOAD_CLASS</code> can always have a possible value to return.
</p>
<dl>
<dt><a name="index-enum-reg_005fclass"></a>Data type: <strong>enum reg_class</strong></dt>
<dd><p>An enumerated type that must be defined with all the register class names
as enumerated values.  <code>NO_REGS</code> must be first.  <code>ALL_REGS</code>
must be the last register class, followed by one more enumerated value,
<code>LIM_REG_CLASSES</code>, which is not a register class but rather
tells how many classes there are.
</p>
<p>Each register class has a number, which is the value of casting
the class name to type <code>int</code>.  The number serves as an index
in many of the tables described below.
</p></dd></dl>

<dl>
<dt><a name="index-N_005fREG_005fCLASSES"></a>Macro: <strong>N_REG_CLASSES</strong></dt>
<dd><p>The number of distinct register classes, defined as follows:
</p>
<div class="smallexample">
<pre class="smallexample">#define N_REG_CLASSES (int) LIM_REG_CLASSES
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-REG_005fCLASS_005fNAMES"></a>Macro: <strong>REG_CLASS_NAMES</strong></dt>
<dd><p>An initializer containing the names of the register classes as C string
constants.  These names are used in writing some of the debugging dumps.
</p></dd></dl>

<dl>
<dt><a name="index-REG_005fCLASS_005fCONTENTS"></a>Macro: <strong>REG_CLASS_CONTENTS</strong></dt>
<dd><p>An initializer containing the contents of the register classes, as integers
which are bit masks.  The <var>n</var>th integer specifies the contents of class
<var>n</var>.  The way the integer <var>mask</var> is interpreted is that
register <var>r</var> is in the class if <code><var>mask</var> &amp; (1 &lt;&lt; <var>r</var>)</code> is 1.
</p>
<p>When the machine has more than 32 registers, an integer does not suffice.
Then the integers are replaced by sub-initializers, braced groupings containing
several integers.  Each sub-initializer must be suitable as an initializer
for the type <code>HARD_REG_SET</code> which is defined in <samp>hard-reg-set.h</samp>.
In this situation, the first integer in each sub-initializer corresponds to
registers 0 through 31, the second integer to registers 32 through 63, and
so on.
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fREG_005fCLASS"></a>Macro: <strong>REGNO_REG_CLASS</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>A C expression whose value is a register class containing hard register
<var>regno</var>.  In general there is more than one such class; choose a class
which is <em>minimal</em>, meaning that no smaller class also contains the
register.
</p></dd></dl>

<dl>
<dt><a name="index-BASE_005fREG_005fCLASS"></a>Macro: <strong>BASE_REG_CLASS</strong></dt>
<dd><p>A macro whose definition is the name of the class to which a valid
base register must belong.  A base register is one used in an address
which is the register value plus a displacement.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fBASE_005fREG_005fCLASS"></a>Macro: <strong>MODE_BASE_REG_CLASS</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>This is a variation of the <code>BASE_REG_CLASS</code> macro which allows
the selection of a base register in a mode dependent manner.  If
<var>mode</var> is VOIDmode then it should return the same value as
<code>BASE_REG_CLASS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fBASE_005fREG_005fREG_005fCLASS"></a>Macro: <strong>MODE_BASE_REG_REG_CLASS</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression whose value is the register class to which a valid
base register must belong in order to be used in a base plus index
register address.  You should define this macro if base plus index
addresses have different requirements than other base register uses.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fCODE_005fBASE_005fREG_005fCLASS"></a>Macro: <strong>MODE_CODE_BASE_REG_CLASS</strong> <em>(<var>mode</var>, <var>address_space</var>, <var>outer_code</var>, <var>index_code</var>)</em></dt>
<dd><p>A C expression whose value is the register class to which a valid
base register for a memory reference in mode <var>mode</var> to address
space <var>address_space</var> must belong.  <var>outer_code</var> and <var>index_code</var>
define the context in which the base register occurs.  <var>outer_code</var> is
the code of the immediately enclosing expression (<code>MEM</code> for the top level
of an address, <code>ADDRESS</code> for something that occurs in an
<code>address_operand</code>).  <var>index_code</var> is the code of the corresponding
index expression if <var>outer_code</var> is <code>PLUS</code>; <code>SCRATCH</code> otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-INDEX_005fREG_005fCLASS"></a>Macro: <strong>INDEX_REG_CLASS</strong></dt>
<dd><p>A macro whose definition is the name of the class to which a valid
index register must belong.  An index register is one used in an
address where its value is either multiplied by a scale factor or
added to another register (as well as added to a displacement).
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fOK_005fFOR_005fBASE_005fP"></a>Macro: <strong>REGNO_OK_FOR_BASE_P</strong> <em>(<var>num</var>)</em></dt>
<dd><p>A C expression which is nonzero if register number <var>num</var> is
suitable for use as a base register in operand addresses.
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fMODE_005fOK_005fFOR_005fBASE_005fP"></a>Macro: <strong>REGNO_MODE_OK_FOR_BASE_P</strong> <em>(<var>num</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression that is just like <code>REGNO_OK_FOR_BASE_P</code>, except that
that expression may examine the mode of the memory reference in
<var>mode</var>.  You should define this macro if the mode of the memory
reference affects whether a register may be used as a base register.  If
you define this macro, the compiler will use it instead of
<code>REGNO_OK_FOR_BASE_P</code>.  The mode may be <code>VOIDmode</code> for
addresses that appear outside a <code>MEM</code>, i.e., as an
<code>address_operand</code>.
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fMODE_005fOK_005fFOR_005fREG_005fBASE_005fP"></a>Macro: <strong>REGNO_MODE_OK_FOR_REG_BASE_P</strong> <em>(<var>num</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression which is nonzero if register number <var>num</var> is suitable for
use as a base register in base plus index operand addresses, accessing
memory in mode <var>mode</var>.  It may be either a suitable hard register or a
pseudo register that has been allocated such a hard register.  You should
define this macro if base plus index addresses have different requirements
than other base register uses.
</p>
<p>Use of this macro is deprecated; please use the more general
<code>REGNO_MODE_CODE_OK_FOR_BASE_P</code>.
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fMODE_005fCODE_005fOK_005fFOR_005fBASE_005fP"></a>Macro: <strong>REGNO_MODE_CODE_OK_FOR_BASE_P</strong> <em>(<var>num</var>, <var>mode</var>, <var>address_space</var>, <var>outer_code</var>, <var>index_code</var>)</em></dt>
<dd><p>A C expression which is nonzero if register number <var>num</var> is
suitable for use as a base register in operand addresses, accessing
memory in mode <var>mode</var> in address space <var>address_space</var>.
This is similar to <code>REGNO_MODE_OK_FOR_BASE_P</code>, except
that that expression may examine the context in which the register
appears in the memory reference.  <var>outer_code</var> is the code of the
immediately enclosing expression (<code>MEM</code> if at the top level of the
address, <code>ADDRESS</code> for something that occurs in an
<code>address_operand</code>).  <var>index_code</var> is the code of the
corresponding index expression if <var>outer_code</var> is <code>PLUS</code>;
<code>SCRATCH</code> otherwise.  The mode may be <code>VOIDmode</code> for addresses
that appear outside a <code>MEM</code>, i.e., as an <code>address_operand</code>.
</p></dd></dl>

<dl>
<dt><a name="index-REGNO_005fOK_005fFOR_005fINDEX_005fP"></a>Macro: <strong>REGNO_OK_FOR_INDEX_P</strong> <em>(<var>num</var>)</em></dt>
<dd><p>A C expression which is nonzero if register number <var>num</var> is
suitable for use as an index register in operand addresses.  It may be
either a suitable hard register or a pseudo register that has been
allocated such a hard register.
</p>
<p>The difference between an index register and a base register is that
the index register may be scaled.  If an address involves the sum of
two registers, neither one of them scaled, then either one may be
labeled the &ldquo;base&rdquo; and the other the &ldquo;index&rdquo;; but whichever
labeling is used must fit the machine&rsquo;s constraints of which registers
may serve in each capacity.  The compiler will try both labelings,
looking for one that is valid, and will reload one or both registers
only if neither labeling works.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPREFERRED_005fRENAME_005fCLASS"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_PREFERRED_RENAME_CLASS</strong> <em>(reg_class_t <var>rclass</var>)</em></dt>
<dd><p>A target hook that places additional preference on the register class to use when it is necessary to rename a register in class <var>rclass</var> to another class, or perhaps <var>NO_REGS</var>, if no preferred register class is found or hook <code>preferred_rename_class</code> is not implemented. Sometimes returning a more restrictive class makes better code.  For example, on ARM, thumb-2 instructions using <code>LO_REGS</code> may be smaller than instructions using <code>GENERIC_REGS</code>.  By returning <code>LO_REGS</code> from <code>preferred_rename_class</code>, code size can be reduced.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPREFERRED_005fRELOAD_005fCLASS"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_PREFERRED_RELOAD_CLASS</strong> <em>(rtx <var>x</var>, reg_class_t <var>rclass</var>)</em></dt>
<dd><p>A target hook that places additional restrictions on the register class
to use when it is necessary to copy value <var>x</var> into a register in class
<var>rclass</var>.  The value is a register class; perhaps <var>rclass</var>, or perhaps
another, smaller class.
</p>
<p>The default version of this hook always returns value of <code>rclass</code> argument.
</p>
<p>Sometimes returning a more restrictive class makes better code.  For
example, on the 68000, when <var>x</var> is an integer constant that is in range
for a &lsquo;<samp>moveq</samp>&rsquo; instruction, the value of this macro is always
<code>DATA_REGS</code> as long as <var>rclass</var> includes the data registers.
Requiring a data register guarantees that a &lsquo;<samp>moveq</samp>&rsquo; will be used.
</p>
<p>One case where <code>TARGET_PREFERRED_RELOAD_CLASS</code> must not return
<var>rclass</var> is if <var>x</var> is a legitimate constant which cannot be
loaded into some register class.  By returning <code>NO_REGS</code> you can
force <var>x</var> into a memory location.  For example, rs6000 can load
immediate values into general-purpose registers, but does not have an
instruction for loading an immediate value into a floating-point
register, so <code>TARGET_PREFERRED_RELOAD_CLASS</code> returns <code>NO_REGS</code> when
<var>x</var> is a floating-point constant.  If the constant can&rsquo;t be loaded
into any kind of register, code generation will be better if
<code>TARGET_LEGITIMATE_CONSTANT_P</code> makes the constant illegitimate instead
of using <code>TARGET_PREFERRED_RELOAD_CLASS</code>.
</p>
<p>If an insn has pseudos in it after register allocation, reload will go
through the alternatives and call repeatedly <code>TARGET_PREFERRED_RELOAD_CLASS</code>
to find the best one.  Returning <code>NO_REGS</code>, in this case, makes
reload add a <code>!</code> in front of the constraint: the x86 back-end uses
this feature to discourage usage of 387 registers when math is done in
the SSE registers (and vice versa).
</p></dd></dl>

<dl>
<dt><a name="index-PREFERRED_005fRELOAD_005fCLASS"></a>Macro: <strong>PREFERRED_RELOAD_CLASS</strong> <em>(<var>x</var>, <var>class</var>)</em></dt>
<dd><p>A C expression that places additional restrictions on the register class
to use when it is necessary to copy value <var>x</var> into a register in class
<var>class</var>.  The value is a register class; perhaps <var>class</var>, or perhaps
another, smaller class.  On many machines, the following definition is
safe:
</p>
<div class="smallexample">
<pre class="smallexample">#define PREFERRED_RELOAD_CLASS(X,CLASS) CLASS
</pre></div>

<p>Sometimes returning a more restrictive class makes better code.  For
example, on the 68000, when <var>x</var> is an integer constant that is in range
for a &lsquo;<samp>moveq</samp>&rsquo; instruction, the value of this macro is always
<code>DATA_REGS</code> as long as <var>class</var> includes the data registers.
Requiring a data register guarantees that a &lsquo;<samp>moveq</samp>&rsquo; will be used.
</p>
<p>One case where <code>PREFERRED_RELOAD_CLASS</code> must not return
<var>class</var> is if <var>x</var> is a legitimate constant which cannot be
loaded into some register class.  By returning <code>NO_REGS</code> you can
force <var>x</var> into a memory location.  For example, rs6000 can load
immediate values into general-purpose registers, but does not have an
instruction for loading an immediate value into a floating-point
register, so <code>PREFERRED_RELOAD_CLASS</code> returns <code>NO_REGS</code> when
<var>x</var> is a floating-point constant.  If the constant can&rsquo;t be loaded
into any kind of register, code generation will be better if
<code>TARGET_LEGITIMATE_CONSTANT_P</code> makes the constant illegitimate instead
of using <code>TARGET_PREFERRED_RELOAD_CLASS</code>.
</p>
<p>If an insn has pseudos in it after register allocation, reload will go
through the alternatives and call repeatedly <code>PREFERRED_RELOAD_CLASS</code>
to find the best one.  Returning <code>NO_REGS</code>, in this case, makes
reload add a <code>!</code> in front of the constraint: the x86 back-end uses
this feature to discourage usage of 387 registers when math is done in
the SSE registers (and vice versa).
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPREFERRED_005fOUTPUT_005fRELOAD_005fCLASS"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_PREFERRED_OUTPUT_RELOAD_CLASS</strong> <em>(rtx <var>x</var>, reg_class_t <var>rclass</var>)</em></dt>
<dd><p>Like <code>TARGET_PREFERRED_RELOAD_CLASS</code>, but for output reloads instead of
input reloads.
</p>
<p>The default version of this hook always returns value of <code>rclass</code>
argument.
</p>
<p>You can also use <code>TARGET_PREFERRED_OUTPUT_RELOAD_CLASS</code> to discourage
reload from using some alternatives, like <code>TARGET_PREFERRED_RELOAD_CLASS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LIMIT_005fRELOAD_005fCLASS"></a>Macro: <strong>LIMIT_RELOAD_CLASS</strong> <em>(<var>mode</var>, <var>class</var>)</em></dt>
<dd><p>A C expression that places additional restrictions on the register class
to use when it is necessary to be able to hold a value of mode
<var>mode</var> in a reload register for which class <var>class</var> would
ordinarily be used.
</p>
<p>Unlike <code>PREFERRED_RELOAD_CLASS</code>, this macro should be used when
there are certain modes that simply can&rsquo;t go in certain reload classes.
</p>
<p>The value is a register class; perhaps <var>class</var>, or perhaps another,
smaller class.
</p>
<p>Don&rsquo;t define this macro unless the target machine has limitations which
require the macro to do something nontrivial.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSECONDARY_005fRELOAD"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_SECONDARY_RELOAD</strong> <em>(bool <var>in_p</var>, rtx <var>x</var>, reg_class_t <var>reload_class</var>, enum machine_mode <var>reload_mode</var>, secondary_reload_info *<var>sri</var>)</em></dt>
<dd><p>Many machines have some registers that cannot be copied directly to or
from memory or even from other types of registers.  An example is the
&lsquo;<samp>MQ</samp>&rsquo; register, which on most machines, can only be copied to or
from general registers, but not memory.  Below, we shall be using the
term &rsquo;intermediate register&rsquo; when a move operation cannot be performed
directly, but has to be done by copying the source into the intermediate
register first, and then copying the intermediate register to the
destination.  An intermediate register always has the same mode as
source and destination.  Since it holds the actual value being copied,
reload might apply optimizations to re-use an intermediate register
and eliding the copy from the source when it can determine that the
intermediate register still holds the required value.
</p>
<p>Another kind of secondary reload is required on some machines which
allow copying all registers to and from memory, but require a scratch
register for stores to some memory locations (e.g., those with symbolic
address on the RT, and those with certain symbolic address on the SPARC
when compiling PIC).  Scratch registers need not have the same mode
as the value being copied, and usually hold a different value than
that being copied.  Special patterns in the md file are needed to
describe how the copy is performed with the help of the scratch register;
these patterns also describe the number, register class(es) and mode(s)
of the scratch register(s).
</p>
<p>In some cases, both an intermediate and a scratch register are required.
</p>
<p>For input reloads, this target hook is called with nonzero <var>in_p</var>,
and <var>x</var> is an rtx that needs to be copied to a register of class
<var>reload_class</var> in <var>reload_mode</var>.  For output reloads, this target
hook is called with zero <var>in_p</var>, and a register of class <var>reload_class</var>
needs to be copied to rtx <var>x</var> in <var>reload_mode</var>.
</p>
<p>If copying a register of <var>reload_class</var> from/to <var>x</var> requires
an intermediate register, the hook <code>secondary_reload</code> should
return the register class required for this intermediate register.
If no intermediate register is required, it should return NO_REGS.
If more than one intermediate register is required, describe the one
that is closest in the copy chain to the reload register.
</p>
<p>If scratch registers are needed, you also have to describe how to
perform the copy from/to the reload register to/from this
closest intermediate register.  Or if no intermediate register is
required, but still a scratch register is needed, describe the
copy  from/to the reload register to/from the reload operand <var>x</var>.
</p>
<p>You do this by setting <code>sri-&gt;icode</code> to the instruction code of a pattern
in the md file which performs the move.  Operands 0 and 1 are the output
and input of this copy, respectively.  Operands from operand 2 onward are
for scratch operands.  These scratch operands must have a mode, and a
single-register-class
output constraint.
</p>
<p>When an intermediate register is used, the <code>secondary_reload</code>
hook will be called again to determine how to copy the intermediate
register to/from the reload operand <var>x</var>, so your hook must also
have code to handle the register class of the intermediate operand.
</p>


<p><var>x</var> might be a pseudo-register or a <code>subreg</code> of a
pseudo-register, which could either be in a hard register or in memory.
Use <code>true_regnum</code> to find out; it will return -1 if the pseudo is
in memory and the hard register number if it is in a register.
</p>
<p>Scratch operands in memory (constraint <code>&quot;=m&quot;</code> / <code>&quot;=&amp;m&quot;</code>) are
currently not supported.  For the time being, you will have to continue
to use <code>SECONDARY_MEMORY_NEEDED</code> for that purpose.
</p>
<p><code>copy_cost</code> also uses this target hook to find out how values are
copied.  If you want it to include some extra cost for the need to allocate
(a) scratch register(s), set <code>sri-&gt;extra_cost</code> to the additional cost.
Or if two dependent moves are supposed to have a lower cost than the sum
of the individual moves due to expected fortuitous scheduling and/or special
forwarding logic, you can set <code>sri-&gt;extra_cost</code> to a negative amount.
</p></dd></dl>

<dl>
<dt><a name="index-SECONDARY_005fRELOAD_005fCLASS"></a>Macro: <strong>SECONDARY_RELOAD_CLASS</strong> <em>(<var>class</var>, <var>mode</var>, <var>x</var>)</em></dt>
<dt><a name="index-SECONDARY_005fINPUT_005fRELOAD_005fCLASS"></a>Macro: <strong>SECONDARY_INPUT_RELOAD_CLASS</strong> <em>(<var>class</var>, <var>mode</var>, <var>x</var>)</em></dt>
<dt><a name="index-SECONDARY_005fOUTPUT_005fRELOAD_005fCLASS"></a>Macro: <strong>SECONDARY_OUTPUT_RELOAD_CLASS</strong> <em>(<var>class</var>, <var>mode</var>, <var>x</var>)</em></dt>
<dd><p>These macros are obsolete, new ports should use the target hook
<code>TARGET_SECONDARY_RELOAD</code> instead.
</p>
<p>These are obsolete macros, replaced by the <code>TARGET_SECONDARY_RELOAD</code>
target hook.  Older ports still define these macros to indicate to the
reload phase that it may
need to allocate at least one register for a reload in addition to the
register to contain the data.  Specifically, if copying <var>x</var> to a
register <var>class</var> in <var>mode</var> requires an intermediate register,
you were supposed to define <code>SECONDARY_INPUT_RELOAD_CLASS</code> to return the
largest register class all of whose registers can be used as
intermediate registers or scratch registers.
</p>
<p>If copying a register <var>class</var> in <var>mode</var> to <var>x</var> requires an
intermediate or scratch register, <code>SECONDARY_OUTPUT_RELOAD_CLASS</code>
was supposed to be defined be defined to return the largest register
class required.  If the
requirements for input and output reloads were the same, the macro
<code>SECONDARY_RELOAD_CLASS</code> should have been used instead of defining both
macros identically.
</p>
<p>The values returned by these macros are often <code>GENERAL_REGS</code>.
Return <code>NO_REGS</code> if no spare register is needed; i.e., if <var>x</var>
can be directly copied to or from a register of <var>class</var> in
<var>mode</var> without requiring a scratch register.  Do not define this
macro if it would always return <code>NO_REGS</code>.
</p>
<p>If a scratch register is required (either with or without an
intermediate register), you were supposed to define patterns for
&lsquo;<samp>reload_in<var>m</var></samp>&rsquo; or &lsquo;<samp>reload_out<var>m</var></samp>&rsquo;, as required
(see <a href="#Standard-Names">Standard Names</a>.  These patterns, which were normally
implemented with a <code>define_expand</code>, should be similar to the
&lsquo;<samp>mov<var>m</var></samp>&rsquo; patterns, except that operand 2 is the scratch
register.
</p>
<p>These patterns need constraints for the reload register and scratch
register that
contain a single register class.  If the original reload register (whose
class is <var>class</var>) can meet the constraint given in the pattern, the
value returned by these macros is used for the class of the scratch
register.  Otherwise, two additional reload registers are required.
Their classes are obtained from the constraints in the insn pattern.
</p>
<p><var>x</var> might be a pseudo-register or a <code>subreg</code> of a
pseudo-register, which could either be in a hard register or in memory.
Use <code>true_regnum</code> to find out; it will return -1 if the pseudo is
in memory and the hard register number if it is in a register.
</p>
<p>These macros should not be used in the case where a particular class of
registers can only be copied to memory and not to another class of
registers.  In that case, secondary reload registers are not needed and
would not be helpful.  Instead, a stack location must be used to perform
the copy and the <code>mov<var>m</var></code> pattern should use memory as an
intermediate storage.  This case often occurs between floating-point and
general registers.
</p></dd></dl>

<dl>
<dt><a name="index-SECONDARY_005fMEMORY_005fNEEDED"></a>Macro: <strong>SECONDARY_MEMORY_NEEDED</strong> <em>(<var>class1</var>, <var>class2</var>, <var>m</var>)</em></dt>
<dd><p>Certain machines have the property that some registers cannot be copied
to some other registers without using memory.  Define this macro on
those machines to be a C expression that is nonzero if objects of mode
<var>m</var> in registers of <var>class1</var> can only be copied to registers of
class <var>class2</var> by storing a register of <var>class1</var> into memory
and loading that memory location into a register of <var>class2</var>.
</p>
<p>Do not define this macro if its value would always be zero.
</p></dd></dl>

<dl>
<dt><a name="index-SECONDARY_005fMEMORY_005fNEEDED_005fRTX"></a>Macro: <strong>SECONDARY_MEMORY_NEEDED_RTX</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>Normally when <code>SECONDARY_MEMORY_NEEDED</code> is defined, the compiler
allocates a stack slot for a memory location needed for register copies.
If this macro is defined, the compiler instead uses the memory location
defined by this macro.
</p>
<p>Do not define this macro if you do not define
<code>SECONDARY_MEMORY_NEEDED</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SECONDARY_005fMEMORY_005fNEEDED_005fMODE"></a>Macro: <strong>SECONDARY_MEMORY_NEEDED_MODE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>When the compiler needs a secondary memory location to copy between two
registers of mode <var>mode</var>, it normally allocates sufficient memory to
hold a quantity of <code>BITS_PER_WORD</code> bits and performs the store and
load operations in a mode that many bits wide and whose class is the
same as that of <var>mode</var>.
</p>
<p>This is right thing to do on most machines because it ensures that all
bits of the register are copied and prevents accesses to the registers
in a narrower mode, which some machines prohibit for floating-point
registers.
</p>
<p>However, this default behavior is not correct on some machines, such as
the DEC Alpha, that store short integers in floating-point registers
differently than in integer registers.  On those machines, the default
widening will not work correctly and you must define this macro to
suppress that widening in some cases.  See the file <samp>alpha.h</samp> for
details.
</p>
<p>Do not define this macro if you do not define
<code>SECONDARY_MEMORY_NEEDED</code> or if widening <var>mode</var> to a mode that
is <code>BITS_PER_WORD</code> bits wide is correct for your machine.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCLASS_005fLIKELY_005fSPILLED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_CLASS_LIKELY_SPILLED_P</strong> <em>(reg_class_t <var>rclass</var>)</em></dt>
<dd><p>A target hook which returns <code>true</code> if pseudos that have been assigned
to registers of class <var>rclass</var> would likely be spilled because
registers of <var>rclass</var> are needed for spill registers.
</p>
<p>The default version of this target hook returns <code>true</code> if <var>rclass</var>
has exactly one register and <code>false</code> otherwise.  On most machines, this
default should be used.  For generally register-starved machines, such as
i386, or machines with right register constraints, such as SH, this hook
can be used to avoid excessive spilling.
</p>
<p>This hook is also used by some of the global intra-procedural code
transformations to throtle code motion, to avoid increasing register
pressure.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCLASS_005fMAX_005fNREGS"></a>Target Hook: <em>unsigned char</em> <strong>TARGET_CLASS_MAX_NREGS</strong> <em>(reg_class_t <var>rclass</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>A target hook returns the maximum number of consecutive registers
of class <var>rclass</var> needed to hold a value of mode <var>mode</var>.
</p>
<p>This is closely related to the macro <code>HARD_REGNO_NREGS</code>.  In fact,
the value returned by <code>TARGET_CLASS_MAX_NREGS (<var>rclass</var>,
<var>mode</var>)</code> target hook should be the maximum value of
<code>HARD_REGNO_NREGS (<var>regno</var>, <var>mode</var>)</code> for all <var>regno</var>
values in the class <var>rclass</var>.
</p>
<p>This target hook helps control the handling of multiple-word values
in the reload pass.
</p>
<p>The default version of this target hook returns the size of <var>mode</var>
in words.
</p></dd></dl>

<dl>
<dt><a name="index-CLASS_005fMAX_005fNREGS"></a>Macro: <strong>CLASS_MAX_NREGS</strong> <em>(<var>class</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression for the maximum number of consecutive registers
of class <var>class</var> needed to hold a value of mode <var>mode</var>.
</p>
<p>This is closely related to the macro <code>HARD_REGNO_NREGS</code>.  In fact,
the value of the macro <code>CLASS_MAX_NREGS (<var>class</var>, <var>mode</var>)</code>
should be the maximum value of <code>HARD_REGNO_NREGS (<var>regno</var>,
<var>mode</var>)</code> for all <var>regno</var> values in the class <var>class</var>.
</p>
<p>This macro helps control the handling of multiple-word values
in the reload pass.
</p></dd></dl>

<dl>
<dt><a name="index-CANNOT_005fCHANGE_005fMODE_005fCLASS"></a>Macro: <strong>CANNOT_CHANGE_MODE_CLASS</strong> <em>(<var>from</var>, <var>to</var>, <var>class</var>)</em></dt>
<dd><p>If defined, a C expression that returns nonzero for a <var>class</var> for which
a change from mode <var>from</var> to mode <var>to</var> is invalid.
</p>
<p>For the example, loading 32-bit integer or floating-point objects into
floating-point registers on the Alpha extends them to 64 bits.
Therefore loading a 64-bit object and then storing it as a 32-bit object
does not store the low-order 32 bits, as would be the case for a normal
register.  Therefore, <samp>alpha.h</samp> defines <code>CANNOT_CHANGE_MODE_CLASS</code>
as below:
</p>
<div class="smallexample">
<pre class="smallexample">#define CANNOT_CHANGE_MODE_CLASS(FROM, TO, CLASS) \
  (GET_MODE_SIZE (FROM) != GET_MODE_SIZE (TO) \
   ? reg_classes_intersect_p (FLOAT_REGS, (CLASS)) : 0)
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-TARGET_005fLRA_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_LRA_P</strong> <em>(void)</em></dt>
<dd><p>A target hook which returns true if we use LRA instead of reload pass.  It means that LRA was ported to the target.    The default version of this target hook returns always false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fREGISTER_005fPRIORITY"></a>Target Hook: <em>int</em> <strong>TARGET_REGISTER_PRIORITY</strong> <em>(int)</em></dt>
<dd><p>A target hook which returns the register priority number to which the  register <var>hard_regno</var> belongs to.  The bigger the number, the  more preferable the hard register usage (when all other conditions are  the same).  This hook can be used to prefer some hard register over  others in LRA.  For example, some x86-64 register usage needs  additional prefix which makes instructions longer.  The hook can  return lower priority number for such registers make them less favorable  and as result making the generated code smaller.    The default version of this target hook returns always zero.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDIFFERENT_005fADDR_005fDISPLACEMENT_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_DIFFERENT_ADDR_DISPLACEMENT_P</strong> <em>(void)</em></dt>
<dd><p>A target hook which returns true if an address with the same structure  can have different maximal legitimate displacement.  For example, the  displacement can depend on memory mode or on operand combinations in  the insn.    The default version of this target hook returns always false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSPILL_005fCLASS"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_SPILL_CLASS</strong> <em>(reg_class_t, enum <var>machine_mode</var>)</em></dt>
<dd><p>This hook defines a class of registers which could be used for spilling  pseudos of the given mode and class, or <code>NO_REGS</code> if only memory  should be used.  Not defining this hook is equivalent to returning  <code>NO_REGS</code> for all inputs.
</p></dd></dl>

<hr>
<a name="Old-Constraints"></a>
<div class="header">
<p>
Next: <a href="#Stack-and-Calling" accesskey="n" rel="next">Stack and Calling</a>, Previous: <a href="#Register-Classes" accesskey="p" rel="prev">Register Classes</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Obsolete-Macros-for-Defining-Constraints"></a>
<h3 class="section">17.9 Obsolete Macros for Defining Constraints</h3>
<a name="index-defining-constraints_002c-obsolete-method"></a>
<a name="index-constraints_002c-defining_002c-obsolete-method"></a>

<p>Machine-specific constraints can be defined with these macros instead
of the machine description constructs described in <a href="#Define-Constraints">Define Constraints</a>.  This mechanism is obsolete.  New ports should not use
it; old ports should convert to the new mechanism.
</p>
<dl>
<dt><a name="index-CONSTRAINT_005fLEN"></a>Macro: <strong>CONSTRAINT_LEN</strong> <em>(<var>char</var>, <var>str</var>)</em></dt>
<dd><p>For the constraint at the start of <var>str</var>, which starts with the letter
<var>c</var>, return the length.  This allows you to have register class /
constant / extra constraints that are longer than a single letter;
you don&rsquo;t need to define this macro if you can do with single-letter
constraints only.  The definition of this macro should use
DEFAULT_CONSTRAINT_LEN for all the characters that you don&rsquo;t want
to handle specially.
There are some sanity checks in genoutput.c that check the constraint lengths
for the md file, so you can also use this macro to help you while you are
transitioning from a byzantine single-letter-constraint scheme: when you
return a negative length for a constraint you want to re-use, genoutput
will complain about every instance where it is used in the md file.
</p></dd></dl>

<dl>
<dt><a name="index-REG_005fCLASS_005fFROM_005fLETTER"></a>Macro: <strong>REG_CLASS_FROM_LETTER</strong> <em>(<var>char</var>)</em></dt>
<dd><p>A C expression which defines the machine-dependent operand constraint
letters for register classes.  If <var>char</var> is such a letter, the
value should be the register class corresponding to it.  Otherwise,
the value should be <code>NO_REGS</code>.  The register letter &lsquo;<samp>r</samp>&rsquo;,
corresponding to class <code>GENERAL_REGS</code>, will not be passed
to this macro; you do not need to handle it.
</p></dd></dl>

<dl>
<dt><a name="index-REG_005fCLASS_005fFROM_005fCONSTRAINT"></a>Macro: <strong>REG_CLASS_FROM_CONSTRAINT</strong> <em>(<var>char</var>, <var>str</var>)</em></dt>
<dd><p>Like <code>REG_CLASS_FROM_LETTER</code>, but you also get the constraint string
passed in <var>str</var>, so that you can use suffixes to distinguish between
different variants.
</p></dd></dl>

<dl>
<dt><a name="index-CONST_005fOK_005fFOR_005fLETTER_005fP"></a>Macro: <strong>CONST_OK_FOR_LETTER_P</strong> <em>(<var>value</var>, <var>c</var>)</em></dt>
<dd><p>A C expression that defines the machine-dependent operand constraint
letters (&lsquo;<samp>I</samp>&rsquo;, &lsquo;<samp>J</samp>&rsquo;, &lsquo;<samp>K</samp>&rsquo;, &hellip; &lsquo;<samp>P</samp>&rsquo;) that specify
particular ranges of integer values.  If <var>c</var> is one of those
letters, the expression should check that <var>value</var>, an integer, is in
the appropriate range and return 1 if so, 0 otherwise.  If <var>c</var> is
not one of those letters, the value should be 0 regardless of
<var>value</var>.
</p></dd></dl>

<dl>
<dt><a name="index-CONST_005fOK_005fFOR_005fCONSTRAINT_005fP"></a>Macro: <strong>CONST_OK_FOR_CONSTRAINT_P</strong> <em>(<var>value</var>, <var>c</var>, <var>str</var>)</em></dt>
<dd><p>Like <code>CONST_OK_FOR_LETTER_P</code>, but you also get the constraint
string passed in <var>str</var>, so that you can use suffixes to distinguish
between different variants.
</p></dd></dl>

<dl>
<dt><a name="index-CONST_005fDOUBLE_005fOK_005fFOR_005fLETTER_005fP"></a>Macro: <strong>CONST_DOUBLE_OK_FOR_LETTER_P</strong> <em>(<var>value</var>, <var>c</var>)</em></dt>
<dd><p>A C expression that defines the machine-dependent operand constraint
letters that specify particular ranges of <code>const_double</code> values
(&lsquo;<samp>G</samp>&rsquo; or &lsquo;<samp>H</samp>&rsquo;).
</p>
<p>If <var>c</var> is one of those letters, the expression should check that
<var>value</var>, an RTX of code <code>const_double</code>, is in the appropriate
range and return 1 if so, 0 otherwise.  If <var>c</var> is not one of those
letters, the value should be 0 regardless of <var>value</var>.
</p>
<p><code>const_double</code> is used for all floating-point constants and for
<code>DImode</code> fixed-point constants.  A given letter can accept either
or both kinds of values.  It can use <code>GET_MODE</code> to distinguish
between these kinds.
</p></dd></dl>

<dl>
<dt><a name="index-CONST_005fDOUBLE_005fOK_005fFOR_005fCONSTRAINT_005fP"></a>Macro: <strong>CONST_DOUBLE_OK_FOR_CONSTRAINT_P</strong> <em>(<var>value</var>, <var>c</var>, <var>str</var>)</em></dt>
<dd><p>Like <code>CONST_DOUBLE_OK_FOR_LETTER_P</code>, but you also get the constraint
string passed in <var>str</var>, so that you can use suffixes to distinguish
between different variants.
</p></dd></dl>

<dl>
<dt><a name="index-EXTRA_005fCONSTRAINT"></a>Macro: <strong>EXTRA_CONSTRAINT</strong> <em>(<var>value</var>, <var>c</var>)</em></dt>
<dd><p>A C expression that defines the optional machine-dependent constraint
letters that can be used to segregate specific types of operands, usually
memory references, for the target machine.  Any letter that is not
elsewhere defined and not matched by <code>REG_CLASS_FROM_LETTER</code> /
<code>REG_CLASS_FROM_CONSTRAINT</code>
may be used.  Normally this macro will not be defined.
</p>
<p>If it is required for a particular target machine, it should return 1
if <var>value</var> corresponds to the operand type represented by the
constraint letter <var>c</var>.  If <var>c</var> is not defined as an extra
constraint, the value returned should be 0 regardless of <var>value</var>.
</p>
<p>For example, on the ROMP, load instructions cannot have their output
in r0 if the memory reference contains a symbolic address.  Constraint
letter &lsquo;<samp>Q</samp>&rsquo; is defined as representing a memory address that does
<em>not</em> contain a symbolic address.  An alternative is specified with
a &lsquo;<samp>Q</samp>&rsquo; constraint on the input and &lsquo;<samp>r</samp>&rsquo; on the output.  The next
alternative specifies &lsquo;<samp>m</samp>&rsquo; on the input and a register class that
does not include r0 on the output.
</p></dd></dl>

<dl>
<dt><a name="index-EXTRA_005fCONSTRAINT_005fSTR"></a>Macro: <strong>EXTRA_CONSTRAINT_STR</strong> <em>(<var>value</var>, <var>c</var>, <var>str</var>)</em></dt>
<dd><p>Like <code>EXTRA_CONSTRAINT</code>, but you also get the constraint string passed
in <var>str</var>, so that you can use suffixes to distinguish between different
variants.
</p></dd></dl>

<dl>
<dt><a name="index-EXTRA_005fMEMORY_005fCONSTRAINT"></a>Macro: <strong>EXTRA_MEMORY_CONSTRAINT</strong> <em>(<var>c</var>, <var>str</var>)</em></dt>
<dd><p>A C expression that defines the optional machine-dependent constraint
letters, amongst those accepted by <code>EXTRA_CONSTRAINT</code>, that should
be treated like memory constraints by the reload pass.
</p>
<p>It should return 1 if the operand type represented by the constraint
at the start of <var>str</var>, the first letter of which is the letter <var>c</var>,
comprises a subset of all memory references including
all those whose address is simply a base register.  This allows the reload
pass to reload an operand, if it does not directly correspond to the operand
type of <var>c</var>, by copying its address into a base register.
</p>
<p>For example, on the S/390, some instructions do not accept arbitrary
memory references, but only those that do not make use of an index
register.  The constraint letter &lsquo;<samp>Q</samp>&rsquo; is defined via
<code>EXTRA_CONSTRAINT</code> as representing a memory address of this type.
If the letter &lsquo;<samp>Q</samp>&rsquo; is marked as <code>EXTRA_MEMORY_CONSTRAINT</code>,
a &lsquo;<samp>Q</samp>&rsquo; constraint can handle any memory operand, because the
reload pass knows it can be reloaded by copying the memory address
into a base register if required.  This is analogous to the way
an &lsquo;<samp>o</samp>&rsquo; constraint can handle any memory operand.
</p></dd></dl>

<dl>
<dt><a name="index-EXTRA_005fADDRESS_005fCONSTRAINT"></a>Macro: <strong>EXTRA_ADDRESS_CONSTRAINT</strong> <em>(<var>c</var>, <var>str</var>)</em></dt>
<dd><p>A C expression that defines the optional machine-dependent constraint
letters, amongst those accepted by <code>EXTRA_CONSTRAINT</code> /
<code>EXTRA_CONSTRAINT_STR</code>, that should
be treated like address constraints by the reload pass.
</p>
<p>It should return 1 if the operand type represented by the constraint
at the start of <var>str</var>, which starts with the letter <var>c</var>, comprises
a subset of all memory addresses including
all those that consist of just a base register.  This allows the reload
pass to reload an operand, if it does not directly correspond to the operand
type of <var>str</var>, by copying it into a base register.
</p>
<p>Any constraint marked as <code>EXTRA_ADDRESS_CONSTRAINT</code> can only
be used with the <code>address_operand</code> predicate.  It is treated
analogously to the &lsquo;<samp>p</samp>&rsquo; constraint.
</p></dd></dl>

<hr>
<a name="Stack-and-Calling"></a>
<div class="header">
<p>
Next: <a href="#Varargs" accesskey="n" rel="next">Varargs</a>, Previous: <a href="#Old-Constraints" accesskey="p" rel="prev">Old Constraints</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Stack-Layout-and-Calling-Conventions"></a>
<h3 class="section">17.10 Stack Layout and Calling Conventions</h3>
<a name="index-calling-conventions"></a>

<p>This describes the stack layout and calling conventions.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Frame-Layout" accesskey="1">Frame Layout</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Exception-Handling" accesskey="2">Exception Handling</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Stack-Checking" accesskey="3">Stack Checking</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Frame-Registers" accesskey="4">Frame Registers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Elimination" accesskey="5">Elimination</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Stack-Arguments" accesskey="6">Stack Arguments</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Register-Arguments" accesskey="7">Register Arguments</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Scalar-Return" accesskey="8">Scalar Return</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Aggregate-Return" accesskey="9">Aggregate Return</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Caller-Saves">Caller Saves</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Function-Entry">Function Entry</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Profiling">Profiling</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Tail-Calls">Tail Calls</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Stack-Smashing-Protection">Stack Smashing Protection</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
</td></tr>
</table>

<hr>
<a name="Frame-Layout"></a>
<div class="header">
<p>
Next: <a href="#Exception-Handling" accesskey="n" rel="next">Exception Handling</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Basic-Stack-Layout"></a>
<h4 class="subsection">17.10.1 Basic Stack Layout</h4>
<a name="index-stack-frame-layout"></a>
<a name="index-frame-layout"></a>

<p>Here is the basic stack layout.
</p>
<dl>
<dt><a name="index-STACK_005fGROWS_005fDOWNWARD"></a>Macro: <strong>STACK_GROWS_DOWNWARD</strong></dt>
<dd><p>Define this macro if pushing a word onto the stack moves the stack
pointer to a smaller address.
</p>
<p>When we say, &ldquo;define this macro if &hellip;&rdquo;, it means that the
compiler checks this macro only with <code>#ifdef</code> so the precise
definition used does not matter.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fPUSH_005fCODE"></a>Macro: <strong>STACK_PUSH_CODE</strong></dt>
<dd><p>This macro defines the operation used when something is pushed
on the stack.  In RTL, a push operation will be
<code>(set (mem (STACK_PUSH_CODE (reg sp))) &hellip;)</code>
</p>
<p>The choices are <code>PRE_DEC</code>, <code>POST_DEC</code>, <code>PRE_INC</code>,
and <code>POST_INC</code>.  Which of these is correct depends on
the stack direction and on whether the stack pointer points
to the last item on the stack or whether it points to the
space for the next item on the stack.
</p>
<p>The default is <code>PRE_DEC</code> when <code>STACK_GROWS_DOWNWARD</code> is
defined, which is almost always right, and <code>PRE_INC</code> otherwise,
which is often wrong.
</p></dd></dl>

<dl>
<dt><a name="index-FRAME_005fGROWS_005fDOWNWARD"></a>Macro: <strong>FRAME_GROWS_DOWNWARD</strong></dt>
<dd><p>Define this macro to nonzero value if the addresses of local variable slots
are at negative offsets from the frame pointer.
</p></dd></dl>

<dl>
<dt><a name="index-ARGS_005fGROW_005fDOWNWARD"></a>Macro: <strong>ARGS_GROW_DOWNWARD</strong></dt>
<dd><p>Define this macro if successive arguments to a function occupy decreasing
addresses on the stack.
</p></dd></dl>

<dl>
<dt><a name="index-STARTING_005fFRAME_005fOFFSET"></a>Macro: <strong>STARTING_FRAME_OFFSET</strong></dt>
<dd><p>Offset from the frame pointer to the first local variable slot to be allocated.
</p>
<p>If <code>FRAME_GROWS_DOWNWARD</code>, find the next slot&rsquo;s offset by
subtracting the first slot&rsquo;s length from <code>STARTING_FRAME_OFFSET</code>.
Otherwise, it is found by adding the length of the first slot to the
value <code>STARTING_FRAME_OFFSET</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fALIGNMENT_005fNEEDED"></a>Macro: <strong>STACK_ALIGNMENT_NEEDED</strong></dt>
<dd><p>Define to zero to disable final alignment of the stack during reload.
The nonzero default for this macro is suitable for most ports.
</p>
<p>On ports where <code>STARTING_FRAME_OFFSET</code> is nonzero or where there
is a register save block following the local block that doesn&rsquo;t require
alignment to <code>STACK_BOUNDARY</code>, it may be beneficial to disable
stack alignment and do it in the backend.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fPOINTER_005fOFFSET"></a>Macro: <strong>STACK_POINTER_OFFSET</strong></dt>
<dd><p>Offset from the stack pointer register to the first location at which
outgoing arguments are placed.  If not specified, the default value of
zero is used.  This is the proper value for most machines.
</p>
<p>If <code>ARGS_GROW_DOWNWARD</code>, this is the offset to the location above
the first location at which outgoing arguments are placed.
</p></dd></dl>

<dl>
<dt><a name="index-FIRST_005fPARM_005fOFFSET"></a>Macro: <strong>FIRST_PARM_OFFSET</strong> <em>(<var>fundecl</var>)</em></dt>
<dd><p>Offset from the argument pointer register to the first argument&rsquo;s
address.  On some machines it may depend on the data type of the
function.
</p>
<p>If <code>ARGS_GROW_DOWNWARD</code>, this is the offset to the location above
the first argument&rsquo;s address.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fDYNAMIC_005fOFFSET"></a>Macro: <strong>STACK_DYNAMIC_OFFSET</strong> <em>(<var>fundecl</var>)</em></dt>
<dd><p>Offset from the stack pointer register to an item dynamically allocated
on the stack, e.g., by <code>alloca</code>.
</p>
<p>The default value for this macro is <code>STACK_POINTER_OFFSET</code> plus the
length of the outgoing arguments.  The default is correct for most
machines.  See <samp>function.c</samp> for details.
</p></dd></dl>

<dl>
<dt><a name="index-INITIAL_005fFRAME_005fADDRESS_005fRTX"></a>Macro: <strong>INITIAL_FRAME_ADDRESS_RTX</strong></dt>
<dd><p>A C expression whose value is RTL representing the address of the initial
stack frame. This address is passed to <code>RETURN_ADDR_RTX</code> and
<code>DYNAMIC_CHAIN_ADDRESS</code>.  If you don&rsquo;t define this macro, a reasonable
default value will be used.  Define this macro in order to make frame pointer
elimination work in the presence of <code>__builtin_frame_address (count)</code> and
<code>__builtin_return_address (count)</code> for <code>count</code> not equal to zero.
</p></dd></dl>

<dl>
<dt><a name="index-DYNAMIC_005fCHAIN_005fADDRESS"></a>Macro: <strong>DYNAMIC_CHAIN_ADDRESS</strong> <em>(<var>frameaddr</var>)</em></dt>
<dd><p>A C expression whose value is RTL representing the address in a stack
frame where the pointer to the caller&rsquo;s frame is stored.  Assume that
<var>frameaddr</var> is an RTL expression for the address of the stack frame
itself.
</p>
<p>If you don&rsquo;t define this macro, the default is to return the value
of <var>frameaddr</var>&mdash;that is, the stack frame address is also the
address of the stack word that points to the previous frame.
</p></dd></dl>

<dl>
<dt><a name="index-SETUP_005fFRAME_005fADDRESSES"></a>Macro: <strong>SETUP_FRAME_ADDRESSES</strong></dt>
<dd><p>If defined, a C expression that produces the machine-specific code to
setup the stack so that arbitrary frames can be accessed.  For example,
on the SPARC, we must flush all of the register windows to the stack
before we can access arbitrary stack frames.  You will seldom need to
define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBUILTIN_005fSETJMP_005fFRAME_005fVALUE"></a>Target Hook: <em>rtx</em> <strong>TARGET_BUILTIN_SETJMP_FRAME_VALUE</strong> <em>(void)</em></dt>
<dd><p>This target hook should return an rtx that is used to store
the address of the current frame into the built in <code>setjmp</code> buffer.
The default value, <code>virtual_stack_vars_rtx</code>, is correct for most
machines.  One reason you may need to define this target hook is if
<code>hard_frame_pointer_rtx</code> is the appropriate value on your machine.
</p></dd></dl>

<dl>
<dt><a name="index-FRAME_005fADDR_005fRTX"></a>Macro: <strong>FRAME_ADDR_RTX</strong> <em>(<var>frameaddr</var>)</em></dt>
<dd><p>A C expression whose value is RTL representing the value of the frame
address for the current frame.  <var>frameaddr</var> is the frame pointer
of the current frame.  This is used for __builtin_frame_address.
You need only define this macro if the frame address is not the same
as the frame pointer.  Most machines do not need to define it.
</p></dd></dl>

<dl>
<dt><a name="index-RETURN_005fADDR_005fRTX"></a>Macro: <strong>RETURN_ADDR_RTX</strong> <em>(<var>count</var>, <var>frameaddr</var>)</em></dt>
<dd><p>A C expression whose value is RTL representing the value of the return
address for the frame <var>count</var> steps up from the current frame, after
the prologue.  <var>frameaddr</var> is the frame pointer of the <var>count</var>
frame, or the frame pointer of the <var>count</var> - 1 frame if
<code>RETURN_ADDR_IN_PREVIOUS_FRAME</code> is defined.
</p>
<p>The value of the expression must always be the correct address when
<var>count</var> is zero, but may be <code>NULL_RTX</code> if there is no way to
determine the return address of other frames.
</p></dd></dl>

<dl>
<dt><a name="index-RETURN_005fADDR_005fIN_005fPREVIOUS_005fFRAME"></a>Macro: <strong>RETURN_ADDR_IN_PREVIOUS_FRAME</strong></dt>
<dd><p>Define this if the return address of a particular stack frame is accessed
from the frame pointer of the previous stack frame.
</p></dd></dl>

<dl>
<dt><a name="index-INCOMING_005fRETURN_005fADDR_005fRTX"></a>Macro: <strong>INCOMING_RETURN_ADDR_RTX</strong></dt>
<dd><p>A C expression whose value is RTL representing the location of the
incoming return address at the beginning of any function, before the
prologue.  This RTL is either a <code>REG</code>, indicating that the return
value is saved in &lsquo;<samp>REG</samp>&rsquo;, or a <code>MEM</code> representing a location in
the stack.
</p>
<p>You only need to define this macro if you want to support call frame
debugging information like that provided by DWARF 2.
</p>
<p>If this RTL is a <code>REG</code>, you should also define
<code>DWARF_FRAME_RETURN_COLUMN</code> to <code>DWARF_FRAME_REGNUM (REGNO)</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fALT_005fFRAME_005fRETURN_005fCOLUMN"></a>Macro: <strong>DWARF_ALT_FRAME_RETURN_COLUMN</strong></dt>
<dd><p>A C expression whose value is an integer giving a DWARF 2 column
number that may be used as an alternative return column.  The column
must not correspond to any gcc hard register (that is, it must not
be in the range of <code>DWARF_FRAME_REGNUM</code>).
</p>
<p>This macro can be useful if <code>DWARF_FRAME_RETURN_COLUMN</code> is set to a
general register, but an alternative column needs to be used for signal
frames.  Some targets have also used different frame return columns
over time.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fZERO_005fREG"></a>Macro: <strong>DWARF_ZERO_REG</strong></dt>
<dd><p>A C expression whose value is an integer giving a DWARF 2 register
number that is considered to always have the value zero.  This should
only be defined if the target has an architected zero register, and
someone decided it was a good idea to use that register number to
terminate the stack backtrace.  New ports should avoid this.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDWARF_005fHANDLE_005fFRAME_005fUNSPEC"></a>Target Hook: <em>void</em> <strong>TARGET_DWARF_HANDLE_FRAME_UNSPEC</strong> <em>(const char *<var>label</var>, rtx <var>pattern</var>, int <var>index</var>)</em></dt>
<dd><p>This target hook allows the backend to emit frame-related insns that
contain UNSPECs or UNSPEC_VOLATILEs.  The DWARF 2 call frame debugging
info engine will invoke it on insns of the form
</p><div class="smallexample">
<pre class="smallexample">(set (reg) (unspec [&hellip;] UNSPEC_INDEX))
</pre></div>
<p>and
</p><div class="smallexample">
<pre class="smallexample">(set (reg) (unspec_volatile [&hellip;] UNSPECV_INDEX)).
</pre></div>
<p>to let the backend emit the call frame instructions.  <var>label</var> is
the CFI label attached to the insn, <var>pattern</var> is the pattern of
the insn and <var>index</var> is <code>UNSPEC_INDEX</code> or <code>UNSPECV_INDEX</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INCOMING_005fFRAME_005fSP_005fOFFSET"></a>Macro: <strong>INCOMING_FRAME_SP_OFFSET</strong></dt>
<dd><p>A C expression whose value is an integer giving the offset, in bytes,
from the value of the stack pointer register to the top of the stack
frame at the beginning of any function, before the prologue.  The top of
the frame is defined to be the value of the stack pointer in the
previous frame, just before the call instruction.
</p>
<p>You only need to define this macro if you want to support call frame
debugging information like that provided by DWARF 2.
</p></dd></dl>

<dl>
<dt><a name="index-ARG_005fPOINTER_005fCFA_005fOFFSET"></a>Macro: <strong>ARG_POINTER_CFA_OFFSET</strong> <em>(<var>fundecl</var>)</em></dt>
<dd><p>A C expression whose value is an integer giving the offset, in bytes,
from the argument pointer to the canonical frame address (cfa).  The
final value should coincide with that calculated by
<code>INCOMING_FRAME_SP_OFFSET</code>.  Which is unfortunately not usable
during virtual register instantiation.
</p>
<p>The default value for this macro is
<code>FIRST_PARM_OFFSET (fundecl) + crtl-&gt;args.pretend_args_size</code>,
which is correct for most machines; in general, the arguments are found
immediately before the stack frame.  Note that this is not the case on
some targets that save registers into the caller&rsquo;s frame, such as SPARC
and rs6000, and so such targets need to define this macro.
</p>
<p>You only need to define this macro if the default is incorrect, and you
want to support call frame debugging information like that provided by
DWARF 2.
</p></dd></dl>

<dl>
<dt><a name="index-FRAME_005fPOINTER_005fCFA_005fOFFSET"></a>Macro: <strong>FRAME_POINTER_CFA_OFFSET</strong> <em>(<var>fundecl</var>)</em></dt>
<dd><p>If defined, a C expression whose value is an integer giving the offset
in bytes from the frame pointer to the canonical frame address (cfa).
The final value should coincide with that calculated by
<code>INCOMING_FRAME_SP_OFFSET</code>.
</p>
<p>Normally the CFA is calculated as an offset from the argument pointer,
via <code>ARG_POINTER_CFA_OFFSET</code>, but if the argument pointer is
variable due to the ABI, this may not be possible.  If this macro is
defined, it implies that the virtual register instantiation should be
based on the frame pointer instead of the argument pointer.  Only one
of <code>FRAME_POINTER_CFA_OFFSET</code> and <code>ARG_POINTER_CFA_OFFSET</code>
should be defined.
</p></dd></dl>

<dl>
<dt><a name="index-CFA_005fFRAME_005fBASE_005fOFFSET"></a>Macro: <strong>CFA_FRAME_BASE_OFFSET</strong> <em>(<var>fundecl</var>)</em></dt>
<dd><p>If defined, a C expression whose value is an integer giving the offset
in bytes from the canonical frame address (cfa) to the frame base used
in DWARF 2 debug information.  The default is zero.  A different value
may reduce the size of debug information on some ports.
</p></dd></dl>

<hr>
<a name="Exception-Handling"></a>
<div class="header">
<p>
Next: <a href="#Stack-Checking" accesskey="n" rel="next">Stack Checking</a>, Previous: <a href="#Frame-Layout" accesskey="p" rel="prev">Frame Layout</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Exception-Handling-Support"></a>
<h4 class="subsection">17.10.2 Exception Handling Support</h4>
<a name="index-exception-handling-1"></a>

<dl>
<dt><a name="index-EH_005fRETURN_005fDATA_005fREGNO"></a>Macro: <strong>EH_RETURN_DATA_REGNO</strong> <em>(<var>N</var>)</em></dt>
<dd><p>A C expression whose value is the <var>N</var>th register number used for
data by exception handlers, or <code>INVALID_REGNUM</code> if fewer than
<var>N</var> registers are usable.
</p>
<p>The exception handling library routines communicate with the exception
handlers via a set of agreed upon registers.  Ideally these registers
should be call-clobbered; it is possible to use call-saved registers,
but may negatively impact code size.  The target must support at least
2 data registers, but should define 4 if there are enough free registers.
</p>
<p>You must define this macro if you want to support call frame exception
handling like that provided by DWARF 2.
</p></dd></dl>

<dl>
<dt><a name="index-EH_005fRETURN_005fSTACKADJ_005fRTX"></a>Macro: <strong>EH_RETURN_STACKADJ_RTX</strong></dt>
<dd><p>A C expression whose value is RTL representing a location in which
to store a stack adjustment to be applied before function return.
This is used to unwind the stack to an exception handler&rsquo;s call frame.
It will be assigned zero on code paths that return normally.
</p>
<p>Typically this is a call-clobbered hard register that is otherwise
untouched by the epilogue, but could also be a stack slot.
</p>
<p>Do not define this macro if the stack pointer is saved and restored
by the regular prolog and epilog code in the call frame itself; in
this case, the exception handling library routines will update the
stack location to be restored in place.  Otherwise, you must define
this macro if you want to support call frame exception handling like
that provided by DWARF 2.
</p></dd></dl>

<dl>
<dt><a name="index-EH_005fRETURN_005fHANDLER_005fRTX"></a>Macro: <strong>EH_RETURN_HANDLER_RTX</strong></dt>
<dd><p>A C expression whose value is RTL representing a location in which
to store the address of an exception handler to which we should
return.  It will not be assigned on code paths that return normally.
</p>
<p>Typically this is the location in the call frame at which the normal
return address is stored.  For targets that return by popping an
address off the stack, this might be a memory address just below
the <em>target</em> call frame rather than inside the current call
frame.  If defined, <code>EH_RETURN_STACKADJ_RTX</code> will have already
been assigned, so it may be used to calculate the location of the
target call frame.
</p>
<p>Some targets have more complex requirements than storing to an
address calculable during initial code generation.  In that case
the <code>eh_return</code> instruction pattern should be used instead.
</p>
<p>If you want to support call frame exception handling, you must
define either this macro or the <code>eh_return</code> instruction pattern.
</p></dd></dl>

<dl>
<dt><a name="index-RETURN_005fADDR_005fOFFSET"></a>Macro: <strong>RETURN_ADDR_OFFSET</strong></dt>
<dd><p>If defined, an integer-valued C expression for which rtl will be generated
to add it to the exception handler address before it is searched in the
exception handling tables, and to subtract it again from the address before
using it to return to the exception handler.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fPREFERRED_005fEH_005fDATA_005fFORMAT"></a>Macro: <strong>ASM_PREFERRED_EH_DATA_FORMAT</strong> <em>(<var>code</var>, <var>global</var>)</em></dt>
<dd><p>This macro chooses the encoding of pointers embedded in the exception
handling sections.  If at all possible, this should be defined such
that the exception handling section will not require dynamic relocations,
and so may be read-only.
</p>
<p><var>code</var> is 0 for data, 1 for code labels, 2 for function pointers.
<var>global</var> is true if the symbol may be affected by dynamic relocations.
The macro should return a combination of the <code>DW_EH_PE_*</code> defines
as found in <samp>dwarf2.h</samp>.
</p>
<p>If this macro is not defined, pointers will not be encoded but
represented directly.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fMAYBE_005fOUTPUT_005fENCODED_005fADDR_005fRTX"></a>Macro: <strong>ASM_MAYBE_OUTPUT_ENCODED_ADDR_RTX</strong> <em>(<var>file</var>, <var>encoding</var>, <var>size</var>, <var>addr</var>, <var>done</var>)</em></dt>
<dd><p>This macro allows the target to emit whatever special magic is required
to represent the encoding chosen by <code>ASM_PREFERRED_EH_DATA_FORMAT</code>.
Generic code takes care of pc-relative and indirect encodings; this must
be defined if the target uses text-relative or data-relative encodings.
</p>
<p>This is a C statement that branches to <var>done</var> if the format was
handled.  <var>encoding</var> is the format chosen, <var>size</var> is the number
of bytes that the format occupies, <var>addr</var> is the <code>SYMBOL_REF</code>
to be emitted.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fFALLBACK_005fFRAME_005fSTATE_005fFOR"></a>Macro: <strong>MD_FALLBACK_FRAME_STATE_FOR</strong> <em>(<var>context</var>, <var>fs</var>)</em></dt>
<dd><p>This macro allows the target to add CPU and operating system specific
code to the call-frame unwinder for use when there is no unwind data
available.  The most common reason to implement this macro is to unwind
through signal frames.
</p>
<p>This macro is called from <code>uw_frame_state_for</code> in
<samp>unwind-dw2.c</samp>, <samp>unwind-dw2-xtensa.c</samp> and
<samp>unwind-ia64.c</samp>.  <var>context</var> is an <code>_Unwind_Context</code>;
<var>fs</var> is an <code>_Unwind_FrameState</code>.  Examine <code>context-&gt;ra</code>
for the address of the code being executed and <code>context-&gt;cfa</code> for
the stack pointer value.  If the frame can be decoded, the register
save addresses should be updated in <var>fs</var> and the macro should
evaluate to <code>_URC_NO_REASON</code>.  If the frame cannot be decoded,
the macro should evaluate to <code>_URC_END_OF_STACK</code>.
</p>
<p>For proper signal handling in Java this macro is accompanied by
<code>MAKE_THROW_FRAME</code>, defined in <samp>libjava/include/*-signal.h</samp> headers.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fHANDLE_005fUNWABI"></a>Macro: <strong>MD_HANDLE_UNWABI</strong> <em>(<var>context</var>, <var>fs</var>)</em></dt>
<dd><p>This macro allows the target to add operating system specific code to the
call-frame unwinder to handle the IA-64 <code>.unwabi</code> unwinding directive,
usually used for signal or interrupt frames.
</p>
<p>This macro is called from <code>uw_update_context</code> in libgcc&rsquo;s
<samp>unwind-ia64.c</samp>.  <var>context</var> is an <code>_Unwind_Context</code>;
<var>fs</var> is an <code>_Unwind_FrameState</code>.  Examine <code>fs-&gt;unwabi</code>
for the abi and context in the <code>.unwabi</code> directive.  If the
<code>.unwabi</code> directive can be handled, the register save addresses should
be updated in <var>fs</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUSES_005fWEAK_005fUNWIND_005fINFO"></a>Macro: <strong>TARGET_USES_WEAK_UNWIND_INFO</strong></dt>
<dd><p>A C expression that evaluates to true if the target requires unwind
info to be given comdat linkage.  Define it to be <code>1</code> if comdat
linkage is necessary.  The default is <code>0</code>.
</p></dd></dl>

<hr>
<a name="Stack-Checking"></a>
<div class="header">
<p>
Next: <a href="#Frame-Registers" accesskey="n" rel="next">Frame Registers</a>, Previous: <a href="#Exception-Handling" accesskey="p" rel="prev">Exception Handling</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Specifying-How-Stack-Checking-is-Done"></a>
<h4 class="subsection">17.10.3 Specifying How Stack Checking is Done</h4>

<p>GCC will check that stack references are within the boundaries of the
stack, if the option <samp>-fstack-check</samp> is specified, in one of
three ways:
</p>
<ol>
<li> If the value of the <code>STACK_CHECK_BUILTIN</code> macro is nonzero, GCC
will assume that you have arranged for full stack checking to be done
at appropriate places in the configuration files.  GCC will not do
other special processing.

</li><li> If <code>STACK_CHECK_BUILTIN</code> is zero and the value of the
<code>STACK_CHECK_STATIC_BUILTIN</code> macro is nonzero, GCC will assume
that you have arranged for static stack checking (checking of the
static stack frame of functions) to be done at appropriate places
in the configuration files.  GCC will only emit code to do dynamic
stack checking (checking on dynamic stack allocations) using the third
approach below.

</li><li> If neither of the above are true, GCC will generate code to periodically
&ldquo;probe&rdquo; the stack pointer using the values of the macros defined below.
</li></ol>

<p>If neither STACK_CHECK_BUILTIN nor STACK_CHECK_STATIC_BUILTIN is defined,
GCC will change its allocation strategy for large objects if the option
<samp>-fstack-check</samp> is specified: they will always be allocated
dynamically if their size exceeds <code>STACK_CHECK_MAX_VAR_SIZE</code> bytes.
</p>
<dl>
<dt><a name="index-STACK_005fCHECK_005fBUILTIN"></a>Macro: <strong>STACK_CHECK_BUILTIN</strong></dt>
<dd><p>A nonzero value if stack checking is done by the configuration files in a
machine-dependent manner.  You should define this macro if stack checking
is required by the ABI of your machine or if you would like to do stack
checking in some more efficient way than the generic approach.  The default
value of this macro is zero.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fSTATIC_005fBUILTIN"></a>Macro: <strong>STACK_CHECK_STATIC_BUILTIN</strong></dt>
<dd><p>A nonzero value if static stack checking is done by the configuration files
in a machine-dependent manner.  You should define this macro if you would
like to do static stack checking in some more efficient way than the generic
approach.  The default value of this macro is zero.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fPROBE_005fINTERVAL_005fEXP"></a>Macro: <strong>STACK_CHECK_PROBE_INTERVAL_EXP</strong></dt>
<dd><p>An integer specifying the interval at which GCC must generate stack probe
instructions, defined as 2 raised to this integer.  You will normally
define this macro so that the interval be no larger than the size of
the &ldquo;guard pages&rdquo; at the end of a stack area.  The default value
of 12 (4096-byte interval) is suitable for most systems.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fMOVING_005fSP"></a>Macro: <strong>STACK_CHECK_MOVING_SP</strong></dt>
<dd><p>An integer which is nonzero if GCC should move the stack pointer page by page
when doing probes.  This can be necessary on systems where the stack pointer
contains the bottom address of the memory area accessible to the executing
thread at any point in time.  In this situation an alternate signal stack
is required in order to be able to recover from a stack overflow.  The
default value of this macro is zero.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fPROTECT"></a>Macro: <strong>STACK_CHECK_PROTECT</strong></dt>
<dd><p>The number of bytes of stack needed to recover from a stack overflow, for
languages where such a recovery is supported.  The default value of 75 words
with the <code>setjmp</code>/<code>longjmp</code>-based exception handling mechanism and
8192 bytes with other exception handling mechanisms should be adequate for
most machines.
</p></dd></dl>

<p>The following macros are relevant only if neither STACK_CHECK_BUILTIN
nor STACK_CHECK_STATIC_BUILTIN is defined; you can omit them altogether
in the opposite case.
</p>
<dl>
<dt><a name="index-STACK_005fCHECK_005fMAX_005fFRAME_005fSIZE"></a>Macro: <strong>STACK_CHECK_MAX_FRAME_SIZE</strong></dt>
<dd><p>The maximum size of a stack frame, in bytes.  GCC will generate probe
instructions in non-leaf functions to ensure at least this many bytes of
stack are available.  If a stack frame is larger than this size, stack
checking will not be reliable and GCC will issue a warning.  The
default is chosen so that GCC only generates one instruction on most
systems.  You should normally not change the default value of this macro.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fFIXED_005fFRAME_005fSIZE"></a>Macro: <strong>STACK_CHECK_FIXED_FRAME_SIZE</strong></dt>
<dd><p>GCC uses this value to generate the above warning message.  It
represents the amount of fixed frame used by a function, not including
space for any callee-saved registers, temporaries and user variables.
You need only specify an upper bound for this amount and will normally
use the default of four words.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fCHECK_005fMAX_005fVAR_005fSIZE"></a>Macro: <strong>STACK_CHECK_MAX_VAR_SIZE</strong></dt>
<dd><p>The maximum size, in bytes, of an object that GCC will place in the
fixed area of the stack frame when the user specifies
<samp>-fstack-check</samp>.
GCC computed the default from the values of the above macros and you will
normally not need to override that default.
</p></dd></dl>

<hr>
<a name="Frame-Registers"></a>
<div class="header">
<p>
Next: <a href="#Elimination" accesskey="n" rel="next">Elimination</a>, Previous: <a href="#Stack-Checking" accesskey="p" rel="prev">Stack Checking</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Registers-That-Address-the-Stack-Frame"></a>
<h4 class="subsection">17.10.4 Registers That Address the Stack Frame</h4>

<p>This discusses registers that address the stack frame.
</p>
<dl>
<dt><a name="index-STACK_005fPOINTER_005fREGNUM"></a>Macro: <strong>STACK_POINTER_REGNUM</strong></dt>
<dd><p>The register number of the stack pointer register, which must also be a
fixed register according to <code>FIXED_REGISTERS</code>.  On most machines,
the hardware determines which register this is.
</p></dd></dl>

<dl>
<dt><a name="index-FRAME_005fPOINTER_005fREGNUM"></a>Macro: <strong>FRAME_POINTER_REGNUM</strong></dt>
<dd><p>The register number of the frame pointer register, which is used to
access automatic variables in the stack frame.  On some machines, the
hardware determines which register this is.  On other machines, you can
choose any register you wish for this purpose.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fFRAME_005fPOINTER_005fREGNUM"></a>Macro: <strong>HARD_FRAME_POINTER_REGNUM</strong></dt>
<dd><p>On some machines the offset between the frame pointer and starting
offset of the automatic variables is not known until after register
allocation has been done (for example, because the saved registers are
between these two locations).  On those machines, define
<code>FRAME_POINTER_REGNUM</code> the number of a special, fixed register to
be used internally until the offset is known, and define
<code>HARD_FRAME_POINTER_REGNUM</code> to be the actual hard register number
used for the frame pointer.
</p>
<p>You should define this macro only in the very rare circumstances when it
is not possible to calculate the offset between the frame pointer and
the automatic variables until after register allocation has been
completed.  When this macro is defined, you must also indicate in your
definition of <code>ELIMINABLE_REGS</code> how to eliminate
<code>FRAME_POINTER_REGNUM</code> into either <code>HARD_FRAME_POINTER_REGNUM</code>
or <code>STACK_POINTER_REGNUM</code>.
</p>
<p>Do not define this macro if it would be the same as
<code>FRAME_POINTER_REGNUM</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ARG_005fPOINTER_005fREGNUM"></a>Macro: <strong>ARG_POINTER_REGNUM</strong></dt>
<dd><p>The register number of the arg pointer register, which is used to access
the function&rsquo;s argument list.  On some machines, this is the same as the
frame pointer register.  On some machines, the hardware determines which
register this is.  On other machines, you can choose any register you
wish for this purpose.  If this is not the same register as the frame
pointer register, then you must mark it as a fixed register according to
<code>FIXED_REGISTERS</code>, or arrange to be able to eliminate it
(see <a href="#Elimination">Elimination</a>).
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fFRAME_005fPOINTER_005fIS_005fFRAME_005fPOINTER"></a>Macro: <strong>HARD_FRAME_POINTER_IS_FRAME_POINTER</strong></dt>
<dd><p>Define this to a preprocessor constant that is nonzero if
<code>hard_frame_pointer_rtx</code> and <code>frame_pointer_rtx</code> should be
the same.  The default definition is &lsquo;<samp>(HARD_FRAME_POINTER_REGNUM
== FRAME_POINTER_REGNUM)</samp>&rsquo;; you only need to define this macro if that
definition is not suitable for use in preprocessor conditionals.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fFRAME_005fPOINTER_005fIS_005fARG_005fPOINTER"></a>Macro: <strong>HARD_FRAME_POINTER_IS_ARG_POINTER</strong></dt>
<dd><p>Define this to a preprocessor constant that is nonzero if
<code>hard_frame_pointer_rtx</code> and <code>arg_pointer_rtx</code> should be the
same.  The default definition is &lsquo;<samp>(HARD_FRAME_POINTER_REGNUM ==
ARG_POINTER_REGNUM)</samp>&rsquo;; you only need to define this macro if that
definition is not suitable for use in preprocessor conditionals.
</p></dd></dl>

<dl>
<dt><a name="index-RETURN_005fADDRESS_005fPOINTER_005fREGNUM"></a>Macro: <strong>RETURN_ADDRESS_POINTER_REGNUM</strong></dt>
<dd><p>The register number of the return address pointer register, which is used to
access the current function&rsquo;s return address from the stack.  On some
machines, the return address is not at a fixed offset from the frame
pointer or stack pointer or argument pointer.  This register can be defined
to point to the return address on the stack, and then be converted by
<code>ELIMINABLE_REGS</code> into either the frame pointer or stack pointer.
</p>
<p>Do not define this macro unless there is no other way to get the return
address from the stack.
</p></dd></dl>

<dl>
<dt><a name="index-STATIC_005fCHAIN_005fREGNUM"></a>Macro: <strong>STATIC_CHAIN_REGNUM</strong></dt>
<dt><a name="index-STATIC_005fCHAIN_005fINCOMING_005fREGNUM"></a>Macro: <strong>STATIC_CHAIN_INCOMING_REGNUM</strong></dt>
<dd><p>Register numbers used for passing a function&rsquo;s static chain pointer.  If
register windows are used, the register number as seen by the called
function is <code>STATIC_CHAIN_INCOMING_REGNUM</code>, while the register
number as seen by the calling function is <code>STATIC_CHAIN_REGNUM</code>.  If
these registers are the same, <code>STATIC_CHAIN_INCOMING_REGNUM</code> need
not be defined.
</p>
<p>The static chain register need not be a fixed register.
</p>
<p>If the static chain is passed in memory, these macros should not be
defined; instead, the <code>TARGET_STATIC_CHAIN</code> hook should be used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTATIC_005fCHAIN"></a>Target Hook: <em>rtx</em> <strong>TARGET_STATIC_CHAIN</strong> <em>(const_tree <var>fndecl</var>, bool <var>incoming_p</var>)</em></dt>
<dd><p>This hook replaces the use of <code>STATIC_CHAIN_REGNUM</code> et al for
targets that may use different static chain locations for different
nested functions.  This may be required if the target has function
attributes that affect the calling conventions of the function and
those calling conventions use different static chain locations.
</p>
<p>The default version of this hook uses <code>STATIC_CHAIN_REGNUM</code> et al.
</p>
<p>If the static chain is passed in memory, this hook should be used to
provide rtx giving <code>mem</code> expressions that denote where they are stored.
Often the <code>mem</code> expression as seen by the caller will be at an offset
from the stack pointer and the <code>mem</code> expression as seen by the callee
will be at an offset from the frame pointer.
<a name="index-stack_005fpointer_005frtx"></a>
<a name="index-frame_005fpointer_005frtx"></a>
<a name="index-arg_005fpointer_005frtx"></a>
The variables <code>stack_pointer_rtx</code>, <code>frame_pointer_rtx</code>, and
<code>arg_pointer_rtx</code> will have been initialized and should be used
to refer to those items.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fFRAME_005fREGISTERS"></a>Macro: <strong>DWARF_FRAME_REGISTERS</strong></dt>
<dd><p>This macro specifies the maximum number of hard registers that can be
saved in a call frame.  This is used to size data structures used in
DWARF2 exception handling.
</p>
<p>Prior to GCC 3.0, this macro was needed in order to establish a stable
exception handling ABI in the face of adding new hard registers for ISA
extensions.  In GCC 3.0 and later, the EH ABI is insulated from changes
in the number of hard registers.  Nevertheless, this macro can still be
used to reduce the runtime memory requirements of the exception handling
routines, which can be substantial if the ISA contains a lot of
registers that are not call-saved.
</p>
<p>If this macro is not defined, it defaults to
<code>FIRST_PSEUDO_REGISTER</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PRE_005fGCC3_005fDWARF_005fFRAME_005fREGISTERS"></a>Macro: <strong>PRE_GCC3_DWARF_FRAME_REGISTERS</strong></dt>
<dd>
<p>This macro is similar to <code>DWARF_FRAME_REGISTERS</code>, but is provided
for backward compatibility in pre GCC 3.0 compiled code.
</p>
<p>If this macro is not defined, it defaults to
<code>DWARF_FRAME_REGISTERS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fREG_005fTO_005fUNWIND_005fCOLUMN"></a>Macro: <strong>DWARF_REG_TO_UNWIND_COLUMN</strong> <em>(<var>regno</var>)</em></dt>
<dd>
<p>Define this macro if the target&rsquo;s representation for dwarf registers
is different than the internal representation for unwind column.
Given a dwarf register, this macro should return the internal unwind
column number to use instead.
</p>
<p>See the PowerPC&rsquo;s SPE target for an example.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fFRAME_005fREGNUM"></a>Macro: <strong>DWARF_FRAME_REGNUM</strong> <em>(<var>regno</var>)</em></dt>
<dd>
<p>Define this macro if the target&rsquo;s representation for dwarf registers
used in .eh_frame or .debug_frame is different from that used in other
debug info sections.  Given a GCC hard register number, this macro
should return the .eh_frame register number.  The default is
<code>DBX_REGISTER_NUMBER (<var>regno</var>)</code>.
</p>
</dd></dl>

<dl>
<dt><a name="index-DWARF2_005fFRAME_005fREG_005fOUT"></a>Macro: <strong>DWARF2_FRAME_REG_OUT</strong> <em>(<var>regno</var>, <var>for_eh</var>)</em></dt>
<dd>
<p>Define this macro to map register numbers held in the call frame info
that GCC has collected using <code>DWARF_FRAME_REGNUM</code> to those that
should be output in .debug_frame (<code><var>for_eh</var></code> is zero) and
.eh_frame (<code><var>for_eh</var></code> is nonzero).  The default is to
return <code><var>regno</var></code>.
</p>
</dd></dl>

<dl>
<dt><a name="index-REG_005fVALUE_005fIN_005fUNWIND_005fCONTEXT"></a>Macro: <strong>REG_VALUE_IN_UNWIND_CONTEXT</strong></dt>
<dd>
<p>Define this macro if the target stores register values as
<code>_Unwind_Word</code> type in unwind context.  It should be defined if
target register size is larger than the size of <code>void *</code>.  The
default is to store register values as <code>void *</code> type.
</p>
</dd></dl>

<dl>
<dt><a name="index-ASSUME_005fEXTENDED_005fUNWIND_005fCONTEXT"></a>Macro: <strong>ASSUME_EXTENDED_UNWIND_CONTEXT</strong></dt>
<dd>
<p>Define this macro to be 1 if the target always uses extended unwind
context with version, args_size and by_value fields.  If it is undefined,
it will be defined to 1 when <code>REG_VALUE_IN_UNWIND_CONTEXT</code> is
defined and 0 otherwise.
</p>
</dd></dl>

<hr>
<a name="Elimination"></a>
<div class="header">
<p>
Next: <a href="#Stack-Arguments" accesskey="n" rel="next">Stack Arguments</a>, Previous: <a href="#Frame-Registers" accesskey="p" rel="prev">Frame Registers</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Eliminating-Frame-Pointer-and-Arg-Pointer"></a>
<h4 class="subsection">17.10.5 Eliminating Frame Pointer and Arg Pointer</h4>

<p>This is about eliminating the frame pointer and arg pointer.
</p>
<dl>
<dt><a name="index-TARGET_005fFRAME_005fPOINTER_005fREQUIRED"></a>Target Hook: <em>bool</em> <strong>TARGET_FRAME_POINTER_REQUIRED</strong> <em>(void)</em></dt>
<dd><p>This target hook should return <code>true</code> if a function must have and use
a frame pointer.  This target hook is called in the reload pass.  If its return
value is <code>true</code> the function will have a frame pointer.
</p>
<p>This target hook can in principle examine the current function and decide
according to the facts, but on most machines the constant <code>false</code> or the
constant <code>true</code> suffices.  Use <code>false</code> when the machine allows code
to be generated with no frame pointer, and doing so saves some time or space.
Use <code>true</code> when there is no possible advantage to avoiding a frame
pointer.
</p>
<p>In certain cases, the compiler does not know how to produce valid code
without a frame pointer.  The compiler recognizes those cases and
automatically gives the function a frame pointer regardless of what
<code>TARGET_FRAME_POINTER_REQUIRED</code> returns.  You don&rsquo;t need to worry about
them.
</p>
<p>In a function that does not require a frame pointer, the frame pointer
register can be allocated for ordinary usage, unless you mark it as a
fixed register.  See <code>FIXED_REGISTERS</code> for more information.
</p>
<p>Default return value is <code>false</code>.
</p></dd></dl>

<a name="index-get_005fframe_005fsize"></a>
<dl>
<dt><a name="index-INITIAL_005fFRAME_005fPOINTER_005fOFFSET"></a>Macro: <strong>INITIAL_FRAME_POINTER_OFFSET</strong> <em>(<var>depth-var</var>)</em></dt>
<dd><p>A C statement to store in the variable <var>depth-var</var> the difference
between the frame pointer and the stack pointer values immediately after
the function prologue.  The value would be computed from information
such as the result of <code>get_frame_size ()</code> and the tables of
registers <code>regs_ever_live</code> and <code>call_used_regs</code>.
</p>
<p>If <code>ELIMINABLE_REGS</code> is defined, this macro will be not be used and
need not be defined.  Otherwise, it must be defined even if
<code>TARGET_FRAME_POINTER_REQUIRED</code> always returns true; in that
case, you may set <var>depth-var</var> to anything.
</p></dd></dl>

<dl>
<dt><a name="index-ELIMINABLE_005fREGS"></a>Macro: <strong>ELIMINABLE_REGS</strong></dt>
<dd><p>If defined, this macro specifies a table of register pairs used to
eliminate unneeded registers that point into the stack frame.  If it is not
defined, the only elimination attempted by the compiler is to replace
references to the frame pointer with references to the stack pointer.
</p>
<p>The definition of this macro is a list of structure initializations, each
of which specifies an original and replacement register.
</p>
<p>On some machines, the position of the argument pointer is not known until
the compilation is completed.  In such a case, a separate hard register
must be used for the argument pointer.  This register can be eliminated by
replacing it with either the frame pointer or the argument pointer,
depending on whether or not the frame pointer has been eliminated.
</p>
<p>In this case, you might specify:
</p><div class="smallexample">
<pre class="smallexample">#define ELIMINABLE_REGS  \
{{ARG_POINTER_REGNUM, STACK_POINTER_REGNUM}, \
 {ARG_POINTER_REGNUM, FRAME_POINTER_REGNUM}, \
 {FRAME_POINTER_REGNUM, STACK_POINTER_REGNUM}}
</pre></div>

<p>Note that the elimination of the argument pointer with the stack pointer is
specified first since that is the preferred elimination.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCAN_005fELIMINATE"></a>Target Hook: <em>bool</em> <strong>TARGET_CAN_ELIMINATE</strong> <em>(const int <var>from_reg</var>, const int <var>to_reg</var>)</em></dt>
<dd><p>This target hook should returns <code>true</code> if the compiler is allowed to
try to replace register number <var>from_reg</var> with register number
<var>to_reg</var>.  This target hook need only be defined if <code>ELIMINABLE_REGS</code>
is defined, and will usually be <code>true</code>, since most of the cases
preventing register elimination are things that the compiler already
knows about.
</p>
<p>Default return value is <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INITIAL_005fELIMINATION_005fOFFSET"></a>Macro: <strong>INITIAL_ELIMINATION_OFFSET</strong> <em>(<var>from-reg</var>, <var>to-reg</var>, <var>offset-var</var>)</em></dt>
<dd><p>This macro is similar to <code>INITIAL_FRAME_POINTER_OFFSET</code>.  It
specifies the initial difference between the specified pair of
registers.  This macro must be defined if <code>ELIMINABLE_REGS</code> is
defined.
</p></dd></dl>

<hr>
<a name="Stack-Arguments"></a>
<div class="header">
<p>
Next: <a href="#Register-Arguments" accesskey="n" rel="next">Register Arguments</a>, Previous: <a href="#Elimination" accesskey="p" rel="prev">Elimination</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Passing-Function-Arguments-on-the-Stack"></a>
<h4 class="subsection">17.10.6 Passing Function Arguments on the Stack</h4>
<a name="index-arguments-on-stack"></a>
<a name="index-stack-arguments"></a>

<p>The macros in this section control how arguments are passed
on the stack.  See the following section for other macros that
control passing certain arguments in registers.
</p>
<dl>
<dt><a name="index-TARGET_005fPROMOTE_005fPROTOTYPES"></a>Target Hook: <em>bool</em> <strong>TARGET_PROMOTE_PROTOTYPES</strong> <em>(const_tree <var>fntype</var>)</em></dt>
<dd><p>This target hook returns <code>true</code> if an argument declared in a
prototype as an integral type smaller than <code>int</code> should actually be
passed as an <code>int</code>.  In addition to avoiding errors in certain
cases of mismatch, it also makes for better code on certain machines.
The default is to not promote prototypes.
</p></dd></dl>

<dl>
<dt><a name="index-PUSH_005fARGS"></a>Macro: <strong>PUSH_ARGS</strong></dt>
<dd><p>A C expression.  If nonzero, push insns will be used to pass
outgoing arguments.
If the target machine does not have a push instruction, set it to zero.
That directs GCC to use an alternate strategy: to
allocate the entire argument block and then store the arguments into
it.  When <code>PUSH_ARGS</code> is nonzero, <code>PUSH_ROUNDING</code> must be defined too.
</p></dd></dl>

<dl>
<dt><a name="index-PUSH_005fARGS_005fREVERSED"></a>Macro: <strong>PUSH_ARGS_REVERSED</strong></dt>
<dd><p>A C expression.  If nonzero, function arguments will be evaluated from
last to first, rather than from first to last.  If this macro is not
defined, it defaults to <code>PUSH_ARGS</code> on targets where the stack
and args grow in opposite directions, and 0 otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-PUSH_005fROUNDING"></a>Macro: <strong>PUSH_ROUNDING</strong> <em>(<var>npushed</var>)</em></dt>
<dd><p>A C expression that is the number of bytes actually pushed onto the
stack when an instruction attempts to push <var>npushed</var> bytes.
</p>
<p>On some machines, the definition
</p>
<div class="smallexample">
<pre class="smallexample">#define PUSH_ROUNDING(BYTES) (BYTES)
</pre></div>

<p>will suffice.  But on other machines, instructions that appear
to push one byte actually push two bytes in an attempt to maintain
alignment.  Then the definition should be
</p>
<div class="smallexample">
<pre class="smallexample">#define PUSH_ROUNDING(BYTES) (((BYTES) + 1) &amp; ~1)
</pre></div>

<p>If the value of this macro has a type, it should be an unsigned type.
</p></dd></dl>

<a name="index-outgoing_005fargs_005fsize"></a>
<a name="index-crtl_002d_003eoutgoing_005fargs_005fsize"></a>
<dl>
<dt><a name="index-ACCUMULATE_005fOUTGOING_005fARGS"></a>Macro: <strong>ACCUMULATE_OUTGOING_ARGS</strong></dt>
<dd><p>A C expression.  If nonzero, the maximum amount of space required for outgoing arguments
will be computed and placed into
<code>crtl-&gt;outgoing_args_size</code>.  No space will be pushed
onto the stack for each call; instead, the function prologue should
increase the stack frame size by this amount.
</p>
<p>Setting both <code>PUSH_ARGS</code> and <code>ACCUMULATE_OUTGOING_ARGS</code>
is not proper.
</p></dd></dl>

<dl>
<dt><a name="index-REG_005fPARM_005fSTACK_005fSPACE"></a>Macro: <strong>REG_PARM_STACK_SPACE</strong> <em>(<var>fndecl</var>)</em></dt>
<dd><p>Define this macro if functions should assume that stack space has been
allocated for arguments even when their values are passed in
registers.
</p>
<p>The value of this macro is the size, in bytes, of the area reserved for
arguments passed in registers for the function represented by <var>fndecl</var>,
which can be zero if GCC is calling a library function.
The argument <var>fndecl</var> can be the FUNCTION_DECL, or the type itself
of the function.
</p>
<p>This space can be allocated by the caller, or be a part of the
machine-dependent stack frame: <code>OUTGOING_REG_PARM_STACK_SPACE</code> says
which.
</p></dd></dl>

<dl>
<dt><a name="index-INCOMING_005fREG_005fPARM_005fSTACK_005fSPACE"></a>Macro: <strong>INCOMING_REG_PARM_STACK_SPACE</strong> <em>(<var>fndecl</var>)</em></dt>
<dd><p>Like <code>REG_PARM_STACK_SPACE</code>, but for incoming register arguments.
Define this macro if space guaranteed when compiling a function body
is different to space required when making a call, a situation that
can arise with K&amp;R style function definitions.
</p></dd></dl>

<dl>
<dt><a name="index-OUTGOING_005fREG_005fPARM_005fSTACK_005fSPACE"></a>Macro: <strong>OUTGOING_REG_PARM_STACK_SPACE</strong> <em>(<var>fntype</var>)</em></dt>
<dd><p>Define this to a nonzero value if it is the responsibility of the
caller to allocate the area reserved for arguments passed in registers
when calling a function of <var>fntype</var>.  <var>fntype</var> may be NULL
if the function called is a library function.
</p>
<p>If <code>ACCUMULATE_OUTGOING_ARGS</code> is defined, this macro controls
whether the space for these arguments counts in the value of
<code>crtl-&gt;outgoing_args_size</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STACK_005fPARMS_005fIN_005fREG_005fPARM_005fAREA"></a>Macro: <strong>STACK_PARMS_IN_REG_PARM_AREA</strong></dt>
<dd><p>Define this macro if <code>REG_PARM_STACK_SPACE</code> is defined, but the
stack parameters don&rsquo;t skip the area specified by it.
</p>
<p>Normally, when a parameter is not passed in registers, it is placed on the
stack beyond the <code>REG_PARM_STACK_SPACE</code> area.  Defining this macro
suppresses this behavior and causes the parameter to be passed on the
stack in its natural location.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fRETURN_005fPOPS_005fARGS"></a>Target Hook: <em>int</em> <strong>TARGET_RETURN_POPS_ARGS</strong> <em>(tree <var>fundecl</var>, tree <var>funtype</var>, int <var>size</var>)</em></dt>
<dd><p>This target hook returns the number of bytes of its own arguments that
a function pops on returning, or 0 if the function pops no arguments
and the caller must therefore pop them all after the function returns.
</p>
<p><var>fundecl</var> is a C variable whose value is a tree node that describes
the function in question.  Normally it is a node of type
<code>FUNCTION_DECL</code> that describes the declaration of the function.
From this you can obtain the <code>DECL_ATTRIBUTES</code> of the function.
</p>
<p><var>funtype</var> is a C variable whose value is a tree node that
describes the function in question.  Normally it is a node of type
<code>FUNCTION_TYPE</code> that describes the data type of the function.
From this it is possible to obtain the data types of the value and
arguments (if known).
</p>
<p>When a call to a library function is being considered, <var>fundecl</var>
will contain an identifier node for the library function.  Thus, if
you need to distinguish among various library functions, you can do so
by their names.  Note that &ldquo;library function&rdquo; in this context means
a function used to perform arithmetic, whose name is known specially
in the compiler and was not mentioned in the C code being compiled.
</p>
<p><var>size</var> is the number of bytes of arguments passed on the
stack.  If a variable number of bytes is passed, it is zero, and
argument popping will always be the responsibility of the calling function.
</p>
<p>On the VAX, all functions always pop their arguments, so the definition
of this macro is <var>size</var>.  On the 68000, using the standard
calling convention, no functions pop their arguments, so the value of
the macro is always 0 in this case.  But an alternative calling
convention is available in which functions that take a fixed number of
arguments pop them but other functions (such as <code>printf</code>) pop
nothing (the caller pops all).  When this convention is in use,
<var>funtype</var> is examined to determine whether a function takes a fixed
number of arguments.
</p></dd></dl>

<dl>
<dt><a name="index-CALL_005fPOPS_005fARGS"></a>Macro: <strong>CALL_POPS_ARGS</strong> <em>(<var>cum</var>)</em></dt>
<dd><p>A C expression that should indicate the number of bytes a call sequence
pops off the stack.  It is added to the value of <code>RETURN_POPS_ARGS</code>
when compiling a function call.
</p>
<p><var>cum</var> is the variable in which all arguments to the called function
have been accumulated.
</p>
<p>On certain architectures, such as the SH5, a call trampoline is used
that pops certain registers off the stack, depending on the arguments
that have been passed to the function.  Since this is a property of the
call site, not of the called function, <code>RETURN_POPS_ARGS</code> is not
appropriate.
</p></dd></dl>

<hr>
<a name="Register-Arguments"></a>
<div class="header">
<p>
Next: <a href="#Scalar-Return" accesskey="n" rel="next">Scalar Return</a>, Previous: <a href="#Stack-Arguments" accesskey="p" rel="prev">Stack Arguments</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Passing-Arguments-in-Registers"></a>
<h4 class="subsection">17.10.7 Passing Arguments in Registers</h4>
<a name="index-arguments-in-registers"></a>
<a name="index-registers-arguments"></a>

<p>This section describes the macros which let you control how various
types of arguments are passed in registers or how they are arranged in
the stack.
</p>
<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fARG"></a>Target Hook: <em>rtx</em> <strong>TARGET_FUNCTION_ARG</strong> <em>(cumulative_args_t <var>ca</var>, enum machine_mode <var>mode</var>, const_tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>Return an RTX indicating whether a function argument is passed in a
register and if so, which register.
</p>
<p>The arguments are <var>ca</var>, which summarizes all the previous
arguments; <var>mode</var>, the machine mode of the argument; <var>type</var>,
the data type of the argument as a tree node or 0 if that is not known
(which happens for C support library functions); and <var>named</var>,
which is <code>true</code> for an ordinary argument and <code>false</code> for
nameless arguments that correspond to &lsquo;<samp>&hellip;</samp>&rsquo; in the called
function&rsquo;s prototype.  <var>type</var> can be an incomplete type if a
syntax error has previously occurred.
</p>
<p>The return value is usually either a <code>reg</code> RTX for the hard
register in which to pass the argument, or zero to pass the argument
on the stack.
</p>
<p>The value of the expression can also be a <code>parallel</code> RTX.  This is
used when an argument is passed in multiple locations.  The mode of the
<code>parallel</code> should be the mode of the entire argument.  The
<code>parallel</code> holds any number of <code>expr_list</code> pairs; each one
describes where part of the argument is passed.  In each
<code>expr_list</code> the first operand must be a <code>reg</code> RTX for the hard
register in which to pass this part of the argument, and the mode of the
register RTX indicates how large this part of the argument is.  The
second operand of the <code>expr_list</code> is a <code>const_int</code> which gives
the offset in bytes into the entire argument of where this part starts.
As a special exception the first <code>expr_list</code> in the <code>parallel</code>
RTX may have a first operand of zero.  This indicates that the entire
argument is also stored on the stack.
</p>
<p>The last time this hook is called, it is called with <code>MODE ==
VOIDmode</code>, and its result is passed to the <code>call</code> or <code>call_value</code>
pattern as operands 2 and 3 respectively.
</p>
<a name="index-stdarg_002eh-and-register-arguments"></a>
<p>The usual way to make the ISO library <samp>stdarg.h</samp> work on a
machine where some arguments are usually passed in registers, is to
cause nameless arguments to be passed on the stack instead.  This is
done by making <code>TARGET_FUNCTION_ARG</code> return 0 whenever
<var>named</var> is <code>false</code>.
</p>
<a name="index-TARGET_005fMUST_005fPASS_005fIN_005fSTACK_002c-and-TARGET_005fFUNCTION_005fARG"></a>
<a name="index-REG_005fPARM_005fSTACK_005fSPACE_002c-and-TARGET_005fFUNCTION_005fARG"></a>
<p>You may use the hook <code>targetm.calls.must_pass_in_stack</code>
in the definition of this macro to determine if this argument is of a
type that must be passed in the stack.  If <code>REG_PARM_STACK_SPACE</code>
is not defined and <code>TARGET_FUNCTION_ARG</code> returns nonzero for such an
argument, the compiler will abort.  If <code>REG_PARM_STACK_SPACE</code> is
defined, the argument will be computed in the stack and then loaded into
a register.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMUST_005fPASS_005fIN_005fSTACK"></a>Target Hook: <em>bool</em> <strong>TARGET_MUST_PASS_IN_STACK</strong> <em>(enum machine_mode <var>mode</var>, const_tree <var>type</var>)</em></dt>
<dd><p>This target hook should return <code>true</code> if we should not pass <var>type</var>
solely in registers.  The file <samp>expr.h</samp> defines a
definition that is usually appropriate, refer to <samp>expr.h</samp> for additional
documentation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fINCOMING_005fARG"></a>Target Hook: <em>rtx</em> <strong>TARGET_FUNCTION_INCOMING_ARG</strong> <em>(cumulative_args_t <var>ca</var>, enum machine_mode <var>mode</var>, const_tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>Define this hook if the target machine has &ldquo;register windows&rdquo;, so
that the register in which a function sees an arguments is not
necessarily the same as the one in which the caller passed the
argument.
</p>
<p>For such machines, <code>TARGET_FUNCTION_ARG</code> computes the register in
which the caller passes the value, and
<code>TARGET_FUNCTION_INCOMING_ARG</code> should be defined in a similar
fashion to tell the function being called where the arguments will
arrive.
</p>
<p>If <code>TARGET_FUNCTION_INCOMING_ARG</code> is not defined,
<code>TARGET_FUNCTION_ARG</code> serves both purposes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fARG_005fPARTIAL_005fBYTES"></a>Target Hook: <em>int</em> <strong>TARGET_ARG_PARTIAL_BYTES</strong> <em>(cumulative_args_t <var>cum</var>, enum machine_mode <var>mode</var>, tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>This target hook returns the number of bytes at the beginning of an
argument that must be put in registers.  The value must be zero for
arguments that are passed entirely in registers or that are entirely
pushed on the stack.
</p>
<p>On some machines, certain arguments must be passed partially in
registers and partially in memory.  On these machines, typically the
first few words of arguments are passed in registers, and the rest
on the stack.  If a multi-word argument (a <code>double</code> or a
structure) crosses that boundary, its first few words must be passed
in registers and the rest must be pushed.  This macro tells the
compiler when this occurs, and how many bytes should go in registers.
</p>
<p><code>TARGET_FUNCTION_ARG</code> for these arguments should return the first
register to be used by the caller for this argument; likewise
<code>TARGET_FUNCTION_INCOMING_ARG</code>, for the called function.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPASS_005fBY_005fREFERENCE"></a>Target Hook: <em>bool</em> <strong>TARGET_PASS_BY_REFERENCE</strong> <em>(cumulative_args_t <var>cum</var>, enum machine_mode <var>mode</var>, const_tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>This target hook should return <code>true</code> if an argument at the
position indicated by <var>cum</var> should be passed by reference.  This
predicate is queried after target independent reasons for being
passed by reference, such as <code>TREE_ADDRESSABLE (type)</code>.
</p>
<p>If the hook returns true, a copy of that argument is made in memory and a
pointer to the argument is passed instead of the argument itself.
The pointer is passed in whatever way is appropriate for passing a pointer
to that type.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCALLEE_005fCOPIES"></a>Target Hook: <em>bool</em> <strong>TARGET_CALLEE_COPIES</strong> <em>(cumulative_args_t <var>cum</var>, enum machine_mode <var>mode</var>, const_tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>The function argument described by the parameters to this hook is
known to be passed by reference.  The hook should return true if the
function argument should be copied by the callee instead of copied
by the caller.
</p>
<p>For any argument for which the hook returns true, if it can be
determined that the argument is not modified, then a copy need
not be generated.
</p>
<p>The default version of this hook always returns false.
</p></dd></dl>

<dl>
<dt><a name="index-CUMULATIVE_005fARGS"></a>Macro: <strong>CUMULATIVE_ARGS</strong></dt>
<dd><p>A C type for declaring a variable that is used as the first argument
of <code>TARGET_FUNCTION_ARG</code> and other related values.  For some
target machines, the type <code>int</code> suffices and can hold the number
of bytes of argument so far.
</p>
<p>There is no need to record in <code>CUMULATIVE_ARGS</code> anything about the
arguments that have been passed on the stack.  The compiler has other
variables to keep track of that.  For target machines on which all
arguments are passed on the stack, there is no need to store anything in
<code>CUMULATIVE_ARGS</code>; however, the data structure must exist and
should not be empty, so use <code>int</code>.
</p></dd></dl>

<dl>
<dt><a name="index-OVERRIDE_005fABI_005fFORMAT"></a>Macro: <strong>OVERRIDE_ABI_FORMAT</strong> <em>(<var>fndecl</var>)</em></dt>
<dd><p>If defined, this macro is called before generating any code for a
function, but after the <var>cfun</var> descriptor for the function has been
created.  The back end may use this macro to update <var>cfun</var> to
reflect an ABI other than that which would normally be used by default.
If the compiler is generating code for a compiler-generated function,
<var>fndecl</var> may be <code>NULL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fCUMULATIVE_005fARGS"></a>Macro: <strong>INIT_CUMULATIVE_ARGS</strong> <em>(<var>cum</var>, <var>fntype</var>, <var>libname</var>, <var>fndecl</var>, <var>n_named_args</var>)</em></dt>
<dd><p>A C statement (sans semicolon) for initializing the variable
<var>cum</var> for the state at the beginning of the argument list.  The
variable has type <code>CUMULATIVE_ARGS</code>.  The value of <var>fntype</var>
is the tree node for the data type of the function which will receive
the args, or 0 if the args are to a compiler support library function.
For direct calls that are not libcalls, <var>fndecl</var> contain the
declaration node of the function.  <var>fndecl</var> is also set when
<code>INIT_CUMULATIVE_ARGS</code> is used to find arguments for the function
being compiled.  <var>n_named_args</var> is set to the number of named
arguments, including a structure return address if it is passed as a
parameter, when making a call.  When processing incoming arguments,
<var>n_named_args</var> is set to -1.
</p>
<p>When processing a call to a compiler support library function,
<var>libname</var> identifies which one.  It is a <code>symbol_ref</code> rtx which
contains the name of the function, as a string.  <var>libname</var> is 0 when
an ordinary C function call is being processed.  Thus, each time this
macro is called, either <var>libname</var> or <var>fntype</var> is nonzero, but
never both of them at once.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fCUMULATIVE_005fLIBCALL_005fARGS"></a>Macro: <strong>INIT_CUMULATIVE_LIBCALL_ARGS</strong> <em>(<var>cum</var>, <var>mode</var>, <var>libname</var>)</em></dt>
<dd><p>Like <code>INIT_CUMULATIVE_ARGS</code> but only used for outgoing libcalls,
it gets a <code>MODE</code> argument instead of <var>fntype</var>, that would be
<code>NULL</code>.  <var>indirect</var> would always be zero, too.  If this macro
is not defined, <code>INIT_CUMULATIVE_ARGS (cum, NULL_RTX, libname,
0)</code> is used instead.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fCUMULATIVE_005fINCOMING_005fARGS"></a>Macro: <strong>INIT_CUMULATIVE_INCOMING_ARGS</strong> <em>(<var>cum</var>, <var>fntype</var>, <var>libname</var>)</em></dt>
<dd><p>Like <code>INIT_CUMULATIVE_ARGS</code> but overrides it for the purposes of
finding the arguments for the function being compiled.  If this macro is
undefined, <code>INIT_CUMULATIVE_ARGS</code> is used instead.
</p>
<p>The value passed for <var>libname</var> is always 0, since library routines
with special calling conventions are never compiled with GCC.  The
argument <var>libname</var> exists for symmetry with
<code>INIT_CUMULATIVE_ARGS</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fARG_005fADVANCE"></a>Target Hook: <em>void</em> <strong>TARGET_FUNCTION_ARG_ADVANCE</strong> <em>(cumulative_args_t <var>ca</var>, enum machine_mode <var>mode</var>, const_tree <var>type</var>, bool <var>named</var>)</em></dt>
<dd><p>This hook updates the summarizer variable pointed to by <var>ca</var> to
advance past an argument in the argument list.  The values <var>mode</var>,
<var>type</var> and <var>named</var> describe that argument.  Once this is done,
the variable <var>cum</var> is suitable for analyzing the <em>following</em>
argument with <code>TARGET_FUNCTION_ARG</code>, etc.
</p>
<p>This hook need not do anything if the argument in question was passed
on the stack.  The compiler knows how to track the amount of stack space
used for arguments without any special help.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fARG_005fOFFSET"></a>Macro: <strong>FUNCTION_ARG_OFFSET</strong> <em>(<var>mode</var>, <var>type</var>)</em></dt>
<dd><p>If defined, a C expression that is the number of bytes to add to the
offset of the argument passed in memory.  This is needed for the SPU,
which passes <code>char</code> and <code>short</code> arguments in the preferred
slot that is in the middle of the quad word instead of starting at the
top.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fARG_005fPADDING"></a>Macro: <strong>FUNCTION_ARG_PADDING</strong> <em>(<var>mode</var>, <var>type</var>)</em></dt>
<dd><p>If defined, a C expression which determines whether, and in which direction,
to pad out an argument with extra space.  The value should be of type
<code>enum direction</code>: either <code>upward</code> to pad above the argument,
<code>downward</code> to pad below, or <code>none</code> to inhibit padding.
</p>
<p>The <em>amount</em> of padding is not controlled by this macro, but by the
target hook <code>TARGET_FUNCTION_ARG_ROUND_BOUNDARY</code>.  It is
always just enough to reach the next multiple of that boundary. 
</p>
<p>This macro has a default definition which is right for most systems.
For little-endian machines, the default is to pad upward.  For
big-endian machines, the default is to pad downward for an argument of
constant size shorter than an <code>int</code>, and upward otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-PAD_005fVARARGS_005fDOWN"></a>Macro: <strong>PAD_VARARGS_DOWN</strong></dt>
<dd><p>If defined, a C expression which determines whether the default
implementation of va_arg will attempt to pad down before reading the
next argument, if that argument is smaller than its aligned space as
controlled by <code>PARM_BOUNDARY</code>.  If this macro is not defined, all such
arguments are padded down if <code>BYTES_BIG_ENDIAN</code> is true.
</p></dd></dl>

<dl>
<dt><a name="index-BLOCK_005fREG_005fPADDING"></a>Macro: <strong>BLOCK_REG_PADDING</strong> <em>(<var>mode</var>, <var>type</var>, <var>first</var>)</em></dt>
<dd><p>Specify padding for the last element of a block move between registers and
memory.  <var>first</var> is nonzero if this is the only element.  Defining this
macro allows better control of register function parameters on big-endian
machines, without using <code>PARALLEL</code> rtl.  In particular,
<code>MUST_PASS_IN_STACK</code> need not test padding and mode of types in
registers, as there is no longer a &quot;wrong&quot; part of a register;  For example,
a three byte aggregate may be passed in the high part of a register if so
required.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fARG_005fBOUNDARY"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_FUNCTION_ARG_BOUNDARY</strong> <em>(enum machine_mode <var>mode</var>, const_tree <var>type</var>)</em></dt>
<dd><p>This hook returns the alignment boundary, in bits, of an argument
with the specified mode and type.  The default hook returns
<code>PARM_BOUNDARY</code> for all arguments.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fARG_005fROUND_005fBOUNDARY"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_FUNCTION_ARG_ROUND_BOUNDARY</strong> <em>(enum machine_mode <var>mode</var>, const_tree <var>type</var>)</em></dt>
<dd><p>Normally, the size of an argument is rounded up to <code>PARM_BOUNDARY</code>,
which is the default value for this hook.  You can define this hook to
return a different value if an argument size must be rounded to a larger
value.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fARG_005fREGNO_005fP"></a>Macro: <strong>FUNCTION_ARG_REGNO_P</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>A C expression that is nonzero if <var>regno</var> is the number of a hard
register in which function arguments are sometimes passed.  This does
<em>not</em> include implicit arguments such as the static chain and
the structure-value address.  On many machines, no registers can be
used for this purpose since all function arguments are pushed on the
stack.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSPLIT_005fCOMPLEX_005fARG"></a>Target Hook: <em>bool</em> <strong>TARGET_SPLIT_COMPLEX_ARG</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>This hook should return true if parameter of type <var>type</var> are passed
as two scalar parameters.  By default, GCC will attempt to pack complex
arguments into the target&rsquo;s word size.  Some ABIs require complex arguments
to be split and treated as their individual components.  For example, on
AIX64, complex floats should be passed in a pair of floating point
registers, even though a complex float would fit in one 64-bit floating
point register.
</p>
<p>The default value of this hook is <code>NULL</code>, which is treated as always
false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBUILD_005fBUILTIN_005fVA_005fLIST"></a>Target Hook: <em>tree</em> <strong>TARGET_BUILD_BUILTIN_VA_LIST</strong> <em>(void)</em></dt>
<dd><p>This hook returns a type node for <code>va_list</code> for the target.
The default version of the hook returns <code>void*</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fENUM_005fVA_005fLIST_005fP"></a>Target Hook: <em>int</em> <strong>TARGET_ENUM_VA_LIST_P</strong> <em>(int <var>idx</var>, const char **<var>pname</var>, tree *<var>ptree</var>)</em></dt>
<dd><p>This target hook is used in function <code>c_common_nodes_and_builtins</code>
to iterate through the target specific builtin types for va_list. The
variable <var>idx</var> is used as iterator. <var>pname</var> has to be a pointer
to a <code>const char *</code> and <var>ptree</var> a pointer to a <code>tree</code> typed
variable.
The arguments <var>pname</var> and <var>ptree</var> are used to store the result of
this macro and are set to the name of the va_list builtin type and its
internal type.
If the return value of this macro is zero, then there is no more element.
Otherwise the <var>IDX</var> should be increased for the next call of this
macro to iterate through all types.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFN_005fABI_005fVA_005fLIST"></a>Target Hook: <em>tree</em> <strong>TARGET_FN_ABI_VA_LIST</strong> <em>(tree <var>fndecl</var>)</em></dt>
<dd><p>This hook returns the va_list type of the calling convention specified by
<var>fndecl</var>.
The default version of this hook returns <code>va_list_type_node</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCANONICAL_005fVA_005fLIST_005fTYPE"></a>Target Hook: <em>tree</em> <strong>TARGET_CANONICAL_VA_LIST_TYPE</strong> <em>(tree <var>type</var>)</em></dt>
<dd><p>This hook returns the va_list type of the calling convention specified by the
type of <var>type</var>. If <var>type</var> is not a valid va_list type, it returns
<code>NULL_TREE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGIMPLIFY_005fVA_005fARG_005fEXPR"></a>Target Hook: <em>tree</em> <strong>TARGET_GIMPLIFY_VA_ARG_EXPR</strong> <em>(tree <var>valist</var>, tree <var>type</var>, gimple_seq *<var>pre_p</var>, gimple_seq *<var>post_p</var>)</em></dt>
<dd><p>This hook performs target-specific gimplification of
<code>VA_ARG_EXPR</code>.  The first two parameters correspond to the
arguments to <code>va_arg</code>; the latter two are as in
<code>gimplify.c:gimplify_expr</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVALID_005fPOINTER_005fMODE"></a>Target Hook: <em>bool</em> <strong>TARGET_VALID_POINTER_MODE</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Define this to return nonzero if the port can handle pointers
with machine mode <var>mode</var>.  The default version of this
hook returns true for both <code>ptr_mode</code> and <code>Pmode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fREF_005fMAY_005fALIAS_005fERRNO"></a>Target Hook: <em>bool</em> <strong>TARGET_REF_MAY_ALIAS_ERRNO</strong> <em>(struct ao_ref_s *<var>ref</var>)</em></dt>
<dd><p>Define this to return nonzero if the memory reference <var>ref</var>  may alias with the system C library errno location.  The default  version of this hook assumes the system C library errno location  is either a declaration of type int or accessed by dereferencing  a pointer to int.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCALAR_005fMODE_005fSUPPORTED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_SCALAR_MODE_SUPPORTED_P</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Define this to return nonzero if the port is prepared to handle
insns involving scalar mode <var>mode</var>.  For a scalar mode to be
considered supported, all the basic arithmetic and comparisons
must work.
</p>
<p>The default version of this hook returns true for any mode
required to handle the basic C types (as defined by the port).
Included here are the double-word arithmetic supported by the
code in <samp>optabs.c</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTOR_005fMODE_005fSUPPORTED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_VECTOR_MODE_SUPPORTED_P</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Define this to return nonzero if the port is prepared to handle
insns involving vector mode <var>mode</var>.  At the very least, it
must have move patterns for this mode.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fARRAY_005fMODE_005fSUPPORTED_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_ARRAY_MODE_SUPPORTED_P</strong> <em>(enum machine_mode <var>mode</var>, unsigned HOST_WIDE_INT <var>nelems</var>)</em></dt>
<dd><p>Return true if GCC should try to use a scalar mode to store an array
of <var>nelems</var> elements, given that each element has mode <var>mode</var>.
Returning true here overrides the usual <code>MAX_FIXED_MODE</code> limit
and allows GCC to use any defined integer mode.
</p>
<p>One use of this hook is to support vector load and store operations
that operate on several homogeneous vectors.  For example, ARM NEON
has operations like:
</p>
<div class="smallexample">
<pre class="smallexample">int8x8x3_t vld3_s8 (const int8_t *)
</pre></div>

<p>where the return type is defined as:
</p>
<div class="smallexample">
<pre class="smallexample">typedef struct int8x8x3_t
{
  int8x8_t val[3];
} int8x8x3_t;
</pre></div>

<p>If this hook allows <code>val</code> to have a scalar mode, then
<code>int8x8x3_t</code> can have the same mode.  GCC can then store
<code>int8x8x3_t</code>s in registers rather than forcing them onto the stack.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSMALL_005fREGISTER_005fCLASSES_005fFOR_005fMODE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_SMALL_REGISTER_CLASSES_FOR_MODE_P</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Define this to return nonzero for machine modes for which the port has
small register classes.  If this target hook returns nonzero for a given
<var>mode</var>, the compiler will try to minimize the lifetime of registers
in <var>mode</var>.  The hook may be called with <code>VOIDmode</code> as argument.
In this case, the hook is expected to return nonzero if it returns nonzero
for any mode.
</p>
<p>On some machines, it is risky to let hard registers live across arbitrary
insns.  Typically, these machines have instructions that require values
to be in specific registers (like an accumulator), and reload will fail
if the required hard register is used for another purpose across such an
insn.
</p>
<p>Passes before reload do not know which hard registers will be used
in an instruction, but the machine modes of the registers set or used in
the instruction are already known.  And for some machines, register
classes are small for, say, integer registers but not for floating point
registers.  For example, the AMD x86-64 architecture requires specific
registers for the legacy x86 integer instructions, but there are many
SSE registers for floating point operations.  On such targets, a good
strategy may be to return nonzero from this hook for <code>INTEGRAL_MODE_P</code>
machine modes but zero for the SSE register classes.
</p>
<p>The default version of this hook returns false for any mode.  It is always
safe to redefine this hook to return with a nonzero value.  But if you
unnecessarily define it, you will reduce the amount of optimizations
that can be performed in some cases.  If you do not define this hook
to return a nonzero value when it is required, the compiler will run out
of spill registers and print a fatal error message.
</p></dd></dl>

<hr>
<a name="Scalar-Return"></a>
<div class="header">
<p>
Next: <a href="#Aggregate-Return" accesskey="n" rel="next">Aggregate Return</a>, Previous: <a href="#Register-Arguments" accesskey="p" rel="prev">Register Arguments</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="How-Scalar-Function-Values-Are-Returned"></a>
<h4 class="subsection">17.10.8 How Scalar Function Values Are Returned</h4>
<a name="index-return-values-in-registers"></a>
<a name="index-values_002c-returned-by-functions"></a>
<a name="index-scalars_002c-returned-as-values"></a>

<p>This section discusses the macros that control returning scalars as
values&mdash;values that can fit in registers.
</p>
<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fVALUE"></a>Target Hook: <em>rtx</em> <strong>TARGET_FUNCTION_VALUE</strong> <em>(const_tree <var>ret_type</var>, const_tree <var>fn_decl_or_type</var>, bool <var>outgoing</var>)</em></dt>
<dd>
<p>Define this to return an RTX representing the place where a function
returns or receives a value of data type <var>ret_type</var>, a tree node
representing a data type.  <var>fn_decl_or_type</var> is a tree node
representing <code>FUNCTION_DECL</code> or <code>FUNCTION_TYPE</code> of a
function being called.  If <var>outgoing</var> is false, the hook should
compute the register in which the caller will see the return value.
Otherwise, the hook should return an RTX representing the place where
a function returns a value.
</p>
<p>On many machines, only <code>TYPE_MODE (<var>ret_type</var>)</code> is relevant.
(Actually, on most machines, scalar values are returned in the same
place regardless of mode.)  The value of the expression is usually a
<code>reg</code> RTX for the hard register where the return value is stored.
The value can also be a <code>parallel</code> RTX, if the return value is in
multiple places.  See <code>TARGET_FUNCTION_ARG</code> for an explanation of the
<code>parallel</code> form.   Note that the callee will populate every
location specified in the <code>parallel</code>, but if the first element of
the <code>parallel</code> contains the whole return value, callers will use
that element as the canonical location and ignore the others.  The m68k
port uses this type of <code>parallel</code> to return pointers in both
&lsquo;<samp>%a0</samp>&rsquo; (the canonical location) and &lsquo;<samp>%d0</samp>&rsquo;.
</p>
<p>If <code>TARGET_PROMOTE_FUNCTION_RETURN</code> returns true, you must apply
the same promotion rules specified in <code>PROMOTE_MODE</code> if
<var>valtype</var> is a scalar type.
</p>
<p>If the precise function being called is known, <var>func</var> is a tree
node (<code>FUNCTION_DECL</code>) for it; otherwise, <var>func</var> is a null
pointer.  This makes it possible to use a different value-returning
convention for specific functions when all their calls are
known.
</p>
<p>Some target machines have &ldquo;register windows&rdquo; so that the register in
which a function returns its value is not the same as the one in which
the caller sees the value.  For such machines, you should return
different RTX depending on <var>outgoing</var>.
</p>
<p><code>TARGET_FUNCTION_VALUE</code> is not used for return values with
aggregate data types, because these are returned in another way.  See
<code>TARGET_STRUCT_VALUE_RTX</code> and related macros, below.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fVALUE"></a>Macro: <strong>FUNCTION_VALUE</strong> <em>(<var>valtype</var>, <var>func</var>)</em></dt>
<dd><p>This macro has been deprecated.  Use <code>TARGET_FUNCTION_VALUE</code> for
a new target instead.
</p></dd></dl>

<dl>
<dt><a name="index-LIBCALL_005fVALUE"></a>Macro: <strong>LIBCALL_VALUE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression to create an RTX representing the place where a library
function returns a value of mode <var>mode</var>.
</p>
<p>Note that &ldquo;library function&rdquo; in this context means a compiler
support routine, used to perform arithmetic, whose name is known
specially by the compiler and was not mentioned in the C code being
compiled.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIBCALL_005fVALUE"></a>Target Hook: <em>rtx</em> <strong>TARGET_LIBCALL_VALUE</strong> <em>(enum machine_mode <var>mode</var>, const_rtx <var>fun</var>)</em></dt>
<dd><p>Define this hook if the back-end needs to know the name of the libcall
function in order to determine where the result should be returned.
</p>
<p>The mode of the result is given by <var>mode</var> and the name of the called
library function is given by <var>fun</var>.  The hook should return an RTX
representing the place where the library function result will be returned.
</p>
<p>If this hook is not defined, then LIBCALL_VALUE will be used.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fVALUE_005fREGNO_005fP"></a>Macro: <strong>FUNCTION_VALUE_REGNO_P</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>A C expression that is nonzero if <var>regno</var> is the number of a hard
register in which the values of called function may come back.
</p>
<p>A register whose use for returning values is limited to serving as the
second of a pair (for a value of type <code>double</code>, say) need not be
recognized by this macro.  So for most machines, this definition
suffices:
</p>
<div class="smallexample">
<pre class="smallexample">#define FUNCTION_VALUE_REGNO_P(N) ((N) == 0)
</pre></div>

<p>If the machine has register windows, so that the caller and the called
function use different registers for the return value, this macro
should recognize only the caller&rsquo;s register numbers.
</p>
<p>This macro has been deprecated.  Use <code>TARGET_FUNCTION_VALUE_REGNO_P</code>
for a new target instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fVALUE_005fREGNO_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_FUNCTION_VALUE_REGNO_P</strong> <em>(const unsigned int <var>regno</var>)</em></dt>
<dd><p>A target hook that return <code>true</code> if <var>regno</var> is the number of a hard
register in which the values of called function may come back.
</p>
<p>A register whose use for returning values is limited to serving as the
second of a pair (for a value of type <code>double</code>, say) need not be
recognized by this target hook.
</p>
<p>If the machine has register windows, so that the caller and the called
function use different registers for the return value, this target hook
should recognize only the caller&rsquo;s register numbers.
</p>
<p>If this hook is not defined, then FUNCTION_VALUE_REGNO_P will be used.
</p></dd></dl>

<dl>
<dt><a name="index-APPLY_005fRESULT_005fSIZE"></a>Macro: <strong>APPLY_RESULT_SIZE</strong></dt>
<dd><p>Define this macro if &lsquo;<samp>untyped_call</samp>&rsquo; and &lsquo;<samp>untyped_return</samp>&rsquo;
need more space than is implied by <code>FUNCTION_VALUE_REGNO_P</code> for
saving and restoring an arbitrary return value.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fRETURN_005fIN_005fMSB"></a>Target Hook: <em>bool</em> <strong>TARGET_RETURN_IN_MSB</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>This hook should return true if values of type <var>type</var> are returned
at the most significant end of a register (in other words, if they are
padded at the least significant end).  You can assume that <var>type</var>
is returned in a register; the caller is required to check this.
</p>
<p>Note that the register provided by <code>TARGET_FUNCTION_VALUE</code> must
be able to hold the complete return value.  For example, if a 1-, 2-
or 3-byte structure is returned at the most significant end of a
4-byte register, <code>TARGET_FUNCTION_VALUE</code> should provide an
<code>SImode</code> rtx.
</p></dd></dl>

<hr>
<a name="Aggregate-Return"></a>
<div class="header">
<p>
Next: <a href="#Caller-Saves" accesskey="n" rel="next">Caller Saves</a>, Previous: <a href="#Scalar-Return" accesskey="p" rel="prev">Scalar Return</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="How-Large-Values-Are-Returned"></a>
<h4 class="subsection">17.10.9 How Large Values Are Returned</h4>
<a name="index-aggregates-as-return-values"></a>
<a name="index-large-return-values"></a>
<a name="index-returning-aggregate-values"></a>
<a name="index-structure-value-address"></a>

<p>When a function value&rsquo;s mode is <code>BLKmode</code> (and in some other
cases), the value is not returned according to
<code>TARGET_FUNCTION_VALUE</code> (see <a href="#Scalar-Return">Scalar Return</a>).  Instead, the
caller passes the address of a block of memory in which the value
should be stored.  This address is called the <em>structure value
address</em>.
</p>
<p>This section describes how to control returning structure values in
memory.
</p>
<dl>
<dt><a name="index-TARGET_005fRETURN_005fIN_005fMEMORY"></a>Target Hook: <em>bool</em> <strong>TARGET_RETURN_IN_MEMORY</strong> <em>(const_tree <var>type</var>, const_tree <var>fntype</var>)</em></dt>
<dd><p>This target hook should return a nonzero value to say to return the
function value in memory, just as large structures are always returned.
Here <var>type</var> will be the data type of the value, and <var>fntype</var>
will be the type of the function doing the returning, or <code>NULL</code> for
libcalls.
</p>
<p>Note that values of mode <code>BLKmode</code> must be explicitly handled
by this function.  Also, the option <samp>-fpcc-struct-return</samp>
takes effect regardless of this macro.  On most systems, it is
possible to leave the hook undefined; this causes a default
definition to be used, whose value is the constant 1 for <code>BLKmode</code>
values, and 0 otherwise.
</p>
<p>Do not use this hook to indicate that structures and unions should always
be returned in memory.  You should instead use <code>DEFAULT_PCC_STRUCT_RETURN</code>
to indicate this.
</p></dd></dl>

<dl>
<dt><a name="index-DEFAULT_005fPCC_005fSTRUCT_005fRETURN"></a>Macro: <strong>DEFAULT_PCC_STRUCT_RETURN</strong></dt>
<dd><p>Define this macro to be 1 if all structure and union return values must be
in memory.  Since this results in slower code, this should be defined
only if needed for compatibility with other compilers or with an ABI.
If you define this macro to be 0, then the conventions used for structure
and union return values are decided by the <code>TARGET_RETURN_IN_MEMORY</code>
target hook.
</p>
<p>If not defined, this defaults to the value 1.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTRUCT_005fVALUE_005fRTX"></a>Target Hook: <em>rtx</em> <strong>TARGET_STRUCT_VALUE_RTX</strong> <em>(tree <var>fndecl</var>, int <var>incoming</var>)</em></dt>
<dd><p>This target hook should return the location of the structure value
address (normally a <code>mem</code> or <code>reg</code>), or 0 if the address is
passed as an &ldquo;invisible&rdquo; first argument.  Note that <var>fndecl</var> may
be <code>NULL</code>, for libcalls.  You do not need to define this target
hook if the address is always passed as an &ldquo;invisible&rdquo; first
argument.
</p>
<p>On some architectures the place where the structure value address
is found by the called function is not the same place that the
caller put it.  This can be due to register windows, or it could
be because the function prologue moves it to a different place.
<var>incoming</var> is <code>1</code> or <code>2</code> when the location is needed in
the context of the called function, and <code>0</code> in the context of
the caller.
</p>
<p>If <var>incoming</var> is nonzero and the address is to be found on the
stack, return a <code>mem</code> which refers to the frame pointer. If
<var>incoming</var> is <code>2</code>, the result is being used to fetch the
structure value address at the beginning of a function.  If you need
to emit adjusting code, you should do it at this point.
</p></dd></dl>

<dl>
<dt><a name="index-PCC_005fSTATIC_005fSTRUCT_005fRETURN"></a>Macro: <strong>PCC_STATIC_STRUCT_RETURN</strong></dt>
<dd><p>Define this macro if the usual system convention on the target machine
for returning structures and unions is for the called function to return
the address of a static variable containing the value.
</p>
<p>Do not define this if the usual system convention is for the caller to
pass an address to the subroutine.
</p>
<p>This macro has effect in <samp>-fpcc-struct-return</samp> mode, but it does
nothing when you use <samp>-freg-struct-return</samp> mode.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGET_005fRAW_005fRESULT_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_GET_RAW_RESULT_MODE</strong> <em>(int <var>regno</var>)</em></dt>
<dd><p>This target hook returns the mode to be used when accessing raw return registers in <code>__builtin_return</code>.  Define this macro if the value in <var>reg_raw_mode</var> is not correct.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGET_005fRAW_005fARG_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_GET_RAW_ARG_MODE</strong> <em>(int <var>regno</var>)</em></dt>
<dd><p>This target hook returns the mode to be used when accessing raw argument registers in <code>__builtin_apply_args</code>.  Define this macro if the value in <var>reg_raw_mode</var> is not correct.
</p></dd></dl>

<hr>
<a name="Caller-Saves"></a>
<div class="header">
<p>
Next: <a href="#Function-Entry" accesskey="n" rel="next">Function Entry</a>, Previous: <a href="#Aggregate-Return" accesskey="p" rel="prev">Aggregate Return</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Caller_002dSaves-Register-Allocation"></a>
<h4 class="subsection">17.10.10 Caller-Saves Register Allocation</h4>

<p>If you enable it, GCC can save registers around function calls.  This
makes it possible to use call-clobbered registers to hold variables that
must live across calls.
</p>
<dl>
<dt><a name="index-CALLER_005fSAVE_005fPROFITABLE"></a>Macro: <strong>CALLER_SAVE_PROFITABLE</strong> <em>(<var>refs</var>, <var>calls</var>)</em></dt>
<dd><p>A C expression to determine whether it is worthwhile to consider placing
a pseudo-register in a call-clobbered hard register and saving and
restoring it around each function call.  The expression should be 1 when
this is worth doing, and 0 otherwise.
</p>
<p>If you don&rsquo;t define this macro, a default is used which is good on most
machines: <code>4 * <var>calls</var> &lt; <var>refs</var></code>.
</p></dd></dl>

<dl>
<dt><a name="index-HARD_005fREGNO_005fCALLER_005fSAVE_005fMODE"></a>Macro: <strong>HARD_REGNO_CALLER_SAVE_MODE</strong> <em>(<var>regno</var>, <var>nregs</var>)</em></dt>
<dd><p>A C expression specifying which mode is required for saving <var>nregs</var>
of a pseudo-register in call-clobbered hard register <var>regno</var>.  If
<var>regno</var> is unsuitable for caller save, <code>VOIDmode</code> should be
returned.  For most machines this macro need not be defined since GCC
will select the smallest suitable mode.
</p></dd></dl>

<hr>
<a name="Function-Entry"></a>
<div class="header">
<p>
Next: <a href="#Profiling" accesskey="n" rel="next">Profiling</a>, Previous: <a href="#Caller-Saves" accesskey="p" rel="prev">Caller Saves</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Function-Entry-and-Exit"></a>
<h4 class="subsection">17.10.11 Function Entry and Exit</h4>
<a name="index-function-entry-and-exit"></a>
<a name="index-prologue"></a>
<a name="index-epilogue"></a>

<p>This section describes the macros that output function entry
(<em>prologue</em>) and exit (<em>epilogue</em>) code.
</p>
<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fPROLOGUE"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FUNCTION_PROLOGUE</strong> <em>(FILE *<var>file</var>, HOST_WIDE_INT <var>size</var>)</em></dt>
<dd><p>If defined, a function that outputs the assembler code for entry to a
function.  The prologue is responsible for setting up the stack frame,
initializing the frame pointer register, saving registers that must be
saved, and allocating <var>size</var> additional bytes of storage for the
local variables.  <var>size</var> is an integer.  <var>file</var> is a stdio
stream to which the assembler code should be output.
</p>
<p>The label for the beginning of the function need not be output by this
macro.  That has already been done when the macro is run.
</p>
<a name="index-regs_005fever_005flive"></a>
<p>To determine which registers to save, the macro can refer to the array
<code>regs_ever_live</code>: element <var>r</var> is nonzero if hard register
<var>r</var> is used anywhere within the function.  This implies the function
prologue should save register <var>r</var>, provided it is not one of the
call-used registers.  (<code>TARGET_ASM_FUNCTION_EPILOGUE</code> must likewise use
<code>regs_ever_live</code>.)
</p>
<p>On machines that have &ldquo;register windows&rdquo;, the function entry code does
not save on the stack the registers that are in the windows, even if
they are supposed to be preserved by function calls; instead it takes
appropriate steps to &ldquo;push&rdquo; the register stack, if any non-call-used
registers are used in the function.
</p>
<a name="index-frame_005fpointer_005fneeded"></a>
<p>On machines where functions may or may not have frame-pointers, the
function entry code must vary accordingly; it must set up the frame
pointer if one is wanted, and not otherwise.  To determine whether a
frame pointer is in wanted, the macro can refer to the variable
<code>frame_pointer_needed</code>.  The variable&rsquo;s value will be 1 at run
time in a function that needs a frame pointer.  See <a href="#Elimination">Elimination</a>.
</p>
<p>The function entry code is responsible for allocating any stack space
required for the function.  This stack space consists of the regions
listed below.  In most cases, these regions are allocated in the
order listed, with the last listed region closest to the top of the
stack (the lowest address if <code>STACK_GROWS_DOWNWARD</code> is defined, and
the highest address if it is not defined).  You can use a different order
for a machine if doing so is more convenient or required for
compatibility reasons.  Except in cases where required by standard
or by a debugger, there is no reason why the stack layout used by GCC
need agree with that used by other compilers for a machine.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fEND_005fPROLOGUE"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FUNCTION_END_PROLOGUE</strong> <em>(FILE *<var>file</var>)</em></dt>
<dd><p>If defined, a function that outputs assembler code at the end of a
prologue.  This should be used when the function prologue is being
emitted as RTL, and you have some extra assembler that needs to be
emitted.  See <a href="#prologue-instruction-pattern">prologue instruction pattern</a>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fBEGIN_005fEPILOGUE"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FUNCTION_BEGIN_EPILOGUE</strong> <em>(FILE *<var>file</var>)</em></dt>
<dd><p>If defined, a function that outputs assembler code at the start of an
epilogue.  This should be used when the function epilogue is being
emitted as RTL, and you have some extra assembler that needs to be
emitted.  See <a href="#epilogue-instruction-pattern">epilogue instruction pattern</a>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fEPILOGUE"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FUNCTION_EPILOGUE</strong> <em>(FILE *<var>file</var>, HOST_WIDE_INT <var>size</var>)</em></dt>
<dd><p>If defined, a function that outputs the assembler code for exit from a
function.  The epilogue is responsible for restoring the saved
registers and stack pointer to their values when the function was
called, and returning control to the caller.  This macro takes the
same arguments as the macro <code>TARGET_ASM_FUNCTION_PROLOGUE</code>, and the
registers to restore are determined from <code>regs_ever_live</code> and
<code>CALL_USED_REGISTERS</code> in the same way.
</p>
<p>On some machines, there is a single instruction that does all the work
of returning from the function.  On these machines, give that
instruction the name &lsquo;<samp>return</samp>&rsquo; and do not define the macro
<code>TARGET_ASM_FUNCTION_EPILOGUE</code> at all.
</p>
<p>Do not define a pattern named &lsquo;<samp>return</samp>&rsquo; if you want the
<code>TARGET_ASM_FUNCTION_EPILOGUE</code> to be used.  If you want the target
switches to control whether return instructions or epilogues are used,
define a &lsquo;<samp>return</samp>&rsquo; pattern with a validity condition that tests the
target switches appropriately.  If the &lsquo;<samp>return</samp>&rsquo; pattern&rsquo;s validity
condition is false, epilogues will be used.
</p>
<p>On machines where functions may or may not have frame-pointers, the
function exit code must vary accordingly.  Sometimes the code for these
two cases is completely different.  To determine whether a frame pointer
is wanted, the macro can refer to the variable
<code>frame_pointer_needed</code>.  The variable&rsquo;s value will be 1 when compiling
a function that needs a frame pointer.
</p>
<p>Normally, <code>TARGET_ASM_FUNCTION_PROLOGUE</code> and
<code>TARGET_ASM_FUNCTION_EPILOGUE</code> must treat leaf functions specially.
The C variable <code>current_function_is_leaf</code> is nonzero for such a
function.  See <a href="#Leaf-Functions">Leaf Functions</a>.
</p>
<p>On some machines, some functions pop their arguments on exit while
others leave that for the caller to do.  For example, the 68020 when
given <samp>-mrtd</samp> pops arguments in functions that take a fixed
number of arguments.
</p>
<a name="index-pops_005fargs"></a>
<a name="index-crtl_002d_003eargs_002epops_005fargs"></a>
<p>Your definition of the macro <code>RETURN_POPS_ARGS</code> decides which
functions pop their own arguments.  <code>TARGET_ASM_FUNCTION_EPILOGUE</code>
needs to know what was decided.  The number of bytes of the current
function&rsquo;s arguments that this function should pop is available in
<code>crtl-&gt;args.pops_args</code>.  See <a href="#Scalar-Return">Scalar Return</a>.
</p></dd></dl>

<ul>
<li> <a name="index-pretend_005fargs_005fsize"></a>
<a name="index-crtl_002d_003eargs_002epretend_005fargs_005fsize"></a>
A region of <code>crtl-&gt;args.pretend_args_size</code> bytes of
uninitialized space just underneath the first argument arriving on the
stack.  (This may not be at the very start of the allocated stack region
if the calling sequence has pushed anything else since pushing the stack
arguments.  But usually, on such machines, nothing else has been pushed
yet, because the function prologue itself does all the pushing.)  This
region is used on machines where an argument may be passed partly in
registers and partly in memory, and, in some cases to support the
features in <code>&lt;stdarg.h&gt;</code>.

</li><li> An area of memory used to save certain registers used by the function.
The size of this area, which may also include space for such things as
the return address and pointers to previous stack frames, is
machine-specific and usually depends on which registers have been used
in the function.  Machines with register windows often do not require
a save area.

</li><li> A region of at least <var>size</var> bytes, possibly rounded up to an allocation
boundary, to contain the local variables of the function.  On some machines,
this region and the save area may occur in the opposite order, with the
save area closer to the top of the stack.

</li><li> <a name="index-ACCUMULATE_005fOUTGOING_005fARGS-and-stack-frames"></a>
Optionally, when <code>ACCUMULATE_OUTGOING_ARGS</code> is defined, a region of
<code>crtl-&gt;outgoing_args_size</code> bytes to be used for outgoing
argument lists of the function.  See <a href="#Stack-Arguments">Stack Arguments</a>.
</li></ul>

<dl>
<dt><a name="index-EXIT_005fIGNORE_005fSTACK"></a>Macro: <strong>EXIT_IGNORE_STACK</strong></dt>
<dd><p>Define this macro as a C expression that is nonzero if the return
instruction or the function epilogue ignores the value of the stack
pointer; in other words, if it is safe to delete an instruction to
adjust the stack pointer before a return from the function.  The
default is 0.
</p>
<p>Note that this macro&rsquo;s value is relevant only for functions for which
frame pointers are maintained.  It is never safe to delete a final
stack adjustment in a function that has no frame pointer, and the
compiler knows this regardless of <code>EXIT_IGNORE_STACK</code>.
</p></dd></dl>

<dl>
<dt><a name="index-EPILOGUE_005fUSES"></a>Macro: <strong>EPILOGUE_USES</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>Define this macro as a C expression that is nonzero for registers that are
used by the epilogue or the &lsquo;<samp>return</samp>&rsquo; pattern.  The stack and frame
pointer registers are already assumed to be used as needed.
</p></dd></dl>

<dl>
<dt><a name="index-EH_005fUSES"></a>Macro: <strong>EH_USES</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>Define this macro as a C expression that is nonzero for registers that are
used by the exception handling mechanism, and so should be considered live
on entry to an exception edge.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fMI_005fTHUNK"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_OUTPUT_MI_THUNK</strong> <em>(FILE *<var>file</var>, tree <var>thunk_fndecl</var>, HOST_WIDE_INT <var>delta</var>, HOST_WIDE_INT <var>vcall_offset</var>, tree <var>function</var>)</em></dt>
<dd><p>A function that outputs the assembler code for a thunk
function, used to implement C++ virtual function calls with multiple
inheritance.  The thunk acts as a wrapper around a virtual function,
adjusting the implicit object parameter before handing control off to
the real function.
</p>
<p>First, emit code to add the integer <var>delta</var> to the location that
contains the incoming first argument.  Assume that this argument
contains a pointer, and is the one used to pass the <code>this</code> pointer
in C++.  This is the incoming argument <em>before</em> the function prologue,
e.g. &lsquo;<samp>%o0</samp>&rsquo; on a sparc.  The addition must preserve the values of
all other incoming arguments.
</p>
<p>Then, if <var>vcall_offset</var> is nonzero, an additional adjustment should be
made after adding <code>delta</code>.  In particular, if <var>p</var> is the
adjusted pointer, the following adjustment should be made:
</p>
<div class="smallexample">
<pre class="smallexample">p += (*((ptrdiff_t **)p))[vcall_offset/sizeof(ptrdiff_t)]
</pre></div>

<p>After the additions, emit code to jump to <var>function</var>, which is a
<code>FUNCTION_DECL</code>.  This is a direct pure jump, not a call, and does
not touch the return address.  Hence returning from <var>FUNCTION</var> will
return to whoever called the current &lsquo;<samp>thunk</samp>&rsquo;.
</p>
<p>The effect must be as if <var>function</var> had been called directly with
the adjusted first argument.  This macro is responsible for emitting all
of the code for a thunk function; <code>TARGET_ASM_FUNCTION_PROLOGUE</code>
and <code>TARGET_ASM_FUNCTION_EPILOGUE</code> are not invoked.
</p>
<p>The <var>thunk_fndecl</var> is redundant.  (<var>delta</var> and <var>function</var>
have already been extracted from it.)  It might possibly be useful on
some targets, but probably not.
</p>
<p>If you do not define this macro, the target-independent code in the C++
front end will generate a less efficient heavyweight thunk that calls
<var>function</var> instead of jumping to it.  The generic approach does
not support varargs.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fCAN_005fOUTPUT_005fMI_005fTHUNK"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_CAN_OUTPUT_MI_THUNK</strong> <em>(const_tree <var>thunk_fndecl</var>, HOST_WIDE_INT <var>delta</var>, HOST_WIDE_INT <var>vcall_offset</var>, const_tree <var>function</var>)</em></dt>
<dd><p>A function that returns true if TARGET_ASM_OUTPUT_MI_THUNK would be able
to output the assembler code for the thunk function specified by the
arguments it is passed, and false otherwise.  In the latter case, the
generic approach will be used by the C++ front end, with the limitations
previously exposed.
</p></dd></dl>

<hr>
<a name="Profiling"></a>
<div class="header">
<p>
Next: <a href="#Tail-Calls" accesskey="n" rel="next">Tail Calls</a>, Previous: <a href="#Function-Entry" accesskey="p" rel="prev">Function Entry</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Generating-Code-for-Profiling"></a>
<h4 class="subsection">17.10.12 Generating Code for Profiling</h4>
<a name="index-profiling_002c-code-generation"></a>

<p>These macros will help you generate code for profiling.
</p>
<dl>
<dt><a name="index-FUNCTION_005fPROFILER"></a>Macro: <strong>FUNCTION_PROFILER</strong> <em>(<var>file</var>, <var>labelno</var>)</em></dt>
<dd><p>A C statement or compound statement to output to <var>file</var> some
assembler code to call the profiling subroutine <code>mcount</code>.
</p>
<a name="index-mcount"></a>
<p>The details of how <code>mcount</code> expects to be called are determined by
your operating system environment, not by GCC.  To figure them out,
compile a small program for profiling using the system&rsquo;s installed C
compiler and look at the assembler code that results.
</p>
<p>Older implementations of <code>mcount</code> expect the address of a counter
variable to be loaded into some register.  The name of this variable is
&lsquo;<samp>LP</samp>&rsquo; followed by the number <var>labelno</var>, so you would generate
the name using &lsquo;<samp>LP%d</samp>&rsquo; in a <code>fprintf</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PROFILE_005fHOOK"></a>Macro: <strong>PROFILE_HOOK</strong></dt>
<dd><p>A C statement or compound statement to output to <var>file</var> some assembly
code to call the profiling subroutine <code>mcount</code> even the target does
not support profiling.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fPROFILE_005fCOUNTERS"></a>Macro: <strong>NO_PROFILE_COUNTERS</strong></dt>
<dd><p>Define this macro to be an expression with a nonzero value if the
<code>mcount</code> subroutine on your system does not need a counter variable
allocated for each function.  This is true for almost all modern
implementations.  If you define this macro, you must not use the
<var>labelno</var> argument to <code>FUNCTION_PROFILER</code>.
</p></dd></dl>

<dl>
<dt><a name="index-PROFILE_005fBEFORE_005fPROLOGUE"></a>Macro: <strong>PROFILE_BEFORE_PROLOGUE</strong></dt>
<dd><p>Define this macro if the code for function profiling should come before
the function prologue.  Normally, the profiling code comes after.
</p></dd></dl>

<hr>
<a name="Tail-Calls"></a>
<div class="header">
<p>
Next: <a href="#Stack-Smashing-Protection" accesskey="n" rel="next">Stack Smashing Protection</a>, Previous: <a href="#Profiling" accesskey="p" rel="prev">Profiling</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Permitting-tail-calls"></a>
<h4 class="subsection">17.10.13 Permitting tail calls</h4>
<a name="index-tail-calls"></a>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fOK_005fFOR_005fSIBCALL"></a>Target Hook: <em>bool</em> <strong>TARGET_FUNCTION_OK_FOR_SIBCALL</strong> <em>(tree <var>decl</var>, tree <var>exp</var>)</em></dt>
<dd><p>True if it is ok to do sibling call optimization for the specified
call expression <var>exp</var>.  <var>decl</var> will be the called function,
or <code>NULL</code> if this is an indirect call.
</p>
<p>It is not uncommon for limitations of calling conventions to prevent
tail calls to functions outside the current unit of translation, or
during PIC compilation.  The hook is used to enforce these restrictions,
as the <code>sibcall</code> md pattern can not fail, or fall over to a
&ldquo;normal&rdquo; call.  The criteria for successful sibling call optimization
may vary greatly between different architectures.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXTRA_005fLIVE_005fON_005fENTRY"></a>Target Hook: <em>void</em> <strong>TARGET_EXTRA_LIVE_ON_ENTRY</strong> <em>(bitmap <var>regs</var>)</em></dt>
<dd><p>Add any hard registers to <var>regs</var> that are live on entry to the
function.  This hook only needs to be defined to provide registers that
cannot be found by examination of FUNCTION_ARG_REGNO_P, the callee saved
registers, STATIC_CHAIN_INCOMING_REGNUM, STATIC_CHAIN_REGNUM,
TARGET_STRUCT_VALUE_RTX, FRAME_POINTER_REGNUM, EH_USES,
FRAME_POINTER_REGNUM, ARG_POINTER_REGNUM, and the PIC_OFFSET_TABLE_REGNUM.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSET_005fUP_005fBY_005fPROLOGUE"></a>Target Hook: <em>void</em> <strong>TARGET_SET_UP_BY_PROLOGUE</strong> <em>(struct hard_reg_set_container *<var></var>)</em></dt>
<dd><p>This hook should add additional registers that are computed by the prologue to the hard regset for shrink-wrapping optimization purposes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fWARN_005fFUNC_005fRETURN"></a>Target Hook: <em>bool</em> <strong>TARGET_WARN_FUNC_RETURN</strong> <em>(tree)</em></dt>
<dd><p>True if a function&rsquo;s return statements should be checked for matching the function&rsquo;s return type.  This includes checking for falling off the end of a non-void function.  Return false if no such check should be made.
</p></dd></dl>

<hr>
<a name="Stack-Smashing-Protection"></a>
<div class="header">
<p>
Previous: <a href="#Tail-Calls" accesskey="p" rel="prev">Tail Calls</a>, Up: <a href="#Stack-and-Calling" accesskey="u" rel="up">Stack and Calling</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Stack-smashing-protection"></a>
<h4 class="subsection">17.10.14 Stack smashing protection</h4>
<a name="index-stack-smashing-protection"></a>

<dl>
<dt><a name="index-TARGET_005fSTACK_005fPROTECT_005fGUARD"></a>Target Hook: <em>tree</em> <strong>TARGET_STACK_PROTECT_GUARD</strong> <em>(void)</em></dt>
<dd><p>This hook returns a <code>DECL</code> node for the external variable to use
for the stack protection guard.  This variable is initialized by the
runtime to some random value and is used to initialize the guard value
that is placed at the top of the local stack frame.  The type of this
variable must be <code>ptr_type_node</code>.
</p>
<p>The default version of this hook creates a variable called
&lsquo;<samp>__stack_chk_guard</samp>&rsquo;, which is normally defined in <samp>libgcc2.c</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTACK_005fPROTECT_005fFAIL"></a>Target Hook: <em>tree</em> <strong>TARGET_STACK_PROTECT_FAIL</strong> <em>(void)</em></dt>
<dd><p>This hook returns a <code>CALL_EXPR</code> that alerts the runtime that the
stack protect guard variable has been modified.  This expression should
involve a call to a <code>noreturn</code> function.
</p>
<p>The default version of this hook invokes a function called
&lsquo;<samp>__stack_chk_fail</samp>&rsquo;, taking no arguments.  This function is
normally defined in <samp>libgcc2.c</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSUPPORTS_005fSPLIT_005fSTACK"></a>Common Target Hook: <em>bool</em> <strong>TARGET_SUPPORTS_SPLIT_STACK</strong> <em>(bool <var>report</var>, struct gcc_options *<var>opts</var>)</em></dt>
<dd><p>Whether this target supports splitting the stack when the options described in <var>opts</var> have been passed.  This is called after options have been parsed, so the target may reject splitting the stack in some configurations.  The default version of this hook returns false.  If <var>report</var> is true, this function may issue a warning or error; if <var>report</var> is false, it must simply return a value
</p></dd></dl>

<hr>
<a name="Varargs"></a>
<div class="header">
<p>
Next: <a href="#Trampolines" accesskey="n" rel="next">Trampolines</a>, Previous: <a href="#Stack-and-Calling" accesskey="p" rel="prev">Stack and Calling</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Implementing-the-Varargs-Macros"></a>
<h3 class="section">17.11 Implementing the Varargs Macros</h3>
<a name="index-varargs-implementation"></a>

<p>GCC comes with an implementation of <code>&lt;varargs.h&gt;</code> and
<code>&lt;stdarg.h&gt;</code> that work without change on machines that pass arguments
on the stack.  Other machines require their own implementations of
varargs, and the two machine independent header files must have
conditionals to include it.
</p>
<p>ISO <code>&lt;stdarg.h&gt;</code> differs from traditional <code>&lt;varargs.h&gt;</code> mainly in
the calling convention for <code>va_start</code>.  The traditional
implementation takes just one argument, which is the variable in which
to store the argument pointer.  The ISO implementation of
<code>va_start</code> takes an additional second argument.  The user is
supposed to write the last named argument of the function here.
</p>
<p>However, <code>va_start</code> should not use this argument.  The way to find
the end of the named arguments is with the built-in functions described
below.
</p>
<dl>
<dt><a name="index-_005f_005fbuiltin_005fsaveregs"></a>Macro: <strong>__builtin_saveregs</strong> <em>()</em></dt>
<dd><p>Use this built-in function to save the argument registers in memory so
that the varargs mechanism can access them.  Both ISO and traditional
versions of <code>va_start</code> must use <code>__builtin_saveregs</code>, unless
you use <code>TARGET_SETUP_INCOMING_VARARGS</code> (see below) instead.
</p>
<p>On some machines, <code>__builtin_saveregs</code> is open-coded under the
control of the target hook <code>TARGET_EXPAND_BUILTIN_SAVEREGS</code>.  On
other machines, it calls a routine written in assembler language,
found in <samp>libgcc2.c</samp>.
</p>
<p>Code generated for the call to <code>__builtin_saveregs</code> appears at the
beginning of the function, as opposed to where the call to
<code>__builtin_saveregs</code> is written, regardless of what the code is.
This is because the registers must be saved before the function starts
to use them for its own purposes.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fbuiltin_005fnext_005farg"></a>Macro: <strong>__builtin_next_arg</strong> <em>(<var>lastarg</var>)</em></dt>
<dd><p>This builtin returns the address of the first anonymous stack
argument, as type <code>void *</code>.  If <code>ARGS_GROW_DOWNWARD</code>, it
returns the address of the location above the first anonymous stack
argument.  Use it in <code>va_start</code> to initialize the pointer for
fetching arguments from the stack.  Also use it in <code>va_start</code> to
verify that the second parameter <var>lastarg</var> is the last named argument
of the current function.
</p></dd></dl>

<dl>
<dt><a name="index-_005f_005fbuiltin_005fclassify_005ftype"></a>Macro: <strong>__builtin_classify_type</strong> <em>(<var>object</var>)</em></dt>
<dd><p>Since each machine has its own conventions for which data types are
passed in which kind of register, your implementation of <code>va_arg</code>
has to embody these conventions.  The easiest way to categorize the
specified data type is to use <code>__builtin_classify_type</code> together
with <code>sizeof</code> and <code>__alignof__</code>.
</p>
<p><code>__builtin_classify_type</code> ignores the value of <var>object</var>,
considering only its data type.  It returns an integer describing what
kind of type that is&mdash;integer, floating, pointer, structure, and so on.
</p>
<p>The file <samp>typeclass.h</samp> defines an enumeration that you can use to
interpret the values of <code>__builtin_classify_type</code>.
</p></dd></dl>

<p>These machine description macros help implement varargs:
</p>
<dl>
<dt><a name="index-TARGET_005fEXPAND_005fBUILTIN_005fSAVEREGS"></a>Target Hook: <em>rtx</em> <strong>TARGET_EXPAND_BUILTIN_SAVEREGS</strong> <em>(void)</em></dt>
<dd><p>If defined, this hook produces the machine-specific code for a call to
<code>__builtin_saveregs</code>.  This code will be moved to the very
beginning of the function, before any parameter access are made.  The
return value of this function should be an RTX that contains the value
to use as the return of <code>__builtin_saveregs</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSETUP_005fINCOMING_005fVARARGS"></a>Target Hook: <em>void</em> <strong>TARGET_SETUP_INCOMING_VARARGS</strong> <em>(cumulative_args_t <var>args_so_far</var>, enum machine_mode <var>mode</var>, tree <var>type</var>, int *<var>pretend_args_size</var>, int <var>second_time</var>)</em></dt>
<dd><p>This target hook offers an alternative to using
<code>__builtin_saveregs</code> and defining the hook
<code>TARGET_EXPAND_BUILTIN_SAVEREGS</code>.  Use it to store the anonymous
register arguments into the stack so that all the arguments appear to
have been passed consecutively on the stack.  Once this is done, you can
use the standard implementation of varargs that works for machines that
pass all their arguments on the stack.
</p>
<p>The argument <var>args_so_far</var> points to the <code>CUMULATIVE_ARGS</code> data
structure, containing the values that are obtained after processing the
named arguments.  The arguments <var>mode</var> and <var>type</var> describe the
last named argument&mdash;its machine mode and its data type as a tree node.
</p>
<p>The target hook should do two things: first, push onto the stack all the
argument registers <em>not</em> used for the named arguments, and second,
store the size of the data thus pushed into the <code>int</code>-valued
variable pointed to by <var>pretend_args_size</var>.  The value that you
store here will serve as additional offset for setting up the stack
frame.
</p>
<p>Because you must generate code to push the anonymous arguments at
compile time without knowing their data types,
<code>TARGET_SETUP_INCOMING_VARARGS</code> is only useful on machines that
have just a single category of argument register and use it uniformly
for all data types.
</p>
<p>If the argument <var>second_time</var> is nonzero, it means that the
arguments of the function are being analyzed for the second time.  This
happens for an inline function, which is not actually compiled until the
end of the source file.  The hook <code>TARGET_SETUP_INCOMING_VARARGS</code> should
not generate any instructions in this case.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTRICT_005fARGUMENT_005fNAMING"></a>Target Hook: <em>bool</em> <strong>TARGET_STRICT_ARGUMENT_NAMING</strong> <em>(cumulative_args_t <var>ca</var>)</em></dt>
<dd><p>Define this hook to return <code>true</code> if the location where a function
argument is passed depends on whether or not it is a named argument.
</p>
<p>This hook controls how the <var>named</var> argument to <code>TARGET_FUNCTION_ARG</code>
is set for varargs and stdarg functions.  If this hook returns
<code>true</code>, the <var>named</var> argument is always true for named
arguments, and false for unnamed arguments.  If it returns <code>false</code>,
but <code>TARGET_PRETEND_OUTGOING_VARARGS_NAMED</code> returns <code>true</code>,
then all arguments are treated as named.  Otherwise, all named arguments
except the last are treated as named.
</p>
<p>You need not define this hook if it always returns <code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPRETEND_005fOUTGOING_005fVARARGS_005fNAMED"></a>Target Hook: <em>bool</em> <strong>TARGET_PRETEND_OUTGOING_VARARGS_NAMED</strong> <em>(cumulative_args_t <var>ca</var>)</em></dt>
<dd><p>If you need to conditionally change ABIs so that one works with
<code>TARGET_SETUP_INCOMING_VARARGS</code>, but the other works like neither
<code>TARGET_SETUP_INCOMING_VARARGS</code> nor <code>TARGET_STRICT_ARGUMENT_NAMING</code> was
defined, then define this hook to return <code>true</code> if
<code>TARGET_SETUP_INCOMING_VARARGS</code> is used, <code>false</code> otherwise.
Otherwise, you should not define this hook.
</p></dd></dl>

<hr>
<a name="Trampolines"></a>
<div class="header">
<p>
Next: <a href="#Library-Calls" accesskey="n" rel="next">Library Calls</a>, Previous: <a href="#Varargs" accesskey="p" rel="prev">Varargs</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Trampolines-for-Nested-Functions"></a>
<h3 class="section">17.12 Trampolines for Nested Functions</h3>
<a name="index-trampolines-for-nested-functions"></a>
<a name="index-nested-functions_002c-trampolines-for"></a>

<p>A <em>trampoline</em> is a small piece of code that is created at run time
when the address of a nested function is taken.  It normally resides on
the stack, in the stack frame of the containing function.  These macros
tell GCC how to generate code to allocate and initialize a
trampoline.
</p>
<p>The instructions in the trampoline must do two things: load a constant
address into the static chain register, and jump to the real address of
the nested function.  On CISC machines such as the m68k, this requires
two instructions, a move immediate and a jump.  Then the two addresses
exist in the trampoline as word-long immediate operands.  On RISC
machines, it is often necessary to load each address into a register in
two parts.  Then pieces of each address form separate immediate
operands.
</p>
<p>The code generated to initialize the trampoline must store the variable
parts&mdash;the static chain value and the function address&mdash;into the
immediate operands of the instructions.  On a CISC machine, this is
simply a matter of copying each address to a memory reference at the
proper offset from the start of the trampoline.  On a RISC machine, it
may be necessary to take out pieces of the address and store them
separately.
</p>
<dl>
<dt><a name="index-TARGET_005fASM_005fTRAMPOLINE_005fTEMPLATE"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_TRAMPOLINE_TEMPLATE</strong> <em>(FILE *<var>f</var>)</em></dt>
<dd><p>This hook is called by <code>assemble_trampoline_template</code> to output,
on the stream <var>f</var>, assembler code for a block of data that contains
the constant parts of a trampoline.  This code should not include a
label&mdash;the label is taken care of automatically.
</p>
<p>If you do not define this hook, it means no template is needed
for the target.  Do not define this hook on systems where the block move
code to copy the trampoline into place would be larger than the code
to generate it on the spot.
</p></dd></dl>

<dl>
<dt><a name="index-TRAMPOLINE_005fSECTION"></a>Macro: <strong>TRAMPOLINE_SECTION</strong></dt>
<dd><p>Return the section into which the trampoline template is to be placed
(see <a href="#Sections">Sections</a>).  The default value is <code>readonly_data_section</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TRAMPOLINE_005fSIZE"></a>Macro: <strong>TRAMPOLINE_SIZE</strong></dt>
<dd><p>A C expression for the size in bytes of the trampoline, as an integer.
</p></dd></dl>

<dl>
<dt><a name="index-TRAMPOLINE_005fALIGNMENT"></a>Macro: <strong>TRAMPOLINE_ALIGNMENT</strong></dt>
<dd><p>Alignment required for trampolines, in bits.
</p>
<p>If you don&rsquo;t define this macro, the value of <code>FUNCTION_ALIGNMENT</code>
is used for aligning trampolines.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fTRAMPOLINE_005fINIT"></a>Target Hook: <em>void</em> <strong>TARGET_TRAMPOLINE_INIT</strong> <em>(rtx <var>m_tramp</var>, tree <var>fndecl</var>, rtx <var>static_chain</var>)</em></dt>
<dd><p>This hook is called to initialize a trampoline.
<var>m_tramp</var> is an RTX for the memory block for the trampoline; <var>fndecl</var>
is the <code>FUNCTION_DECL</code> for the nested function; <var>static_chain</var> is an
RTX for the static chain value that should be passed to the function
when it is called.
</p>
<p>If the target defines <code>TARGET_ASM_TRAMPOLINE_TEMPLATE</code>, then the
first thing this hook should do is emit a block move into <var>m_tramp</var>
from the memory block returned by <code>assemble_trampoline_template</code>.
Note that the block move need only cover the constant parts of the
trampoline.  If the target isolates the variable parts of the trampoline
to the end, not all <code>TRAMPOLINE_SIZE</code> bytes need be copied.
</p>
<p>If the target requires any other actions, such as flushing caches or
enabling stack execution, these actions should be performed after
initializing the trampoline proper.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fTRAMPOLINE_005fADJUST_005fADDRESS"></a>Target Hook: <em>rtx</em> <strong>TARGET_TRAMPOLINE_ADJUST_ADDRESS</strong> <em>(rtx <var>addr</var>)</em></dt>
<dd><p>This hook should perform any machine-specific adjustment in
the address of the trampoline.  Its argument contains the address of the
memory block that was passed to <code>TARGET_TRAMPOLINE_INIT</code>.  In case
the address to be used for a function call should be different from the
address at which the template was stored, the different address should
be returned; otherwise <var>addr</var> should be returned unchanged.
If this hook is not defined, <var>addr</var> will be used for function calls.
</p></dd></dl>

<p>Implementing trampolines is difficult on many machines because they have
separate instruction and data caches.  Writing into a stack location
fails to clear the memory in the instruction cache, so when the program
jumps to that location, it executes the old contents.
</p>
<p>Here are two possible solutions.  One is to clear the relevant parts of
the instruction cache whenever a trampoline is set up.  The other is to
make all trampolines identical, by having them jump to a standard
subroutine.  The former technique makes trampoline execution faster; the
latter makes initialization faster.
</p>
<p>To clear the instruction cache when a trampoline is initialized, define
the following macro.
</p>
<dl>
<dt><a name="index-CLEAR_005fINSN_005fCACHE"></a>Macro: <strong>CLEAR_INSN_CACHE</strong> <em>(<var>beg</var>, <var>end</var>)</em></dt>
<dd><p>If defined, expands to a C expression clearing the <em>instruction
cache</em> in the specified interval.  The definition of this macro would
typically be a series of <code>asm</code> statements.  Both <var>beg</var> and
<var>end</var> are both pointer expressions.
</p></dd></dl>

<p>To use a standard subroutine, define the following macro.  In addition,
you must make sure that the instructions in a trampoline fill an entire
cache line with identical instructions, or else ensure that the
beginning of the trampoline code is always aligned at the same point in
its cache line.  Look in <samp>m68k.h</samp> as a guide.
</p>
<dl>
<dt><a name="index-TRANSFER_005fFROM_005fTRAMPOLINE"></a>Macro: <strong>TRANSFER_FROM_TRAMPOLINE</strong></dt>
<dd><p>Define this macro if trampolines need a special subroutine to do their
work.  The macro should expand to a series of <code>asm</code> statements
which will be compiled with GCC.  They go in a library function named
<code>__transfer_from_trampoline</code>.
</p>
<p>If you need to avoid executing the ordinary prologue code of a compiled
C function when you jump to the subroutine, you can do so by placing a
special label of your own in the assembler code.  Use one <code>asm</code>
statement to generate an assembler label, and another to make the label
global.  Then trampolines can use that label to jump directly to your
special assembler code.
</p></dd></dl>

<hr>
<a name="Library-Calls"></a>
<div class="header">
<p>
Next: <a href="#Addressing-Modes" accesskey="n" rel="next">Addressing Modes</a>, Previous: <a href="#Trampolines" accesskey="p" rel="prev">Trampolines</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Implicit-Calls-to-Library-Routines"></a>
<h3 class="section">17.13 Implicit Calls to Library Routines</h3>
<a name="index-library-subroutine-names"></a>
<a name="index-libgcc_002ea"></a>

<p>Here is an explanation of implicit calls to library routines.
</p>
<dl>
<dt><a name="index-DECLARE_005fLIBRARY_005fRENAMES"></a>Macro: <strong>DECLARE_LIBRARY_RENAMES</strong></dt>
<dd><p>This macro, if defined, should expand to a piece of C code that will get
expanded when compiling functions for libgcc.a.  It can be used to
provide alternate names for GCC&rsquo;s internal library functions if there
are ABI-mandated names that the compiler should provide.
</p></dd></dl>

<a name="index-set_005foptab_005flibfunc"></a>
<a name="index-init_005fone_005flibfunc"></a>
<dl>
<dt><a name="index-TARGET_005fINIT_005fLIBFUNCS"></a>Target Hook: <em>void</em> <strong>TARGET_INIT_LIBFUNCS</strong> <em>(void)</em></dt>
<dd><p>This hook should declare additional library routines or rename
existing ones, using the functions <code>set_optab_libfunc</code> and
<code>init_one_libfunc</code> defined in <samp>optabs.c</samp>.
<code>init_optabs</code> calls this macro after initializing all the normal
library routines.
</p>
<p>The default is to do nothing.  Most ports don&rsquo;t need to define this hook.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIBFUNC_005fGNU_005fPREFIX"></a>Target Hook: <em>bool</em> <strong>TARGET_LIBFUNC_GNU_PREFIX</strong></dt>
<dd><p>If false (the default), internal library routines start with two
underscores.  If set to true, these routines start with <code>__gnu_</code>
instead.  E.g., <code>__muldi3</code> changes to <code>__gnu_muldi3</code>.  This
currently only affects functions defined in <samp>libgcc2.c</samp>.  If this
is set to true, the <samp>tm.h</samp> file must also
<code>#define LIBGCC2_GNU_PREFIX</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FLOAT_005fLIB_005fCOMPARE_005fRETURNS_005fBOOL"></a>Macro: <strong>FLOAT_LIB_COMPARE_RETURNS_BOOL</strong> <em>(<var>mode</var>, <var>comparison</var>)</em></dt>
<dd><p>This macro should return <code>true</code> if the library routine that
implements the floating point comparison operator <var>comparison</var> in
mode <var>mode</var> will return a boolean, and <var>false</var> if it will
return a tristate.
</p>
<p>GCC&rsquo;s own floating point libraries return tristates from the
comparison operators, so the default returns false always.  Most ports
don&rsquo;t need to define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIB_005fINT_005fCMP_005fBIASED"></a>Macro: <strong>TARGET_LIB_INT_CMP_BIASED</strong></dt>
<dd><p>This macro should evaluate to <code>true</code> if the integer comparison
functions (like <code>__cmpdi2</code>) return 0 to indicate that the first
operand is smaller than the second, 1 to indicate that they are equal,
and 2 to indicate that the first operand is greater than the second.
If this macro evaluates to <code>false</code> the comparison functions return
-1, 0, and 1 instead of 0, 1, and 2.  If the target uses the routines
in <samp>libgcc.a</samp>, you do not need to define this macro.
</p></dd></dl>

<a name="index-EDOM_002c-implicit-usage"></a>
<a name="index-matherr"></a>
<dl>
<dt><a name="index-TARGET_005fEDOM"></a>Macro: <strong>TARGET_EDOM</strong></dt>
<dd><p>The value of <code>EDOM</code> on the target machine, as a C integer constant
expression.  If you don&rsquo;t define this macro, GCC does not attempt to
deposit the value of <code>EDOM</code> into <code>errno</code> directly.  Look in
<samp>/usr/include/errno.h</samp> to find the value of <code>EDOM</code> on your
system.
</p>
<p>If you do not define <code>TARGET_EDOM</code>, then compiled code reports
domain errors by calling the library function and letting it report the
error.  If mathematical functions on your system use <code>matherr</code> when
there is an error, then you should leave <code>TARGET_EDOM</code> undefined so
that <code>matherr</code> is used normally.
</p></dd></dl>

<a name="index-errno_002c-implicit-usage"></a>
<dl>
<dt><a name="index-GEN_005fERRNO_005fRTX"></a>Macro: <strong>GEN_ERRNO_RTX</strong></dt>
<dd><p>Define this macro as a C expression to create an rtl expression that
refers to the global &ldquo;variable&rdquo; <code>errno</code>.  (On certain systems,
<code>errno</code> may not actually be a variable.)  If you don&rsquo;t define this
macro, a reasonable default is used.
</p></dd></dl>

<a name="index-C99-math-functions_002c-implicit-usage"></a>
<dl>
<dt><a name="index-TARGET_005fC99_005fFUNCTIONS"></a>Macro: <strong>TARGET_C99_FUNCTIONS</strong></dt>
<dd><p>When this macro is nonzero, GCC will implicitly optimize <code>sin</code> calls into
<code>sinf</code> and similarly for other functions defined by C99 standard.  The
default is zero because a number of existing systems lack support for these
functions in their runtime so this macro needs to be redefined to one on
systems that do support the C99 runtime.
</p></dd></dl>

<a name="index-sincos-math-function_002c-implicit-usage"></a>
<dl>
<dt><a name="index-TARGET_005fHAS_005fSINCOS"></a>Macro: <strong>TARGET_HAS_SINCOS</strong></dt>
<dd><p>When this macro is nonzero, GCC will implicitly optimize calls to <code>sin</code>
and <code>cos</code> with the same argument to a call to <code>sincos</code>.  The
default is zero.  The target has to provide the following functions:
</p><div class="smallexample">
<pre class="smallexample">void sincos(double x, double *sin, double *cos);
void sincosf(float x, float *sin, float *cos);
void sincosl(long double x, long double *sin, long double *cos);
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-NEXT_005fOBJC_005fRUNTIME"></a>Macro: <strong>NEXT_OBJC_RUNTIME</strong></dt>
<dd><p>Set this macro to 1 to use the &quot;NeXT&quot; Objective-C message sending conventions
by default.  This calling convention involves passing the object, the selector
and the method arguments all at once to the method-lookup library function.
This is the usual setting when targeting Darwin/Mac OS X systems, which have
the NeXT runtime installed.
</p>
<p>If the macro is set to 0, the &quot;GNU&quot; Objective-C message sending convention
will be used by default.  This convention passes just the object and the
selector to the method-lookup function, which returns a pointer to the method.
</p>
<p>In either case, it remains possible to select code-generation for the alternate
scheme, by means of compiler command line switches.
</p></dd></dl>

<hr>
<a name="Addressing-Modes"></a>
<div class="header">
<p>
Next: <a href="#Anchored-Addresses" accesskey="n" rel="next">Anchored Addresses</a>, Previous: <a href="#Library-Calls" accesskey="p" rel="prev">Library Calls</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Addressing-Modes-1"></a>
<h3 class="section">17.14 Addressing Modes</h3>
<a name="index-addressing-modes"></a>

<p>This is about addressing modes.
</p>
<dl>
<dt><a name="index-HAVE_005fPRE_005fINCREMENT"></a>Macro: <strong>HAVE_PRE_INCREMENT</strong></dt>
<dt><a name="index-HAVE_005fPRE_005fDECREMENT"></a>Macro: <strong>HAVE_PRE_DECREMENT</strong></dt>
<dt><a name="index-HAVE_005fPOST_005fINCREMENT"></a>Macro: <strong>HAVE_POST_INCREMENT</strong></dt>
<dt><a name="index-HAVE_005fPOST_005fDECREMENT"></a>Macro: <strong>HAVE_POST_DECREMENT</strong></dt>
<dd><p>A C expression that is nonzero if the machine supports pre-increment,
pre-decrement, post-increment, or post-decrement addressing respectively.
</p></dd></dl>

<dl>
<dt><a name="index-HAVE_005fPRE_005fMODIFY_005fDISP"></a>Macro: <strong>HAVE_PRE_MODIFY_DISP</strong></dt>
<dt><a name="index-HAVE_005fPOST_005fMODIFY_005fDISP"></a>Macro: <strong>HAVE_POST_MODIFY_DISP</strong></dt>
<dd><p>A C expression that is nonzero if the machine supports pre- or
post-address side-effect generation involving constants other than
the size of the memory operand.
</p></dd></dl>

<dl>
<dt><a name="index-HAVE_005fPRE_005fMODIFY_005fREG"></a>Macro: <strong>HAVE_PRE_MODIFY_REG</strong></dt>
<dt><a name="index-HAVE_005fPOST_005fMODIFY_005fREG"></a>Macro: <strong>HAVE_POST_MODIFY_REG</strong></dt>
<dd><p>A C expression that is nonzero if the machine supports pre- or
post-address side-effect generation involving a register displacement.
</p></dd></dl>

<dl>
<dt><a name="index-CONSTANT_005fADDRESS_005fP"></a>Macro: <strong>CONSTANT_ADDRESS_P</strong> <em>(<var>x</var>)</em></dt>
<dd><p>A C expression that is 1 if the RTX <var>x</var> is a constant which
is a valid address.  On most machines the default definition of
<code>(CONSTANT_P (<var>x</var>) &amp;&amp; GET_CODE (<var>x</var>) != CONST_DOUBLE)</code>
is acceptable, but a few machines are more restrictive as to which
constant addresses are supported.
</p></dd></dl>

<dl>
<dt><a name="index-CONSTANT_005fP"></a>Macro: <strong>CONSTANT_P</strong> <em>(<var>x</var>)</em></dt>
<dd><p><code>CONSTANT_P</code>, which is defined by target-independent code,
accepts integer-values expressions whose values are not explicitly
known, such as <code>symbol_ref</code>, <code>label_ref</code>, and <code>high</code>
expressions and <code>const</code> arithmetic expressions, in addition to
<code>const_int</code> and <code>const_double</code> expressions.
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fREGS_005fPER_005fADDRESS"></a>Macro: <strong>MAX_REGS_PER_ADDRESS</strong></dt>
<dd><p>A number, the maximum number of registers that can appear in a valid
memory address.  Note that it is up to you to specify a value equal to
the maximum number that <code>TARGET_LEGITIMATE_ADDRESS_P</code> would ever
accept.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLEGITIMATE_005fADDRESS_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_LEGITIMATE_ADDRESS_P</strong> <em>(enum machine_mode <var>mode</var>, rtx <var>x</var>, bool <var>strict</var>)</em></dt>
<dd><p>A function that returns whether <var>x</var> (an RTX) is a legitimate memory
address on the target machine for a memory operand of mode <var>mode</var>.
</p>
<p>Legitimate addresses are defined in two variants: a strict variant and a
non-strict one.  The <var>strict</var> parameter chooses which variant is
desired by the caller.
</p>
<p>The strict variant is used in the reload pass.  It must be defined so
that any pseudo-register that has not been allocated a hard register is
considered a memory reference.  This is because in contexts where some
kind of register is required, a pseudo-register with no hard register
must be rejected.  For non-hard registers, the strict variant should look
up the <code>reg_renumber</code> array; it should then proceed using the hard
register number in the array, or treat the pseudo as a memory reference
if the array holds <code>-1</code>.
</p>
<p>The non-strict variant is used in other passes.  It must be defined to
accept all pseudo-registers in every context where some kind of
register is required.
</p>
<p>Normally, constant addresses which are the sum of a <code>symbol_ref</code>
and an integer are stored inside a <code>const</code> RTX to mark them as
constant.  Therefore, there is no need to recognize such sums
specifically as legitimate addresses.  Normally you would simply
recognize any <code>const</code> as legitimate.
</p>
<p>Usually <code>PRINT_OPERAND_ADDRESS</code> is not prepared to handle constant
sums that are not marked with  <code>const</code>.  It assumes that a naked
<code>plus</code> indicates indexing.  If so, then you <em>must</em> reject such
naked constant sums as illegitimate addresses, so that none of them will
be given to <code>PRINT_OPERAND_ADDRESS</code>.
</p>
<a name="index-TARGET_005fENCODE_005fSECTION_005fINFO-and-address-validation"></a>
<p>On some machines, whether a symbolic address is legitimate depends on
the section that the address refers to.  On these machines, define the
target hook <code>TARGET_ENCODE_SECTION_INFO</code> to store the information
into the <code>symbol_ref</code>, and then check for it here.  When you see a
<code>const</code>, you will have to look inside it to find the
<code>symbol_ref</code> in order to determine the section.  See <a href="#Assembler-Format">Assembler Format</a>.
</p>
<a name="index-GO_005fIF_005fLEGITIMATE_005fADDRESS"></a>
<p>Some ports are still using a deprecated legacy substitute for
this hook, the <code>GO_IF_LEGITIMATE_ADDRESS</code> macro.  This macro
has this syntax:
</p>
<div class="example">
<pre class="example">#define GO_IF_LEGITIMATE_ADDRESS (<var>mode</var>, <var>x</var>, <var>label</var>)
</pre></div>

<p>and should <code>goto <var>label</var></code> if the address <var>x</var> is a valid
address on the target machine for a memory operand of mode <var>mode</var>.
</p>
<a name="index-REG_005fOK_005fSTRICT"></a>
<p>Compiler source files that want to use the strict variant of this
macro define the macro <code>REG_OK_STRICT</code>.  You should use an
<code>#ifdef REG_OK_STRICT</code> conditional to define the strict variant in
that case and the non-strict variant otherwise.
</p>
<p>Using the hook is usually simpler because it limits the number of
files that are recompiled when changes are made.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMEM_005fCONSTRAINT"></a>Macro: <strong>TARGET_MEM_CONSTRAINT</strong></dt>
<dd><p>A single character to be used instead of the default <code>'m'</code>
character for general memory addresses.  This defines the constraint
letter which matches the memory addresses accepted by
<code>TARGET_LEGITIMATE_ADDRESS_P</code>.  Define this macro if you want to
support new address formats in your back end without changing the
semantics of the <code>'m'</code> constraint.  This is necessary in order to
preserve functionality of inline assembly constructs using the
<code>'m'</code> constraint.
</p></dd></dl>

<dl>
<dt><a name="index-FIND_005fBASE_005fTERM"></a>Macro: <strong>FIND_BASE_TERM</strong> <em>(<var>x</var>)</em></dt>
<dd><p>A C expression to determine the base term of address <var>x</var>,
or to provide a simplified version of <var>x</var> from which <samp>alias.c</samp>
can easily find the base term.  This macro is used in only two places:
<code>find_base_value</code> and <code>find_base_term</code> in <samp>alias.c</samp>.
</p>
<p>It is always safe for this macro to not be defined.  It exists so
that alias analysis can understand machine-dependent addresses.
</p>
<p>The typical use of this macro is to handle addresses containing
a label_ref or symbol_ref within an UNSPEC.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLEGITIMIZE_005fADDRESS"></a>Target Hook: <em>rtx</em> <strong>TARGET_LEGITIMIZE_ADDRESS</strong> <em>(rtx <var>x</var>, rtx <var>oldx</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>This hook is given an invalid memory address <var>x</var> for an
operand of mode <var>mode</var> and should try to return a valid memory
address.
</p>
<a name="index-break_005fout_005fmemory_005frefs"></a>
<p><var>x</var> will always be the result of a call to <code>break_out_memory_refs</code>,
and <var>oldx</var> will be the operand that was given to that function to produce
<var>x</var>.
</p>
<p>The code of the hook should not alter the substructure of
<var>x</var>.  If it transforms <var>x</var> into a more legitimate form, it
should return the new <var>x</var>.
</p>
<p>It is not necessary for this hook to come up with a legitimate address,
with the exception of native TLS addresses (see <a href="#Emulated-TLS">Emulated TLS</a>).
The compiler has standard ways of doing so in all cases.  In fact, if
the target supports only emulated TLS, it
is safe to omit this hook or make it return <var>x</var> if it cannot find
a valid way to legitimize the address.  But often a machine-dependent
strategy can generate better code.
</p></dd></dl>

<dl>
<dt><a name="index-LEGITIMIZE_005fRELOAD_005fADDRESS"></a>Macro: <strong>LEGITIMIZE_RELOAD_ADDRESS</strong> <em>(<var>x</var>, <var>mode</var>, <var>opnum</var>, <var>type</var>, <var>ind_levels</var>, <var>win</var>)</em></dt>
<dd><p>A C compound statement that attempts to replace <var>x</var>, which is an address
that needs reloading, with a valid memory address for an operand of mode
<var>mode</var>.  <var>win</var> will be a C statement label elsewhere in the code.
It is not necessary to define this macro, but it might be useful for
performance reasons.
</p>
<p>For example, on the i386, it is sometimes possible to use a single
reload register instead of two by reloading a sum of two pseudo
registers into a register.  On the other hand, for number of RISC
processors offsets are limited so that often an intermediate address
needs to be generated in order to address a stack slot.  By defining
<code>LEGITIMIZE_RELOAD_ADDRESS</code> appropriately, the intermediate addresses
generated for adjacent some stack slots can be made identical, and thus
be shared.
</p>
<p><em>Note</em>: This macro should be used with caution.  It is necessary
to know something of how reload works in order to effectively use this,
and it is quite easy to produce macros that build in too much knowledge
of reload internals.
</p>
<p><em>Note</em>: This macro must be able to reload an address created by a
previous invocation of this macro.  If it fails to handle such addresses
then the compiler may generate incorrect code or abort.
</p>
<a name="index-push_005freload"></a>
<p>The macro definition should use <code>push_reload</code> to indicate parts that
need reloading; <var>opnum</var>, <var>type</var> and <var>ind_levels</var> are usually
suitable to be passed unaltered to <code>push_reload</code>.
</p>
<p>The code generated by this macro must not alter the substructure of
<var>x</var>.  If it transforms <var>x</var> into a more legitimate form, it
should assign <var>x</var> (which will always be a C variable) a new value.
This also applies to parts that you change indirectly by calling
<code>push_reload</code>.
</p>
<a name="index-strict_005fmemory_005faddress_005fp"></a>
<p>The macro definition may use <code>strict_memory_address_p</code> to test if
the address has become legitimate.
</p>
<a name="index-copy_005frtx"></a>
<p>If you want to change only a part of <var>x</var>, one standard way of doing
this is to use <code>copy_rtx</code>.  Note, however, that it unshares only a
single level of rtl.  Thus, if the part to be changed is not at the
top level, you&rsquo;ll need to replace first the top level.
It is not necessary for this macro to come up with a legitimate
address;  but often a machine-dependent strategy can generate better code.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMODE_005fDEPENDENT_005fADDRESS_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_MODE_DEPENDENT_ADDRESS_P</strong> <em>(const_rtx <var>addr</var>, addr_space_t <var>addrspace</var>)</em></dt>
<dd><p>This hook returns <code>true</code> if memory address <var>addr</var> in address
space <var>addrspace</var> can have
different meanings depending on the machine mode of the memory
reference it is used for or if the address is valid for some modes
but not others.
</p>
<p>Autoincrement and autodecrement addresses typically have mode-dependent
effects because the amount of the increment or decrement is the size
of the operand being addressed.  Some machines have other mode-dependent
addresses.  Many RISC machines have no mode-dependent addresses.
</p>
<p>You may assume that <var>addr</var> is a valid address for the machine.
</p>
<p>The default version of this hook returns <code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLEGITIMATE_005fCONSTANT_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_LEGITIMATE_CONSTANT_P</strong> <em>(enum machine_mode <var>mode</var>, rtx <var>x</var>)</em></dt>
<dd><p>This hook returns true if <var>x</var> is a legitimate constant for a
<var>mode</var>-mode immediate operand on the target machine.  You can assume that
<var>x</var> satisfies <code>CONSTANT_P</code>, so you need not check this.
</p>
<p>The default definition returns true.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDELEGITIMIZE_005fADDRESS"></a>Target Hook: <em>rtx</em> <strong>TARGET_DELEGITIMIZE_ADDRESS</strong> <em>(rtx <var>x</var>)</em></dt>
<dd><p>This hook is used to undo the possibly obfuscating effects of the
<code>LEGITIMIZE_ADDRESS</code> and <code>LEGITIMIZE_RELOAD_ADDRESS</code> target
macros.  Some backend implementations of these macros wrap symbol
references inside an <code>UNSPEC</code> rtx to represent PIC or similar
addressing modes.  This target hook allows GCC&rsquo;s optimizers to understand
the semantics of these opaque <code>UNSPEC</code>s by converting them back
into their original form.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCONST_005fNOT_005fOK_005fFOR_005fDEBUG_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_CONST_NOT_OK_FOR_DEBUG_P</strong> <em>(rtx <var>x</var>)</em></dt>
<dd><p>This hook should return true if <var>x</var> should not be emitted into
debug sections.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCANNOT_005fFORCE_005fCONST_005fMEM"></a>Target Hook: <em>bool</em> <strong>TARGET_CANNOT_FORCE_CONST_MEM</strong> <em>(enum machine_mode <var>mode</var>, rtx <var>x</var>)</em></dt>
<dd><p>This hook should return true if <var>x</var> is of a form that cannot (or
should not) be spilled to the constant pool.  <var>mode</var> is the mode
of <var>x</var>.
</p>
<p>The default version of this hook returns false.
</p>
<p>The primary reason to define this hook is to prevent reload from
deciding that a non-legitimate constant would be better reloaded
from the constant pool instead of spilling and reloading a register
holding the constant.  This restriction is often true of addresses
of TLS symbols for various targets.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUSE_005fBLOCKS_005fFOR_005fCONSTANT_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_USE_BLOCKS_FOR_CONSTANT_P</strong> <em>(enum machine_mode <var>mode</var>, const_rtx <var>x</var>)</em></dt>
<dd><p>This hook should return true if pool entries for constant <var>x</var> can
be placed in an <code>object_block</code> structure.  <var>mode</var> is the mode
of <var>x</var>.
</p>
<p>The default version returns false for all constants.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUSE_005fBLOCKS_005fFOR_005fDECL_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_USE_BLOCKS_FOR_DECL_P</strong> <em>(const_tree <var>decl</var>)</em></dt>
<dd><p>This hook should return true if pool entries for <var>decl</var> should
be placed in an <code>object_block</code> structure.
</p>
<p>The default version returns true for all decls.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBUILTIN_005fRECIPROCAL"></a>Target Hook: <em>tree</em> <strong>TARGET_BUILTIN_RECIPROCAL</strong> <em>(unsigned <var>fn</var>, bool <var>md_fn</var>, bool <var>sqrt</var>)</em></dt>
<dd><p>This hook should return the DECL of a function that implements reciprocal of
the builtin function with builtin function code <var>fn</var>, or
<code>NULL_TREE</code> if such a function is not available.  <var>md_fn</var> is true
when <var>fn</var> is a code of a machine-dependent builtin function.  When
<var>sqrt</var> is true, additional optimizations that apply only to the reciprocal
of a square root function are performed, and only reciprocals of <code>sqrt</code>
function are valid.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fMASK_005fFOR_005fLOAD"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_MASK_FOR_LOAD</strong> <em>(void)</em></dt>
<dd><p>This hook should return the DECL of a function <var>f</var> that given an
address <var>addr</var> as an argument returns a mask <var>m</var> that can be
used to extract from two vectors the relevant data that resides in
<var>addr</var> in case <var>addr</var> is not properly aligned.
</p>
<p>The autovectorizer, when vectorizing a load operation from an address
<var>addr</var> that may be unaligned, will generate two vector loads from
the two aligned addresses around <var>addr</var>. It then generates a
<code>REALIGN_LOAD</code> operation to extract the relevant data from the
two loaded vectors. The first two arguments to <code>REALIGN_LOAD</code>,
<var>v1</var> and <var>v2</var>, are the two vectors, each of size <var>VS</var>, and
the third argument, <var>OFF</var>, defines how the data will be extracted
from these two vectors: if <var>OFF</var> is 0, then the returned vector is
<var>v2</var>; otherwise, the returned vector is composed from the last
<var>VS</var>-<var>OFF</var> elements of <var>v1</var> concatenated to the first
<var>OFF</var> elements of <var>v2</var>.
</p>
<p>If this hook is defined, the autovectorizer will generate a call
to <var>f</var> (using the DECL tree that this hook returns) and will
use the return value of <var>f</var> as the argument <var>OFF</var> to
<code>REALIGN_LOAD</code>. Therefore, the mask <var>m</var> returned by <var>f</var>
should comply with the semantics expected by <code>REALIGN_LOAD</code>
described above.
If this hook is not defined, then <var>addr</var> will be used as
the argument <var>OFF</var> to <code>REALIGN_LOAD</code>, in which case the low
log2(<var>VS</var>) - 1 bits of <var>addr</var> will be considered.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fVECTORIZATION_005fCOST"></a>Target Hook: <em>int</em> <strong>TARGET_VECTORIZE_BUILTIN_VECTORIZATION_COST</strong> <em>(enum vect_cost_for_stmt <var>type_of_cost</var>, tree <var>vectype</var>, int <var>misalign</var>)</em></dt>
<dd><p>Returns cost of different scalar or vector statements for vectorization cost model.
For vector memory operations the cost may depend on type (<var>vectype</var>) and
misalignment value (<var>misalign</var>).
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fVECTOR_005fALIGNMENT_005fREACHABLE"></a>Target Hook: <em>bool</em> <strong>TARGET_VECTORIZE_VECTOR_ALIGNMENT_REACHABLE</strong> <em>(const_tree <var>type</var>, bool <var>is_packed</var>)</em></dt>
<dd><p>Return true if vector alignment is reachable (by peeling N iterations) for the given type.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fVEC_005fPERM_005fCONST_005fOK"></a>Target Hook: <em>bool</em> <strong>TARGET_VECTORIZE_VEC_PERM_CONST_OK</strong> <em>(enum <var>machine_mode</var>, const unsigned char *<var>sel</var>)</em></dt>
<dd><p>Return true if a vector created for <code>vec_perm_const</code> is valid.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fCONVERSION"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_CONVERSION</strong> <em>(unsigned <var>code</var>, tree <var>dest_type</var>, tree <var>src_type</var>)</em></dt>
<dd><p>This hook should return the DECL of a function that implements conversion of the
input vector of type <var>src_type</var> to type <var>dest_type</var>.
The value of <var>code</var> is one of the enumerators in <code>enum tree_code</code> and
specifies how the conversion is to be applied
(truncation, rounding, etc.).
</p>
<p>If this hook is defined, the autovectorizer will use the
<code>TARGET_VECTORIZE_BUILTIN_CONVERSION</code> target hook when vectorizing
conversion. Otherwise, it will return <code>NULL_TREE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fVECTORIZED_005fFUNCTION"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_VECTORIZED_FUNCTION</strong> <em>(tree <var>fndecl</var>, tree <var>vec_type_out</var>, tree <var>vec_type_in</var>)</em></dt>
<dd><p>This hook should return the decl of a function that implements the
vectorized variant of the builtin function with builtin function code
<var>code</var> or <code>NULL_TREE</code> if such a function is not available.
The value of <var>fndecl</var> is the builtin function declaration.  The
return type of the vectorized function shall be of vector type
<var>vec_type_out</var> and the argument types should be <var>vec_type_in</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fSUPPORT_005fVECTOR_005fMISALIGNMENT"></a>Target Hook: <em>bool</em> <strong>TARGET_VECTORIZE_SUPPORT_VECTOR_MISALIGNMENT</strong> <em>(enum machine_mode <var>mode</var>, const_tree <var>type</var>, int <var>misalignment</var>, bool <var>is_packed</var>)</em></dt>
<dd><p>This hook should return true if the target supports misaligned vector
store/load of a specific factor denoted in the <var>misalignment</var>
parameter.  The vector store/load should be of machine mode <var>mode</var> and
the elements in the vectors should be of type <var>type</var>.  <var>is_packed</var>
parameter is true if the memory access is defined in a packed struct.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fPREFERRED_005fSIMD_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_VECTORIZE_PREFERRED_SIMD_MODE</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>This hook should return the preferred mode for vectorizing scalar
mode <var>mode</var>.  The default is
equal to <code>word_mode</code>, because the vectorizer can do some
transformations even in absence of specialized <acronym>SIMD</acronym> hardware.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fAUTOVECTORIZE_005fVECTOR_005fSIZES"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_VECTORIZE_AUTOVECTORIZE_VECTOR_SIZES</strong> <em>(void)</em></dt>
<dd><p>This hook should return a mask of sizes that should be iterated over
after trying to autovectorize using the vector size derived from the
mode returned by <code>TARGET_VECTORIZE_PREFERRED_SIMD_MODE</code>.
The default is zero which means to not iterate over other vector sizes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fINIT_005fCOST"></a>Target Hook: <em>void *</em> <strong>TARGET_VECTORIZE_INIT_COST</strong> <em>(struct loop *<var>loop_info</var>)</em></dt>
<dd><p>This hook should initialize target-specific data structures in preparation for modeling the costs of vectorizing a loop or basic block.  The default allocates three unsigned integers for accumulating costs for the prologue, body, and epilogue of the loop or basic block.  If <var>loop_info</var> is non-NULL, it identifies the loop being vectorized; otherwise a single block is being vectorized.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fADD_005fSTMT_005fCOST"></a>Target Hook: <em>unsigned</em> <strong>TARGET_VECTORIZE_ADD_STMT_COST</strong> <em>(void *<var>data</var>, int <var>count</var>, enum vect_cost_for_stmt <var>kind</var>, struct _stmt_vec_info *<var>stmt_info</var>, int <var>misalign</var>, enum vect_cost_model_location <var>where</var>)</em></dt>
<dd><p>This hook should update the target-specific <var>data</var> in response to adding <var>count</var> copies of the given <var>kind</var> of statement to a loop or basic block.  The default adds the builtin vectorizer cost for the copies of the statement to the accumulator specified by <var>where</var>, (the prologue, body, or epilogue) and returns the amount added.  The return value should be viewed as a tentative cost that may later be revised.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fFINISH_005fCOST"></a>Target Hook: <em>void</em> <strong>TARGET_VECTORIZE_FINISH_COST</strong> <em>(void *<var>data</var>, unsigned *<var>prologue_cost</var>, unsigned *<var>body_cost</var>, unsigned *<var>epilogue_cost</var>)</em></dt>
<dd><p>This hook should complete calculations of the cost of vectorizing a loop or basic block based on <var>data</var>, and return the prologue, body, and epilogue costs as unsigned integers.  The default returns the value of the three accumulators.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fDESTROY_005fCOST_005fDATA"></a>Target Hook: <em>void</em> <strong>TARGET_VECTORIZE_DESTROY_COST_DATA</strong> <em>(void *<var>data</var>)</em></dt>
<dd><p>This hook should release <var>data</var> and any related data structures allocated by TARGET_VECTORIZE_INIT_COST.  The default releases the accumulator.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fTM_005fLOAD"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_TM_LOAD</strong> <em>(tree)</em></dt>
<dd><p>This hook should return the built-in decl needed to load a vector of the given type within a transaction.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fTM_005fSTORE"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_TM_STORE</strong> <em>(tree)</em></dt>
<dd><p>This hook should return the built-in decl needed to store a vector of the given type within a transaction.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVECTORIZE_005fBUILTIN_005fGATHER"></a>Target Hook: <em>tree</em> <strong>TARGET_VECTORIZE_BUILTIN_GATHER</strong> <em>(const_tree <var>mem_vectype</var>, const_tree <var>index_type</var>, int <var>scale</var>)</em></dt>
<dd><p>Target builtin that implements vector gather operation.  <var>mem_vectype</var>
is the vector type of the load and <var>index_type</var> is scalar type of
the index, scaled by <var>scale</var>.
The default is <code>NULL_TREE</code> which means to not vectorize gather
loads.
</p></dd></dl>

<hr>
<a name="Anchored-Addresses"></a>
<div class="header">
<p>
Next: <a href="#Condition-Code" accesskey="n" rel="next">Condition Code</a>, Previous: <a href="#Addressing-Modes" accesskey="p" rel="prev">Addressing Modes</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Anchored-Addresses-1"></a>
<h3 class="section">17.15 Anchored Addresses</h3>
<a name="index-anchored-addresses"></a>
<a name="index-_002dfsection_002danchors-1"></a>

<p>GCC usually addresses every static object as a separate entity.
For example, if we have:
</p>
<div class="smallexample">
<pre class="smallexample">static int a, b, c;
int foo (void) { return a + b + c; }
</pre></div>

<p>the code for <code>foo</code> will usually calculate three separate symbolic
addresses: those of <code>a</code>, <code>b</code> and <code>c</code>.  On some targets,
it would be better to calculate just one symbolic address and access
the three variables relative to it.  The equivalent pseudocode would
be something like:
</p>
<div class="smallexample">
<pre class="smallexample">int foo (void)
{
  register int *xr = &amp;x;
  return xr[&amp;a - &amp;x] + xr[&amp;b - &amp;x] + xr[&amp;c - &amp;x];
}
</pre></div>

<p>(which isn&rsquo;t valid C).  We refer to shared addresses like <code>x</code> as
&ldquo;section anchors&rdquo;.  Their use is controlled by <samp>-fsection-anchors</samp>.
</p>
<p>The hooks below describe the target properties that GCC needs to know
in order to make effective use of section anchors.  It won&rsquo;t use
section anchors at all unless either <code>TARGET_MIN_ANCHOR_OFFSET</code>
or <code>TARGET_MAX_ANCHOR_OFFSET</code> is set to a nonzero value.
</p>
<dl>
<dt><a name="index-TARGET_005fMIN_005fANCHOR_005fOFFSET"></a>Target Hook: <em>HOST_WIDE_INT</em> <strong>TARGET_MIN_ANCHOR_OFFSET</strong></dt>
<dd><p>The minimum offset that should be applied to a section anchor.
On most targets, it should be the smallest offset that can be
applied to a base register while still giving a legitimate address
for every mode.  The default value is 0.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMAX_005fANCHOR_005fOFFSET"></a>Target Hook: <em>HOST_WIDE_INT</em> <strong>TARGET_MAX_ANCHOR_OFFSET</strong></dt>
<dd><p>Like <code>TARGET_MIN_ANCHOR_OFFSET</code>, but the maximum (inclusive)
offset that should be applied to section anchors.  The default
value is 0.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fANCHOR"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_OUTPUT_ANCHOR</strong> <em>(rtx <var>x</var>)</em></dt>
<dd><p>Write the assembly code to define section anchor <var>x</var>, which is a
<code>SYMBOL_REF</code> for which &lsquo;<samp>SYMBOL_REF_ANCHOR_P (<var>x</var>)</samp>&rsquo; is true.
The hook is called with the assembly output position set to the beginning
of <code>SYMBOL_REF_BLOCK (<var>x</var>)</code>.
</p>
<p>If <code>ASM_OUTPUT_DEF</code> is available, the hook&rsquo;s default definition uses
it to define the symbol as &lsquo;<samp>. + SYMBOL_REF_BLOCK_OFFSET (<var>x</var>)</samp>&rsquo;.
If <code>ASM_OUTPUT_DEF</code> is not available, the hook&rsquo;s default definition
is <code>NULL</code>, which disables the use of section anchors altogether.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUSE_005fANCHORS_005fFOR_005fSYMBOL_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_USE_ANCHORS_FOR_SYMBOL_P</strong> <em>(const_rtx <var>x</var>)</em></dt>
<dd><p>Return true if GCC should attempt to use anchors to access <code>SYMBOL_REF</code>
<var>x</var>.  You can assume &lsquo;<samp>SYMBOL_REF_HAS_BLOCK_INFO_P (<var>x</var>)</samp>&rsquo; and
&lsquo;<samp>!SYMBOL_REF_ANCHOR_P (<var>x</var>)</samp>&rsquo;.
</p>
<p>The default version is correct for most targets, but you might need to
intercept this hook to handle things like target-specific attributes
or target-specific sections.
</p></dd></dl>

<hr>
<a name="Condition-Code"></a>
<div class="header">
<p>
Next: <a href="#Costs" accesskey="n" rel="next">Costs</a>, Previous: <a href="#Anchored-Addresses" accesskey="p" rel="prev">Anchored Addresses</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Condition-Code-Status"></a>
<h3 class="section">17.16 Condition Code Status</h3>
<a name="index-condition-code-status"></a>

<p>The macros in this section can be split in two families, according to the
two ways of representing condition codes in GCC.
</p>
<p>The first representation is the so called <code>(cc0)</code> representation
(see <a href="#Jump-Patterns">Jump Patterns</a>), where all instructions can have an implicit
clobber of the condition codes.  The second is the condition code
register representation, which provides better schedulability for
architectures that do have a condition code register, but on which
most instructions do not affect it.  The latter category includes
most RISC machines.
</p>
<p>The implicit clobbering poses a strong restriction on the placement of
the definition and use of the condition code, which need to be in adjacent
insns for machines using <code>(cc0)</code>.  This can prevent important
optimizations on some machines.  For example, on the IBM RS/6000, there
is a delay for taken branches unless the condition code register is set
three instructions earlier than the conditional branch.  The instruction
scheduler cannot perform this optimization if it is not permitted to
separate the definition and use of the condition code register.
</p>
<p>For this reason, it is possible and suggested to use a register to
represent the condition code for new ports.  If there is a specific
condition code register in the machine, use a hard register.  If the
condition code or comparison result can be placed in any general register,
or if there are multiple condition registers, use a pseudo register.
Registers used to store the condition code value will usually have a mode
that is in class <code>MODE_CC</code>.
</p>
<p>Alternatively, you can use <code>BImode</code> if the comparison operator is
specified already in the compare instruction.  In this case, you are not
interested in most macros in this section.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#CC0-Condition-Codes" accesskey="1">CC0 Condition Codes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Old style representation of condition codes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#MODE_005fCC-Condition-Codes" accesskey="2">MODE_CC Condition Codes</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Modern representation of condition codes.
</td></tr>
</table>

<hr>
<a name="CC0-Condition-Codes"></a>
<div class="header">
<p>
Next: <a href="#MODE_005fCC-Condition-Codes" accesskey="n" rel="next">MODE_CC Condition Codes</a>, Up: <a href="#Condition-Code" accesskey="u" rel="up">Condition Code</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Representation-of-condition-codes-using-_0028cc0_0029"></a>
<h4 class="subsection">17.16.1 Representation of condition codes using <code>(cc0)</code></h4>
<a name="index-cc0-1"></a>

<a name="index-cc_005fstatus"></a>
<p>The file <samp>conditions.h</samp> defines a variable <code>cc_status</code> to
describe how the condition code was computed (in case the interpretation of
the condition code depends on the instruction that it was set by).  This
variable contains the RTL expressions on which the condition code is
currently based, and several standard flags.
</p>
<p>Sometimes additional machine-specific flags must be defined in the machine
description header file.  It can also add additional machine-specific
information by defining <code>CC_STATUS_MDEP</code>.
</p>
<dl>
<dt><a name="index-CC_005fSTATUS_005fMDEP"></a>Macro: <strong>CC_STATUS_MDEP</strong></dt>
<dd><p>C code for a data type which is used for declaring the <code>mdep</code>
component of <code>cc_status</code>.  It defaults to <code>int</code>.
</p>
<p>This macro is not used on machines that do not use <code>cc0</code>.
</p></dd></dl>

<dl>
<dt><a name="index-CC_005fSTATUS_005fMDEP_005fINIT"></a>Macro: <strong>CC_STATUS_MDEP_INIT</strong></dt>
<dd><p>A C expression to initialize the <code>mdep</code> field to &ldquo;empty&rdquo;.
The default definition does nothing, since most machines don&rsquo;t use
the field anyway.  If you want to use the field, you should probably
define this macro to initialize it.
</p>
<p>This macro is not used on machines that do not use <code>cc0</code>.
</p></dd></dl>

<dl>
<dt><a name="index-NOTICE_005fUPDATE_005fCC"></a>Macro: <strong>NOTICE_UPDATE_CC</strong> <em>(<var>exp</var>, <var>insn</var>)</em></dt>
<dd><p>A C compound statement to set the components of <code>cc_status</code>
appropriately for an insn <var>insn</var> whose body is <var>exp</var>.  It is
this macro&rsquo;s responsibility to recognize insns that set the condition
code as a byproduct of other activity as well as those that explicitly
set <code>(cc0)</code>.
</p>
<p>This macro is not used on machines that do not use <code>cc0</code>.
</p>
<p>If there are insns that do not set the condition code but do alter
other machine registers, this macro must check to see whether they
invalidate the expressions that the condition code is recorded as
reflecting.  For example, on the 68000, insns that store in address
registers do not set the condition code, which means that usually
<code>NOTICE_UPDATE_CC</code> can leave <code>cc_status</code> unaltered for such
insns.  But suppose that the previous insn set the condition code
based on location &lsquo;<samp>a4@(102)</samp>&rsquo; and the current insn stores a new
value in &lsquo;<samp>a4</samp>&rsquo;.  Although the condition code is not changed by
this, it will no longer be true that it reflects the contents of
&lsquo;<samp>a4@(102)</samp>&rsquo;.  Therefore, <code>NOTICE_UPDATE_CC</code> must alter
<code>cc_status</code> in this case to say that nothing is known about the
condition code value.
</p>
<p>The definition of <code>NOTICE_UPDATE_CC</code> must be prepared to deal
with the results of peephole optimization: insns whose patterns are
<code>parallel</code> RTXs containing various <code>reg</code>, <code>mem</code> or
constants which are just the operands.  The RTL structure of these
insns is not sufficient to indicate what the insns actually do.  What
<code>NOTICE_UPDATE_CC</code> should do when it sees one is just to run
<code>CC_STATUS_INIT</code>.
</p>
<p>A possible definition of <code>NOTICE_UPDATE_CC</code> is to call a function
that looks at an attribute (see <a href="#Insn-Attributes">Insn Attributes</a>) named, for example,
&lsquo;<samp>cc</samp>&rsquo;.  This avoids having detailed information about patterns in
two places, the <samp>md</samp> file and in <code>NOTICE_UPDATE_CC</code>.
</p></dd></dl>

<hr>
<a name="MODE_005fCC-Condition-Codes"></a>
<div class="header">
<p>
Previous: <a href="#CC0-Condition-Codes" accesskey="p" rel="prev">CC0 Condition Codes</a>, Up: <a href="#Condition-Code" accesskey="u" rel="up">Condition Code</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Representation-of-condition-codes-using-registers"></a>
<h4 class="subsection">17.16.2 Representation of condition codes using registers</h4>
<a name="index-CCmode-1"></a>
<a name="index-MODE_005fCC-1"></a>

<dl>
<dt><a name="index-SELECT_005fCC_005fMODE"></a>Macro: <strong>SELECT_CC_MODE</strong> <em>(<var>op</var>, <var>x</var>, <var>y</var>)</em></dt>
<dd><p>On many machines, the condition code may be produced by other instructions
than compares, for example the branch can use directly the condition
code set by a subtract instruction.  However, on some machines
when the condition code is set this way some bits (such as the overflow
bit) are not set in the same way as a test instruction, so that a different
branch instruction must be used for some conditional branches.  When
this happens, use the machine mode of the condition code register to
record different formats of the condition code register.  Modes can
also be used to record which compare instruction (e.g. a signed or an
unsigned comparison) produced the condition codes.
</p>
<p>If other modes than <code>CCmode</code> are required, add them to
<samp><var>machine</var>-modes.def</samp> and define <code>SELECT_CC_MODE</code> to choose
a mode given an operand of a compare.  This is needed because the modes
have to be chosen not only during RTL generation but also, for example,
by instruction combination.  The result of <code>SELECT_CC_MODE</code> should
be consistent with the mode used in the patterns; for example to support
the case of the add on the SPARC discussed above, we have the pattern
</p>
<div class="smallexample">
<pre class="smallexample">(define_insn &quot;&quot;
  [(set (reg:CC_NOOV 0)
        (compare:CC_NOOV
          (plus:SI (match_operand:SI 0 &quot;register_operand&quot; &quot;%r&quot;)
                   (match_operand:SI 1 &quot;arith_operand&quot; &quot;rI&quot;))
          (const_int 0)))]
  &quot;&quot;
  &quot;&hellip;&quot;)
</pre></div>

<p>together with a <code>SELECT_CC_MODE</code> that returns <code>CC_NOOVmode</code>
for comparisons whose argument is a <code>plus</code>:
</p>
<div class="smallexample">
<pre class="smallexample">#define SELECT_CC_MODE(OP,X,Y) \
  (GET_MODE_CLASS (GET_MODE (X)) == MODE_FLOAT           \
   ? ((OP == LT || OP == LE || OP == GT || OP == GE)     \
      ? CCFPEmode : CCFPmode)                            \
   : ((GET_CODE (X) == PLUS || GET_CODE (X) == MINUS     \
       || GET_CODE (X) == NEG || GET_CODE (x) == ASHIFT) \
      ? CC_NOOVmode : CCmode))
</pre></div>

<p>Another reason to use modes is to retain information on which operands
were used by the comparison; see <code>REVERSIBLE_CC_MODE</code> later in
this section.
</p>
<p>You should define this macro if and only if you define extra CC modes
in <samp><var>machine</var>-modes.def</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCANONICALIZE_005fCOMPARISON"></a>Target Hook: <em>void</em> <strong>TARGET_CANONICALIZE_COMPARISON</strong> <em>(int *<var>code</var>, rtx *<var>op0</var>, rtx *<var>op1</var>, bool <var>op0_preserve_value</var>) (<var>code</var>, <var>op0</var>, <var>op1</var>, <var>op0_preserve_value</var>)</em></dt>
<dd><p>On some machines not all possible comparisons are defined, but you can
convert an invalid comparison into a valid one.  For example, the Alpha
does not have a <code>GT</code> comparison, but you can use an <code>LT</code>
comparison instead and swap the order of the operands.
</p>
<p>On such machines, implement this hook to do any required conversions.
<var>code</var> is the initial comparison code and <var>op0</var> and <var>op1</var>
are the left and right operands of the comparison, respectively.  If
<var>op0_preserve_value</var> is <code>true</code> the implementation is not
allowed to change the value of <var>op0</var> since the value might be used
in RTXs which aren&rsquo;t comparisons.  E.g. the implementation is not
allowed to swap operands in that case.
</p>
<p>GCC will not assume that the comparison resulting from this macro is
valid but will see if the resulting insn matches a pattern in the
<samp>md</samp> file.
</p>
<p>You need not to implement this hook if it would never change the
comparison code or operands.
</p></dd></dl>

<dl>
<dt><a name="index-REVERSIBLE_005fCC_005fMODE"></a>Macro: <strong>REVERSIBLE_CC_MODE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression whose value is one if it is always safe to reverse a
comparison whose mode is <var>mode</var>.  If <code>SELECT_CC_MODE</code>
can ever return <var>mode</var> for a floating-point inequality comparison,
then <code>REVERSIBLE_CC_MODE (<var>mode</var>)</code> must be zero.
</p>
<p>You need not define this macro if it would always returns zero or if the
floating-point format is anything other than <code>IEEE_FLOAT_FORMAT</code>.
For example, here is the definition used on the SPARC, where floating-point
inequality comparisons are given either <code>CCFPEmode</code> or <code>CCFPmode</code>:
</p>
<div class="smallexample">
<pre class="smallexample">#define REVERSIBLE_CC_MODE(MODE) \
   ((MODE) != CCFPEmode &amp;&amp; (MODE) != CCFPmode)
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-REVERSE_005fCONDITION"></a>Macro: <strong>REVERSE_CONDITION</strong> <em>(<var>code</var>, <var>mode</var>)</em></dt>
<dd><p>A C expression whose value is reversed condition code of the <var>code</var> for
comparison done in CC_MODE <var>mode</var>.  The macro is used only in case
<code>REVERSIBLE_CC_MODE (<var>mode</var>)</code> is nonzero.  Define this macro in case
machine has some non-standard way how to reverse certain conditionals.  For
instance in case all floating point conditions are non-trapping, compiler may
freely convert unordered compares to ordered ones.  Then definition may look
like:
</p>
<div class="smallexample">
<pre class="smallexample">#define REVERSE_CONDITION(CODE, MODE) \
   ((MODE) != CCFPmode ? reverse_condition (CODE) \
    : reverse_condition_maybe_unordered (CODE))
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-TARGET_005fFIXED_005fCONDITION_005fCODE_005fREGS"></a>Target Hook: <em>bool</em> <strong>TARGET_FIXED_CONDITION_CODE_REGS</strong> <em>(unsigned int *<var>p1</var>, unsigned int *<var>p2</var>)</em></dt>
<dd><p>On targets which do not use <code>(cc0)</code>, and which use a hard
register rather than a pseudo-register to hold condition codes, the
regular CSE passes are often not able to identify cases in which the
hard register is set to a common value.  Use this hook to enable a
small pass which optimizes such cases.  This hook should return true
to enable this pass, and it should set the integers to which its
arguments point to the hard register numbers used for condition codes.
When there is only one such register, as is true on most systems, the
integer pointed to by <var>p2</var> should be set to
<code>INVALID_REGNUM</code>.
</p>
<p>The default version of this hook returns false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCC_005fMODES_005fCOMPATIBLE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_CC_MODES_COMPATIBLE</strong> <em>(enum machine_mode <var>m1</var>, enum machine_mode <var>m2</var>)</em></dt>
<dd><p>On targets which use multiple condition code modes in class
<code>MODE_CC</code>, it is sometimes the case that a comparison can be
validly done in more than one mode.  On such a system, define this
target hook to take two mode arguments and to return a mode in which
both comparisons may be validly done.  If there is no such mode,
return <code>VOIDmode</code>.
</p>
<p>The default version of this hook checks whether the modes are the
same.  If they are, it returns that mode.  If they are different, it
returns <code>VOIDmode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFLAGS_005fREGNUM"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_FLAGS_REGNUM</strong></dt>
<dd><p>If the target has a dedicated flags register, and it needs to use the post-reload comparison elimination pass, then this value should be set appropriately.
</p></dd></dl>

<hr>
<a name="Costs"></a>
<div class="header">
<p>
Next: <a href="#Scheduling" accesskey="n" rel="next">Scheduling</a>, Previous: <a href="#Condition-Code" accesskey="p" rel="prev">Condition Code</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Describing-Relative-Costs-of-Operations"></a>
<h3 class="section">17.17 Describing Relative Costs of Operations</h3>
<a name="index-costs-of-instructions"></a>
<a name="index-relative-costs"></a>
<a name="index-speed-of-instructions"></a>

<p>These macros let you describe the relative speed of various operations
on the target machine.
</p>
<dl>
<dt><a name="index-REGISTER_005fMOVE_005fCOST"></a>Macro: <strong>REGISTER_MOVE_COST</strong> <em>(<var>mode</var>, <var>from</var>, <var>to</var>)</em></dt>
<dd><p>A C expression for the cost of moving data of mode <var>mode</var> from a
register in class <var>from</var> to one in class <var>to</var>.  The classes are
expressed using the enumeration values such as <code>GENERAL_REGS</code>.  A
value of 2 is the default; other values are interpreted relative to
that.
</p>
<p>It is not required that the cost always equal 2 when <var>from</var> is the
same as <var>to</var>; on some machines it is expensive to move between
registers if they are not general registers.
</p>
<p>If reload sees an insn consisting of a single <code>set</code> between two
hard registers, and if <code>REGISTER_MOVE_COST</code> applied to their
classes returns a value of 2, reload does not check to ensure that the
constraints of the insn are met.  Setting a cost of other than 2 will
allow reload to verify that the constraints are met.  You should do this
if the &lsquo;<samp>mov<var>m</var></samp>&rsquo; pattern&rsquo;s constraints do not allow such copying.
</p>
<p>These macros are obsolete, new ports should use the target hook
<code>TARGET_REGISTER_MOVE_COST</code> instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fREGISTER_005fMOVE_005fCOST"></a>Target Hook: <em>int</em> <strong>TARGET_REGISTER_MOVE_COST</strong> <em>(enum machine_mode <var>mode</var>, reg_class_t <var>from</var>, reg_class_t <var>to</var>)</em></dt>
<dd><p>This target hook should return the cost of moving data of mode <var>mode</var>
from a register in class <var>from</var> to one in class <var>to</var>.  The classes
are expressed using the enumeration values such as <code>GENERAL_REGS</code>.
A value of 2 is the default; other values are interpreted relative to
that.
</p>
<p>It is not required that the cost always equal 2 when <var>from</var> is the
same as <var>to</var>; on some machines it is expensive to move between
registers if they are not general registers.
</p>
<p>If reload sees an insn consisting of a single <code>set</code> between two
hard registers, and if <code>TARGET_REGISTER_MOVE_COST</code> applied to their
classes returns a value of 2, reload does not check to ensure that the
constraints of the insn are met.  Setting a cost of other than 2 will
allow reload to verify that the constraints are met.  You should do this
if the &lsquo;<samp>mov<var>m</var></samp>&rsquo; pattern&rsquo;s constraints do not allow such copying.
</p>
<p>The default version of this function returns 2.
</p></dd></dl>

<dl>
<dt><a name="index-MEMORY_005fMOVE_005fCOST"></a>Macro: <strong>MEMORY_MOVE_COST</strong> <em>(<var>mode</var>, <var>class</var>, <var>in</var>)</em></dt>
<dd><p>A C expression for the cost of moving data of mode <var>mode</var> between a
register of class <var>class</var> and memory; <var>in</var> is zero if the value
is to be written to memory, nonzero if it is to be read in.  This cost
is relative to those in <code>REGISTER_MOVE_COST</code>.  If moving between
registers and memory is more expensive than between two registers, you
should define this macro to express the relative cost.
</p>
<p>If you do not define this macro, GCC uses a default cost of 4 plus
the cost of copying via a secondary reload register, if one is
needed.  If your machine requires a secondary reload register to copy
between memory and a register of <var>class</var> but the reload mechanism is
more complex than copying via an intermediate, define this macro to
reflect the actual cost of the move.
</p>
<p>GCC defines the function <code>memory_move_secondary_cost</code> if
secondary reloads are needed.  It computes the costs due to copying via
a secondary register.  If your machine copies from memory using a
secondary register in the conventional way but the default base value of
4 is not correct for your machine, define this macro to add some other
value to the result of that function.  The arguments to that function
are the same as to this macro.
</p>
<p>These macros are obsolete, new ports should use the target hook
<code>TARGET_MEMORY_MOVE_COST</code> instead.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMEMORY_005fMOVE_005fCOST"></a>Target Hook: <em>int</em> <strong>TARGET_MEMORY_MOVE_COST</strong> <em>(enum machine_mode <var>mode</var>, reg_class_t <var>rclass</var>, bool <var>in</var>)</em></dt>
<dd><p>This target hook should return the cost of moving data of mode <var>mode</var>
between a register of class <var>rclass</var> and memory; <var>in</var> is <code>false</code>
if the value is to be written to memory, <code>true</code> if it is to be read in.
This cost is relative to those in <code>TARGET_REGISTER_MOVE_COST</code>.
If moving between registers and memory is more expensive than between two
registers, you should add this target hook to express the relative cost.
</p>
<p>If you do not add this target hook, GCC uses a default cost of 4 plus
the cost of copying via a secondary reload register, if one is
needed.  If your machine requires a secondary reload register to copy
between memory and a register of <var>rclass</var> but the reload mechanism is
more complex than copying via an intermediate, use this target hook to
reflect the actual cost of the move.
</p>
<p>GCC defines the function <code>memory_move_secondary_cost</code> if
secondary reloads are needed.  It computes the costs due to copying via
a secondary register.  If your machine copies from memory using a
secondary register in the conventional way but the default base value of
4 is not correct for your machine, use this target hook to add some other
value to the result of that function.  The arguments to that function
are the same as to this target hook.
</p></dd></dl>

<dl>
<dt><a name="index-BRANCH_005fCOST"></a>Macro: <strong>BRANCH_COST</strong> <em>(<var>speed_p</var>, <var>predictable_p</var>)</em></dt>
<dd><p>A C expression for the cost of a branch instruction.  A value of 1 is
the default; other values are interpreted relative to that. Parameter
<var>speed_p</var> is true when the branch in question should be optimized
for speed.  When it is false, <code>BRANCH_COST</code> should return a value
optimal for code size rather than performance.  <var>predictable_p</var> is
true for well-predicted branches. On many architectures the
<code>BRANCH_COST</code> can be reduced then.
</p></dd></dl>

<p>Here are additional macros which do not specify precise relative costs,
but only that certain actions are more expensive than GCC would
ordinarily expect.
</p>
<dl>
<dt><a name="index-SLOW_005fBYTE_005fACCESS"></a>Macro: <strong>SLOW_BYTE_ACCESS</strong></dt>
<dd><p>Define this macro as a C expression which is nonzero if accessing less
than a word of memory (i.e. a <code>char</code> or a <code>short</code>) is no
faster than accessing a word of memory, i.e., if such access
require more than one instruction or if there is no difference in cost
between byte and (aligned) word loads.
</p>
<p>When this macro is not defined, the compiler will access a field by
finding the smallest containing object; when it is defined, a fullword
load will be used if alignment permits.  Unless bytes accesses are
faster than word accesses, using word accesses is preferable since it
may eliminate subsequent memory access if subsequent accesses occur to
other fields in the same word of the structure, but to different bytes.
</p></dd></dl>

<dl>
<dt><a name="index-SLOW_005fUNALIGNED_005fACCESS"></a>Macro: <strong>SLOW_UNALIGNED_ACCESS</strong> <em>(<var>mode</var>, <var>alignment</var>)</em></dt>
<dd><p>Define this macro to be the value 1 if memory accesses described by the
<var>mode</var> and <var>alignment</var> parameters have a cost many times greater
than aligned accesses, for example if they are emulated in a trap
handler.
</p>
<p>When this macro is nonzero, the compiler will act as if
<code>STRICT_ALIGNMENT</code> were nonzero when generating code for block
moves.  This can cause significantly more instructions to be produced.
Therefore, do not set this macro nonzero if unaligned accesses only add a
cycle or two to the time for a memory access.
</p>
<p>If the value of this macro is always zero, it need not be defined.  If
this macro is defined, it should produce a nonzero value when
<code>STRICT_ALIGNMENT</code> is nonzero.
</p></dd></dl>

<dl>
<dt><a name="index-MOVE_005fRATIO"></a>Macro: <strong>MOVE_RATIO</strong> <em>(<var>speed</var>)</em></dt>
<dd><p>The threshold of number of scalar memory-to-memory move insns, <em>below</em>
which a sequence of insns should be generated instead of a
string move insn or a library call.  Increasing the value will always
make code faster, but eventually incurs high cost in increased code size.
</p>
<p>Note that on machines where the corresponding move insn is a
<code>define_expand</code> that emits a sequence of insns, this macro counts
the number of such sequences.
</p>
<p>The parameter <var>speed</var> is true if the code is currently being
optimized for speed rather than size.
</p>
<p>If you don&rsquo;t define this, a reasonable default is used.
</p></dd></dl>

<dl>
<dt><a name="index-MOVE_005fBY_005fPIECES_005fP"></a>Macro: <strong>MOVE_BY_PIECES_P</strong> <em>(<var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>A C expression used to determine whether <code>move_by_pieces</code> will be used to
copy a chunk of memory, or whether some other block move mechanism
will be used.  Defaults to 1 if <code>move_by_pieces_ninsns</code> returns less
than <code>MOVE_RATIO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-MOVE_005fMAX_005fPIECES"></a>Macro: <strong>MOVE_MAX_PIECES</strong></dt>
<dd><p>A C expression used by <code>move_by_pieces</code> to determine the largest unit
a load or store used to copy memory is.  Defaults to <code>MOVE_MAX</code>.
</p></dd></dl>

<dl>
<dt><a name="index-CLEAR_005fRATIO"></a>Macro: <strong>CLEAR_RATIO</strong> <em>(<var>speed</var>)</em></dt>
<dd><p>The threshold of number of scalar move insns, <em>below</em> which a sequence
of insns should be generated to clear memory instead of a string clear insn
or a library call.  Increasing the value will always make code faster, but
eventually incurs high cost in increased code size.
</p>
<p>The parameter <var>speed</var> is true if the code is currently being
optimized for speed rather than size.
</p>
<p>If you don&rsquo;t define this, a reasonable default is used.
</p></dd></dl>

<dl>
<dt><a name="index-CLEAR_005fBY_005fPIECES_005fP"></a>Macro: <strong>CLEAR_BY_PIECES_P</strong> <em>(<var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>A C expression used to determine whether <code>clear_by_pieces</code> will be used
to clear a chunk of memory, or whether some other block clear mechanism
will be used.  Defaults to 1 if <code>move_by_pieces_ninsns</code> returns less
than <code>CLEAR_RATIO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SET_005fRATIO"></a>Macro: <strong>SET_RATIO</strong> <em>(<var>speed</var>)</em></dt>
<dd><p>The threshold of number of scalar move insns, <em>below</em> which a sequence
of insns should be generated to set memory to a constant value, instead of
a block set insn or a library call.
Increasing the value will always make code faster, but
eventually incurs high cost in increased code size.
</p>
<p>The parameter <var>speed</var> is true if the code is currently being
optimized for speed rather than size.
</p>
<p>If you don&rsquo;t define this, it defaults to the value of <code>MOVE_RATIO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SET_005fBY_005fPIECES_005fP"></a>Macro: <strong>SET_BY_PIECES_P</strong> <em>(<var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>A C expression used to determine whether <code>store_by_pieces</code> will be
used to set a chunk of memory to a constant value, or whether some
other mechanism will be used.  Used by <code>__builtin_memset</code> when
storing values other than constant zero.
Defaults to 1 if <code>move_by_pieces_ninsns</code> returns less
than <code>SET_RATIO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STORE_005fBY_005fPIECES_005fP"></a>Macro: <strong>STORE_BY_PIECES_P</strong> <em>(<var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>A C expression used to determine whether <code>store_by_pieces</code> will be
used to set a chunk of memory to a constant string value, or whether some
other mechanism will be used.  Used by <code>__builtin_strcpy</code> when
called with a constant source string.
Defaults to 1 if <code>move_by_pieces_ninsns</code> returns less
than <code>MOVE_RATIO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fLOAD_005fPOST_005fINCREMENT"></a>Macro: <strong>USE_LOAD_POST_INCREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a load postincrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_POST_INCREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fLOAD_005fPOST_005fDECREMENT"></a>Macro: <strong>USE_LOAD_POST_DECREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a load postdecrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_POST_DECREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fLOAD_005fPRE_005fINCREMENT"></a>Macro: <strong>USE_LOAD_PRE_INCREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a load preincrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_PRE_INCREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fLOAD_005fPRE_005fDECREMENT"></a>Macro: <strong>USE_LOAD_PRE_DECREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a load predecrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_PRE_DECREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fSTORE_005fPOST_005fINCREMENT"></a>Macro: <strong>USE_STORE_POST_INCREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a store postincrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_POST_INCREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fSTORE_005fPOST_005fDECREMENT"></a>Macro: <strong>USE_STORE_POST_DECREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression used to determine whether a store postdecrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_POST_DECREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fSTORE_005fPRE_005fINCREMENT"></a>Macro: <strong>USE_STORE_PRE_INCREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>This macro is used to determine whether a store preincrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_PRE_INCREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fSTORE_005fPRE_005fDECREMENT"></a>Macro: <strong>USE_STORE_PRE_DECREMENT</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>This macro is used to determine whether a store predecrement is a good
thing to use for a given mode.  Defaults to the value of
<code>HAVE_PRE_DECREMENT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fFUNCTION_005fCSE"></a>Macro: <strong>NO_FUNCTION_CSE</strong></dt>
<dd><p>Define this macro if it is as good or better to call a constant
function address than to call an address kept in a register.
</p></dd></dl>

<dl>
<dt><a name="index-LOGICAL_005fOP_005fNON_005fSHORT_005fCIRCUIT"></a>Macro: <strong>LOGICAL_OP_NON_SHORT_CIRCUIT</strong></dt>
<dd><p>Define this macro if a non-short-circuit operation produced by
&lsquo;<samp>fold_range_test ()</samp>&rsquo; is optimal.  This macro defaults to true if
<code>BRANCH_COST</code> is greater than or equal to the value 2.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fRTX_005fCOSTS"></a>Target Hook: <em>bool</em> <strong>TARGET_RTX_COSTS</strong> <em>(rtx <var>x</var>, int <var>code</var>, int <var>outer_code</var>, int <var>opno</var>, int *<var>total</var>, bool <var>speed</var>)</em></dt>
<dd><p>This target hook describes the relative costs of RTL expressions.
</p>
<p>The cost may depend on the precise form of the expression, which is
available for examination in <var>x</var>, and the fact that <var>x</var> appears
as operand <var>opno</var> of an expression with rtx code <var>outer_code</var>.
That is, the hook can assume that there is some rtx <var>y</var> such
that &lsquo;<samp>GET_CODE (<var>y</var>) == <var>outer_code</var></samp>&rsquo; and such that
either (a) &lsquo;<samp>XEXP (<var>y</var>, <var>opno</var>) == <var>x</var></samp>&rsquo; or
(b) &lsquo;<samp>XVEC (<var>y</var>, <var>opno</var>)</samp>&rsquo; contains <var>x</var>.
</p>
<p><var>code</var> is <var>x</var>&rsquo;s expression code&mdash;redundant, since it can be
obtained with <code>GET_CODE (<var>x</var>)</code>.
</p>
<p>In implementing this hook, you can use the construct
<code>COSTS_N_INSNS (<var>n</var>)</code> to specify a cost equal to <var>n</var> fast
instructions.
</p>
<p>On entry to the hook, <code>*<var>total</var></code> contains a default estimate
for the cost of the expression.  The hook should modify this value as
necessary.  Traditionally, the default costs are <code>COSTS_N_INSNS (5)</code>
for multiplications, <code>COSTS_N_INSNS (7)</code> for division and modulus
operations, and <code>COSTS_N_INSNS (1)</code> for all other operations.
</p>
<p>When optimizing for code size, i.e. when <code>speed</code> is
false, this target hook should be used to estimate the relative
size cost of an expression, again relative to <code>COSTS_N_INSNS</code>.
</p>
<p>The hook returns true when all subexpressions of <var>x</var> have been
processed, and false when <code>rtx_cost</code> should recurse.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDRESS_005fCOST"></a>Target Hook: <em>int</em> <strong>TARGET_ADDRESS_COST</strong> <em>(rtx <var>address</var>, enum machine_mode <var>mode</var>, addr_space_t <var>as</var>, bool <var>speed</var>)</em></dt>
<dd><p>This hook computes the cost of an addressing mode that contains
<var>address</var>.  If not defined, the cost is computed from
the <var>address</var> expression and the <code>TARGET_RTX_COST</code> hook.
</p>
<p>For most CISC machines, the default cost is a good approximation of the
true cost of the addressing mode.  However, on RISC machines, all
instructions normally have the same length and execution time.  Hence
all addresses will have equal costs.
</p>
<p>In cases where more than one form of an address is known, the form with
the lowest cost will be used.  If multiple forms have the same, lowest,
cost, the one that is the most complex will be used.
</p>
<p>For example, suppose an address that is equal to the sum of a register
and a constant is used twice in the same basic block.  When this macro
is not defined, the address will be computed in a register and memory
references will be indirect through that register.  On machines where
the cost of the addressing mode containing the sum is no higher than
that of a simple indirect reference, this will produce an additional
instruction and possibly require an additional register.  Proper
specification of this macro eliminates this overhead for such machines.
</p>
<p>This hook is never called with an invalid address.
</p>
<p>On machines where an address involving more than one register is as
cheap as an address computation involving only one register, defining
<code>TARGET_ADDRESS_COST</code> to reflect this can cause two registers to
be live over a region of code where only one would have been if
<code>TARGET_ADDRESS_COST</code> were not defined in that manner.  This effect
should be considered in the definition of this macro.  Equivalent costs
should probably only be given to addresses with different numbers of
registers on machines with lots of registers.
</p></dd></dl>

<hr>
<a name="Scheduling"></a>
<div class="header">
<p>
Next: <a href="#Sections" accesskey="n" rel="next">Sections</a>, Previous: <a href="#Costs" accesskey="p" rel="prev">Costs</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Adjusting-the-Instruction-Scheduler"></a>
<h3 class="section">17.18 Adjusting the Instruction Scheduler</h3>

<p>The instruction scheduler may need a fair amount of machine-specific
adjustment in order to produce good code.  GCC provides several target
hooks for this purpose.  It is usually enough to define just a few of
them: try the first ones in this list first.
</p>
<dl>
<dt><a name="index-TARGET_005fSCHED_005fISSUE_005fRATE"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_ISSUE_RATE</strong> <em>(void)</em></dt>
<dd><p>This hook returns the maximum number of instructions that can ever
issue at the same time on the target machine.  The default is one.
Although the insn scheduler can define itself the possibility of issue
an insn on the same cycle, the value can serve as an additional
constraint to issue insns on the same simulated processor cycle (see
hooks &lsquo;<samp>TARGET_SCHED_REORDER</samp>&rsquo; and &lsquo;<samp>TARGET_SCHED_REORDER2</samp>&rsquo;).
This value must be constant over the entire compilation.  If you need
it to vary depending on what the instructions are, you must use
&lsquo;<samp>TARGET_SCHED_VARIABLE_ISSUE</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fVARIABLE_005fISSUE"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_VARIABLE_ISSUE</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>, rtx <var>insn</var>, int <var>more</var>)</em></dt>
<dd><p>This hook is executed by the scheduler after it has scheduled an insn
from the ready list.  It should return the number of insns which can
still be issued in the current cycle.  The default is
&lsquo;<samp><var>more</var>&nbsp;<span class="nolinebreak">-</span>&nbsp;1<!-- /@w --></samp>&rsquo; for insns other than <code>CLOBBER</code> and
<code>USE</code>, which normally are not counted against the issue rate.
You should define this hook if some insns take more machine resources
than others, so that fewer insns can follow them in the same cycle.
<var>file</var> is either a null pointer, or a stdio stream to write any
debug output to.  <var>verbose</var> is the verbose level provided by
<samp>-fsched-verbose-<var>n</var></samp>.  <var>insn</var> is the instruction that
was scheduled.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fADJUST_005fCOST"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_ADJUST_COST</strong> <em>(rtx <var>insn</var>, rtx <var>link</var>, rtx <var>dep_insn</var>, int <var>cost</var>)</em></dt>
<dd><p>This function corrects the value of <var>cost</var> based on the
relationship between <var>insn</var> and <var>dep_insn</var> through the
dependence <var>link</var>.  It should return the new value.  The default
is to make no adjustment to <var>cost</var>.  This can be used for example
to specify to the scheduler using the traditional pipeline description
that an output- or anti-dependence does not incur the same cost as a
data-dependence.  If the scheduler using the automaton based pipeline
description, the cost of anti-dependence is zero and the cost of
output-dependence is maximum of one and the difference of latency
times of the first and the second insns.  If these values are not
acceptable, you could use the hook to modify them too.  See also
see <a href="#Processor-pipeline-description">Processor pipeline description</a>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fADJUST_005fPRIORITY"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_ADJUST_PRIORITY</strong> <em>(rtx <var>insn</var>, int <var>priority</var>)</em></dt>
<dd><p>This hook adjusts the integer scheduling priority <var>priority</var> of
<var>insn</var>.  It should return the new priority.  Increase the priority to
execute <var>insn</var> earlier, reduce the priority to execute <var>insn</var>
later.  Do not define this hook if you do not need to adjust the
scheduling priorities of insns.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fREORDER"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_REORDER</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>, rtx *<var>ready</var>, int *<var>n_readyp</var>, int <var>clock</var>)</em></dt>
<dd><p>This hook is executed by the scheduler after it has scheduled the ready
list, to allow the machine description to reorder it (for example to
combine two small instructions together on &lsquo;<samp>VLIW</samp>&rsquo; machines).
<var>file</var> is either a null pointer, or a stdio stream to write any
debug output to.  <var>verbose</var> is the verbose level provided by
<samp>-fsched-verbose-<var>n</var></samp>.  <var>ready</var> is a pointer to the ready
list of instructions that are ready to be scheduled.  <var>n_readyp</var> is
a pointer to the number of elements in the ready list.  The scheduler
reads the ready list in reverse order, starting with
<var>ready</var>[<var>*n_readyp</var> - 1] and going to <var>ready</var>[0].  <var>clock</var>
is the timer tick of the scheduler.  You may modify the ready list and
the number of ready insns.  The return value is the number of insns that
can issue this cycle; normally this is just <code>issue_rate</code>.  See also
&lsquo;<samp>TARGET_SCHED_REORDER2</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fREORDER2"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_REORDER2</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>, rtx *<var>ready</var>, int *<var>n_readyp</var>, int <var>clock</var>)</em></dt>
<dd><p>Like &lsquo;<samp>TARGET_SCHED_REORDER</samp>&rsquo;, but called at a different time.  That
function is called whenever the scheduler starts a new cycle.  This one
is called once per iteration over a cycle, immediately after
&lsquo;<samp>TARGET_SCHED_VARIABLE_ISSUE</samp>&rsquo;; it can reorder the ready list and
return the number of insns to be scheduled in the same cycle.  Defining
this hook can be useful if there are frequent situations where
scheduling one insn causes other insns to become ready in the same
cycle.  These other insns can then be taken into account properly.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDEPENDENCIES_005fEVALUATION_005fHOOK"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_DEPENDENCIES_EVALUATION_HOOK</strong> <em>(rtx <var>head</var>, rtx <var>tail</var>)</em></dt>
<dd><p>This hook is called after evaluation forward dependencies of insns in
chain given by two parameter values (<var>head</var> and <var>tail</var>
correspondingly) but before insns scheduling of the insn chain.  For
example, it can be used for better insn classification if it requires
analysis of dependencies.  This hook can use backward and forward
dependencies of the insn scheduler because they are already
calculated.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fINIT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_INIT</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>, int <var>max_ready</var>)</em></dt>
<dd><p>This hook is executed by the scheduler at the beginning of each block of
instructions that are to be scheduled.  <var>file</var> is either a null
pointer, or a stdio stream to write any debug output to.  <var>verbose</var>
is the verbose level provided by <samp>-fsched-verbose-<var>n</var></samp>.
<var>max_ready</var> is the maximum number of insns in the current scheduling
region that can be live at the same time.  This can be used to allocate
scratch space if it is needed, e.g. by &lsquo;<samp>TARGET_SCHED_REORDER</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFINISH"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FINISH</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>)</em></dt>
<dd><p>This hook is executed by the scheduler at the end of each block of
instructions that are to be scheduled.  It can be used to perform
cleanup of any actions done by the other scheduling hooks.  <var>file</var>
is either a null pointer, or a stdio stream to write any debug output
to.  <var>verbose</var> is the verbose level provided by
<samp>-fsched-verbose-<var>n</var></samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fINIT_005fGLOBAL"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_INIT_GLOBAL</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>, int <var>old_max_uid</var>)</em></dt>
<dd><p>This hook is executed by the scheduler after function level initializations.
<var>file</var> is either a null pointer, or a stdio stream to write any debug output to.
<var>verbose</var> is the verbose level provided by <samp>-fsched-verbose-<var>n</var></samp>.
<var>old_max_uid</var> is the maximum insn uid when scheduling begins.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFINISH_005fGLOBAL"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FINISH_GLOBAL</strong> <em>(FILE *<var>file</var>, int <var>verbose</var>)</em></dt>
<dd><p>This is the cleanup hook corresponding to <code>TARGET_SCHED_INIT_GLOBAL</code>.
<var>file</var> is either a null pointer, or a stdio stream to write any debug output to.
<var>verbose</var> is the verbose level provided by <samp>-fsched-verbose-<var>n</var></samp>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDFA_005fPRE_005fCYCLE_005fINSN"></a>Target Hook: <em>rtx</em> <strong>TARGET_SCHED_DFA_PRE_CYCLE_INSN</strong> <em>(void)</em></dt>
<dd><p>The hook returns an RTL insn.  The automaton state used in the
pipeline hazard recognizer is changed as if the insn were scheduled
when the new simulated processor cycle starts.  Usage of the hook may
simplify the automaton pipeline description for some <acronym>VLIW</acronym>
processors.  If the hook is defined, it is used only for the automaton
based pipeline description.  The default is not to change the state
when the new simulated processor cycle starts.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fINIT_005fDFA_005fPRE_005fCYCLE_005fINSN"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_INIT_DFA_PRE_CYCLE_INSN</strong> <em>(void)</em></dt>
<dd><p>The hook can be used to initialize data used by the previous hook.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDFA_005fPOST_005fCYCLE_005fINSN"></a>Target Hook: <em>rtx</em> <strong>TARGET_SCHED_DFA_POST_CYCLE_INSN</strong> <em>(void)</em></dt>
<dd><p>The hook is analogous to &lsquo;<samp>TARGET_SCHED_DFA_PRE_CYCLE_INSN</samp>&rsquo; but used
to changed the state as if the insn were scheduled when the new
simulated processor cycle finishes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fINIT_005fDFA_005fPOST_005fCYCLE_005fINSN"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_INIT_DFA_POST_CYCLE_INSN</strong> <em>(void)</em></dt>
<dd><p>The hook is analogous to &lsquo;<samp>TARGET_SCHED_INIT_DFA_PRE_CYCLE_INSN</samp>&rsquo; but
used to initialize data used by the previous hook.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDFA_005fPRE_005fADVANCE_005fCYCLE"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_DFA_PRE_ADVANCE_CYCLE</strong> <em>(void)</em></dt>
<dd><p>The hook to notify target that the current simulated cycle is about to finish.
The hook is analogous to &lsquo;<samp>TARGET_SCHED_DFA_PRE_CYCLE_INSN</samp>&rsquo; but used
to change the state in more complicated situations - e.g., when advancing
state on a single insn is not enough.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDFA_005fPOST_005fADVANCE_005fCYCLE"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_DFA_POST_ADVANCE_CYCLE</strong> <em>(void)</em></dt>
<dd><p>The hook to notify target that new simulated cycle has just started.
The hook is analogous to &lsquo;<samp>TARGET_SCHED_DFA_POST_CYCLE_INSN</samp>&rsquo; but used
to change the state in more complicated situations - e.g., when advancing
state on a single insn is not enough.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD</strong> <em>(void)</em></dt>
<dd><p>This hook controls better choosing an insn from the ready insn queue
for the <acronym>DFA</acronym>-based insn scheduler.  Usually the scheduler
chooses the first insn from the queue.  If the hook returns a positive
value, an additional scheduler code tries all permutations of
&lsquo;<samp>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD ()</samp>&rsquo;
subsequent ready insns to choose an insn whose issue will result in
maximal number of issued insns on the same cycle.  For the
<acronym>VLIW</acronym> processor, the code could actually solve the problem of
packing simple insns into the <acronym>VLIW</acronym> insn.  Of course, if the
rules of <acronym>VLIW</acronym> packing are described in the automaton.
</p>
<p>This code also could be used for superscalar <acronym>RISC</acronym>
processors.  Let us consider a superscalar <acronym>RISC</acronym> processor
with 3 pipelines.  Some insns can be executed in pipelines <var>A</var> or
<var>B</var>, some insns can be executed only in pipelines <var>B</var> or
<var>C</var>, and one insn can be executed in pipeline <var>B</var>.  The
processor may issue the 1st insn into <var>A</var> and the 2nd one into
<var>B</var>.  In this case, the 3rd insn will wait for freeing <var>B</var>
until the next cycle.  If the scheduler issues the 3rd insn the first,
the processor could issue all 3 insns per cycle.
</p>
<p>Actually this code demonstrates advantages of the automaton based
pipeline hazard recognizer.  We try quickly and easy many insn
schedules to choose the best one.
</p>
<p>The default is no multipass scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD_005fGUARD"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD_GUARD</strong> <em>(rtx <var>insn</var>)</em></dt>
<dd>
<p>This hook controls what insns from the ready insn queue will be
considered for the multipass insn scheduling.  If the hook returns
zero for <var>insn</var>, the insn will be not chosen to
be issued.
</p>
<p>The default is that any ready insns can be chosen to be issued.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fBEGIN"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_BEGIN</strong> <em>(void *<var>data</var>, char *<var>ready_try</var>, int <var>n_ready</var>, bool <var>first_cycle_insn_p</var>)</em></dt>
<dd><p>This hook prepares the target backend for a new round of multipass
scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fISSUE"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_ISSUE</strong> <em>(void *<var>data</var>, char *<var>ready_try</var>, int <var>n_ready</var>, rtx <var>insn</var>, const void *<var>prev_data</var>)</em></dt>
<dd><p>This hook is called when multipass scheduling evaluates instruction INSN.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fBACKTRACK"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_BACKTRACK</strong> <em>(const void *<var>data</var>, char *<var>ready_try</var>, int <var>n_ready</var>)</em></dt>
<dd><p>This is called when multipass scheduling backtracks from evaluation of
an instruction.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fEND"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_END</strong> <em>(const void *<var>data</var>)</em></dt>
<dd><p>This hook notifies the target about the result of the concluded current
round of multipass scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fINIT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_INIT</strong> <em>(void *<var>data</var>)</em></dt>
<dd><p>This hook initializes target-specific data used in multipass scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fFINI"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_FINI</strong> <em>(void *<var>data</var>)</em></dt>
<dd><p>This hook finalizes target-specific data used in multipass scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDFA_005fNEW_005fCYCLE"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_DFA_NEW_CYCLE</strong> <em>(FILE *<var>dump</var>, int <var>verbose</var>, rtx <var>insn</var>, int <var>last_clock</var>, int <var>clock</var>, int *<var>sort_p</var>)</em></dt>
<dd><p>This hook is called by the insn scheduler before issuing <var>insn</var>
on cycle <var>clock</var>.  If the hook returns nonzero,
<var>insn</var> is not issued on this processor cycle.  Instead,
the processor cycle is advanced.  If *<var>sort_p</var>
is zero, the insn ready queue is not sorted on the new cycle
start as usually.  <var>dump</var> and <var>verbose</var> specify the file and
verbosity level to use for debugging output.
<var>last_clock</var> and <var>clock</var> are, respectively, the
processor cycle on which the previous insn has been issued,
and the current processor cycle.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fIS_005fCOSTLY_005fDEPENDENCE"></a>Target Hook: <em>bool</em> <strong>TARGET_SCHED_IS_COSTLY_DEPENDENCE</strong> <em>(struct _dep *<var>_dep</var>, int <var>cost</var>, int <var>distance</var>)</em></dt>
<dd><p>This hook is used to define which dependences are considered costly by
the target, so costly that it is not advisable to schedule the insns that
are involved in the dependence too close to one another.  The parameters
to this hook are as follows:  The first parameter <var>_dep</var> is the dependence
being evaluated.  The second parameter <var>cost</var> is the cost of the
dependence as estimated by the scheduler, and the third
parameter <var>distance</var> is the distance in cycles between the two insns.
The hook returns <code>true</code> if considering the distance between the two
insns the dependence between them is considered costly by the target,
and <code>false</code> otherwise.
</p>
<p>Defining this hook can be useful in multiple-issue out-of-order machines,
where (a) it&rsquo;s practically hopeless to predict the actual data/resource
delays, however: (b) there&rsquo;s a better chance to predict the actual grouping
that will be formed, and (c) correctly emulating the grouping can be very
important.  In such targets one may want to allow issuing dependent insns
closer to one another&mdash;i.e., closer than the dependence distance;  however,
not in cases of &ldquo;costly dependences&rdquo;, which this hooks allows to define.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fH_005fI_005fD_005fEXTENDED"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_H_I_D_EXTENDED</strong> <em>(void)</em></dt>
<dd><p>This hook is called by the insn scheduler after emitting a new instruction to
the instruction stream.  The hook notifies a target backend to extend its
per instruction data structures.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fALLOC_005fSCHED_005fCONTEXT"></a>Target Hook: <em>void *</em> <strong>TARGET_SCHED_ALLOC_SCHED_CONTEXT</strong> <em>(void)</em></dt>
<dd><p>Return a pointer to a store large enough to hold target scheduling context.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fINIT_005fSCHED_005fCONTEXT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_INIT_SCHED_CONTEXT</strong> <em>(void *<var>tc</var>, bool <var>clean_p</var>)</em></dt>
<dd><p>Initialize store pointed to by <var>tc</var> to hold target scheduling context.
It <var>clean_p</var> is true then initialize <var>tc</var> as if scheduler is at the
beginning of the block.  Otherwise, copy the current context into <var>tc</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fSET_005fSCHED_005fCONTEXT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_SET_SCHED_CONTEXT</strong> <em>(void *<var>tc</var>)</em></dt>
<dd><p>Copy target scheduling context pointed to by <var>tc</var> to the current context.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fCLEAR_005fSCHED_005fCONTEXT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_CLEAR_SCHED_CONTEXT</strong> <em>(void *<var>tc</var>)</em></dt>
<dd><p>Deallocate internal data in target scheduling context pointed to by <var>tc</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFREE_005fSCHED_005fCONTEXT"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_FREE_SCHED_CONTEXT</strong> <em>(void *<var>tc</var>)</em></dt>
<dd><p>Deallocate a store for target scheduling context pointed to by <var>tc</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fSPECULATE_005fINSN"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_SPECULATE_INSN</strong> <em>(rtx <var>insn</var>, int <var>request</var>, rtx *<var>new_pat</var>)</em></dt>
<dd><p>This hook is called by the insn scheduler when <var>insn</var> has only
speculative dependencies and therefore can be scheduled speculatively.
The hook is used to check if the pattern of <var>insn</var> has a speculative
version and, in case of successful check, to generate that speculative
pattern.  The hook should return 1, if the instruction has a speculative form,
or -1, if it doesn&rsquo;t.  <var>request</var> describes the type of requested
speculation.  If the return value equals 1 then <var>new_pat</var> is assigned
the generated speculative pattern.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fNEEDS_005fBLOCK_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_SCHED_NEEDS_BLOCK_P</strong> <em>(int <var>dep_status</var>)</em></dt>
<dd><p>This hook is called by the insn scheduler during generation of recovery code
for <var>insn</var>.  It should return <code>true</code>, if the corresponding check
instruction should branch to recovery code, or <code>false</code> otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fGEN_005fSPEC_005fCHECK"></a>Target Hook: <em>rtx</em> <strong>TARGET_SCHED_GEN_SPEC_CHECK</strong> <em>(rtx <var>insn</var>, rtx <var>label</var>, int <var>mutate_p</var>)</em></dt>
<dd><p>This hook is called by the insn scheduler to generate a pattern for recovery
check instruction.  If <var>mutate_p</var> is zero, then <var>insn</var> is a
speculative instruction for which the check should be generated.
<var>label</var> is either a label of a basic block, where recovery code should
be emitted, or a null pointer, when requested check doesn&rsquo;t branch to
recovery code (a simple check).  If <var>mutate_p</var> is nonzero, then
a pattern for a branchy check corresponding to a simple check denoted by
<var>insn</var> should be generated.  In this case <var>label</var> can&rsquo;t be null.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD_005fGUARD_005fSPEC"></a>Target Hook: <em>bool</em> <strong>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD_GUARD_SPEC</strong> <em>(const_rtx <var>insn</var>)</em></dt>
<dd><p>This hook is used as a workaround for
&lsquo;<samp>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD_GUARD</samp>&rsquo; not being
called on the first instruction of the ready list.  The hook is used to
discard speculative instructions that stand first in the ready list from
being scheduled on the current cycle.  If the hook returns <code>false</code>,
<var>insn</var> will not be chosen to be issued.
For non-speculative instructions,
the hook should always return <code>true</code>.  For example, in the ia64 backend
the hook is used to cancel data speculative insns when the ALAT table
is nearly full.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fSET_005fSCHED_005fFLAGS"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_SET_SCHED_FLAGS</strong> <em>(struct spec_info_def *<var>spec_info</var>)</em></dt>
<dd><p>This hook is used by the insn scheduler to find out what features should be
enabled/used.
The structure *<var>spec_info</var> should be filled in by the target.
The structure describes speculation types that can be used in the scheduler.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fSMS_005fRES_005fMII"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_SMS_RES_MII</strong> <em>(struct ddg *<var>g</var>)</em></dt>
<dd><p>This hook is called by the swing modulo scheduler to calculate a
resource-based lower bound which is based on the resources available in
the machine and the resources required by each instruction.  The target
backend can use <var>g</var> to calculate such bound.  A very simple lower
bound will be used in case this hook is not implemented: the total number
of instructions divided by the issue rate.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDISPATCH"></a>Target Hook: <em>bool</em> <strong>TARGET_SCHED_DISPATCH</strong> <em>(rtx <var>insn</var>, int <var>x</var>)</em></dt>
<dd><p>This hook is called by Haifa Scheduler.  It returns true if dispatch scheduling
is supported in hardware and the condition specified in the parameter is true.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fDISPATCH_005fDO"></a>Target Hook: <em>void</em> <strong>TARGET_SCHED_DISPATCH_DO</strong> <em>(rtx <var>insn</var>, int <var>x</var>)</em></dt>
<dd><p>This hook is called by Haifa Scheduler.  It performs the operation specified
in its second parameter.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fEXPOSED_005fPIPELINE"></a>Target Hook: <em>bool</em> <strong>TARGET_SCHED_EXPOSED_PIPELINE</strong></dt>
<dd><p>True if the processor has an exposed pipeline, which means that not just
the order of instructions is important for correctness when scheduling, but
also the latencies of operations.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSCHED_005fREASSOCIATION_005fWIDTH"></a>Target Hook: <em>int</em> <strong>TARGET_SCHED_REASSOCIATION_WIDTH</strong> <em>(unsigned int <var>opc</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>This hook is called by tree reassociator to determine a level of
parallelism required in output calculations chain.
</p></dd></dl>

<hr>
<a name="Sections"></a>
<div class="header">
<p>
Next: <a href="#PIC" accesskey="n" rel="next">PIC</a>, Previous: <a href="#Scheduling" accesskey="p" rel="prev">Scheduling</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Dividing-the-Output-into-Sections-_0028Texts_002c-Data_002c-_2026_0029"></a>
<h3 class="section">17.19 Dividing the Output into Sections (Texts, Data, &hellip;)</h3>

<p>An object file is divided into sections containing different types of
data.  In the most common case, there are three sections: the <em>text
section</em>, which holds instructions and read-only data; the <em>data
section</em>, which holds initialized writable data; and the <em>bss
section</em>, which holds uninitialized data.  Some systems have other kinds
of sections.
</p>
<p><samp>varasm.c</samp> provides several well-known sections, such as
<code>text_section</code>, <code>data_section</code> and <code>bss_section</code>.
The normal way of controlling a <code><var>foo</var>_section</code> variable
is to define the associated <code><var>FOO</var>_SECTION_ASM_OP</code> macro,
as described below.  The macros are only read once, when <samp>varasm.c</samp>
initializes itself, so their values must be run-time constants.
They may however depend on command-line flags.
</p>
<p><em>Note:</em> Some run-time files, such <samp>crtstuff.c</samp>, also make
use of the <code><var>FOO</var>_SECTION_ASM_OP</code> macros, and expect them
to be string literals.
</p>
<p>Some assemblers require a different string to be written every time a
section is selected.  If your assembler falls into this category, you
should define the <code>TARGET_ASM_INIT_SECTIONS</code> hook and use
<code>get_unnamed_section</code> to set up the sections.
</p>
<p>You must always create a <code>text_section</code>, either by defining
<code>TEXT_SECTION_ASM_OP</code> or by initializing <code>text_section</code>
in <code>TARGET_ASM_INIT_SECTIONS</code>.  The same is true of
<code>data_section</code> and <code>DATA_SECTION_ASM_OP</code>.  If you do not
create a distinct <code>readonly_data_section</code>, the default is to
reuse <code>text_section</code>.
</p>
<p>All the other <samp>varasm.c</samp> sections are optional, and are null
if the target does not provide them.
</p>
<dl>
<dt><a name="index-TEXT_005fSECTION_005fASM_005fOP"></a>Macro: <strong>TEXT_SECTION_ASM_OP</strong></dt>
<dd><p>A C expression whose value is a string, including spacing, containing the
assembler operation that should precede instructions and read-only data.
Normally <code>&quot;\t.text&quot;</code> is right.
</p></dd></dl>

<dl>
<dt><a name="index-HOT_005fTEXT_005fSECTION_005fNAME"></a>Macro: <strong>HOT_TEXT_SECTION_NAME</strong></dt>
<dd><p>If defined, a C string constant for the name of the section containing most
frequently executed functions of the program.  If not defined, GCC will provide
a default definition if the target supports named sections.
</p></dd></dl>

<dl>
<dt><a name="index-UNLIKELY_005fEXECUTED_005fTEXT_005fSECTION_005fNAME"></a>Macro: <strong>UNLIKELY_EXECUTED_TEXT_SECTION_NAME</strong></dt>
<dd><p>If defined, a C string constant for the name of the section containing unlikely
executed functions in the program.
</p></dd></dl>

<dl>
<dt><a name="index-DATA_005fSECTION_005fASM_005fOP"></a>Macro: <strong>DATA_SECTION_ASM_OP</strong></dt>
<dd><p>A C expression whose value is a string, including spacing, containing the
assembler operation to identify the following data as writable initialized
data.  Normally <code>&quot;\t.data&quot;</code> is right.
</p></dd></dl>

<dl>
<dt><a name="index-SDATA_005fSECTION_005fASM_005fOP"></a>Macro: <strong>SDATA_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
initialized, writable small data.
</p></dd></dl>

<dl>
<dt><a name="index-READONLY_005fDATA_005fSECTION_005fASM_005fOP"></a>Macro: <strong>READONLY_DATA_SECTION_ASM_OP</strong></dt>
<dd><p>A C expression whose value is a string, including spacing, containing the
assembler operation to identify the following data as read-only initialized
data.
</p></dd></dl>

<dl>
<dt><a name="index-BSS_005fSECTION_005fASM_005fOP"></a>Macro: <strong>BSS_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
uninitialized global data.  If not defined, and
<code>ASM_OUTPUT_ALIGNED_BSS</code> not defined,
uninitialized global data will be output in the data section if
<samp>-fno-common</samp> is passed, otherwise <code>ASM_OUTPUT_COMMON</code> will be
used.
</p></dd></dl>

<dl>
<dt><a name="index-SBSS_005fSECTION_005fASM_005fOP"></a>Macro: <strong>SBSS_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
uninitialized, writable small data.
</p></dd></dl>

<dl>
<dt><a name="index-TLS_005fCOMMON_005fASM_005fOP"></a>Macro: <strong>TLS_COMMON_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string containing the
assembler operation to identify the following data as thread-local
common data.  The default is <code>&quot;.tls_common&quot;</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TLS_005fSECTION_005fASM_005fFLAG"></a>Macro: <strong>TLS_SECTION_ASM_FLAG</strong></dt>
<dd><p>If defined, a C expression whose value is a character constant
containing the flag used to mark a section as a TLS section.  The
default is <code>'T'</code>.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fSECTION_005fASM_005fOP"></a>Macro: <strong>INIT_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
initialization code.  If not defined, GCC will assume such a section does
not exist.  This section has no corresponding <code>init_section</code>
variable; it is used entirely in runtime code.
</p></dd></dl>

<dl>
<dt><a name="index-FINI_005fSECTION_005fASM_005fOP"></a>Macro: <strong>FINI_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
finalization code.  If not defined, GCC will assume such a section does
not exist.  This section has no corresponding <code>fini_section</code>
variable; it is used entirely in runtime code.
</p></dd></dl>

<dl>
<dt><a name="index-INIT_005fARRAY_005fSECTION_005fASM_005fOP"></a>Macro: <strong>INIT_ARRAY_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
part of the <code>.init_array</code> (or equivalent) section.  If not
defined, GCC will assume such a section does not exist.  Do not define
both this macro and <code>INIT_SECTION_ASM_OP</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FINI_005fARRAY_005fSECTION_005fASM_005fOP"></a>Macro: <strong>FINI_ARRAY_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C expression whose value is a string, including spacing,
containing the assembler operation to identify the following data as
part of the <code>.fini_array</code> (or equivalent) section.  If not
defined, GCC will assume such a section does not exist.  Do not define
both this macro and <code>FINI_SECTION_ASM_OP</code>.
</p></dd></dl>

<dl>
<dt><a name="index-CRT_005fCALL_005fSTATIC_005fFUNCTION"></a>Macro: <strong>CRT_CALL_STATIC_FUNCTION</strong> <em>(<var>section_op</var>, <var>function</var>)</em></dt>
<dd><p>If defined, an ASM statement that switches to a different section
via <var>section_op</var>, calls <var>function</var>, and switches back to
the text section.  This is used in <samp>crtstuff.c</samp> if
<code>INIT_SECTION_ASM_OP</code> or <code>FINI_SECTION_ASM_OP</code> to calls
to initialization and finalization functions from the init and fini
sections.  By default, this macro uses a simple function call.  Some
ports need hand-crafted assembly code to avoid dependencies on
registers initialized in the function prologue or to ensure that
constant pools don&rsquo;t end up too far way in the text section.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLIBGCC_005fSDATA_005fSECTION"></a>Macro: <strong>TARGET_LIBGCC_SDATA_SECTION</strong></dt>
<dd><p>If defined, a string which names the section into which small
variables defined in crtstuff and libgcc should go.  This is useful
when the target has options for optimizing access to small data, and
you want the crtstuff and libgcc routines to be conservative in what
they expect of your application yet liberal in what your application
expects.  For example, for targets with a <code>.sdata</code> section (like
MIPS), you could compile crtstuff with <code>-G 0</code> so that it doesn&rsquo;t
require small data support from your application, but use this macro
to put small data into <code>.sdata</code> so that your application can
access these variables whether it uses small data or not.
</p></dd></dl>

<dl>
<dt><a name="index-FORCE_005fCODE_005fSECTION_005fALIGN"></a>Macro: <strong>FORCE_CODE_SECTION_ALIGN</strong></dt>
<dd><p>If defined, an ASM statement that aligns a code section to some
arbitrary boundary.  This is used to force all fragments of the
<code>.init</code> and <code>.fini</code> sections to have to same alignment
and thus prevent the linker from having to add any padding.
</p></dd></dl>

<dl>
<dt><a name="index-JUMP_005fTABLES_005fIN_005fTEXT_005fSECTION"></a>Macro: <strong>JUMP_TABLES_IN_TEXT_SECTION</strong></dt>
<dd><p>Define this macro to be an expression with a nonzero value if jump
tables (for <code>tablejump</code> insns) should be output in the text
section, along with the assembler instructions.  Otherwise, the
readonly data section is used.
</p>
<p>This macro is irrelevant if there is no separate readonly data section.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fINIT_005fSECTIONS"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_INIT_SECTIONS</strong> <em>(void)</em></dt>
<dd><p>Define this hook if you need to do something special to set up the
<samp>varasm.c</samp> sections, or if your target has some special sections
of its own that you need to create.
</p>
<p>GCC calls this hook after processing the command line, but before writing
any assembly code, and before calling any of the section-returning hooks
described below.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fRELOC_005fRW_005fMASK"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_RELOC_RW_MASK</strong> <em>(void)</em></dt>
<dd><p>Return a mask describing how relocations should be treated when
selecting sections.  Bit 1 should be set if global relocations
should be placed in a read-write section; bit 0 should be set if
local relocations should be placed in a read-write section.
</p>
<p>The default version of this function returns 3 when <samp>-fpic</samp>
is in effect, and 0 otherwise.  The hook is typically redefined
when the target cannot support (some kinds of) dynamic relocations
in read-only sections even in executables.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fSELECT_005fSECTION"></a>Target Hook: <em>section *</em> <strong>TARGET_ASM_SELECT_SECTION</strong> <em>(tree <var>exp</var>, int <var>reloc</var>, unsigned HOST_WIDE_INT <var>align</var>)</em></dt>
<dd><p>Return the section into which <var>exp</var> should be placed.  You can
assume that <var>exp</var> is either a <code>VAR_DECL</code> node or a constant of
some sort.  <var>reloc</var> indicates whether the initial value of <var>exp</var>
requires link-time relocations.  Bit 0 is set when variable contains
local relocations only, while bit 1 is set for global relocations.
<var>align</var> is the constant alignment in bits.
</p>
<p>The default version of this function takes care of putting read-only
variables in <code>readonly_data_section</code>.
</p>
<p>See also <var>USE_SELECT_SECTION_FOR_FUNCTIONS</var>.
</p></dd></dl>

<dl>
<dt><a name="index-USE_005fSELECT_005fSECTION_005fFOR_005fFUNCTIONS"></a>Macro: <strong>USE_SELECT_SECTION_FOR_FUNCTIONS</strong></dt>
<dd><p>Define this macro if you wish TARGET_ASM_SELECT_SECTION to be called
for <code>FUNCTION_DECL</code>s as well as for variables and constants.
</p>
<p>In the case of a <code>FUNCTION_DECL</code>, <var>reloc</var> will be zero if the
function has been determined to be likely to be called, and nonzero if
it is unlikely to be called.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fUNIQUE_005fSECTION"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_UNIQUE_SECTION</strong> <em>(tree <var>decl</var>, int <var>reloc</var>)</em></dt>
<dd><p>Build up a unique section name, expressed as a <code>STRING_CST</code> node,
and assign it to &lsquo;<samp>DECL_SECTION_NAME (<var>decl</var>)</samp>&rsquo;.
As with <code>TARGET_ASM_SELECT_SECTION</code>, <var>reloc</var> indicates whether
the initial value of <var>exp</var> requires link-time relocations.
</p>
<p>The default version of this function appends the symbol name to the
ELF section name that would normally be used for the symbol.  For
example, the function <code>foo</code> would be placed in <code>.text.foo</code>.
Whatever the actual target object format, this is often good enough.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fRODATA_005fSECTION"></a>Target Hook: <em>section *</em> <strong>TARGET_ASM_FUNCTION_RODATA_SECTION</strong> <em>(tree <var>decl</var>)</em></dt>
<dd><p>Return the readonly data section associated with
&lsquo;<samp>DECL_SECTION_NAME (<var>decl</var>)</samp>&rsquo;.
The default version of this function selects <code>.gnu.linkonce.r.name</code> if
the function&rsquo;s section is <code>.gnu.linkonce.t.name</code>, <code>.rodata.name</code>
if function is in <code>.text.name</code>, and the normal readonly-data section
otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fMERGEABLE_005fRODATA_005fPREFIX"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_MERGEABLE_RODATA_PREFIX</strong></dt>
<dd><p>Usually, the compiler uses the prefix <code>&quot;.rodata&quot;</code> to construct
section names for mergeable constant data.  Define this macro to override
the string if a different section name should be used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fTM_005fCLONE_005fTABLE_005fSECTION"></a>Target Hook: <em>section *</em> <strong>TARGET_ASM_TM_CLONE_TABLE_SECTION</strong> <em>(void)</em></dt>
<dd><p>Return the section that should be used for transactional memory clone  tables.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fSELECT_005fRTX_005fSECTION"></a>Target Hook: <em>section *</em> <strong>TARGET_ASM_SELECT_RTX_SECTION</strong> <em>(enum machine_mode <var>mode</var>, rtx <var>x</var>, unsigned HOST_WIDE_INT <var>align</var>)</em></dt>
<dd><p>Return the section into which a constant <var>x</var>, of mode <var>mode</var>,
should be placed.  You can assume that <var>x</var> is some kind of
constant in RTL.  The argument <var>mode</var> is redundant except in the
case of a <code>const_int</code> rtx.  <var>align</var> is the constant alignment
in bits.
</p>
<p>The default version of this function takes care of putting symbolic
constants in <code>flag_pic</code> mode in <code>data_section</code> and everything
else in <code>readonly_data_section</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMANGLE_005fDECL_005fASSEMBLER_005fNAME"></a>Target Hook: <em>tree</em> <strong>TARGET_MANGLE_DECL_ASSEMBLER_NAME</strong> <em>(tree <var>decl</var>, tree <var>id</var>)</em></dt>
<dd><p>Define this hook if you need to postprocess the assembler name generated
by target-independent code.  The <var>id</var> provided to this hook will be
the computed name (e.g., the macro <code>DECL_NAME</code> of the <var>decl</var> in C,
or the mangled name of the <var>decl</var> in C++).  The return value of the
hook is an <code>IDENTIFIER_NODE</code> for the appropriate mangled name on
your target system.  The default implementation of this hook just
returns the <var>id</var> provided.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fENCODE_005fSECTION_005fINFO"></a>Target Hook: <em>void</em> <strong>TARGET_ENCODE_SECTION_INFO</strong> <em>(tree <var>decl</var>, rtx <var>rtl</var>, int <var>new_decl_p</var>)</em></dt>
<dd><p>Define this hook if references to a symbol or a constant must be
treated differently depending on something about the variable or
function named by the symbol (such as what section it is in).
</p>
<p>The hook is executed immediately after rtl has been created for
<var>decl</var>, which may be a variable or function declaration or
an entry in the constant pool.  In either case, <var>rtl</var> is the
rtl in question.  Do <em>not</em> use <code>DECL_RTL (<var>decl</var>)</code>
in this hook; that field may not have been initialized yet.
</p>
<p>In the case of a constant, it is safe to assume that the rtl is
a <code>mem</code> whose address is a <code>symbol_ref</code>.  Most decls
will also have this form, but that is not guaranteed.  Global
register variables, for instance, will have a <code>reg</code> for their
rtl.  (Normally the right thing to do with such unusual rtl is
leave it alone.)
</p>
<p>The <var>new_decl_p</var> argument will be true if this is the first time
that <code>TARGET_ENCODE_SECTION_INFO</code> has been invoked on this decl.  It will
be false for subsequent invocations, which will happen for duplicate
declarations.  Whether or not anything must be done for the duplicate
declaration depends on whether the hook examines <code>DECL_ATTRIBUTES</code>.
<var>new_decl_p</var> is always true when the hook is called for a constant.
</p>
<a name="index-SYMBOL_005fREF_005fFLAG_002c-in-TARGET_005fENCODE_005fSECTION_005fINFO"></a>
<p>The usual thing for this hook to do is to record flags in the
<code>symbol_ref</code>, using <code>SYMBOL_REF_FLAG</code> or <code>SYMBOL_REF_FLAGS</code>.
Historically, the name string was modified if it was necessary to
encode more than one bit of information, but this practice is now
discouraged; use <code>SYMBOL_REF_FLAGS</code>.
</p>
<p>The default definition of this hook, <code>default_encode_section_info</code>
in <samp>varasm.c</samp>, sets a number of commonly-useful bits in
<code>SYMBOL_REF_FLAGS</code>.  Check whether the default does what you need
before overriding it.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSTRIP_005fNAME_005fENCODING"></a>Target Hook: <em>const char *</em> <strong>TARGET_STRIP_NAME_ENCODING</strong> <em>(const char *<var>name</var>)</em></dt>
<dd><p>Decode <var>name</var> and return the real name part, sans
the characters that <code>TARGET_ENCODE_SECTION_INFO</code>
may have added.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fIN_005fSMALL_005fDATA_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_IN_SMALL_DATA_P</strong> <em>(const_tree <var>exp</var>)</em></dt>
<dd><p>Returns true if <var>exp</var> should be placed into a &ldquo;small data&rdquo; section.
The default version of this hook always returns false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHAVE_005fSRODATA_005fSECTION"></a>Target Hook: <em>bool</em> <strong>TARGET_HAVE_SRODATA_SECTION</strong></dt>
<dd><p>Contains the value true if the target places read-only
&ldquo;small data&rdquo; into a separate section.  The default value is false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPROFILE_005fBEFORE_005fPROLOGUE"></a>Target Hook: <em>bool</em> <strong>TARGET_PROFILE_BEFORE_PROLOGUE</strong> <em>(void)</em></dt>
<dd><p>It returns true if target wants profile code emitted before prologue.
</p>
<p>The default version of this hook use the target macro
<code>PROFILE_BEFORE_PROLOGUE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBINDS_005fLOCAL_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_BINDS_LOCAL_P</strong> <em>(const_tree <var>exp</var>)</em></dt>
<dd><p>Returns true if <var>exp</var> names an object for which name resolution
rules must resolve to the current &ldquo;module&rdquo; (dynamic shared library
or executable image).
</p>
<p>The default version of this hook implements the name resolution rules
for ELF, which has a looser model of global name binding than other
currently supported object file formats.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHAVE_005fTLS"></a>Target Hook: <em>bool</em> <strong>TARGET_HAVE_TLS</strong></dt>
<dd><p>Contains the value true if the target supports thread-local storage.
The default value is false.
</p></dd></dl>


<hr>
<a name="PIC"></a>
<div class="header">
<p>
Next: <a href="#Assembler-Format" accesskey="n" rel="next">Assembler Format</a>, Previous: <a href="#Sections" accesskey="p" rel="prev">Sections</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Position-Independent-Code"></a>
<h3 class="section">17.20 Position Independent Code</h3>
<a name="index-position-independent-code"></a>
<a name="index-PIC"></a>

<p>This section describes macros that help implement generation of position
independent code.  Simply defining these macros is not enough to
generate valid PIC; you must also add support to the hook
<code>TARGET_LEGITIMATE_ADDRESS_P</code> and to the macro
<code>PRINT_OPERAND_ADDRESS</code>, as well as <code>LEGITIMIZE_ADDRESS</code>.  You
must modify the definition of &lsquo;<samp>movsi</samp>&rsquo; to do something appropriate
when the source operand contains a symbolic address.  You may also
need to alter the handling of switch statements so that they use
relative addresses.
</p>
<dl>
<dt><a name="index-PIC_005fOFFSET_005fTABLE_005fREGNUM"></a>Macro: <strong>PIC_OFFSET_TABLE_REGNUM</strong></dt>
<dd><p>The register number of the register used to address a table of static
data addresses in memory.  In some cases this register is defined by a
processor&rsquo;s &ldquo;application binary interface&rdquo; (ABI).  When this macro
is defined, RTL is generated for this register once, as with the stack
pointer and frame pointer registers.  If this macro is not defined, it
is up to the machine-dependent files to allocate such a register (if
necessary).  Note that this register must be fixed when in use (e.g.
when <code>flag_pic</code> is true).
</p></dd></dl>

<dl>
<dt><a name="index-PIC_005fOFFSET_005fTABLE_005fREG_005fCALL_005fCLOBBERED"></a>Macro: <strong>PIC_OFFSET_TABLE_REG_CALL_CLOBBERED</strong></dt>
<dd><p>A C expression that is nonzero if the register defined by
<code>PIC_OFFSET_TABLE_REGNUM</code> is clobbered by calls.  If not defined,
the default is zero.  Do not define
this macro if <code>PIC_OFFSET_TABLE_REGNUM</code> is not defined.
</p></dd></dl>

<dl>
<dt><a name="index-LEGITIMATE_005fPIC_005fOPERAND_005fP"></a>Macro: <strong>LEGITIMATE_PIC_OPERAND_P</strong> <em>(<var>x</var>)</em></dt>
<dd><p>A C expression that is nonzero if <var>x</var> is a legitimate immediate
operand on the target machine when generating position independent code.
You can assume that <var>x</var> satisfies <code>CONSTANT_P</code>, so you need not
check this.  You can also assume <var>flag_pic</var> is true, so you need not
check it either.  You need not define this macro if all constants
(including <code>SYMBOL_REF</code>) can be immediate operands when generating
position independent code.
</p></dd></dl>

<hr>
<a name="Assembler-Format"></a>
<div class="header">
<p>
Next: <a href="#Debugging-Info" accesskey="n" rel="next">Debugging Info</a>, Previous: <a href="#PIC" accesskey="p" rel="prev">PIC</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-the-Output-Assembler-Language"></a>
<h3 class="section">17.21 Defining the Output Assembler Language</h3>

<p>This section describes macros whose principal purpose is to describe how
to write instructions in assembler language&mdash;rather than what the
instructions do.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#File-Framework" accesskey="1">File Framework</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Structural information for the assembler file.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Data-Output" accesskey="2">Data Output</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output of constants (numbers, strings, addresses).
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Uninitialized-Data" accesskey="3">Uninitialized Data</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output of uninitialized variables.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Label-Output" accesskey="4">Label Output</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output and generation of labels.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Initialization" accesskey="5">Initialization</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">General principles of initialization
                         and termination routines.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Macros-for-Initialization" accesskey="6">Macros for Initialization</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">
                         Specific macros that control the handling of
                         initialization and termination routines.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Instruction-Output" accesskey="7">Instruction Output</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output of actual instructions.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Dispatch-Tables" accesskey="8">Dispatch Tables</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output of jump tables.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Exception-Region-Output" accesskey="9">Exception Region Output</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Output of exception region code.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Alignment-Output">Alignment Output</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Pseudo ops for alignment and skipping data.
</td></tr>
</table>

<hr>
<a name="File-Framework"></a>
<div class="header">
<p>
Next: <a href="#Data-Output" accesskey="n" rel="next">Data Output</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Overall-Framework-of-an-Assembler-File"></a>
<h4 class="subsection">17.21.1 The Overall Framework of an Assembler File</h4>
<a name="index-assembler-format"></a>
<a name="index-output-of-assembler-code"></a>

<p>This describes the overall framework of an assembly file.
</p>
<a name="index-default_005ffile_005fstart"></a>
<dl>
<dt><a name="index-TARGET_005fASM_005fFILE_005fSTART"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FILE_START</strong> <em>(void)</em></dt>
<dd><p>Output to <code>asm_out_file</code> any text which the assembler expects to
find at the beginning of a file.  The default behavior is controlled
by two flags, documented below.  Unless your target&rsquo;s assembler is
quite unusual, if you override the default, you should call
<code>default_file_start</code> at some point in your target hook.  This
lets other target files rely on these variables.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFILE_005fSTART_005fAPP_005fOFF"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_FILE_START_APP_OFF</strong></dt>
<dd><p>If this flag is true, the text of the macro <code>ASM_APP_OFF</code> will be
printed as the very first line in the assembly file, unless
<samp>-fverbose-asm</samp> is in effect.  (If that macro has been defined
to the empty string, this variable has no effect.)  With the normal
definition of <code>ASM_APP_OFF</code>, the effect is to notify the GNU
assembler that it need not bother stripping comments or extra
whitespace from its input.  This allows it to work a bit faster.
</p>
<p>The default is false.  You should not set it to true unless you have
verified that your port does not generate any extra whitespace or
comments that will cause GAS to issue errors in NO_APP mode.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFILE_005fSTART_005fFILE_005fDIRECTIVE"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_FILE_START_FILE_DIRECTIVE</strong></dt>
<dd><p>If this flag is true, <code>output_file_directive</code> will be called
for the primary source file, immediately after printing
<code>ASM_APP_OFF</code> (if that is enabled).  Most ELF assemblers expect
this to be done.  The default is false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFILE_005fEND"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FILE_END</strong> <em>(void)</em></dt>
<dd><p>Output to <code>asm_out_file</code> any text which the assembler expects
to find at the end of a file.  The default is to output nothing.
</p></dd></dl>

<dl>
<dt><a name="index-file_005fend_005findicate_005fexec_005fstack"></a>Function: <em>void</em> <strong>file_end_indicate_exec_stack</strong> <em>()</em></dt>
<dd><p>Some systems use a common convention, the &lsquo;<samp>.note.GNU-stack</samp>&rsquo;
special section, to indicate whether or not an object file relies on
the stack being executable.  If your system uses this convention, you
should define <code>TARGET_ASM_FILE_END</code> to this function.  If you
need to do other things in that hook, have your hook function call
this function.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fLTO_005fSTART"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_LTO_START</strong> <em>(void)</em></dt>
<dd><p>Output to <code>asm_out_file</code> any text which the assembler expects
to find at the start of an LTO section.  The default is to output
nothing.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fLTO_005fEND"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_LTO_END</strong> <em>(void)</em></dt>
<dd><p>Output to <code>asm_out_file</code> any text which the assembler expects
to find at the end of an LTO section.  The default is to output
nothing.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fCODE_005fEND"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_CODE_END</strong> <em>(void)</em></dt>
<dd><p>Output to <code>asm_out_file</code> any text which is needed before emitting
unwind info and debug info at the end of a file.  Some targets emit
here PIC setup thunks that cannot be emitted at the end of file,
because they couldn&rsquo;t have unwind info then.  The default is to output
nothing.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fCOMMENT_005fSTART"></a>Macro: <strong>ASM_COMMENT_START</strong></dt>
<dd><p>A C string constant describing how to begin a comment in the target
assembler language.  The compiler assumes that the comment will end at
the end of the line.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fAPP_005fON"></a>Macro: <strong>ASM_APP_ON</strong></dt>
<dd><p>A C string constant for text to be output before each <code>asm</code>
statement or group of consecutive ones.  Normally this is
<code>&quot;#APP&quot;</code>, which is a comment that has no effect on most
assemblers but tells the GNU assembler that it must check the lines
that follow for all valid assembler constructs.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fAPP_005fOFF"></a>Macro: <strong>ASM_APP_OFF</strong></dt>
<dd><p>A C string constant for text to be output after each <code>asm</code>
statement or group of consecutive ones.  Normally this is
<code>&quot;#NO_APP&quot;</code>, which tells the GNU assembler to resume making the
time-saving assumptions that are valid for ordinary compiler output.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fSOURCE_005fFILENAME"></a>Macro: <strong>ASM_OUTPUT_SOURCE_FILENAME</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement to output COFF information or DWARF debugging information
which indicates that filename <var>name</var> is the current source file to
the stdio stream <var>stream</var>.
</p>
<p>This macro need not be defined if the standard form of output
for the file format in use is appropriate.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fSOURCE_005fFILENAME"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_OUTPUT_SOURCE_FILENAME</strong> <em>(FILE *<var>file</var>, const char *<var>name</var>)</em></dt>
<dd><p>Output COFF information or DWARF debugging information which indicates that filename <var>name</var> is the current source file to the stdio stream <var>file</var>.
</p> 
<p>This target hook need not be defined if the standard form of output for the file format in use is appropriate.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fIDENT"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_OUTPUT_IDENT</strong> <em>(const char *<var>name</var>)</em></dt>
<dd><p>Output a string based on <var>name</var>, suitable for the &lsquo;<samp>#ident</samp>&rsquo;  directive, or the equivalent directive or pragma in non-C-family languages.  If this hook is not defined, nothing is output for the &lsquo;<samp>#ident</samp>&rsquo;  directive.
</p></dd></dl>

<dl>
<dt><a name="index-OUTPUT_005fQUOTED_005fSTRING"></a>Macro: <strong>OUTPUT_QUOTED_STRING</strong> <em>(<var>stream</var>, <var>string</var>)</em></dt>
<dd><p>A C statement to output the string <var>string</var> to the stdio stream
<var>stream</var>.  If you do not call the function <code>output_quoted_string</code>
in your config files, GCC will only call it to output filenames to
the assembler source.  So you can use it to canonicalize the format
of the filename using this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fNAMED_005fSECTION"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_NAMED_SECTION</strong> <em>(const char *<var>name</var>, unsigned int <var>flags</var>, tree <var>decl</var>)</em></dt>
<dd><p>Output assembly directives to switch to section <var>name</var>.  The section
should have attributes as specified by <var>flags</var>, which is a bit mask
of the <code>SECTION_*</code> flags defined in <samp>output.h</samp>.  If <var>decl</var>
is non-NULL, it is the <code>VAR_DECL</code> or <code>FUNCTION_DECL</code> with which
this section is associated.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fSECTION"></a>Target Hook: <em>section *</em> <strong>TARGET_ASM_FUNCTION_SECTION</strong> <em>(tree <var>decl</var>, enum node_frequency <var>freq</var>, bool <var>startup</var>, bool <var>exit</var>)</em></dt>
<dd><p>Return preferred text (sub)section for function <var>decl</var>.
Main purpose of this function is to separate cold, normal and hot
functions. <var>startup</var> is true when function is known to be used only
at startup (from static constructors or it is <code>main()</code>).
<var>exit</var> is true when function is known to be used only at exit
(from static destructors).
Return NULL if function should go to default text section.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFUNCTION_005fSWITCHED_005fTEXT_005fSECTIONS"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FUNCTION_SWITCHED_TEXT_SECTIONS</strong> <em>(FILE *<var>file</var>, tree <var>decl</var>, bool <var>new_is_cold</var>)</em></dt>
<dd><p>Used by the target to emit any assembler directives or additional  labels needed when a function is partitioned between different  sections.  Output should be written to <var>file</var>.  The function  decl is available as <var>decl</var> and the new section is &lsquo;cold&rsquo; if  <var>new_is_cold</var> is <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHAVE_005fNAMED_005fSECTIONS"></a>Common Target Hook: <em>bool</em> <strong>TARGET_HAVE_NAMED_SECTIONS</strong></dt>
<dd><p>This flag is true if the target supports <code>TARGET_ASM_NAMED_SECTION</code>.
It must not be modified by command-line option processing.
</p></dd></dl>

<a name="TARGET_005fHAVE_005fSWITCHABLE_005fBSS_005fSECTIONS"></a><dl>
<dt><a name="index-TARGET_005fHAVE_005fSWITCHABLE_005fBSS_005fSECTIONS"></a>Target Hook: <em>bool</em> <strong>TARGET_HAVE_SWITCHABLE_BSS_SECTIONS</strong></dt>
<dd><p>This flag is true if we can create zeroed data by switching to a BSS
section and then using <code>ASM_OUTPUT_SKIP</code> to allocate the space.
This is true on most ELF targets.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSECTION_005fTYPE_005fFLAGS"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_SECTION_TYPE_FLAGS</strong> <em>(tree <var>decl</var>, const char *<var>name</var>, int <var>reloc</var>)</em></dt>
<dd><p>Choose a set of section attributes for use by <code>TARGET_ASM_NAMED_SECTION</code>
based on a variable or function decl, a section name, and whether or not the
declaration&rsquo;s initializer may contain runtime relocations.  <var>decl</var> may be
null, in which case read-write data should be assumed.
</p>
<p>The default version of this function handles choosing code vs data,
read-only vs read-write data, and <code>flag_pic</code>.  You should only
need to override this if your target has special flags that might be
set via <code>__attribute__</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fRECORD_005fGCC_005fSWITCHES"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_RECORD_GCC_SWITCHES</strong> <em>(print_switch_type <var>type</var>, const char *<var>text</var>)</em></dt>
<dd><p>Provides the target with the ability to record the gcc command line
switches that have been passed to the compiler, and options that are
enabled.  The <var>type</var> argument specifies what is being recorded.
It can take the following values:
</p>
<dl compact="compact">
<dt><code>SWITCH_TYPE_PASSED</code></dt>
<dd><p><var>text</var> is a command line switch that has been set by the user.
</p>
</dd>
<dt><code>SWITCH_TYPE_ENABLED</code></dt>
<dd><p><var>text</var> is an option which has been enabled.  This might be as a
direct result of a command line switch, or because it is enabled by
default or because it has been enabled as a side effect of a different
command line switch.  For example, the <samp>-O2</samp> switch enables
various different individual optimization passes.
</p>
</dd>
<dt><code>SWITCH_TYPE_DESCRIPTIVE</code></dt>
<dd><p><var>text</var> is either NULL or some descriptive text which should be
ignored.  If <var>text</var> is NULL then it is being used to warn the
target hook that either recording is starting or ending.  The first
time <var>type</var> is SWITCH_TYPE_DESCRIPTIVE and <var>text</var> is NULL, the
warning is for start up and the second time the warning is for
wind down.  This feature is to allow the target hook to make any
necessary preparations before it starts to record switches and to
perform any necessary tidying up after it has finished recording
switches.
</p>
</dd>
<dt><code>SWITCH_TYPE_LINE_START</code></dt>
<dd><p>This option can be ignored by this target hook.
</p>
</dd>
<dt><code>SWITCH_TYPE_LINE_END</code></dt>
<dd><p>This option can be ignored by this target hook.
</p></dd>
</dl>

<p>The hook&rsquo;s return value must be zero.  Other return values may be
supported in the future.
</p>
<p>By default this hook is set to NULL, but an example implementation is
provided for ELF based targets.  Called <var>elf_record_gcc_switches</var>,
it records the switches as ASCII text inside a new, string mergeable
section in the assembler output file.  The name of the new section is
provided by the <code>TARGET_ASM_RECORD_GCC_SWITCHES_SECTION</code> target
hook.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fRECORD_005fGCC_005fSWITCHES_005fSECTION"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_RECORD_GCC_SWITCHES_SECTION</strong></dt>
<dd><p>This is the name of the section that will be created by the example
ELF implementation of the <code>TARGET_ASM_RECORD_GCC_SWITCHES</code> target
hook.
</p></dd></dl>

<hr>
<a name="Data-Output"></a>
<div class="header">
<p>
Next: <a href="#Uninitialized-Data" accesskey="n" rel="next">Uninitialized Data</a>, Previous: <a href="#File-Framework" accesskey="p" rel="prev">File Framework</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-of-Data"></a>
<h4 class="subsection">17.21.2 Output of Data</h4>


<dl>
<dt><a name="index-TARGET_005fASM_005fBYTE_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_BYTE_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fALIGNED_005fHI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_ALIGNED_HI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fALIGNED_005fSI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_ALIGNED_SI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fALIGNED_005fDI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_ALIGNED_DI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fALIGNED_005fTI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_ALIGNED_TI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fUNALIGNED_005fHI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_UNALIGNED_HI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fUNALIGNED_005fSI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_UNALIGNED_SI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fUNALIGNED_005fDI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_UNALIGNED_DI_OP</strong></dt>
<dt><a name="index-TARGET_005fASM_005fUNALIGNED_005fTI_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_UNALIGNED_TI_OP</strong></dt>
<dd><p>These hooks specify assembly directives for creating certain kinds
of integer object.  The <code>TARGET_ASM_BYTE_OP</code> directive creates a
byte-sized object, the <code>TARGET_ASM_ALIGNED_HI_OP</code> one creates an
aligned two-byte object, and so on.  Any of the hooks may be
<code>NULL</code>, indicating that no suitable directive is available.
</p>
<p>The compiler will print these strings at the start of a new line,
followed immediately by the object&rsquo;s initial value.  In most cases,
the string should contain a tab, a pseudo-op, and then another tab.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fINTEGER"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_INTEGER</strong> <em>(rtx <var>x</var>, unsigned int <var>size</var>, int <var>aligned_p</var>)</em></dt>
<dd><p>The <code>assemble_integer</code> function uses this hook to output an
integer object.  <var>x</var> is the object&rsquo;s value, <var>size</var> is its size
in bytes and <var>aligned_p</var> indicates whether it is aligned.  The
function should return <code>true</code> if it was able to output the
object.  If it returns false, <code>assemble_integer</code> will try to
split the object into smaller parts.
</p>
<p>The default implementation of this hook will use the
<code>TARGET_ASM_BYTE_OP</code> family of strings, returning <code>false</code>
when the relevant string is <code>NULL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fADDR_005fCONST_005fEXTRA"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_OUTPUT_ADDR_CONST_EXTRA</strong> <em>(FILE *<var>file</var>, rtx <var>x</var>)</em></dt>
<dd><p>A target hook to recognize <var>rtx</var> patterns that <code>output_addr_const</code>
can&rsquo;t deal with, and output assembly code to <var>file</var> corresponding to
the pattern <var>x</var>.  This may be used to allow machine-dependent
<code>UNSPEC</code>s to appear within constants.
</p>
<p>If target hook fails to recognize a pattern, it must return <code>false</code>,
so that a standard error message is printed.  If it prints an error message
itself, by calling, for example, <code>output_operand_lossage</code>, it may just
return <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fASCII"></a>Macro: <strong>ASM_OUTPUT_ASCII</strong> <em>(<var>stream</var>, <var>ptr</var>, <var>len</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> an assembler
instruction to assemble a string constant containing the <var>len</var>
bytes at <var>ptr</var>.  <var>ptr</var> will be a C expression of type
<code>char *</code> and <var>len</var> a C expression of type <code>int</code>.
</p>
<p>If the assembler has a <code>.ascii</code> pseudo-op as found in the
Berkeley Unix assembler, do not define the macro
<code>ASM_OUTPUT_ASCII</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fFDESC"></a>Macro: <strong>ASM_OUTPUT_FDESC</strong> <em>(<var>stream</var>, <var>decl</var>, <var>n</var>)</em></dt>
<dd><p>A C statement to output word <var>n</var> of a function descriptor for
<var>decl</var>.  This must be defined if <code>TARGET_VTABLE_USES_DESCRIPTORS</code>
is defined, and is otherwise unused.
</p></dd></dl>

<dl>
<dt><a name="index-CONSTANT_005fPOOL_005fBEFORE_005fFUNCTION"></a>Macro: <strong>CONSTANT_POOL_BEFORE_FUNCTION</strong></dt>
<dd><p>You may define this macro as a C expression.  You should define the
expression to have a nonzero value if GCC should output the constant
pool for a function before the code for the function, or a zero value if
GCC should output the constant pool after the function.  If you do
not define this macro, the usual case, GCC will output the constant
pool before the function.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fPOOL_005fPROLOGUE"></a>Macro: <strong>ASM_OUTPUT_POOL_PROLOGUE</strong> <em>(<var>file</var>, <var>funname</var>, <var>fundecl</var>, <var>size</var>)</em></dt>
<dd><p>A C statement to output assembler commands to define the start of the
constant pool for a function.  <var>funname</var> is a string giving
the name of the function.  Should the return type of the function
be required, it can be obtained via <var>fundecl</var>.  <var>size</var>
is the size, in bytes, of the constant pool that will be written
immediately after this call.
</p>
<p>If no constant-pool prefix is required, the usual case, this macro need
not be defined.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fSPECIAL_005fPOOL_005fENTRY"></a>Macro: <strong>ASM_OUTPUT_SPECIAL_POOL_ENTRY</strong> <em>(<var>file</var>, <var>x</var>, <var>mode</var>, <var>align</var>, <var>labelno</var>, <var>jumpto</var>)</em></dt>
<dd><p>A C statement (with or without semicolon) to output a constant in the
constant pool, if it needs special treatment.  (This macro need not do
anything for RTL expressions that can be output normally.)
</p>
<p>The argument <var>file</var> is the standard I/O stream to output the
assembler code on.  <var>x</var> is the RTL expression for the constant to
output, and <var>mode</var> is the machine mode (in case <var>x</var> is a
&lsquo;<samp>const_int</samp>&rsquo;).  <var>align</var> is the required alignment for the value
<var>x</var>; you should output an assembler directive to force this much
alignment.
</p>
<p>The argument <var>labelno</var> is a number to use in an internal label for
the address of this pool entry.  The definition of this macro is
responsible for outputting the label definition at the proper place.
Here is how to do this:
</p>
<div class="smallexample">
<pre class="smallexample"><code>(*targetm.asm_out.internal_label)</code> (<var>file</var>, &quot;LC&quot;, <var>labelno</var>);
</pre></div>

<p>When you output a pool entry specially, you should end with a
<code>goto</code> to the label <var>jumpto</var>.  This will prevent the same pool
entry from being output a second time in the usual manner.
</p>
<p>You need not define this macro if it would do nothing.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fPOOL_005fEPILOGUE"></a>Macro: <strong>ASM_OUTPUT_POOL_EPILOGUE</strong> <em>(<var>file</var> <var>funname</var> <var>fundecl</var> <var>size</var>)</em></dt>
<dd><p>A C statement to output assembler commands to at the end of the constant
pool for a function.  <var>funname</var> is a string giving the name of the
function.  Should the return type of the function be required, you can
obtain it via <var>fundecl</var>.  <var>size</var> is the size, in bytes, of the
constant pool that GCC wrote immediately before this call.
</p>
<p>If no constant-pool epilogue is required, the usual case, you need not
define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-IS_005fASM_005fLOGICAL_005fLINE_005fSEPARATOR"></a>Macro: <strong>IS_ASM_LOGICAL_LINE_SEPARATOR</strong> <em>(<var>C</var>, <var>STR</var>)</em></dt>
<dd><p>Define this macro as a C expression which is nonzero if <var>C</var> is
used as a logical line separator by the assembler.  <var>STR</var> points
to the position in the string where <var>C</var> was found; this can be used if
a line separator uses multiple characters.
</p>
<p>If you do not define this macro, the default is that only
the character &lsquo;<samp>;</samp>&rsquo; is treated as a logical line separator.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOPEN_005fPAREN"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_OPEN_PAREN</strong></dt>
<dt><a name="index-TARGET_005fASM_005fCLOSE_005fPAREN"></a>Target Hook: <em>const char *</em> <strong>TARGET_ASM_CLOSE_PAREN</strong></dt>
<dd><p>These target hooks are C string constants, describing the syntax in the
assembler for grouping arithmetic expressions.  If not overridden, they
default to normal parentheses, which is correct for most assemblers.
</p></dd></dl>

<p>These macros are provided by <samp>real.h</samp> for writing the definitions
of <code>ASM_OUTPUT_DOUBLE</code> and the like:
</p>
<dl>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fSINGLE"></a>Macro: <strong>REAL_VALUE_TO_TARGET_SINGLE</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fDOUBLE"></a>Macro: <strong>REAL_VALUE_TO_TARGET_DOUBLE</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fLONG_005fDOUBLE"></a>Macro: <strong>REAL_VALUE_TO_TARGET_LONG_DOUBLE</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL32"></a>Macro: <strong>REAL_VALUE_TO_TARGET_DECIMAL32</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL64"></a>Macro: <strong>REAL_VALUE_TO_TARGET_DECIMAL64</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dt><a name="index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL128"></a>Macro: <strong>REAL_VALUE_TO_TARGET_DECIMAL128</strong> <em>(<var>x</var>, <var>l</var>)</em></dt>
<dd><p>These translate <var>x</var>, of type <code>REAL_VALUE_TYPE</code>, to the
target&rsquo;s floating point representation, and store its bit pattern in
the variable <var>l</var>.  For <code>REAL_VALUE_TO_TARGET_SINGLE</code> and
<code>REAL_VALUE_TO_TARGET_DECIMAL32</code>, this variable should be a
simple <code>long int</code>.  For the others, it should be an array of
<code>long int</code>.  The number of elements in this array is determined
by the size of the desired target floating point data type: 32 bits of
it go in each <code>long int</code> array element.  Each array element holds
32 bits of the result, even if <code>long int</code> is wider than 32 bits
on the host machine.
</p>
<p>The array element values are designed so that you can print them out
using <code>fprintf</code> in the order they should appear in the target
machine&rsquo;s memory.
</p></dd></dl>

<hr>
<a name="Uninitialized-Data"></a>
<div class="header">
<p>
Next: <a href="#Label-Output" accesskey="n" rel="next">Label Output</a>, Previous: <a href="#Data-Output" accesskey="p" rel="prev">Data Output</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-of-Uninitialized-Variables"></a>
<h4 class="subsection">17.21.3 Output of Uninitialized Variables</h4>

<p>Each of the macros in this section is used to do the whole job of
outputting a single uninitialized variable.
</p>
<dl>
<dt><a name="index-ASM_005fOUTPUT_005fCOMMON"></a>Macro: <strong>ASM_OUTPUT_COMMON</strong> <em>(<var>stream</var>, <var>name</var>, <var>size</var>, <var>rounded</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> the assembler definition of a common-label named
<var>name</var> whose size is <var>size</var> bytes.  The variable <var>rounded</var>
is the size rounded up to whatever alignment the caller wants.  It is
possible that <var>size</var> may be zero, for instance if a struct with no
other member than a zero-length array is defined.  In this case, the
backend must output a symbol definition that allocates at least one
byte, both so that the address of the resulting object does not compare
equal to any other, and because some object formats cannot even express
the concept of a zero-sized common symbol, as that is how they represent
an ordinary undefined external.
</p>
<p>Use the expression <code>assemble_name (<var>stream</var>, <var>name</var>)</code> to
output the name itself; before and after that, output the additional
assembler syntax for defining the name, and a newline.
</p>
<p>This macro controls how the assembler definitions of uninitialized
common global variables are output.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGNED_005fCOMMON"></a>Macro: <strong>ASM_OUTPUT_ALIGNED_COMMON</strong> <em>(<var>stream</var>, <var>name</var>, <var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>Like <code>ASM_OUTPUT_COMMON</code> except takes the required alignment as a
separate, explicit argument.  If you define this macro, it is used in
place of <code>ASM_OUTPUT_COMMON</code>, and gives you more flexibility in
handling the required alignment of the variable.  The alignment is specified
as the number of bits.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGNED_005fDECL_005fCOMMON"></a>Macro: <strong>ASM_OUTPUT_ALIGNED_DECL_COMMON</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>, <var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>Like <code>ASM_OUTPUT_ALIGNED_COMMON</code> except that <var>decl</var> of the
variable to be output, if there is one, or <code>NULL_TREE</code> if there
is no corresponding variable.  If you define this macro, GCC will use it
in place of both <code>ASM_OUTPUT_COMMON</code> and
<code>ASM_OUTPUT_ALIGNED_COMMON</code>.  Define this macro when you need to see
the variable&rsquo;s decl in order to chose what to output.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGNED_005fBSS"></a>Macro: <strong>ASM_OUTPUT_ALIGNED_BSS</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>, <var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> the assembler definition of uninitialized global <var>decl</var> named
<var>name</var> whose size is <var>size</var> bytes.  The variable <var>alignment</var>
is the alignment specified as the number of bits.
</p>
<p>Try to use function <code>asm_output_aligned_bss</code> defined in file
<samp>varasm.c</samp> when defining this macro.  If unable, use the expression
<code>assemble_name (<var>stream</var>, <var>name</var>)</code> to output the name itself;
before and after that, output the additional assembler syntax for defining
the name, and a newline.
</p>
<p>There are two ways of handling global BSS.  One is to define this macro.
The other is to have <code>TARGET_ASM_SELECT_SECTION</code> return a
switchable BSS section (see <a href="#TARGET_005fHAVE_005fSWITCHABLE_005fBSS_005fSECTIONS">TARGET_HAVE_SWITCHABLE_BSS_SECTIONS</a>).
You do not need to do both.
</p>
<p>Some languages do not have <code>common</code> data, and require a
non-common form of global BSS in order to handle uninitialized globals
efficiently.  C++ is one example of this.  However, if the target does
not support global BSS, the front end may choose to make globals
common in order to save space in the object file.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fLOCAL"></a>Macro: <strong>ASM_OUTPUT_LOCAL</strong> <em>(<var>stream</var>, <var>name</var>, <var>size</var>, <var>rounded</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> the assembler definition of a local-common-label named
<var>name</var> whose size is <var>size</var> bytes.  The variable <var>rounded</var>
is the size rounded up to whatever alignment the caller wants.
</p>
<p>Use the expression <code>assemble_name (<var>stream</var>, <var>name</var>)</code> to
output the name itself; before and after that, output the additional
assembler syntax for defining the name, and a newline.
</p>
<p>This macro controls how the assembler definitions of uninitialized
static variables are output.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGNED_005fLOCAL"></a>Macro: <strong>ASM_OUTPUT_ALIGNED_LOCAL</strong> <em>(<var>stream</var>, <var>name</var>, <var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>Like <code>ASM_OUTPUT_LOCAL</code> except takes the required alignment as a
separate, explicit argument.  If you define this macro, it is used in
place of <code>ASM_OUTPUT_LOCAL</code>, and gives you more flexibility in
handling the required alignment of the variable.  The alignment is specified
as the number of bits.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGNED_005fDECL_005fLOCAL"></a>Macro: <strong>ASM_OUTPUT_ALIGNED_DECL_LOCAL</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>, <var>size</var>, <var>alignment</var>)</em></dt>
<dd><p>Like <code>ASM_OUTPUT_ALIGNED_DECL</code> except that <var>decl</var> of the
variable to be output, if there is one, or <code>NULL_TREE</code> if there
is no corresponding variable.  If you define this macro, GCC will use it
in place of both <code>ASM_OUTPUT_DECL</code> and
<code>ASM_OUTPUT_ALIGNED_DECL</code>.  Define this macro when you need to see
the variable&rsquo;s decl in order to chose what to output.
</p></dd></dl>

<hr>
<a name="Label-Output"></a>
<div class="header">
<p>
Next: <a href="#Initialization" accesskey="n" rel="next">Initialization</a>, Previous: <a href="#Uninitialized-Data" accesskey="p" rel="prev">Uninitialized Data</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-and-Generation-of-Labels"></a>
<h4 class="subsection">17.21.4 Output and Generation of Labels</h4>

<p>This is about outputting labels.
</p>
<a name="index-assemble_005fname"></a>
<dl>
<dt><a name="index-ASM_005fOUTPUT_005fLABEL"></a>Macro: <strong>ASM_OUTPUT_LABEL</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> the assembler definition of a label named <var>name</var>.
Use the expression <code>assemble_name (<var>stream</var>, <var>name</var>)</code> to
output the name itself; before and after that, output the additional
assembler syntax for defining the name, and a newline.  A default
definition of this macro is provided which is correct for most systems.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fFUNCTION_005fLABEL"></a>Macro: <strong>ASM_OUTPUT_FUNCTION_LABEL</strong> <em>(<var>stream</var>, <var>name</var>, <var>decl</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> the assembler definition of a label named <var>name</var> of
a function.
Use the expression <code>assemble_name (<var>stream</var>, <var>name</var>)</code> to
output the name itself; before and after that, output the additional
assembler syntax for defining the name, and a newline.  A default
definition of this macro is provided which is correct for most systems.
</p>
<p>If this macro is not defined, then the function name is defined in the
usual manner as a label (by means of <code>ASM_OUTPUT_LABEL</code>).
</p></dd></dl>

<a name="index-assemble_005fname_005fraw"></a>
<dl>
<dt><a name="index-ASM_005fOUTPUT_005fINTERNAL_005fLABEL"></a>Macro: <strong>ASM_OUTPUT_INTERNAL_LABEL</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>Identical to <code>ASM_OUTPUT_LABEL</code>, except that <var>name</var> is known
to refer to a compiler-generated label.  The default definition uses
<code>assemble_name_raw</code>, which is like <code>assemble_name</code> except
that it is more efficient.
</p></dd></dl>

<dl>
<dt><a name="index-SIZE_005fASM_005fOP"></a>Macro: <strong>SIZE_ASM_OP</strong></dt>
<dd><p>A C string containing the appropriate assembler directive to specify the
size of a symbol, without any arguments.  On systems that use ELF, the
default (in <samp>config/elfos.h</samp>) is &lsquo;<samp>&quot;\t.size\t&quot;</samp>&rsquo;; on other
systems, the default is not to define this macro.
</p>
<p>Define this macro only if it is correct to use the default definitions
of <code>ASM_OUTPUT_SIZE_DIRECTIVE</code> and <code>ASM_OUTPUT_MEASURED_SIZE</code>
for your system.  If you need your own custom definitions of those
macros, or if you do not need explicit symbol sizes at all, do not
define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fSIZE_005fDIRECTIVE"></a>Macro: <strong>ASM_OUTPUT_SIZE_DIRECTIVE</strong> <em>(<var>stream</var>, <var>name</var>, <var>size</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> a directive telling the assembler that the size of the
symbol <var>name</var> is <var>size</var>.  <var>size</var> is a <code>HOST_WIDE_INT</code>.
If you define <code>SIZE_ASM_OP</code>, a default definition of this macro is
provided.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fMEASURED_005fSIZE"></a>Macro: <strong>ASM_OUTPUT_MEASURED_SIZE</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> a directive telling the assembler to calculate the size of
the symbol <var>name</var> by subtracting its address from the current
address.
</p>
<p>If you define <code>SIZE_ASM_OP</code>, a default definition of this macro is
provided.  The default assumes that the assembler recognizes a special
&lsquo;<samp>.</samp>&rsquo; symbol as referring to the current address, and can calculate
the difference between this and another symbol.  If your assembler does
not recognize &lsquo;<samp>.</samp>&rsquo; or cannot do calculations with it, you will need
to redefine <code>ASM_OUTPUT_MEASURED_SIZE</code> to use some other technique.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDOLLAR_005fIN_005fLABEL"></a>Macro: <strong>NO_DOLLAR_IN_LABEL</strong></dt>
<dd><p>Define this macro if the assembler does not accept the character
&lsquo;<samp>$</samp>&rsquo; in label names.  By default constructors and destructors in
G++ have &lsquo;<samp>$</samp>&rsquo; in the identifiers.  If this macro is defined,
&lsquo;<samp>.</samp>&rsquo; is used instead.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDOT_005fIN_005fLABEL"></a>Macro: <strong>NO_DOT_IN_LABEL</strong></dt>
<dd><p>Define this macro if the assembler does not accept the character
&lsquo;<samp>.</samp>&rsquo; in label names.  By default constructors and destructors in G++
have names that use &lsquo;<samp>.</samp>&rsquo;.  If this macro is defined, these names
are rewritten to avoid &lsquo;<samp>.</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TYPE_005fASM_005fOP"></a>Macro: <strong>TYPE_ASM_OP</strong></dt>
<dd><p>A C string containing the appropriate assembler directive to specify the
type of a symbol, without any arguments.  On systems that use ELF, the
default (in <samp>config/elfos.h</samp>) is &lsquo;<samp>&quot;\t.type\t&quot;</samp>&rsquo;; on other
systems, the default is not to define this macro.
</p>
<p>Define this macro only if it is correct to use the default definition of
<code>ASM_OUTPUT_TYPE_DIRECTIVE</code> for your system.  If you need your own
custom definition of this macro, or if you do not need explicit symbol
types at all, do not define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TYPE_005fOPERAND_005fFMT"></a>Macro: <strong>TYPE_OPERAND_FMT</strong></dt>
<dd><p>A C string which specifies (using <code>printf</code> syntax) the format of
the second operand to <code>TYPE_ASM_OP</code>.  On systems that use ELF, the
default (in <samp>config/elfos.h</samp>) is &lsquo;<samp>&quot;@%s&quot;</samp>&rsquo;; on other systems,
the default is not to define this macro.
</p>
<p>Define this macro only if it is correct to use the default definition of
<code>ASM_OUTPUT_TYPE_DIRECTIVE</code> for your system.  If you need your own
custom definition of this macro, or if you do not need explicit symbol
types at all, do not define this macro.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fTYPE_005fDIRECTIVE"></a>Macro: <strong>ASM_OUTPUT_TYPE_DIRECTIVE</strong> <em>(<var>stream</var>, <var>type</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> a directive telling the assembler that the type of the
symbol <var>name</var> is <var>type</var>.  <var>type</var> is a C string; currently,
that string is always either &lsquo;<samp>&quot;function&quot;</samp>&rsquo; or &lsquo;<samp>&quot;object&quot;</samp>&rsquo;, but
you should not count on this.
</p>
<p>If you define <code>TYPE_ASM_OP</code> and <code>TYPE_OPERAND_FMT</code>, a default
definition of this macro is provided.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fDECLARE_005fFUNCTION_005fNAME"></a>Macro: <strong>ASM_DECLARE_FUNCTION_NAME</strong> <em>(<var>stream</var>, <var>name</var>, <var>decl</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> any text necessary for declaring the name <var>name</var> of a
function which is being defined.  This macro is responsible for
outputting the label definition (perhaps using
<code>ASM_OUTPUT_FUNCTION_LABEL</code>).  The argument <var>decl</var> is the
<code>FUNCTION_DECL</code> tree node representing the function.
</p>
<p>If this macro is not defined, then the function name is defined in the
usual manner as a label (by means of <code>ASM_OUTPUT_FUNCTION_LABEL</code>).
</p>
<p>You may wish to use <code>ASM_OUTPUT_TYPE_DIRECTIVE</code> in the definition
of this macro.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fDECLARE_005fFUNCTION_005fSIZE"></a>Macro: <strong>ASM_DECLARE_FUNCTION_SIZE</strong> <em>(<var>stream</var>, <var>name</var>, <var>decl</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> any text necessary for declaring the size of a function
which is being defined.  The argument <var>name</var> is the name of the
function.  The argument <var>decl</var> is the <code>FUNCTION_DECL</code> tree node
representing the function.
</p>
<p>If this macro is not defined, then the function size is not defined.
</p>
<p>You may wish to use <code>ASM_OUTPUT_MEASURED_SIZE</code> in the definition
of this macro.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fDECLARE_005fOBJECT_005fNAME"></a>Macro: <strong>ASM_DECLARE_OBJECT_NAME</strong> <em>(<var>stream</var>, <var>name</var>, <var>decl</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> any text necessary for declaring the name <var>name</var> of an
initialized variable which is being defined.  This macro must output the
label definition (perhaps using <code>ASM_OUTPUT_LABEL</code>).  The argument
<var>decl</var> is the <code>VAR_DECL</code> tree node representing the variable.
</p>
<p>If this macro is not defined, then the variable name is defined in the
usual manner as a label (by means of <code>ASM_OUTPUT_LABEL</code>).
</p>
<p>You may wish to use <code>ASM_OUTPUT_TYPE_DIRECTIVE</code> and/or
<code>ASM_OUTPUT_SIZE_DIRECTIVE</code> in the definition of this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fDECLARE_005fCONSTANT_005fNAME"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_DECLARE_CONSTANT_NAME</strong> <em>(FILE *<var>file</var>, const char *<var>name</var>, const_tree <var>expr</var>, HOST_WIDE_INT <var>size</var>)</em></dt>
<dd><p>A target hook to output to the stdio stream <var>file</var> any text necessary
for declaring the name <var>name</var> of a constant which is being defined.  This
target hook is responsible for outputting the label definition (perhaps using
<code>assemble_label</code>).  The argument <var>exp</var> is the value of the constant,
and <var>size</var> is the size of the constant in bytes.  The <var>name</var>
will be an internal label.
</p>
<p>The default version of this target hook, define the <var>name</var> in the
usual manner as a label (by means of <code>assemble_label</code>).
</p>
<p>You may wish to use <code>ASM_OUTPUT_TYPE_DIRECTIVE</code> in this target hook.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fDECLARE_005fREGISTER_005fGLOBAL"></a>Macro: <strong>ASM_DECLARE_REGISTER_GLOBAL</strong> <em>(<var>stream</var>, <var>decl</var>, <var>regno</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> any text necessary for claiming a register <var>regno</var>
for a global variable <var>decl</var> with name <var>name</var>.
</p>
<p>If you don&rsquo;t define this macro, that is equivalent to defining it to do
nothing.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fFINISH_005fDECLARE_005fOBJECT"></a>Macro: <strong>ASM_FINISH_DECLARE_OBJECT</strong> <em>(<var>stream</var>, <var>decl</var>, <var>toplevel</var>, <var>atend</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to finish up declaring a variable name
once the compiler has processed its initializer fully and thus has had a
chance to determine the size of an array when controlled by an
initializer.  This is used on systems where it&rsquo;s necessary to declare
something about the size of the object.
</p>
<p>If you don&rsquo;t define this macro, that is equivalent to defining it to do
nothing.
</p>
<p>You may wish to use <code>ASM_OUTPUT_SIZE_DIRECTIVE</code> and/or
<code>ASM_OUTPUT_MEASURED_SIZE</code> in the definition of this macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fGLOBALIZE_005fLABEL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_GLOBALIZE_LABEL</strong> <em>(FILE *<var>stream</var>, const char *<var>name</var>)</em></dt>
<dd><p>This target hook is a function to output to the stdio stream
<var>stream</var> some commands that will make the label <var>name</var> global;
that is, available for reference from other files.
</p>
<p>The default implementation relies on a proper definition of
<code>GLOBAL_ASM_OP</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fGLOBALIZE_005fDECL_005fNAME"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_GLOBALIZE_DECL_NAME</strong> <em>(FILE *<var>stream</var>, tree <var>decl</var>)</em></dt>
<dd><p>This target hook is a function to output to the stdio stream
<var>stream</var> some commands that will make the name associated with <var>decl</var>
global; that is, available for reference from other files.
</p>
<p>The default implementation uses the TARGET_ASM_GLOBALIZE_LABEL target hook.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fWEAKEN_005fLABEL"></a>Macro: <strong>ASM_WEAKEN_LABEL</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> some commands that will make the label <var>name</var> weak;
that is, available for reference from other files but only used if
no other definition is available.  Use the expression
<code>assemble_name (<var>stream</var>, <var>name</var>)</code> to output the name
itself; before and after that, output the additional assembler syntax
for making that name weak, and a newline.
</p>
<p>If you don&rsquo;t define this macro or <code>ASM_WEAKEN_DECL</code>, GCC will not
support weak symbols and you should not define the <code>SUPPORTS_WEAK</code>
macro.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fWEAKEN_005fDECL"></a>Macro: <strong>ASM_WEAKEN_DECL</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>, <var>value</var>)</em></dt>
<dd><p>Combines (and replaces) the function of <code>ASM_WEAKEN_LABEL</code> and
<code>ASM_OUTPUT_WEAK_ALIAS</code>, allowing access to the associated function
or variable decl.  If <var>value</var> is not <code>NULL</code>, this C statement
should output to the stdio stream <var>stream</var> assembler code which
defines (equates) the weak symbol <var>name</var> to have the value
<var>value</var>.  If <var>value</var> is <code>NULL</code>, it should output commands
to make <var>name</var> weak.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fWEAKREF"></a>Macro: <strong>ASM_OUTPUT_WEAKREF</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>, <var>value</var>)</em></dt>
<dd><p>Outputs a directive that enables <var>name</var> to be used to refer to
symbol <var>value</var> with weak-symbol semantics.  <code>decl</code> is the
declaration of <code>name</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SUPPORTS_005fWEAK"></a>Macro: <strong>SUPPORTS_WEAK</strong></dt>
<dd><p>A preprocessor constant expression which evaluates to true if the target
supports weak symbols.
</p>
<p>If you don&rsquo;t define this macro, <samp>defaults.h</samp> provides a default
definition.  If either <code>ASM_WEAKEN_LABEL</code> or <code>ASM_WEAKEN_DECL</code>
is defined, the default definition is &lsquo;<samp>1</samp>&rsquo;; otherwise, it is &lsquo;<samp>0</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSUPPORTS_005fWEAK"></a>Macro: <strong>TARGET_SUPPORTS_WEAK</strong></dt>
<dd><p>A C expression which evaluates to true if the target supports weak symbols.
</p>
<p>If you don&rsquo;t define this macro, <samp>defaults.h</samp> provides a default
definition.  The default definition is &lsquo;<samp>(SUPPORTS_WEAK)</samp>&rsquo;.  Define
this macro if you want to control weak symbol support with a compiler
flag such as <samp>-melf</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-MAKE_005fDECL_005fONE_005fONLY"></a>Macro: <strong>MAKE_DECL_ONE_ONLY</strong> <em>(<var>decl</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to mark <var>decl</var> to be emitted as a
public symbol such that extra copies in multiple translation units will
be discarded by the linker.  Define this macro if your object file
format provides support for this concept, such as the &lsquo;<samp>COMDAT</samp>&rsquo;
section flags in the Microsoft Windows PE/COFF format, and this support
requires changes to <var>decl</var>, such as putting it in a separate section.
</p></dd></dl>

<dl>
<dt><a name="index-SUPPORTS_005fONE_005fONLY"></a>Macro: <strong>SUPPORTS_ONE_ONLY</strong></dt>
<dd><p>A C expression which evaluates to true if the target supports one-only
semantics.
</p>
<p>If you don&rsquo;t define this macro, <samp>varasm.c</samp> provides a default
definition.  If <code>MAKE_DECL_ONE_ONLY</code> is defined, the default
definition is &lsquo;<samp>1</samp>&rsquo;; otherwise, it is &lsquo;<samp>0</samp>&rsquo;.  Define this macro if
you want to control one-only symbol support with a compiler flag, or if
setting the <code>DECL_ONE_ONLY</code> flag is enough to mark a declaration to
be emitted as one-only.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fASSEMBLE_005fVISIBILITY"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_ASSEMBLE_VISIBILITY</strong> <em>(tree <var>decl</var>, int <var>visibility</var>)</em></dt>
<dd><p>This target hook is a function to output to <var>asm_out_file</var> some
commands that will make the symbol(s) associated with <var>decl</var> have
hidden, protected or internal visibility as specified by <var>visibility</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fWEAK_005fNOT_005fIN_005fARCHIVE_005fTOC"></a>Macro: <strong>TARGET_WEAK_NOT_IN_ARCHIVE_TOC</strong></dt>
<dd><p>A C expression that evaluates to true if the target&rsquo;s linker expects
that weak symbols do not appear in a static archive&rsquo;s table of contents.
The default is <code>0</code>.
</p>
<p>Leaving weak symbols out of an archive&rsquo;s table of contents means that,
if a symbol will only have a definition in one translation unit and
will have undefined references from other translation units, that
symbol should not be weak.  Defining this macro to be nonzero will
thus have the effect that certain symbols that would normally be weak
(explicit template instantiations, and vtables for polymorphic classes
with noninline key methods) will instead be nonweak.
</p>
<p>The C++ ABI requires this macro to be zero.  Define this macro for
targets where full C++ ABI compliance is impossible and where linker
restrictions require weak symbols to be left out of a static archive&rsquo;s
table of contents.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fEXTERNAL"></a>Macro: <strong>ASM_OUTPUT_EXTERNAL</strong> <em>(<var>stream</var>, <var>decl</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> any text necessary for declaring the name of an external
symbol named <var>name</var> which is referenced in this compilation but
not defined.  The value of <var>decl</var> is the tree node for the
declaration.
</p>
<p>This macro need not be defined if it does not need to output anything.
The GNU assembler and most Unix assemblers don&rsquo;t require anything.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fEXTERNAL_005fLIBCALL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_EXTERNAL_LIBCALL</strong> <em>(rtx <var>symref</var>)</em></dt>
<dd><p>This target hook is a function to output to <var>asm_out_file</var> an assembler
pseudo-op to declare a library function name external.  The name of the
library function is given by <var>symref</var>, which is a <code>symbol_ref</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fMARK_005fDECL_005fPRESERVED"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_MARK_DECL_PRESERVED</strong> <em>(const char *<var>symbol</var>)</em></dt>
<dd><p>This target hook is a function to output to <var>asm_out_file</var> an assembler
directive to annotate <var>symbol</var> as used.  The Darwin target uses the
.no_dead_code_strip directive.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fLABELREF"></a>Macro: <strong>ASM_OUTPUT_LABELREF</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output to the stdio stream
<var>stream</var> a reference in assembler syntax to a label named
<var>name</var>.  This should add &lsquo;<samp>_</samp>&rsquo; to the front of the name, if that
is customary on your operating system, as it is in most Berkeley Unix
systems.  This macro is used in <code>assemble_name</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMANGLE_005fASSEMBLER_005fNAME"></a>Target Hook: <em>tree</em> <strong>TARGET_MANGLE_ASSEMBLER_NAME</strong> <em>(const char *<var>name</var>)</em></dt>
<dd><p>Given a symbol <var>name</var>, perform same mangling as <code>varasm.c</code>&rsquo;s <code>assemble_name</code>, but in memory rather than to a file stream, returning result as an <code>IDENTIFIER_NODE</code>.  Required for correct LTO symtabs.  The default implementation calls the <code>TARGET_STRIP_NAME_ENCODING</code> hook and then prepends the <code>USER_LABEL_PREFIX</code>, if any.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fSYMBOL_005fREF"></a>Macro: <strong>ASM_OUTPUT_SYMBOL_REF</strong> <em>(<var>stream</var>, <var>sym</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output a reference to
<code>SYMBOL_REF</code> <var>sym</var>.  If not defined, <code>assemble_name</code>
will be used to output the name of the symbol.  This macro may be used
to modify the way a symbol is referenced depending on information
encoded by <code>TARGET_ENCODE_SECTION_INFO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fLABEL_005fREF"></a>Macro: <strong>ASM_OUTPUT_LABEL_REF</strong> <em>(<var>stream</var>, <var>buf</var>)</em></dt>
<dd><p>A C statement (sans semicolon) to output a reference to <var>buf</var>, the
result of <code>ASM_GENERATE_INTERNAL_LABEL</code>.  If not defined,
<code>assemble_name</code> will be used to output the name of the symbol.
This macro is not used by <code>output_asm_label</code>, or the <code>%l</code>
specifier that calls it; the intention is that this macro should be set
when it is necessary to output a label differently when its address is
being taken.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fINTERNAL_005fLABEL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_INTERNAL_LABEL</strong> <em>(FILE *<var>stream</var>, const char *<var>prefix</var>, unsigned long <var>labelno</var>)</em></dt>
<dd><p>A function to output to the stdio stream <var>stream</var> a label whose
name is made from the string <var>prefix</var> and the number <var>labelno</var>.
</p>
<p>It is absolutely essential that these labels be distinct from the labels
used for user-level functions and variables.  Otherwise, certain programs
will have name conflicts with internal labels.
</p>
<p>It is desirable to exclude internal labels from the symbol table of the
object file.  Most assemblers have a naming convention for labels that
should be excluded; on many systems, the letter &lsquo;<samp>L</samp>&rsquo; at the
beginning of a label has this effect.  You should find out what
convention your system uses, and follow it.
</p>
<p>The default version of this function utilizes <code>ASM_GENERATE_INTERNAL_LABEL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDEBUG_005fLABEL"></a>Macro: <strong>ASM_OUTPUT_DEBUG_LABEL</strong> <em>(<var>stream</var>, <var>prefix</var>, <var>num</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> a debug info
label whose name is made from the string <var>prefix</var> and the number
<var>num</var>.  This is useful for VLIW targets, where debug info labels
may need to be treated differently than branch target labels.  On some
systems, branch target labels must be at the beginning of instruction
bundles, but debug info labels can occur in the middle of instruction
bundles.
</p>
<p>If this macro is not defined, then <code>(*targetm.asm_out.internal_label)</code> will be
used.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fGENERATE_005fINTERNAL_005fLABEL"></a>Macro: <strong>ASM_GENERATE_INTERNAL_LABEL</strong> <em>(<var>string</var>, <var>prefix</var>, <var>num</var>)</em></dt>
<dd><p>A C statement to store into the string <var>string</var> a label whose name
is made from the string <var>prefix</var> and the number <var>num</var>.
</p>
<p>This string, when output subsequently by <code>assemble_name</code>, should
produce the output that <code>(*targetm.asm_out.internal_label)</code> would produce
with the same <var>prefix</var> and <var>num</var>.
</p>
<p>If the string begins with &lsquo;<samp>*</samp>&rsquo;, then <code>assemble_name</code> will
output the rest of the string unchanged.  It is often convenient for
<code>ASM_GENERATE_INTERNAL_LABEL</code> to use &lsquo;<samp>*</samp>&rsquo; in this way.  If the
string doesn&rsquo;t start with &lsquo;<samp>*</samp>&rsquo;, then <code>ASM_OUTPUT_LABELREF</code> gets
to output the string, and may change it.  (Of course,
<code>ASM_OUTPUT_LABELREF</code> is also part of your machine description, so
you should know what it does on your machine.)
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fFORMAT_005fPRIVATE_005fNAME"></a>Macro: <strong>ASM_FORMAT_PRIVATE_NAME</strong> <em>(<var>outvar</var>, <var>name</var>, <var>number</var>)</em></dt>
<dd><p>A C expression to assign to <var>outvar</var> (which is a variable of type
<code>char *</code>) a newly allocated string made from the string
<var>name</var> and the number <var>number</var>, with some suitable punctuation
added.  Use <code>alloca</code> to get space for the string.
</p>
<p>The string will be used as an argument to <code>ASM_OUTPUT_LABELREF</code> to
produce an assembler label for an internal static variable whose name is
<var>name</var>.  Therefore, the string must be such as to result in valid
assembler code.  The argument <var>number</var> is different each time this
macro is executed; it prevents conflicts between similarly-named
internal static variables in different scopes.
</p>
<p>Ideally this string should not be a valid C identifier, to prevent any
conflict with the user&rsquo;s own symbols.  Most assemblers allow periods
or percent signs in assembler symbols; putting at least one of these
between the name and the number will suffice.
</p>
<p>If this macro is not defined, a default definition will be provided
which is correct for most systems.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDEF"></a>Macro: <strong>ASM_OUTPUT_DEF</strong> <em>(<var>stream</var>, <var>name</var>, <var>value</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> assembler code
which defines (equates) the symbol <var>name</var> to have the value <var>value</var>.
</p>
<a name="index-SET_005fASM_005fOP"></a>
<p>If <code>SET_ASM_OP</code> is defined, a default definition is provided which is
correct for most systems.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDEF_005fFROM_005fDECLS"></a>Macro: <strong>ASM_OUTPUT_DEF_FROM_DECLS</strong> <em>(<var>stream</var>, <var>decl_of_name</var>, <var>decl_of_value</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> assembler code
which defines (equates) the symbol whose tree node is <var>decl_of_name</var>
to have the value of the tree node <var>decl_of_value</var>.  This macro will
be used in preference to &lsquo;<samp>ASM_OUTPUT_DEF</samp>&rsquo; if it is defined and if
the tree nodes are available.
</p>
<a name="index-SET_005fASM_005fOP-1"></a>
<p>If <code>SET_ASM_OP</code> is defined, a default definition is provided which is
correct for most systems.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDEFERRED_005fOUTPUT_005fDEFS"></a>Macro: <strong>TARGET_DEFERRED_OUTPUT_DEFS</strong> <em>(<var>decl_of_name</var>, <var>decl_of_value</var>)</em></dt>
<dd><p>A C statement that evaluates to true if the assembler code which defines
(equates) the symbol whose tree node is <var>decl_of_name</var> to have the value
of the tree node <var>decl_of_value</var> should be emitted near the end of the
current compilation unit.  The default is to not defer output of defines.
This macro affects defines output by &lsquo;<samp>ASM_OUTPUT_DEF</samp>&rsquo; and
&lsquo;<samp>ASM_OUTPUT_DEF_FROM_DECLS</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fWEAK_005fALIAS"></a>Macro: <strong>ASM_OUTPUT_WEAK_ALIAS</strong> <em>(<var>stream</var>, <var>name</var>, <var>value</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> assembler code
which defines (equates) the weak symbol <var>name</var> to have the value
<var>value</var>.  If <var>value</var> is <code>NULL</code>, it defines <var>name</var> as
an undefined weak symbol.
</p>
<p>Define this macro if the target only supports weak aliases; define
<code>ASM_OUTPUT_DEF</code> instead if possible.
</p></dd></dl>

<dl>
<dt><a name="index-OBJC_005fGEN_005fMETHOD_005fLABEL"></a>Macro: <strong>OBJC_GEN_METHOD_LABEL</strong> <em>(<var>buf</var>, <var>is_inst</var>, <var>class_name</var>, <var>cat_name</var>, <var>sel_name</var>)</em></dt>
<dd><p>Define this macro to override the default assembler names used for
Objective-C methods.
</p>
<p>The default name is a unique method number followed by the name of the
class (e.g. &lsquo;<samp>_1_Foo</samp>&rsquo;).  For methods in categories, the name of
the category is also included in the assembler name (e.g.
&lsquo;<samp>_1_Foo_Bar</samp>&rsquo;).
</p>
<p>These names are safe on most systems, but make debugging difficult since
the method&rsquo;s selector is not present in the name.  Therefore, particular
systems define other ways of computing names.
</p>
<p><var>buf</var> is an expression of type <code>char *</code> which gives you a
buffer in which to store the name; its length is as long as
<var>class_name</var>, <var>cat_name</var> and <var>sel_name</var> put together, plus
50 characters extra.
</p>
<p>The argument <var>is_inst</var> specifies whether the method is an instance
method or a class method; <var>class_name</var> is the name of the class;
<var>cat_name</var> is the name of the category (or <code>NULL</code> if the method is not
in a category); and <var>sel_name</var> is the name of the selector.
</p>
<p>On systems where the assembler can handle quoted names, you can use this
macro to provide more human-readable names.
</p></dd></dl>

<hr>
<a name="Initialization"></a>
<div class="header">
<p>
Next: <a href="#Macros-for-Initialization" accesskey="n" rel="next">Macros for Initialization</a>, Previous: <a href="#Label-Output" accesskey="p" rel="prev">Label Output</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="How-Initialization-Functions-Are-Handled"></a>
<h4 class="subsection">17.21.5 How Initialization Functions Are Handled</h4>
<a name="index-initialization-routines"></a>
<a name="index-termination-routines"></a>
<a name="index-constructors_002c-output-of"></a>
<a name="index-destructors_002c-output-of"></a>

<p>The compiled code for certain languages includes <em>constructors</em>
(also called <em>initialization routines</em>)&mdash;functions to initialize
data in the program when the program is started.  These functions need
to be called before the program is &ldquo;started&rdquo;&mdash;that is to say, before
<code>main</code> is called.
</p>
<p>Compiling some languages generates <em>destructors</em> (also called
<em>termination routines</em>) that should be called when the program
terminates.
</p>
<p>To make the initialization and termination functions work, the compiler
must output something in the assembler code to cause those functions to
be called at the appropriate time.  When you port the compiler to a new
system, you need to specify how to do this.
</p>
<p>There are two major ways that GCC currently supports the execution of
initialization and termination functions.  Each way has two variants.
Much of the structure is common to all four variations.
</p>
<a name="index-_005f_005fCTOR_005fLIST_005f_005f"></a>
<a name="index-_005f_005fDTOR_005fLIST_005f_005f"></a>
<p>The linker must build two lists of these functions&mdash;a list of
initialization functions, called <code>__CTOR_LIST__</code>, and a list of
termination functions, called <code>__DTOR_LIST__</code>.
</p>
<p>Each list always begins with an ignored function pointer (which may hold
0, -1, or a count of the function pointers after it, depending on
the environment).  This is followed by a series of zero or more function
pointers to constructors (or destructors), followed by a function
pointer containing zero.
</p>
<p>Depending on the operating system and its executable file format, either
<samp>crtstuff.c</samp> or <samp>libgcc2.c</samp> traverses these lists at startup
time and exit time.  Constructors are called in reverse order of the
list; destructors in forward order.
</p>
<p>The best way to handle static constructors works only for object file
formats which provide arbitrarily-named sections.  A section is set
aside for a list of constructors, and another for a list of destructors.
Traditionally these are called &lsquo;<samp>.ctors</samp>&rsquo; and &lsquo;<samp>.dtors</samp>&rsquo;.  Each
object file that defines an initialization function also puts a word in
the constructor section to point to that function.  The linker
accumulates all these words into one contiguous &lsquo;<samp>.ctors</samp>&rsquo; section.
Termination functions are handled similarly.
</p>
<p>This method will be chosen as the default by <samp>target-def.h</samp> if
<code>TARGET_ASM_NAMED_SECTION</code> is defined.  A target that does not
support arbitrary sections, but does support special designated
constructor and destructor sections may define <code>CTORS_SECTION_ASM_OP</code>
and <code>DTORS_SECTION_ASM_OP</code> to achieve the same effect.
</p>
<p>When arbitrary sections are available, there are two variants, depending
upon how the code in <samp>crtstuff.c</samp> is called.  On systems that
support a <em>.init</em> section which is executed at program startup,
parts of <samp>crtstuff.c</samp> are compiled into that section.  The
program is linked by the <code>gcc</code> driver like this:
</p>
<div class="smallexample">
<pre class="smallexample">ld -o <var>output_file</var> crti.o crtbegin.o &hellip; -lgcc crtend.o crtn.o
</pre></div>

<p>The prologue of a function (<code>__init</code>) appears in the <code>.init</code>
section of <samp>crti.o</samp>; the epilogue appears in <samp>crtn.o</samp>.  Likewise
for the function <code>__fini</code> in the <em>.fini</em> section.  Normally these
files are provided by the operating system or by the GNU C library, but
are provided by GCC for a few targets.
</p>
<p>The objects <samp>crtbegin.o</samp> and <samp>crtend.o</samp> are (for most targets)
compiled from <samp>crtstuff.c</samp>.  They contain, among other things, code
fragments within the <code>.init</code> and <code>.fini</code> sections that branch
to routines in the <code>.text</code> section.  The linker will pull all parts
of a section together, which results in a complete <code>__init</code> function
that invokes the routines we need at startup.
</p>
<p>To use this variant, you must define the <code>INIT_SECTION_ASM_OP</code>
macro properly.
</p>
<p>If no init section is available, when GCC compiles any function called
<code>main</code> (or more accurately, any function designated as a program
entry point by the language front end calling <code>expand_main_function</code>),
it inserts a procedure call to <code>__main</code> as the first executable code
after the function prologue.  The <code>__main</code> function is defined
in <samp>libgcc2.c</samp> and runs the global constructors.
</p>
<p>In file formats that don&rsquo;t support arbitrary sections, there are again
two variants.  In the simplest variant, the GNU linker (GNU <code>ld</code>)
and an &lsquo;a.out&rsquo; format must be used.  In this case,
<code>TARGET_ASM_CONSTRUCTOR</code> is defined to produce a <code>.stabs</code>
entry of type &lsquo;<samp>N_SETT</samp>&rsquo;, referencing the name <code>__CTOR_LIST__</code>,
and with the address of the void function containing the initialization
code as its value.  The GNU linker recognizes this as a request to add
the value to a <em>set</em>; the values are accumulated, and are eventually
placed in the executable as a vector in the format described above, with
a leading (ignored) count and a trailing zero element.
<code>TARGET_ASM_DESTRUCTOR</code> is handled similarly.  Since no init
section is available, the absence of <code>INIT_SECTION_ASM_OP</code> causes
the compilation of <code>main</code> to call <code>__main</code> as above, starting
the initialization process.
</p>
<p>The last variant uses neither arbitrary sections nor the GNU linker.
This is preferable when you want to do dynamic linking and when using
file formats which the GNU linker does not support, such as &lsquo;ECOFF&rsquo;.  In
this case, <code>TARGET_HAVE_CTORS_DTORS</code> is false, initialization and
termination functions are recognized simply by their names.  This requires
an extra program in the linkage step, called <code>collect2</code>.  This program
pretends to be the linker, for use with GCC; it does its job by running
the ordinary linker, but also arranges to include the vectors of
initialization and termination functions.  These functions are called
via <code>__main</code> as described above.  In order to use this method,
<code>use_collect2</code> must be defined in the target in <samp>config.gcc</samp>.
</p>

<hr>
<a name="Macros-for-Initialization"></a>
<div class="header">
<p>
Next: <a href="#Instruction-Output" accesskey="n" rel="next">Instruction Output</a>, Previous: <a href="#Initialization" accesskey="p" rel="prev">Initialization</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Macros-Controlling-Initialization-Routines"></a>
<h4 class="subsection">17.21.6 Macros Controlling Initialization Routines</h4>

<p>Here are the macros that control how the compiler handles initialization
and termination functions:
</p>
<dl>
<dt><a name="index-INIT_005fSECTION_005fASM_005fOP-1"></a>Macro: <strong>INIT_SECTION_ASM_OP</strong></dt>
<dd><p>If defined, a C string constant, including spacing, for the assembler
operation to identify the following data as initialization code.  If not
defined, GCC will assume such a section does not exist.  When you are
using special sections for initialization and termination functions, this
macro also controls how <samp>crtstuff.c</samp> and <samp>libgcc2.c</samp> arrange to
run the initialization functions.
</p></dd></dl>

<dl>
<dt><a name="index-HAS_005fINIT_005fSECTION"></a>Macro: <strong>HAS_INIT_SECTION</strong></dt>
<dd><p>If defined, <code>main</code> will not call <code>__main</code> as described above.
This macro should be defined for systems that control start-up code
on a symbol-by-symbol basis, such as OSF/1, and should not
be defined explicitly for systems that support <code>INIT_SECTION_ASM_OP</code>.
</p></dd></dl>

<dl>
<dt><a name="index-LD_005fINIT_005fSWITCH"></a>Macro: <strong>LD_INIT_SWITCH</strong></dt>
<dd><p>If defined, a C string constant for a switch that tells the linker that
the following symbol is an initialization routine.
</p></dd></dl>

<dl>
<dt><a name="index-LD_005fFINI_005fSWITCH"></a>Macro: <strong>LD_FINI_SWITCH</strong></dt>
<dd><p>If defined, a C string constant for a switch that tells the linker that
the following symbol is a finalization routine.
</p></dd></dl>

<dl>
<dt><a name="index-COLLECT_005fSHARED_005fINIT_005fFUNC"></a>Macro: <strong>COLLECT_SHARED_INIT_FUNC</strong> <em>(<var>stream</var>, <var>func</var>)</em></dt>
<dd><p>If defined, a C statement that will write a function that can be
automatically called when a shared library is loaded.  The function
should call <var>func</var>, which takes no arguments.  If not defined, and
the object format requires an explicit initialization function, then a
function called <code>_GLOBAL__DI</code> will be generated.
</p>
<p>This function and the following one are used by collect2 when linking a
shared library that needs constructors or destructors, or has DWARF2
exception tables embedded in the code.
</p></dd></dl>

<dl>
<dt><a name="index-COLLECT_005fSHARED_005fFINI_005fFUNC"></a>Macro: <strong>COLLECT_SHARED_FINI_FUNC</strong> <em>(<var>stream</var>, <var>func</var>)</em></dt>
<dd><p>If defined, a C statement that will write a function that can be
automatically called when a shared library is unloaded.  The function
should call <var>func</var>, which takes no arguments.  If not defined, and
the object format requires an explicit finalization function, then a
function called <code>_GLOBAL__DD</code> will be generated.
</p></dd></dl>

<dl>
<dt><a name="index-INVOKE_005f_005fmain"></a>Macro: <strong>INVOKE__main</strong></dt>
<dd><p>If defined, <code>main</code> will call <code>__main</code> despite the presence of
<code>INIT_SECTION_ASM_OP</code>.  This macro should be defined for systems
where the init section is not actually run automatically, but is still
useful for collecting the lists of constructors and destructors.
</p></dd></dl>

<dl>
<dt><a name="index-SUPPORTS_005fINIT_005fPRIORITY"></a>Macro: <strong>SUPPORTS_INIT_PRIORITY</strong></dt>
<dd><p>If nonzero, the C++ <code>init_priority</code> attribute is supported and the
compiler should emit instructions to control the order of initialization
of objects.  If zero, the compiler will issue an error message upon
encountering an <code>init_priority</code> attribute.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHAVE_005fCTORS_005fDTORS"></a>Target Hook: <em>bool</em> <strong>TARGET_HAVE_CTORS_DTORS</strong></dt>
<dd><p>This value is true if the target supports some &ldquo;native&rdquo; method of
collecting constructors and destructors to be run at startup and exit.
It is false if we must use <code>collect2</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fCONSTRUCTOR"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_CONSTRUCTOR</strong> <em>(rtx <var>symbol</var>, int <var>priority</var>)</em></dt>
<dd><p>If defined, a function that outputs assembler code to arrange to call
the function referenced by <var>symbol</var> at initialization time.
</p>
<p>Assume that <var>symbol</var> is a <code>SYMBOL_REF</code> for a function taking
no arguments and with no return value.  If the target supports initialization
priorities, <var>priority</var> is a value between 0 and <code>MAX_INIT_PRIORITY</code>;
otherwise it must be <code>DEFAULT_INIT_PRIORITY</code>.
</p>
<p>If this macro is not defined by the target, a suitable default will
be chosen if (1) the target supports arbitrary section names, (2) the
target defines <code>CTORS_SECTION_ASM_OP</code>, or (3) <code>USE_COLLECT2</code>
is not defined.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fDESTRUCTOR"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_DESTRUCTOR</strong> <em>(rtx <var>symbol</var>, int <var>priority</var>)</em></dt>
<dd><p>This is like <code>TARGET_ASM_CONSTRUCTOR</code> but used for termination
functions rather than initialization functions.
</p></dd></dl>

<p>If <code>TARGET_HAVE_CTORS_DTORS</code> is true, the initialization routine
generated for the generated object file will have static linkage.
</p>
<p>If your system uses <code>collect2</code> as the means of processing
constructors, then that program normally uses <code>nm</code> to scan
an object file for constructor functions to be called.
</p>
<p>On certain kinds of systems, you can define this macro to make
<code>collect2</code> work faster (and, in some cases, make it work at all):
</p>
<dl>
<dt><a name="index-OBJECT_005fFORMAT_005fCOFF"></a>Macro: <strong>OBJECT_FORMAT_COFF</strong></dt>
<dd><p>Define this macro if the system uses COFF (Common Object File Format)
object files, so that <code>collect2</code> can assume this format and scan
object files directly for dynamic constructor/destructor functions.
</p>
<p>This macro is effective only in a native compiler; <code>collect2</code> as
part of a cross compiler always uses <code>nm</code> for the target machine.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fNM_005fFILE_005fNAME"></a>Macro: <strong>REAL_NM_FILE_NAME</strong></dt>
<dd><p>Define this macro as a C string constant containing the file name to use
to execute <code>nm</code>.  The default is to search the path normally for
<code>nm</code>.
</p></dd></dl>

<dl>
<dt><a name="index-NM_005fFLAGS"></a>Macro: <strong>NM_FLAGS</strong></dt>
<dd><p><code>collect2</code> calls <code>nm</code> to scan object files for static
constructors and destructors and LTO info.  By default, <samp>-n</samp> is
passed.  Define <code>NM_FLAGS</code> to a C string constant if other options
are needed to get the same output format as GNU <code>nm -n</code>
produces.
</p></dd></dl>

<p>If your system supports shared libraries and has a program to list the
dynamic dependencies of a given library or executable, you can define
these macros to enable support for running initialization and
termination functions in shared libraries:
</p>
<dl>
<dt><a name="index-LDD_005fSUFFIX"></a>Macro: <strong>LDD_SUFFIX</strong></dt>
<dd><p>Define this macro to a C string constant containing the name of the program
which lists dynamic dependencies, like <code>ldd</code> under SunOS 4.
</p></dd></dl>

<dl>
<dt><a name="index-PARSE_005fLDD_005fOUTPUT"></a>Macro: <strong>PARSE_LDD_OUTPUT</strong> <em>(<var>ptr</var>)</em></dt>
<dd><p>Define this macro to be C code that extracts filenames from the output
of the program denoted by <code>LDD_SUFFIX</code>.  <var>ptr</var> is a variable
of type <code>char *</code> that points to the beginning of a line of output
from <code>LDD_SUFFIX</code>.  If the line lists a dynamic dependency, the
code must advance <var>ptr</var> to the beginning of the filename on that
line.  Otherwise, it must set <var>ptr</var> to <code>NULL</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SHLIB_005fSUFFIX"></a>Macro: <strong>SHLIB_SUFFIX</strong></dt>
<dd><p>Define this macro to a C string constant containing the default shared
library extension of the target (e.g., &lsquo;<samp>&quot;.so&quot;</samp>&rsquo;).  <code>collect2</code>
strips version information after this suffix when generating global
constructor and destructor names.  This define is only needed on targets
that use <code>collect2</code> to process constructors and destructors.
</p></dd></dl>

<hr>
<a name="Instruction-Output"></a>
<div class="header">
<p>
Next: <a href="#Dispatch-Tables" accesskey="n" rel="next">Dispatch Tables</a>, Previous: <a href="#Macros-for-Initialization" accesskey="p" rel="prev">Macros for Initialization</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-of-Assembler-Instructions"></a>
<h4 class="subsection">17.21.7 Output of Assembler Instructions</h4>

<p>This describes assembler instruction output.
</p>
<dl>
<dt><a name="index-REGISTER_005fNAMES"></a>Macro: <strong>REGISTER_NAMES</strong></dt>
<dd><p>A C initializer containing the assembler&rsquo;s names for the machine
registers, each one as a C string constant.  This is what translates
register numbers in the compiler into assembler language.
</p></dd></dl>

<dl>
<dt><a name="index-ADDITIONAL_005fREGISTER_005fNAMES"></a>Macro: <strong>ADDITIONAL_REGISTER_NAMES</strong></dt>
<dd><p>If defined, a C initializer for an array of structures containing a name
and a register number.  This macro defines additional names for hard
registers, thus allowing the <code>asm</code> option in declarations to refer
to registers using alternate names.
</p></dd></dl>

<dl>
<dt><a name="index-OVERLAPPING_005fREGISTER_005fNAMES"></a>Macro: <strong>OVERLAPPING_REGISTER_NAMES</strong></dt>
<dd><p>If defined, a C initializer for an array of structures containing a
name, a register number and a count of the number of consecutive
machine registers the name overlaps.  This macro defines additional
names for hard registers, thus allowing the <code>asm</code> option in
declarations to refer to registers using alternate names.  Unlike
<code>ADDITIONAL_REGISTER_NAMES</code>, this macro should be used when the
register name implies multiple underlying registers.
</p>
<p>This macro should be used when it is important that a clobber in an
<code>asm</code> statement clobbers all the underlying values implied by the
register name.  For example, on ARM, clobbering the double-precision
VFP register &ldquo;d0&rdquo; implies clobbering both single-precision registers
&ldquo;s0&rdquo; and &ldquo;s1&rdquo;.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fOPCODE"></a>Macro: <strong>ASM_OUTPUT_OPCODE</strong> <em>(<var>stream</var>, <var>ptr</var>)</em></dt>
<dd><p>Define this macro if you are using an unusual assembler that
requires different names for the machine instructions.
</p>
<p>The definition is a C statement or statements which output an
assembler instruction opcode to the stdio stream <var>stream</var>.  The
macro-operand <var>ptr</var> is a variable of type <code>char *</code> which
points to the opcode name in its &ldquo;internal&rdquo; form&mdash;the form that is
written in the machine description.  The definition should output the
opcode name to <var>stream</var>, performing any translation you desire, and
increment the variable <var>ptr</var> to point at the end of the opcode
so that it will not be output twice.
</p>
<p>In fact, your macro definition may process less than the entire opcode
name, or more than the opcode name; but if you want to process text
that includes &lsquo;<samp>%</samp>&rsquo;-sequences to substitute operands, you must take
care of the substitution yourself.  Just be sure to increment
<var>ptr</var> over whatever text should not be output normally.
</p>
<a name="index-recog_005fdata_002eoperand"></a>
<p>If you need to look at the operand values, they can be found as the
elements of <code>recog_data.operand</code>.
</p>
<p>If the macro definition does nothing, the instruction is output
in the usual way.
</p></dd></dl>

<dl>
<dt><a name="index-FINAL_005fPRESCAN_005fINSN"></a>Macro: <strong>FINAL_PRESCAN_INSN</strong> <em>(<var>insn</var>, <var>opvec</var>, <var>noperands</var>)</em></dt>
<dd><p>If defined, a C statement to be executed just prior to the output of
assembler code for <var>insn</var>, to modify the extracted operands so
they will be output differently.
</p>
<p>Here the argument <var>opvec</var> is the vector containing the operands
extracted from <var>insn</var>, and <var>noperands</var> is the number of
elements of the vector which contain meaningful data for this insn.
The contents of this vector are what will be used to convert the insn
template into assembler code, so you can change the assembler output
by changing the contents of the vector.
</p>
<p>This macro is useful when various assembler syntaxes share a single
file of instruction patterns; by defining this macro differently, you
can cause a large class of instructions to be output differently (such
as with rearranged operands).  Naturally, variations in assembler
syntax affecting individual insn patterns ought to be handled by
writing conditional output routines in those patterns.
</p>
<p>If this macro is not defined, it is equivalent to a null statement.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fFINAL_005fPOSTSCAN_005fINSN"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_FINAL_POSTSCAN_INSN</strong> <em>(FILE *<var>file</var>, rtx <var>insn</var>, rtx *<var>opvec</var>, int <var>noperands</var>)</em></dt>
<dd><p>If defined, this target hook is a function which is executed just after the
output of assembler code for <var>insn</var>, to change the mode of the assembler
if necessary.
</p>
<p>Here the argument <var>opvec</var> is the vector containing the operands
extracted from <var>insn</var>, and <var>noperands</var> is the number of
elements of the vector which contain meaningful data for this insn.
The contents of this vector are what was used to convert the insn
template into assembler code, so you can change the assembler mode
by checking the contents of the vector.
</p></dd></dl>

<dl>
<dt><a name="index-PRINT_005fOPERAND"></a>Macro: <strong>PRINT_OPERAND</strong> <em>(<var>stream</var>, <var>x</var>, <var>code</var>)</em></dt>
<dd><p>A C compound statement to output to stdio stream <var>stream</var> the
assembler syntax for an instruction operand <var>x</var>.  <var>x</var> is an
RTL expression.
</p>
<p><var>code</var> is a value that can be used to specify one of several ways
of printing the operand.  It is used when identical operands must be
printed differently depending on the context.  <var>code</var> comes from
the &lsquo;<samp>%</samp>&rsquo; specification that was used to request printing of the
operand.  If the specification was just &lsquo;<samp>%<var>digit</var></samp>&rsquo; then
<var>code</var> is 0; if the specification was &lsquo;<samp>%<var>ltr</var>
<var>digit</var></samp>&rsquo; then <var>code</var> is the ASCII code for <var>ltr</var>.
</p>
<a name="index-reg_005fnames-1"></a>
<p>If <var>x</var> is a register, this macro should print the register&rsquo;s name.
The names can be found in an array <code>reg_names</code> whose type is
<code>char *[]</code>.  <code>reg_names</code> is initialized from
<code>REGISTER_NAMES</code>.
</p>
<p>When the machine description has a specification &lsquo;<samp>%<var>punct</var></samp>&rsquo;
(a &lsquo;<samp>%</samp>&rsquo; followed by a punctuation character), this macro is called
with a null pointer for <var>x</var> and the punctuation character for
<var>code</var>.
</p></dd></dl>

<dl>
<dt><a name="index-PRINT_005fOPERAND_005fPUNCT_005fVALID_005fP"></a>Macro: <strong>PRINT_OPERAND_PUNCT_VALID_P</strong> <em>(<var>code</var>)</em></dt>
<dd><p>A C expression which evaluates to true if <var>code</var> is a valid
punctuation character for use in the <code>PRINT_OPERAND</code> macro.  If
<code>PRINT_OPERAND_PUNCT_VALID_P</code> is not defined, it means that no
punctuation characters (except for the standard one, &lsquo;<samp>%</samp>&rsquo;) are used
in this way.
</p></dd></dl>

<dl>
<dt><a name="index-PRINT_005fOPERAND_005fADDRESS"></a>Macro: <strong>PRINT_OPERAND_ADDRESS</strong> <em>(<var>stream</var>, <var>x</var>)</em></dt>
<dd><p>A C compound statement to output to stdio stream <var>stream</var> the
assembler syntax for an instruction operand that is a memory reference
whose address is <var>x</var>.  <var>x</var> is an RTL expression.
</p>
<a name="index-TARGET_005fENCODE_005fSECTION_005fINFO-usage"></a>
<p>On some machines, the syntax for a symbolic address depends on the
section that the address refers to.  On these machines, define the hook
<code>TARGET_ENCODE_SECTION_INFO</code> to store the information into the
<code>symbol_ref</code>, and then check for it here.  See <a href="#Assembler-Format">Assembler Format</a>.
</p></dd></dl>

<a name="index-dbr_005fsequence_005flength"></a>
<dl>
<dt><a name="index-DBR_005fOUTPUT_005fSEQEND"></a>Macro: <strong>DBR_OUTPUT_SEQEND</strong> <em>(<var>file</var>)</em></dt>
<dd><p>A C statement, to be executed after all slot-filler instructions have
been output.  If necessary, call <code>dbr_sequence_length</code> to
determine the number of slots filled in a sequence (zero if not
currently outputting a sequence), to decide how many no-ops to output,
or whatever.
</p>
<p>Don&rsquo;t define this macro if it has nothing to do, but it is helpful in
reading assembly output if the extent of the delay sequence is made
explicit (e.g. with white space).
</p></dd></dl>

<a name="index-final_005fsequence"></a>
<p>Note that output routines for instructions with delay slots must be
prepared to deal with not being output as part of a sequence
(i.e. when the scheduling pass is not run, or when no slot fillers could be
found.)  The variable <code>final_sequence</code> is null when not
processing a sequence, otherwise it contains the <code>sequence</code> rtx
being output.
</p>
<a name="index-asm_005ffprintf"></a>
<dl>
<dt><a name="index-REGISTER_005fPREFIX"></a>Macro: <strong>REGISTER_PREFIX</strong></dt>
<dt><a name="index-LOCAL_005fLABEL_005fPREFIX"></a>Macro: <strong>LOCAL_LABEL_PREFIX</strong></dt>
<dt><a name="index-USER_005fLABEL_005fPREFIX"></a>Macro: <strong>USER_LABEL_PREFIX</strong></dt>
<dt><a name="index-IMMEDIATE_005fPREFIX"></a>Macro: <strong>IMMEDIATE_PREFIX</strong></dt>
<dd><p>If defined, C string expressions to be used for the &lsquo;<samp>%R</samp>&rsquo;, &lsquo;<samp>%L</samp>&rsquo;,
&lsquo;<samp>%U</samp>&rsquo;, and &lsquo;<samp>%I</samp>&rsquo; options of <code>asm_fprintf</code> (see
<samp>final.c</samp>).  These are useful when a single <samp>md</samp> file must
support multiple assembler formats.  In that case, the various <samp>tm.h</samp>
files can define these macros differently.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fFPRINTF_005fEXTENSIONS"></a>Macro: <strong>ASM_FPRINTF_EXTENSIONS</strong> <em>(<var>file</var>, <var>argptr</var>, <var>format</var>)</em></dt>
<dd><p>If defined this macro should expand to a series of <code>case</code>
statements which will be parsed inside the <code>switch</code> statement of
the <code>asm_fprintf</code> function.  This allows targets to define extra
printf formats which may useful when generating their assembler
statements.  Note that uppercase letters are reserved for future
generic extensions to asm_fprintf, and so are not available to target
specific code.  The output file is given by the parameter <var>file</var>.
The varargs input pointer is <var>argptr</var> and the rest of the format
string, starting the character after the one that is being switched
upon, is pointed to by <var>format</var>.
</p></dd></dl>

<dl>
<dt><a name="index-ASSEMBLER_005fDIALECT"></a>Macro: <strong>ASSEMBLER_DIALECT</strong></dt>
<dd><p>If your target supports multiple dialects of assembler language (such as
different opcodes), define this macro as a C expression that gives the
numeric index of the assembler language dialect to use, with zero as the
first variant.
</p>
<p>If this macro is defined, you may use constructs of the form
</p><div class="smallexample">
<pre class="smallexample">&lsquo;<samp>{option0|option1|option2&hellip;}</samp>&rsquo;
</pre></div>
<p>in the output templates of patterns (see <a href="#Output-Template">Output Template</a>) or in the
first argument of <code>asm_fprintf</code>.  This construct outputs
&lsquo;<samp>option0</samp>&rsquo;, &lsquo;<samp>option1</samp>&rsquo;, &lsquo;<samp>option2</samp>&rsquo;, etc., if the value of
<code>ASSEMBLER_DIALECT</code> is zero, one, two, etc.  Any special characters
within these strings retain their usual meaning.  If there are fewer
alternatives within the braces than the value of
<code>ASSEMBLER_DIALECT</code>, the construct outputs nothing.
</p>
<p>If you do not define this macro, the characters &lsquo;<samp>{</samp>&rsquo;, &lsquo;<samp>|</samp>&rsquo; and
&lsquo;<samp>}</samp>&rsquo; do not have any special meaning when used in templates or
operands to <code>asm_fprintf</code>.
</p>
<p>Define the macros <code>REGISTER_PREFIX</code>, <code>LOCAL_LABEL_PREFIX</code>,
<code>USER_LABEL_PREFIX</code> and <code>IMMEDIATE_PREFIX</code> if you can express
the variations in assembler language syntax with that mechanism.  Define
<code>ASSEMBLER_DIALECT</code> and use the &lsquo;<samp>{option0|option1}</samp>&rsquo; syntax
if the syntax variant are larger and involve such things as different
opcodes or operand order.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fREG_005fPUSH"></a>Macro: <strong>ASM_OUTPUT_REG_PUSH</strong> <em>(<var>stream</var>, <var>regno</var>)</em></dt>
<dd><p>A C expression to output to <var>stream</var> some assembler code
which will push hard register number <var>regno</var> onto the stack.
The code need not be optimal, since this macro is used only when
profiling.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fREG_005fPOP"></a>Macro: <strong>ASM_OUTPUT_REG_POP</strong> <em>(<var>stream</var>, <var>regno</var>)</em></dt>
<dd><p>A C expression to output to <var>stream</var> some assembler code
which will pop hard register number <var>regno</var> off of the stack.
The code need not be optimal, since this macro is used only when
profiling.
</p></dd></dl>

<hr>
<a name="Dispatch-Tables"></a>
<div class="header">
<p>
Next: <a href="#Exception-Region-Output" accesskey="n" rel="next">Exception Region Output</a>, Previous: <a href="#Instruction-Output" accesskey="p" rel="prev">Instruction Output</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Output-of-Dispatch-Tables"></a>
<h4 class="subsection">17.21.8 Output of Dispatch Tables</h4>

<p>This concerns dispatch tables.
</p>
<a name="index-dispatch-table"></a>
<dl>
<dt><a name="index-ASM_005fOUTPUT_005fADDR_005fDIFF_005fELT"></a>Macro: <strong>ASM_OUTPUT_ADDR_DIFF_ELT</strong> <em>(<var>stream</var>, <var>body</var>, <var>value</var>, <var>rel</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> an assembler
pseudo-instruction to generate a difference between two labels.
<var>value</var> and <var>rel</var> are the numbers of two internal labels.  The
definitions of these labels are output using
<code>(*targetm.asm_out.internal_label)</code>, and they must be printed in the same
way here.  For example,
</p>
<div class="smallexample">
<pre class="smallexample">fprintf (<var>stream</var>, &quot;\t.word L%d-L%d\n&quot;,
         <var>value</var>, <var>rel</var>)
</pre></div>

<p>You must provide this macro on machines where the addresses in a
dispatch table are relative to the table&rsquo;s own address.  If defined, GCC
will also use this macro on all machines when producing PIC.
<var>body</var> is the body of the <code>ADDR_DIFF_VEC</code>; it is provided so that the
mode and flags can be read.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fADDR_005fVEC_005fELT"></a>Macro: <strong>ASM_OUTPUT_ADDR_VEC_ELT</strong> <em>(<var>stream</var>, <var>value</var>)</em></dt>
<dd><p>This macro should be provided on machines where the addresses
in a dispatch table are absolute.
</p>
<p>The definition should be a C statement to output to the stdio stream
<var>stream</var> an assembler pseudo-instruction to generate a reference to
a label.  <var>value</var> is the number of an internal label whose
definition is output using <code>(*targetm.asm_out.internal_label)</code>.
For example,
</p>
<div class="smallexample">
<pre class="smallexample">fprintf (<var>stream</var>, &quot;\t.word L%d\n&quot;, <var>value</var>)
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fCASE_005fLABEL"></a>Macro: <strong>ASM_OUTPUT_CASE_LABEL</strong> <em>(<var>stream</var>, <var>prefix</var>, <var>num</var>, <var>table</var>)</em></dt>
<dd><p>Define this if the label before a jump-table needs to be output
specially.  The first three arguments are the same as for
<code>(*targetm.asm_out.internal_label)</code>; the fourth argument is the
jump-table which follows (a <code>jump_insn</code> containing an
<code>addr_vec</code> or <code>addr_diff_vec</code>).
</p>
<p>This feature is used on system V to output a <code>swbeg</code> statement
for the table.
</p>
<p>If this macro is not defined, these labels are output with
<code>(*targetm.asm_out.internal_label)</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fCASE_005fEND"></a>Macro: <strong>ASM_OUTPUT_CASE_END</strong> <em>(<var>stream</var>, <var>num</var>, <var>table</var>)</em></dt>
<dd><p>Define this if something special must be output at the end of a
jump-table.  The definition should be a C statement to be executed
after the assembler code for the table is written.  It should write
the appropriate code to stdio stream <var>stream</var>.  The argument
<var>table</var> is the jump-table insn, and <var>num</var> is the label-number
of the preceding label.
</p>
<p>If this macro is not defined, nothing special is output at the end of
the jump-table.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fEMIT_005fUNWIND_005fLABEL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_EMIT_UNWIND_LABEL</strong> <em>(FILE *<var>stream</var>, tree <var>decl</var>, int <var>for_eh</var>, int <var>empty</var>)</em></dt>
<dd><p>This target hook emits a label at the beginning of each FDE.  It
should be defined on targets where FDEs need special labels, and it
should write the appropriate label, for the FDE associated with the
function declaration <var>decl</var>, to the stdio stream <var>stream</var>.
The third argument, <var>for_eh</var>, is a boolean: true if this is for an
exception table.  The fourth argument, <var>empty</var>, is a boolean:
true if this is a placeholder label for an omitted FDE.
</p>
<p>The default is that FDEs are not given nonlocal labels.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fEMIT_005fEXCEPT_005fTABLE_005fLABEL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_EMIT_EXCEPT_TABLE_LABEL</strong> <em>(FILE *<var>stream</var>)</em></dt>
<dd><p>This target hook emits a label at the beginning of the exception table.
It should be defined on targets where it is desirable for the table
to be broken up according to function.
</p>
<p>The default is that no label is emitted.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fEMIT_005fEXCEPT_005fPERSONALITY"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_EMIT_EXCEPT_PERSONALITY</strong> <em>(rtx <var>personality</var>)</em></dt>
<dd><p>If the target implements <code>TARGET_ASM_UNWIND_EMIT</code>, this hook may be used to emit a directive to install a personality hook into the unwind info.  This hook should not be used if dwarf2 unwind info is used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fUNWIND_005fEMIT"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_UNWIND_EMIT</strong> <em>(FILE *<var>stream</var>, rtx <var>insn</var>)</em></dt>
<dd><p>This target hook emits assembly directives required to unwind the
given instruction.  This is only used when <code>TARGET_EXCEPT_UNWIND_INFO</code>
returns <code>UI_TARGET</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fUNWIND_005fEMIT_005fBEFORE_005fINSN"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_UNWIND_EMIT_BEFORE_INSN</strong></dt>
<dd><p>True if the <code>TARGET_ASM_UNWIND_EMIT</code> hook should be called before the assembly for <var>insn</var> has been emitted, false if the hook should be called afterward.
</p></dd></dl>

<hr>
<a name="Exception-Region-Output"></a>
<div class="header">
<p>
Next: <a href="#Alignment-Output" accesskey="n" rel="next">Alignment Output</a>, Previous: <a href="#Dispatch-Tables" accesskey="p" rel="prev">Dispatch Tables</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Assembler-Commands-for-Exception-Regions"></a>
<h4 class="subsection">17.21.9 Assembler Commands for Exception Regions</h4>


<p>This describes commands marking the start and the end of an exception
region.
</p>
<dl>
<dt><a name="index-EH_005fFRAME_005fSECTION_005fNAME"></a>Macro: <strong>EH_FRAME_SECTION_NAME</strong></dt>
<dd><p>If defined, a C string constant for the name of the section containing
exception handling frame unwind information.  If not defined, GCC will
provide a default definition if the target supports named sections.
<samp>crtstuff.c</samp> uses this macro to switch to the appropriate section.
</p>
<p>You should define this symbol if your target supports DWARF 2 frame
unwind information and the default definition does not work.
</p></dd></dl>

<dl>
<dt><a name="index-EH_005fFRAME_005fIN_005fDATA_005fSECTION"></a>Macro: <strong>EH_FRAME_IN_DATA_SECTION</strong></dt>
<dd><p>If defined, DWARF 2 frame unwind information will be placed in the
data section even though the target supports named sections.  This
might be necessary, for instance, if the system linker does garbage
collection and sections cannot be marked as not to be collected.
</p>
<p>Do not define this macro unless <code>TARGET_ASM_NAMED_SECTION</code> is
also defined.
</p></dd></dl>

<dl>
<dt><a name="index-EH_005fTABLES_005fCAN_005fBE_005fREAD_005fONLY"></a>Macro: <strong>EH_TABLES_CAN_BE_READ_ONLY</strong></dt>
<dd><p>Define this macro to 1 if your target is such that no frame unwind
information encoding used with non-PIC code will ever require a
runtime relocation, but the linker may not support merging read-only
and read-write sections into a single read-write section.
</p></dd></dl>

<dl>
<dt><a name="index-MASK_005fRETURN_005fADDR"></a>Macro: <strong>MASK_RETURN_ADDR</strong></dt>
<dd><p>An rtx used to mask the return address found via <code>RETURN_ADDR_RTX</code>, so
that it does not contain any extraneous set bits in it.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF2_005fUNWIND_005fINFO"></a>Macro: <strong>DWARF2_UNWIND_INFO</strong></dt>
<dd><p>Define this macro to 0 if your target supports DWARF 2 frame unwind
information, but it does not yet work with exception handling.
Otherwise, if your target supports this information (if it defines
<code>INCOMING_RETURN_ADDR_RTX</code> and <code>OBJECT_FORMAT_ELF</code>),
GCC will provide a default definition of 1.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXCEPT_005fUNWIND_005fINFO"></a>Common Target Hook: <em>enum unwind_info_type</em> <strong>TARGET_EXCEPT_UNWIND_INFO</strong> <em>(struct gcc_options *<var>opts</var>)</em></dt>
<dd><p>This hook defines the mechanism that will be used for exception handling
by the target.  If the target has ABI specified unwind tables, the hook
should return <code>UI_TARGET</code>.  If the target is to use the
<code>setjmp</code>/<code>longjmp</code>-based exception handling scheme, the hook
should return <code>UI_SJLJ</code>.  If the target supports DWARF 2 frame unwind
information, the hook should return <code>UI_DWARF2</code>.
</p>
<p>A target may, if exceptions are disabled, choose to return <code>UI_NONE</code>.
This may end up simplifying other parts of target-specific code.  The
default implementation of this hook never returns <code>UI_NONE</code>.
</p>
<p>Note that the value returned by this hook should be constant.  It should
not depend on anything except the command-line switches described by
<var>opts</var>.  In particular, the
setting <code>UI_SJLJ</code> must be fixed at compiler start-up as C pre-processor
macros and builtin functions related to exception handling are set up
depending on this setting.
</p>
<p>The default implementation of the hook first honors the
<samp>--enable-sjlj-exceptions</samp> configure option, then
<code>DWARF2_UNWIND_INFO</code>, and finally defaults to <code>UI_SJLJ</code>.  If
<code>DWARF2_UNWIND_INFO</code> depends on command-line options, the target
must define this hook so that <var>opts</var> is used correctly.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUNWIND_005fTABLES_005fDEFAULT"></a>Common Target Hook: <em>bool</em> <strong>TARGET_UNWIND_TABLES_DEFAULT</strong></dt>
<dd><p>This variable should be set to <code>true</code> if the target ABI requires unwinding
tables even when exceptions are not used.  It must not be modified by
command-line option processing.
</p></dd></dl>

<dl>
<dt><a name="index-DONT_005fUSE_005fBUILTIN_005fSETJMP"></a>Macro: <strong>DONT_USE_BUILTIN_SETJMP</strong></dt>
<dd><p>Define this macro to 1 if the <code>setjmp</code>/<code>longjmp</code>-based scheme
should use the <code>setjmp</code>/<code>longjmp</code> functions from the C library
instead of the <code>__builtin_setjmp</code>/<code>__builtin_longjmp</code> machinery.
</p></dd></dl>

<dl>
<dt><a name="index-JMP_005fBUF_005fSIZE"></a>Macro: <strong>JMP_BUF_SIZE</strong></dt>
<dd><p>This macro has no effect unless <code>DONT_USE_BUILTIN_SETJMP</code> is also
defined.  Define this macro if the default size of <code>jmp_buf</code> buffer
for the <code>setjmp</code>/<code>longjmp</code>-based exception handling mechanism
is not large enough, or if it is much too large.
The default size is <code>FIRST_PSEUDO_REGISTER * sizeof(void *)</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF_005fCIE_005fDATA_005fALIGNMENT"></a>Macro: <strong>DWARF_CIE_DATA_ALIGNMENT</strong></dt>
<dd><p>This macro need only be defined if the target might save registers in the
function prologue at an offset to the stack pointer that is not aligned to
<code>UNITS_PER_WORD</code>.  The definition should be the negative minimum
alignment if <code>STACK_GROWS_DOWNWARD</code> is defined, and the positive
minimum alignment otherwise.  See <a href="#SDB-and-DWARF">SDB and DWARF</a>.  Only applicable if
the target supports DWARF 2 frame unwind information.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fTERMINATE_005fDW2_005fEH_005fFRAME_005fINFO"></a>Target Hook: <em>bool</em> <strong>TARGET_TERMINATE_DW2_EH_FRAME_INFO</strong></dt>
<dd><p>Contains the value true if the target should add a zero word onto the
end of a Dwarf-2 frame info section when used for exception handling.
Default value is false if <code>EH_FRAME_SECTION_NAME</code> is defined, and
true otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDWARF_005fREGISTER_005fSPAN"></a>Target Hook: <em>rtx</em> <strong>TARGET_DWARF_REGISTER_SPAN</strong> <em>(rtx <var>reg</var>)</em></dt>
<dd><p>Given a register, this hook should return a parallel of registers to
represent where to find the register pieces.  Define this hook if the
register and its mode are represented in Dwarf in non-contiguous
locations, or if the register should be represented in more than one
register in Dwarf.  Otherwise, this hook should return <code>NULL_RTX</code>.
If not defined, the default is to return <code>NULL_RTX</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINIT_005fDWARF_005fREG_005fSIZES_005fEXTRA"></a>Target Hook: <em>void</em> <strong>TARGET_INIT_DWARF_REG_SIZES_EXTRA</strong> <em>(tree <var>address</var>)</em></dt>
<dd><p>If some registers are represented in Dwarf-2 unwind information in
multiple pieces, define this hook to fill in information about the
sizes of those pieces in the table used by the unwinder at runtime.
It will be called by <code>expand_builtin_init_dwarf_reg_sizes</code> after
filling in a single size corresponding to each hard register;
<var>address</var> is the address of the table.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fTTYPE"></a>Target Hook: <em>bool</em> <strong>TARGET_ASM_TTYPE</strong> <em>(rtx <var>sym</var>)</em></dt>
<dd><p>This hook is used to output a reference from a frame unwinding table to
the type_info object identified by <var>sym</var>.  It should return <code>true</code>
if the reference was output.  Returning <code>false</code> will cause the
reference to be output using the normal Dwarf2 routines.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fARM_005fEABI_005fUNWINDER"></a>Target Hook: <em>bool</em> <strong>TARGET_ARM_EABI_UNWINDER</strong></dt>
<dd><p>This flag should be set to <code>true</code> on targets that use an ARM EABI
based unwinding library, and <code>false</code> on other targets.  This effects
the format of unwinding tables, and how the unwinder in entered after
running a cleanup.  The default is <code>false</code>.
</p></dd></dl>

<hr>
<a name="Alignment-Output"></a>
<div class="header">
<p>
Previous: <a href="#Exception-Region-Output" accesskey="p" rel="prev">Exception Region Output</a>, Up: <a href="#Assembler-Format" accesskey="u" rel="up">Assembler Format</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Assembler-Commands-for-Alignment"></a>
<h4 class="subsection">17.21.10 Assembler Commands for Alignment</h4>

<p>This describes commands for alignment.
</p>
<dl>
<dt><a name="index-JUMP_005fALIGN"></a>Macro: <strong>JUMP_ALIGN</strong> <em>(<var>label</var>)</em></dt>
<dd><p>The alignment (log base 2) to put in front of <var>label</var>, which is
a common destination of jumps and has no fallthru incoming edge.
</p>
<p>This macro need not be defined if you don&rsquo;t want any special alignment
to be done at such a time.  Most machine descriptions do not currently
define the macro.
</p>
<p>Unless it&rsquo;s necessary to inspect the <var>label</var> parameter, it is better
to set the variable <var>align_jumps</var> in the target&rsquo;s
<code>TARGET_OPTION_OVERRIDE</code>.  Otherwise, you should try to honor the user&rsquo;s
selection in <var>align_jumps</var> in a <code>JUMP_ALIGN</code> implementation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fJUMP_005fALIGN_005fMAX_005fSKIP"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_JUMP_ALIGN_MAX_SKIP</strong> <em>(rtx <var>label</var>)</em></dt>
<dd><p>The maximum number of bytes to skip before <var>label</var> when applying
<code>JUMP_ALIGN</code>.  This works only if
<code>ASM_OUTPUT_MAX_SKIP_ALIGN</code> is defined.
</p></dd></dl>

<dl>
<dt><a name="index-LABEL_005fALIGN_005fAFTER_005fBARRIER"></a>Macro: <strong>LABEL_ALIGN_AFTER_BARRIER</strong> <em>(<var>label</var>)</em></dt>
<dd><p>The alignment (log base 2) to put in front of <var>label</var>, which follows
a <code>BARRIER</code>.
</p>
<p>This macro need not be defined if you don&rsquo;t want any special alignment
to be done at such a time.  Most machine descriptions do not currently
define the macro.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fLABEL_005fALIGN_005fAFTER_005fBARRIER_005fMAX_005fSKIP"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_LABEL_ALIGN_AFTER_BARRIER_MAX_SKIP</strong> <em>(rtx <var>label</var>)</em></dt>
<dd><p>The maximum number of bytes to skip before <var>label</var> when applying
<code>LABEL_ALIGN_AFTER_BARRIER</code>.  This works only if
<code>ASM_OUTPUT_MAX_SKIP_ALIGN</code> is defined.
</p></dd></dl>

<dl>
<dt><a name="index-LOOP_005fALIGN"></a>Macro: <strong>LOOP_ALIGN</strong> <em>(<var>label</var>)</em></dt>
<dd><p>The alignment (log base 2) to put in front of <var>label</var>, which follows
a <code>NOTE_INSN_LOOP_BEG</code> note.
</p>
<p>This macro need not be defined if you don&rsquo;t want any special alignment
to be done at such a time.  Most machine descriptions do not currently
define the macro.
</p>
<p>Unless it&rsquo;s necessary to inspect the <var>label</var> parameter, it is better
to set the variable <code>align_loops</code> in the target&rsquo;s
<code>TARGET_OPTION_OVERRIDE</code>.  Otherwise, you should try to honor the user&rsquo;s
selection in <code>align_loops</code> in a <code>LOOP_ALIGN</code> implementation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fLOOP_005fALIGN_005fMAX_005fSKIP"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_LOOP_ALIGN_MAX_SKIP</strong> <em>(rtx <var>label</var>)</em></dt>
<dd><p>The maximum number of bytes to skip when applying <code>LOOP_ALIGN</code> to
<var>label</var>.  This works only if <code>ASM_OUTPUT_MAX_SKIP_ALIGN</code> is
defined.
</p></dd></dl>

<dl>
<dt><a name="index-LABEL_005fALIGN"></a>Macro: <strong>LABEL_ALIGN</strong> <em>(<var>label</var>)</em></dt>
<dd><p>The alignment (log base 2) to put in front of <var>label</var>.
If <code>LABEL_ALIGN_AFTER_BARRIER</code> / <code>LOOP_ALIGN</code> specify a different alignment,
the maximum of the specified values is used.
</p>
<p>Unless it&rsquo;s necessary to inspect the <var>label</var> parameter, it is better
to set the variable <code>align_labels</code> in the target&rsquo;s
<code>TARGET_OPTION_OVERRIDE</code>.  Otherwise, you should try to honor the user&rsquo;s
selection in <code>align_labels</code> in a <code>LABEL_ALIGN</code> implementation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fLABEL_005fALIGN_005fMAX_005fSKIP"></a>Target Hook: <em>int</em> <strong>TARGET_ASM_LABEL_ALIGN_MAX_SKIP</strong> <em>(rtx <var>label</var>)</em></dt>
<dd><p>The maximum number of bytes to skip when applying <code>LABEL_ALIGN</code>
to <var>label</var>.  This works only if <code>ASM_OUTPUT_MAX_SKIP_ALIGN</code>
is defined.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fSKIP"></a>Macro: <strong>ASM_OUTPUT_SKIP</strong> <em>(<var>stream</var>, <var>nbytes</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> an assembler
instruction to advance the location counter by <var>nbytes</var> bytes.
Those bytes should be zero when loaded.  <var>nbytes</var> will be a C
expression of type <code>unsigned HOST_WIDE_INT</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fNO_005fSKIP_005fIN_005fTEXT"></a>Macro: <strong>ASM_NO_SKIP_IN_TEXT</strong></dt>
<dd><p>Define this macro if <code>ASM_OUTPUT_SKIP</code> should not be used in the
text section because it fails to put zeros in the bytes that are skipped.
This is true on many Unix systems, where the pseudo&ndash;op to skip bytes
produces no-op instructions rather than zeros when used in the text
section.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGN"></a>Macro: <strong>ASM_OUTPUT_ALIGN</strong> <em>(<var>stream</var>, <var>power</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> an assembler
command to advance the location counter to a multiple of 2 to the
<var>power</var> bytes.  <var>power</var> will be a C expression of type <code>int</code>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fALIGN_005fWITH_005fNOP"></a>Macro: <strong>ASM_OUTPUT_ALIGN_WITH_NOP</strong> <em>(<var>stream</var>, <var>power</var>)</em></dt>
<dd><p>Like <code>ASM_OUTPUT_ALIGN</code>, except that the &ldquo;nop&rdquo; instruction is used
for padding, if necessary.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fMAX_005fSKIP_005fALIGN"></a>Macro: <strong>ASM_OUTPUT_MAX_SKIP_ALIGN</strong> <em>(<var>stream</var>, <var>power</var>, <var>max_skip</var>)</em></dt>
<dd><p>A C statement to output to the stdio stream <var>stream</var> an assembler
command to advance the location counter to a multiple of 2 to the
<var>power</var> bytes, but only if <var>max_skip</var> or fewer bytes are needed to
satisfy the alignment request.  <var>power</var> and <var>max_skip</var> will be
a C expression of type <code>int</code>.
</p></dd></dl>

<hr>
<a name="Debugging-Info"></a>
<div class="header">
<p>
Next: <a href="#Floating-Point" accesskey="n" rel="next">Floating Point</a>, Previous: <a href="#Assembler-Format" accesskey="p" rel="prev">Assembler Format</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Controlling-Debugging-Information-Format"></a>
<h3 class="section">17.22 Controlling Debugging Information Format</h3>

<p>This describes how to specify debugging information.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#All-Debuggers" accesskey="1">All Debuggers</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros that affect all debugging formats uniformly.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#DBX-Options" accesskey="2">DBX Options</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros enabling specific options in DBX format.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#DBX-Hooks" accesskey="3">DBX Hooks</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Hook macros for varying DBX format.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#File-Names-and-DBX" accesskey="4">File Names and DBX</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros controlling output of file names in DBX format.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#SDB-and-DWARF" accesskey="5">SDB and DWARF</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros for SDB (COFF) and DWARF formats.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#VMS-Debug" accesskey="6">VMS Debug</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Macros for VMS debug format.
</td></tr>
</table>

<hr>
<a name="All-Debuggers"></a>
<div class="header">
<p>
Next: <a href="#DBX-Options" accesskey="n" rel="next">DBX Options</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Macros-Affecting-All-Debugging-Formats"></a>
<h4 class="subsection">17.22.1 Macros Affecting All Debugging Formats</h4>

<p>These macros affect all debugging formats.
</p>
<dl>
<dt><a name="index-DBX_005fREGISTER_005fNUMBER"></a>Macro: <strong>DBX_REGISTER_NUMBER</strong> <em>(<var>regno</var>)</em></dt>
<dd><p>A C expression that returns the DBX register number for the compiler
register number <var>regno</var>.  In the default macro provided, the value
of this expression will be <var>regno</var> itself.  But sometimes there are
some registers that the compiler knows about and DBX does not, or vice
versa.  In such cases, some register may need to have one number in the
compiler and another for DBX.
</p>
<p>If two registers have consecutive numbers inside GCC, and they can be
used as a pair to hold a multiword value, then they <em>must</em> have
consecutive numbers after renumbering with <code>DBX_REGISTER_NUMBER</code>.
Otherwise, debuggers will be unable to access such a pair, because they
expect register pairs to be consecutive in their own numbering scheme.
</p>
<p>If you find yourself defining <code>DBX_REGISTER_NUMBER</code> in way that
does not preserve register pairs, then what you must do instead is
redefine the actual register numbering scheme.
</p></dd></dl>

<dl>
<dt><a name="index-DEBUGGER_005fAUTO_005fOFFSET"></a>Macro: <strong>DEBUGGER_AUTO_OFFSET</strong> <em>(<var>x</var>)</em></dt>
<dd><p>A C expression that returns the integer offset value for an automatic
variable having address <var>x</var> (an RTL expression).  The default
computation assumes that <var>x</var> is based on the frame-pointer and
gives the offset from the frame-pointer.  This is required for targets
that produce debugging output for DBX or COFF-style debugging output
for SDB and allow the frame-pointer to be eliminated when the
<samp>-g</samp> options is used.
</p></dd></dl>

<dl>
<dt><a name="index-DEBUGGER_005fARG_005fOFFSET"></a>Macro: <strong>DEBUGGER_ARG_OFFSET</strong> <em>(<var>offset</var>, <var>x</var>)</em></dt>
<dd><p>A C expression that returns the integer offset value for an argument
having address <var>x</var> (an RTL expression).  The nominal offset is
<var>offset</var>.
</p></dd></dl>

<dl>
<dt><a name="index-PREFERRED_005fDEBUGGING_005fTYPE"></a>Macro: <strong>PREFERRED_DEBUGGING_TYPE</strong></dt>
<dd><p>A C expression that returns the type of debugging output GCC should
produce when the user specifies just <samp>-g</samp>.  Define
this if you have arranged for GCC to support more than one format of
debugging output.  Currently, the allowable values are <code>DBX_DEBUG</code>,
<code>SDB_DEBUG</code>, <code>DWARF_DEBUG</code>, <code>DWARF2_DEBUG</code>,
<code>XCOFF_DEBUG</code>, <code>VMS_DEBUG</code>, and <code>VMS_AND_DWARF2_DEBUG</code>.
</p>
<p>When the user specifies <samp>-ggdb</samp>, GCC normally also uses the
value of this macro to select the debugging output format, but with two
exceptions.  If <code>DWARF2_DEBUGGING_INFO</code> is defined, GCC uses the
value <code>DWARF2_DEBUG</code>.  Otherwise, if <code>DBX_DEBUGGING_INFO</code> is
defined, GCC uses <code>DBX_DEBUG</code>.
</p>
<p>The value of this macro only affects the default debugging output; the
user can always get a specific type of output by using <samp>-gstabs</samp>,
<samp>-gcoff</samp>, <samp>-gdwarf-2</samp>, <samp>-gxcoff</samp>, or <samp>-gvms</samp>.
</p></dd></dl>

<hr>
<a name="DBX-Options"></a>
<div class="header">
<p>
Next: <a href="#DBX-Hooks" accesskey="n" rel="next">DBX Hooks</a>, Previous: <a href="#All-Debuggers" accesskey="p" rel="prev">All Debuggers</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Specific-Options-for-DBX-Output"></a>
<h4 class="subsection">17.22.2 Specific Options for DBX Output</h4>

<p>These are specific options for DBX output.
</p>
<dl>
<dt><a name="index-DBX_005fDEBUGGING_005fINFO"></a>Macro: <strong>DBX_DEBUGGING_INFO</strong></dt>
<dd><p>Define this macro if GCC should produce debugging output for DBX
in response to the <samp>-g</samp> option.
</p></dd></dl>

<dl>
<dt><a name="index-XCOFF_005fDEBUGGING_005fINFO"></a>Macro: <strong>XCOFF_DEBUGGING_INFO</strong></dt>
<dd><p>Define this macro if GCC should produce XCOFF format debugging output
in response to the <samp>-g</samp> option.  This is a variant of DBX format.
</p></dd></dl>

<dl>
<dt><a name="index-DEFAULT_005fGDB_005fEXTENSIONS"></a>Macro: <strong>DEFAULT_GDB_EXTENSIONS</strong></dt>
<dd><p>Define this macro to control whether GCC should by default generate
GDB&rsquo;s extended version of DBX debugging information (assuming DBX-format
debugging information is enabled at all).  If you don&rsquo;t define the
macro, the default is 1: always generate the extended information
if there is any occasion to.
</p></dd></dl>

<dl>
<dt><a name="index-DEBUG_005fSYMS_005fTEXT"></a>Macro: <strong>DEBUG_SYMS_TEXT</strong></dt>
<dd><p>Define this macro if all <code>.stabs</code> commands should be output while
in the text section.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fSTABS_005fOP"></a>Macro: <strong>ASM_STABS_OP</strong></dt>
<dd><p>A C string constant, including spacing, naming the assembler pseudo op to
use instead of <code>&quot;\t.stabs\t&quot;</code> to define an ordinary debugging symbol.
If you don&rsquo;t define this macro, <code>&quot;\t.stabs\t&quot;</code> is used.  This macro
applies only to DBX debugging information format.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fSTABD_005fOP"></a>Macro: <strong>ASM_STABD_OP</strong></dt>
<dd><p>A C string constant, including spacing, naming the assembler pseudo op to
use instead of <code>&quot;\t.stabd\t&quot;</code> to define a debugging symbol whose
value is the current location.  If you don&rsquo;t define this macro,
<code>&quot;\t.stabd\t&quot;</code> is used.  This macro applies only to DBX debugging
information format.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fSTABN_005fOP"></a>Macro: <strong>ASM_STABN_OP</strong></dt>
<dd><p>A C string constant, including spacing, naming the assembler pseudo op to
use instead of <code>&quot;\t.stabn\t&quot;</code> to define a debugging symbol with no
name.  If you don&rsquo;t define this macro, <code>&quot;\t.stabn\t&quot;</code> is used.  This
macro applies only to DBX debugging information format.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fNO_005fXREFS"></a>Macro: <strong>DBX_NO_XREFS</strong></dt>
<dd><p>Define this macro if DBX on your system does not support the construct
&lsquo;<samp>xs<var>tagname</var></samp>&rsquo;.  On some systems, this construct is used to
describe a forward reference to a structure named <var>tagname</var>.
On other systems, this construct is not supported at all.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fCONTIN_005fLENGTH"></a>Macro: <strong>DBX_CONTIN_LENGTH</strong></dt>
<dd><p>A symbol name in DBX-format debugging information is normally
continued (split into two separate <code>.stabs</code> directives) when it
exceeds a certain length (by default, 80 characters).  On some
operating systems, DBX requires this splitting; on others, splitting
must not be done.  You can inhibit splitting by defining this macro
with the value zero.  You can override the default splitting-length by
defining this macro as an expression for the length you desire.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fCONTIN_005fCHAR"></a>Macro: <strong>DBX_CONTIN_CHAR</strong></dt>
<dd><p>Normally continuation is indicated by adding a &lsquo;<samp>\</samp>&rsquo; character to
the end of a <code>.stabs</code> string when a continuation follows.  To use
a different character instead, define this macro as a character
constant for the character you want to use.  Do not define this macro
if backslash is correct for your system.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fSTATIC_005fSTAB_005fDATA_005fSECTION"></a>Macro: <strong>DBX_STATIC_STAB_DATA_SECTION</strong></dt>
<dd><p>Define this macro if it is necessary to go to the data section before
outputting the &lsquo;<samp>.stabs</samp>&rsquo; pseudo-op for a non-global static
variable.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fTYPE_005fDECL_005fSTABS_005fCODE"></a>Macro: <strong>DBX_TYPE_DECL_STABS_CODE</strong></dt>
<dd><p>The value to use in the &ldquo;code&rdquo; field of the <code>.stabs</code> directive
for a typedef.  The default is <code>N_LSYM</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fSTATIC_005fCONST_005fVAR_005fCODE"></a>Macro: <strong>DBX_STATIC_CONST_VAR_CODE</strong></dt>
<dd><p>The value to use in the &ldquo;code&rdquo; field of the <code>.stabs</code> directive
for a static variable located in the text section.  DBX format does not
provide any &ldquo;right&rdquo; way to do this.  The default is <code>N_FUN</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fREGPARM_005fSTABS_005fCODE"></a>Macro: <strong>DBX_REGPARM_STABS_CODE</strong></dt>
<dd><p>The value to use in the &ldquo;code&rdquo; field of the <code>.stabs</code> directive
for a parameter passed in registers.  DBX format does not provide any
&ldquo;right&rdquo; way to do this.  The default is <code>N_RSYM</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fREGPARM_005fSTABS_005fLETTER"></a>Macro: <strong>DBX_REGPARM_STABS_LETTER</strong></dt>
<dd><p>The letter to use in DBX symbol data to identify a symbol as a parameter
passed in registers.  DBX format does not customarily provide any way to
do this.  The default is <code>'P'</code>.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fFUNCTION_005fFIRST"></a>Macro: <strong>DBX_FUNCTION_FIRST</strong></dt>
<dd><p>Define this macro if the DBX information for a function and its
arguments should precede the assembler code for the function.  Normally,
in DBX format, the debugging information entirely follows the assembler
code.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fBLOCKS_005fFUNCTION_005fRELATIVE"></a>Macro: <strong>DBX_BLOCKS_FUNCTION_RELATIVE</strong></dt>
<dd><p>Define this macro, with value 1, if the value of a symbol describing
the scope of a block (<code>N_LBRAC</code> or <code>N_RBRAC</code>) should be
relative to the start of the enclosing function.  Normally, GCC uses
an absolute address.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fLINES_005fFUNCTION_005fRELATIVE"></a>Macro: <strong>DBX_LINES_FUNCTION_RELATIVE</strong></dt>
<dd><p>Define this macro, with value 1, if the value of a symbol indicating
the current line number (<code>N_SLINE</code>) should be relative to the
start of the enclosing function.  Normally, GCC uses an absolute address.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fUSE_005fBINCL"></a>Macro: <strong>DBX_USE_BINCL</strong></dt>
<dd><p>Define this macro if GCC should generate <code>N_BINCL</code> and
<code>N_EINCL</code> stabs for included header files, as on Sun systems.  This
macro also directs GCC to output a type number as a pair of a file
number and a type number within the file.  Normally, GCC does not
generate <code>N_BINCL</code> or <code>N_EINCL</code> stabs, and it outputs a single
number for a type number.
</p></dd></dl>

<hr>
<a name="DBX-Hooks"></a>
<div class="header">
<p>
Next: <a href="#File-Names-and-DBX" accesskey="n" rel="next">File Names and DBX</a>, Previous: <a href="#DBX-Options" accesskey="p" rel="prev">DBX Options</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Open_002dEnded-Hooks-for-DBX-Format"></a>
<h4 class="subsection">17.22.3 Open-Ended Hooks for DBX Format</h4>

<p>These are hooks for DBX format.
</p>
<dl>
<dt><a name="index-DBX_005fOUTPUT_005fSOURCE_005fLINE"></a>Macro: <strong>DBX_OUTPUT_SOURCE_LINE</strong> <em>(<var>stream</var>, <var>line</var>, <var>counter</var>)</em></dt>
<dd><p>A C statement to output DBX debugging information before code for line
number <var>line</var> of the current source file to the stdio stream
<var>stream</var>.  <var>counter</var> is the number of time the macro was
invoked, including the current invocation; it is intended to generate
unique labels in the assembly output.
</p>
<p>This macro should not be defined if the default output is correct, or
if it can be made correct by defining <code>DBX_LINES_FUNCTION_RELATIVE</code>.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDBX_005fFUNCTION_005fEND"></a>Macro: <strong>NO_DBX_FUNCTION_END</strong></dt>
<dd><p>Some stabs encapsulation formats (in particular ECOFF), cannot handle the
<code>.stabs &quot;&quot;,N_FUN,,0,0,Lscope-function-1</code> gdb dbx extension construct.
On those machines, define this macro to turn this feature off without
disturbing the rest of the gdb extensions.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDBX_005fBNSYM_005fENSYM"></a>Macro: <strong>NO_DBX_BNSYM_ENSYM</strong></dt>
<dd><p>Some assemblers cannot handle the <code>.stabd BNSYM/ENSYM,0,0</code> gdb dbx
extension construct.  On those machines, define this macro to turn this
feature off without disturbing the rest of the gdb extensions.
</p></dd></dl>

<hr>
<a name="File-Names-and-DBX"></a>
<div class="header">
<p>
Next: <a href="#SDB-and-DWARF" accesskey="n" rel="next">SDB and DWARF</a>, Previous: <a href="#DBX-Hooks" accesskey="p" rel="prev">DBX Hooks</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="File-Names-in-DBX-Format"></a>
<h4 class="subsection">17.22.4 File Names in DBX Format</h4>

<p>This describes file names in DBX format.
</p>
<dl>
<dt><a name="index-DBX_005fOUTPUT_005fMAIN_005fSOURCE_005fFILENAME"></a>Macro: <strong>DBX_OUTPUT_MAIN_SOURCE_FILENAME</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement to output DBX debugging information to the stdio stream
<var>stream</var>, which indicates that file <var>name</var> is the main source
file&mdash;the file specified as the input file for compilation.
This macro is called only once, at the beginning of compilation.
</p>
<p>This macro need not be defined if the standard form of output
for DBX debugging information is appropriate.
</p>
<p>It may be necessary to refer to a label equal to the beginning of the
text section.  You can use &lsquo;<samp>assemble_name (stream, ltext_label_name)</samp>&rsquo;
to do so.  If you do this, you must also set the variable
<var>used_ltext_label_name</var> to <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDBX_005fMAIN_005fSOURCE_005fDIRECTORY"></a>Macro: <strong>NO_DBX_MAIN_SOURCE_DIRECTORY</strong></dt>
<dd><p>Define this macro, with value 1, if GCC should not emit an indication
of the current directory for compilation and current source language at
the beginning of the file.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fDBX_005fGCC_005fMARKER"></a>Macro: <strong>NO_DBX_GCC_MARKER</strong></dt>
<dd><p>Define this macro, with value 1, if GCC should not emit an indication
that this object file was compiled by GCC.  The default is to emit
an <code>N_OPT</code> stab at the beginning of every source file, with
&lsquo;<samp>gcc2_compiled.</samp>&rsquo; for the string and value 0.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fOUTPUT_005fMAIN_005fSOURCE_005fFILE_005fEND"></a>Macro: <strong>DBX_OUTPUT_MAIN_SOURCE_FILE_END</strong> <em>(<var>stream</var>, <var>name</var>)</em></dt>
<dd><p>A C statement to output DBX debugging information at the end of
compilation of the main source file <var>name</var>.  Output should be
written to the stdio stream <var>stream</var>.
</p>
<p>If you don&rsquo;t define this macro, nothing special is output at the end
of compilation, which is correct for most machines.
</p></dd></dl>

<dl>
<dt><a name="index-DBX_005fOUTPUT_005fNULL_005fN_005fSO_005fAT_005fMAIN_005fSOURCE_005fFILE_005fEND"></a>Macro: <strong>DBX_OUTPUT_NULL_N_SO_AT_MAIN_SOURCE_FILE_END</strong></dt>
<dd><p>Define this macro <em>instead of</em> defining
<code>DBX_OUTPUT_MAIN_SOURCE_FILE_END</code>, if what needs to be output at
the end of compilation is an <code>N_SO</code> stab with an empty string,
whose value is the highest absolute text address in the file.
</p></dd></dl>

<hr>
<a name="SDB-and-DWARF"></a>
<div class="header">
<p>
Next: <a href="#VMS-Debug" accesskey="n" rel="next">VMS Debug</a>, Previous: <a href="#File-Names-and-DBX" accesskey="p" rel="prev">File Names and DBX</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Macros-for-SDB-and-DWARF-Output"></a>
<h4 class="subsection">17.22.5 Macros for SDB and DWARF Output</h4>

<p>Here are macros for SDB and DWARF output.
</p>
<dl>
<dt><a name="index-SDB_005fDEBUGGING_005fINFO"></a>Macro: <strong>SDB_DEBUGGING_INFO</strong></dt>
<dd><p>Define this macro if GCC should produce COFF-style debugging output
for SDB in response to the <samp>-g</samp> option.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF2_005fDEBUGGING_005fINFO"></a>Macro: <strong>DWARF2_DEBUGGING_INFO</strong></dt>
<dd><p>Define this macro if GCC should produce dwarf version 2 format
debugging output in response to the <samp>-g</samp> option.
</p>
<dl>
<dt><a name="index-TARGET_005fDWARF_005fCALLING_005fCONVENTION"></a>Target Hook: <em>int</em> <strong>TARGET_DWARF_CALLING_CONVENTION</strong> <em>(const_tree <var>function</var>)</em></dt>
<dd><p>Define this to enable the dwarf attribute <code>DW_AT_calling_convention</code> to
be emitted for each function.  Instead of an integer return the enum
value for the <code>DW_CC_</code> tag.
</p></dd></dl>

<p>To support optional call frame debugging information, you must also
define <code>INCOMING_RETURN_ADDR_RTX</code> and either set
<code>RTX_FRAME_RELATED_P</code> on the prologue insns if you use RTL for the
prologue, or call <code>dwarf2out_def_cfa</code> and <code>dwarf2out_reg_save</code>
as appropriate from <code>TARGET_ASM_FUNCTION_PROLOGUE</code> if you don&rsquo;t.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF2_005fFRAME_005fINFO"></a>Macro: <strong>DWARF2_FRAME_INFO</strong></dt>
<dd><p>Define this macro to a nonzero value if GCC should always output
Dwarf 2 frame information.  If <code>TARGET_EXCEPT_UNWIND_INFO</code>
(see <a href="#Exception-Region-Output">Exception Region Output</a>) returns <code>UI_DWARF2</code>, and
exceptions are enabled, GCC will output this information not matter
how you define <code>DWARF2_FRAME_INFO</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDEBUG_005fUNWIND_005fINFO"></a>Target Hook: <em>enum unwind_info_type</em> <strong>TARGET_DEBUG_UNWIND_INFO</strong> <em>(void)</em></dt>
<dd><p>This hook defines the mechanism that will be used for describing frame
unwind information to the debugger.  Normally the hook will return
<code>UI_DWARF2</code> if DWARF 2 debug information is enabled, and
return <code>UI_NONE</code> otherwise.
</p>
<p>A target may return <code>UI_DWARF2</code> even when DWARF 2 debug information
is disabled in order to always output DWARF 2 frame information.
</p>
<p>A target may return <code>UI_TARGET</code> if it has ABI specified unwind tables.
This will suppress generation of the normal debug frame unwind information.
</p></dd></dl>

<dl>
<dt><a name="index-DWARF2_005fASM_005fLINE_005fDEBUG_005fINFO"></a>Macro: <strong>DWARF2_ASM_LINE_DEBUG_INFO</strong></dt>
<dd><p>Define this macro to be a nonzero value if the assembler can generate Dwarf 2
line debug info sections.  This will result in much more compact line number
tables, and hence is desirable if it works.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fWANT_005fDEBUG_005fPUB_005fSECTIONS"></a>Target Hook: <em>bool</em> <strong>TARGET_WANT_DEBUG_PUB_SECTIONS</strong></dt>
<dd><p>True if the <code>.debug_pubtypes</code> and <code>.debug_pubnames</code> sections should be emitted.  These sections are not used on most platforms, and in particular GDB does not use them.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFORCE_005fAT_005fCOMP_005fDIR"></a>Target Hook: <em>bool</em> <strong>TARGET_FORCE_AT_COMP_DIR</strong></dt>
<dd><p>True if the <code>DW_AT_comp_dir</code> attribute should be emitted for each  compilation unit.  This attribute is required for the darwin linker  to emit debug information.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDELAY_005fSCHED2"></a>Target Hook: <em>bool</em> <strong>TARGET_DELAY_SCHED2</strong></dt>
<dd><p>True if sched2 is not to be run at its normal place.  This usually means it will be run as part of machine-specific reorg.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDELAY_005fVARTRACK"></a>Target Hook: <em>bool</em> <strong>TARGET_DELAY_VARTRACK</strong></dt>
<dd><p>True if vartrack is not to be run at its normal place.  This usually means it will be run as part of machine-specific reorg.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDWARF_005fDELTA"></a>Macro: <strong>ASM_OUTPUT_DWARF_DELTA</strong> <em>(<var>stream</var>, <var>size</var>, <var>label1</var>, <var>label2</var>)</em></dt>
<dd><p>A C statement to issue assembly directives that create a difference
<var>lab1</var> minus <var>lab2</var>, using an integer of the given <var>size</var>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDWARF_005fVMS_005fDELTA"></a>Macro: <strong>ASM_OUTPUT_DWARF_VMS_DELTA</strong> <em>(<var>stream</var>, <var>size</var>, <var>label1</var>, <var>label2</var>)</em></dt>
<dd><p>A C statement to issue assembly directives that create a difference
between the two given labels in system defined units, e.g. instruction
slots on IA64 VMS, using an integer of the given size.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDWARF_005fOFFSET"></a>Macro: <strong>ASM_OUTPUT_DWARF_OFFSET</strong> <em>(<var>stream</var>, <var>size</var>, <var>label</var>, <var>section</var>)</em></dt>
<dd><p>A C statement to issue assembly directives that create a
section-relative reference to the given <var>label</var>, using an integer of the
given <var>size</var>.  The label is known to be defined in the given <var>section</var>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDWARF_005fPCREL"></a>Macro: <strong>ASM_OUTPUT_DWARF_PCREL</strong> <em>(<var>stream</var>, <var>size</var>, <var>label</var>)</em></dt>
<dd><p>A C statement to issue assembly directives that create a self-relative
reference to the given <var>label</var>, using an integer of the given <var>size</var>.
</p></dd></dl>

<dl>
<dt><a name="index-ASM_005fOUTPUT_005fDWARF_005fTABLE_005fREF"></a>Macro: <strong>ASM_OUTPUT_DWARF_TABLE_REF</strong> <em>(<var>label</var>)</em></dt>
<dd><p>A C statement to issue assembly directives that create a reference to
the DWARF table identifier <var>label</var> from the current section.  This
is used on some systems to avoid garbage collecting a DWARF table which
is referenced by a function.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASM_005fOUTPUT_005fDWARF_005fDTPREL"></a>Target Hook: <em>void</em> <strong>TARGET_ASM_OUTPUT_DWARF_DTPREL</strong> <em>(FILE *<var>file</var>, int <var>size</var>, rtx <var>x</var>)</em></dt>
<dd><p>If defined, this target hook is a function which outputs a DTP-relative
reference to the given TLS symbol of the specified size.
</p></dd></dl>

<dl>
<dt><a name="index-PUT_005fSDB_005f"></a>Macro: <strong>PUT_SDB_</strong> <em>&hellip;</em></dt>
<dd><p>Define these macros to override the assembler syntax for the special
SDB assembler directives.  See <samp>sdbout.c</samp> for a list of these
macros and their arguments.  If the standard syntax is used, you need
not define them yourself.
</p></dd></dl>

<dl>
<dt><a name="index-SDB_005fDELIM"></a>Macro: <strong>SDB_DELIM</strong></dt>
<dd><p>Some assemblers do not support a semicolon as a delimiter, even between
SDB assembler directives.  In that case, define this macro to be the
delimiter to use (usually &lsquo;<samp>\n</samp>&rsquo;).  It is not necessary to define
a new set of <code>PUT_SDB_<var>op</var></code> macros if this is the only change
required.
</p></dd></dl>

<dl>
<dt><a name="index-SDB_005fALLOW_005fUNKNOWN_005fREFERENCES"></a>Macro: <strong>SDB_ALLOW_UNKNOWN_REFERENCES</strong></dt>
<dd><p>Define this macro to allow references to unknown structure,
union, or enumeration tags to be emitted.  Standard COFF does not
allow handling of unknown references, MIPS ECOFF has support for
it.
</p></dd></dl>

<dl>
<dt><a name="index-SDB_005fALLOW_005fFORWARD_005fREFERENCES"></a>Macro: <strong>SDB_ALLOW_FORWARD_REFERENCES</strong></dt>
<dd><p>Define this macro to allow references to structure, union, or
enumeration tags that have not yet been seen to be handled.  Some
assemblers choke if forward tags are used, while some require it.
</p></dd></dl>

<dl>
<dt><a name="index-SDB_005fOUTPUT_005fSOURCE_005fLINE"></a>Macro: <strong>SDB_OUTPUT_SOURCE_LINE</strong> <em>(<var>stream</var>, <var>line</var>)</em></dt>
<dd><p>A C statement to output SDB debugging information before code for line
number <var>line</var> of the current source file to the stdio stream
<var>stream</var>.  The default is to emit an <code>.ln</code> directive.
</p></dd></dl>

<hr>
<a name="VMS-Debug"></a>
<div class="header">
<p>
Previous: <a href="#SDB-and-DWARF" accesskey="p" rel="prev">SDB and DWARF</a>, Up: <a href="#Debugging-Info" accesskey="u" rel="up">Debugging Info</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Macros-for-VMS-Debug-Format"></a>
<h4 class="subsection">17.22.6 Macros for VMS Debug Format</h4>

<p>Here are macros for VMS debug format.
</p>
<dl>
<dt><a name="index-VMS_005fDEBUGGING_005fINFO"></a>Macro: <strong>VMS_DEBUGGING_INFO</strong></dt>
<dd><p>Define this macro if GCC should produce debugging output for VMS
in response to the <samp>-g</samp> option.  The default behavior for VMS
is to generate minimal debug info for a traceback in the absence of
<samp>-g</samp> unless explicitly overridden with <samp>-g0</samp>.  This
behavior is controlled by <code>TARGET_OPTION_OPTIMIZATION</code> and
<code>TARGET_OPTION_OVERRIDE</code>.
</p></dd></dl>

<hr>
<a name="Floating-Point"></a>
<div class="header">
<p>
Next: <a href="#Mode-Switching" accesskey="n" rel="next">Mode Switching</a>, Previous: <a href="#Debugging-Info" accesskey="p" rel="prev">Debugging Info</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Cross-Compilation-and-Floating-Point"></a>
<h3 class="section">17.23 Cross Compilation and Floating Point</h3>
<a name="index-cross-compilation-and-floating-point"></a>
<a name="index-floating-point-and-cross-compilation"></a>

<p>While all modern machines use twos-complement representation for integers,
there are a variety of representations for floating point numbers.  This
means that in a cross-compiler the representation of floating point numbers
in the compiled program may be different from that used in the machine
doing the compilation.
</p>
<p>Because different representation systems may offer different amounts of
range and precision, all floating point constants must be represented in
the target machine&rsquo;s format.  Therefore, the cross compiler cannot
safely use the host machine&rsquo;s floating point arithmetic; it must emulate
the target&rsquo;s arithmetic.  To ensure consistency, GCC always uses
emulation to work with floating point values, even when the host and
target floating point formats are identical.
</p>
<p>The following macros are provided by <samp>real.h</samp> for the compiler to
use.  All parts of the compiler which generate or optimize
floating-point calculations must use these macros.  They may evaluate
their operands more than once, so operands must not have side effects.
</p>
<dl>
<dt><a name="index-REAL_005fVALUE_005fTYPE"></a>Macro: <strong>REAL_VALUE_TYPE</strong></dt>
<dd><p>The C data type to be used to hold a floating point value in the target
machine&rsquo;s format.  Typically this is a <code>struct</code> containing an
array of <code>HOST_WIDE_INT</code>, but all code should treat it as an opaque
quantity.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUES_005fEQUAL"></a>Macro: <em>int</em> <strong>REAL_VALUES_EQUAL</strong> <em>(REAL_VALUE_TYPE <var>x</var>, REAL_VALUE_TYPE <var>y</var>)</em></dt>
<dd><p>Compares for equality the two values, <var>x</var> and <var>y</var>.  If the target
floating point format supports negative zeroes and/or NaNs,
&lsquo;<samp>REAL_VALUES_EQUAL (-0.0, 0.0)</samp>&rsquo; is true, and
&lsquo;<samp>REAL_VALUES_EQUAL (NaN, NaN)</samp>&rsquo; is false.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUES_005fLESS"></a>Macro: <em>int</em> <strong>REAL_VALUES_LESS</strong> <em>(REAL_VALUE_TYPE <var>x</var>, REAL_VALUE_TYPE <var>y</var>)</em></dt>
<dd><p>Tests whether <var>x</var> is less than <var>y</var>.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fFIX"></a>Macro: <em>HOST_WIDE_INT</em> <strong>REAL_VALUE_FIX</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Truncates <var>x</var> to a signed integer, rounding toward zero.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fUNSIGNED_005fFIX"></a>Macro: <em>unsigned HOST_WIDE_INT</em> <strong>REAL_VALUE_UNSIGNED_FIX</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Truncates <var>x</var> to an unsigned integer, rounding toward zero.  If
<var>x</var> is negative, returns zero.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fATOF"></a>Macro: <em>REAL_VALUE_TYPE</em> <strong>REAL_VALUE_ATOF</strong> <em>(const char *<var>string</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Converts <var>string</var> into a floating point number in the target machine&rsquo;s
representation for mode <var>mode</var>.  This routine can handle both
decimal and hexadecimal floating point constants, using the syntax
defined by the C language for both.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fNEGATIVE"></a>Macro: <em>int</em> <strong>REAL_VALUE_NEGATIVE</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Returns 1 if <var>x</var> is negative (including negative zero), 0 otherwise.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fISINF"></a>Macro: <em>int</em> <strong>REAL_VALUE_ISINF</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Determines whether <var>x</var> represents infinity (positive or negative).
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fISNAN"></a>Macro: <em>int</em> <strong>REAL_VALUE_ISNAN</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Determines whether <var>x</var> represents a &ldquo;NaN&rdquo; (not-a-number).
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fARITHMETIC"></a>Macro: <em>void</em> <strong>REAL_ARITHMETIC</strong> <em>(REAL_VALUE_TYPE <var>output</var>, enum tree_code <var>code</var>, REAL_VALUE_TYPE <var>x</var>, REAL_VALUE_TYPE <var>y</var>)</em></dt>
<dd><p>Calculates an arithmetic operation on the two floating point values
<var>x</var> and <var>y</var>, storing the result in <var>output</var> (which must be a
variable).
</p>
<p>The operation to be performed is specified by <var>code</var>.  Only the
following codes are supported: <code>PLUS_EXPR</code>, <code>MINUS_EXPR</code>,
<code>MULT_EXPR</code>, <code>RDIV_EXPR</code>, <code>MAX_EXPR</code>, <code>MIN_EXPR</code>.
</p>
<p>If <code>REAL_ARITHMETIC</code> is asked to evaluate division by zero and the
target&rsquo;s floating point format cannot represent infinity, it will call
<code>abort</code>.  Callers should check for this situation first, using
<code>MODE_HAS_INFINITIES</code>.  See <a href="#Storage-Layout">Storage Layout</a>.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fNEGATE"></a>Macro: <em>REAL_VALUE_TYPE</em> <strong>REAL_VALUE_NEGATE</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Returns the negative of the floating point value <var>x</var>.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fABS"></a>Macro: <em>REAL_VALUE_TYPE</em> <strong>REAL_VALUE_ABS</strong> <em>(REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Returns the absolute value of <var>x</var>.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fTO_005fINT"></a>Macro: <em>void</em> <strong>REAL_VALUE_TO_INT</strong> <em>(HOST_WIDE_INT <var>low</var>, HOST_WIDE_INT <var>high</var>, REAL_VALUE_TYPE <var>x</var>)</em></dt>
<dd><p>Converts a floating point value <var>x</var> into a double-precision integer
which is then stored into <var>low</var> and <var>high</var>.  If the value is not
integral, it is truncated.
</p></dd></dl>

<dl>
<dt><a name="index-REAL_005fVALUE_005fFROM_005fINT"></a>Macro: <em>void</em> <strong>REAL_VALUE_FROM_INT</strong> <em>(REAL_VALUE_TYPE <var>x</var>, HOST_WIDE_INT <var>low</var>, HOST_WIDE_INT <var>high</var>, enum machine_mode <var>mode</var>)</em></dt>
<dd><p>Converts a double-precision integer found in <var>low</var> and <var>high</var>,
into a floating point value which is then stored into <var>x</var>.  The
value is truncated to fit in mode <var>mode</var>.
</p></dd></dl>

<hr>
<a name="Mode-Switching"></a>
<div class="header">
<p>
Next: <a href="#Target-Attributes" accesskey="n" rel="next">Target Attributes</a>, Previous: <a href="#Floating-Point" accesskey="p" rel="prev">Floating Point</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Mode-Switching-Instructions"></a>
<h3 class="section">17.24 Mode Switching Instructions</h3>
<a name="index-mode-switching"></a>
<p>The following macros control mode switching optimizations:
</p>
<dl>
<dt><a name="index-OPTIMIZE_005fMODE_005fSWITCHING"></a>Macro: <strong>OPTIMIZE_MODE_SWITCHING</strong> <em>(<var>entity</var>)</em></dt>
<dd><p>Define this macro if the port needs extra instructions inserted for mode
switching in an optimizing compilation.
</p>
<p>For an example, the SH4 can perform both single and double precision
floating point operations, but to perform a single precision operation,
the FPSCR PR bit has to be cleared, while for a double precision
operation, this bit has to be set.  Changing the PR bit requires a general
purpose register as a scratch register, hence these FPSCR sets have to
be inserted before reload, i.e. you can&rsquo;t put this into instruction emitting
or <code>TARGET_MACHINE_DEPENDENT_REORG</code>.
</p>
<p>You can have multiple entities that are mode-switched, and select at run time
which entities actually need it.  <code>OPTIMIZE_MODE_SWITCHING</code> should
return nonzero for any <var>entity</var> that needs mode-switching.
If you define this macro, you also have to define
<code>NUM_MODES_FOR_MODE_SWITCHING</code>, <code>MODE_NEEDED</code>,
<code>MODE_PRIORITY_TO_MODE</code> and <code>EMIT_MODE_SET</code>.
<code>MODE_AFTER</code>, <code>MODE_ENTRY</code>, and <code>MODE_EXIT</code>
are optional.
</p></dd></dl>

<dl>
<dt><a name="index-NUM_005fMODES_005fFOR_005fMODE_005fSWITCHING"></a>Macro: <strong>NUM_MODES_FOR_MODE_SWITCHING</strong></dt>
<dd><p>If you define <code>OPTIMIZE_MODE_SWITCHING</code>, you have to define this as
initializer for an array of integers.  Each initializer element
N refers to an entity that needs mode switching, and specifies the number
of different modes that might need to be set for this entity.
The position of the initializer in the initializer&mdash;starting counting at
zero&mdash;determines the integer that is used to refer to the mode-switched
entity in question.
In macros that take mode arguments / yield a mode result, modes are
represented as numbers 0 &hellip; N - 1.  N is used to specify that no mode
switch is needed / supplied.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fNEEDED"></a>Macro: <strong>MODE_NEEDED</strong> <em>(<var>entity</var>, <var>insn</var>)</em></dt>
<dd><p><var>entity</var> is an integer specifying a mode-switched entity.  If
<code>OPTIMIZE_MODE_SWITCHING</code> is defined, you must define this macro to
return an integer value not larger than the corresponding element in
<code>NUM_MODES_FOR_MODE_SWITCHING</code>, to denote the mode that <var>entity</var> must
be switched into prior to the execution of <var>insn</var>.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fAFTER"></a>Macro: <strong>MODE_AFTER</strong> <em>(<var>entity</var>, <var>mode</var>, <var>insn</var>)</em></dt>
<dd><p><var>entity</var> is an integer specifying a mode-switched entity.  If
this macro is defined, it is evaluated for every <var>insn</var> during
mode switching.  It determines the mode that an insn results in (if
different from the incoming mode).
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fENTRY"></a>Macro: <strong>MODE_ENTRY</strong> <em>(<var>entity</var>)</em></dt>
<dd><p>If this macro is defined, it is evaluated for every <var>entity</var> that needs
mode switching.  It should evaluate to an integer, which is a mode that
<var>entity</var> is assumed to be switched to at function entry.  If <code>MODE_ENTRY</code>
is defined then <code>MODE_EXIT</code> must be defined.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fEXIT"></a>Macro: <strong>MODE_EXIT</strong> <em>(<var>entity</var>)</em></dt>
<dd><p>If this macro is defined, it is evaluated for every <var>entity</var> that needs
mode switching.  It should evaluate to an integer, which is a mode that
<var>entity</var> is assumed to be switched to at function exit.  If <code>MODE_EXIT</code>
is defined then <code>MODE_ENTRY</code> must be defined.
</p></dd></dl>

<dl>
<dt><a name="index-MODE_005fPRIORITY_005fTO_005fMODE"></a>Macro: <strong>MODE_PRIORITY_TO_MODE</strong> <em>(<var>entity</var>, <var>n</var>)</em></dt>
<dd><p>This macro specifies the order in which modes for <var>entity</var> are processed.
0 is the highest priority, <code>NUM_MODES_FOR_MODE_SWITCHING[<var>entity</var>] - 1</code> the
lowest.  The value of the macro should be an integer designating a mode
for <var>entity</var>.  For any fixed <var>entity</var>, <code>mode_priority_to_mode</code>
(<var>entity</var>, <var>n</var>) shall be a bijection in 0 &hellip;
<code>num_modes_for_mode_switching[<var>entity</var>] - 1</code>.
</p></dd></dl>

<dl>
<dt><a name="index-EMIT_005fMODE_005fSET"></a>Macro: <strong>EMIT_MODE_SET</strong> <em>(<var>entity</var>, <var>mode</var>, <var>hard_regs_live</var>)</em></dt>
<dd><p>Generate one or more insns to set <var>entity</var> to <var>mode</var>.
<var>hard_reg_live</var> is the set of hard registers live at the point where
the insn(s) are to be inserted.
</p></dd></dl>

<hr>
<a name="Target-Attributes"></a>
<div class="header">
<p>
Next: <a href="#Emulated-TLS" accesskey="n" rel="next">Emulated TLS</a>, Previous: <a href="#Mode-Switching" accesskey="p" rel="prev">Mode Switching</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-target_002dspecific-uses-of-_005f_005fattribute_005f_005f"></a>
<h3 class="section">17.25 Defining target-specific uses of <code>__attribute__</code></h3>
<a name="index-target-attributes"></a>
<a name="index-machine-attributes"></a>
<a name="index-attributes_002c-target_002dspecific"></a>

<p>Target-specific attributes may be defined for functions, data and types.
These are described using the following target hooks; they also need to
be documented in <samp>extend.texi</samp>.
</p>
<dl>
<dt><a name="index-TARGET_005fATTRIBUTE_005fTABLE"></a>Target Hook: <em>const struct attribute_spec *</em> <strong>TARGET_ATTRIBUTE_TABLE</strong></dt>
<dd><p>If defined, this target hook points to an array of &lsquo;<samp>struct
attribute_spec</samp>&rsquo; (defined in <samp>tree.h</samp>) specifying the machine
specific attributes for this target and some of the restrictions on the
entities to which these attributes are applied and the arguments they
take.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fATTRIBUTE_005fTAKES_005fIDENTIFIER_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_ATTRIBUTE_TAKES_IDENTIFIER_P</strong> <em>(const_tree <var>name</var>)</em></dt>
<dd><p>If defined, this target hook is a function which returns true if the
machine-specific attribute named <var>name</var> expects an identifier
given as its first argument to be passed on as a plain identifier, not
subjected to name lookup.  If this is not defined, the default is
false for all machine-specific attributes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCOMP_005fTYPE_005fATTRIBUTES"></a>Target Hook: <em>int</em> <strong>TARGET_COMP_TYPE_ATTRIBUTES</strong> <em>(const_tree <var>type1</var>, const_tree <var>type2</var>)</em></dt>
<dd><p>If defined, this target hook is a function which returns zero if the attributes on
<var>type1</var> and <var>type2</var> are incompatible, one if they are compatible,
and two if they are nearly compatible (which causes a warning to be
generated).  If this is not defined, machine-specific attributes are
supposed always to be compatible.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSET_005fDEFAULT_005fTYPE_005fATTRIBUTES"></a>Target Hook: <em>void</em> <strong>TARGET_SET_DEFAULT_TYPE_ATTRIBUTES</strong> <em>(tree <var>type</var>)</em></dt>
<dd><p>If defined, this target hook is a function which assigns default attributes to
the newly defined <var>type</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMERGE_005fTYPE_005fATTRIBUTES"></a>Target Hook: <em>tree</em> <strong>TARGET_MERGE_TYPE_ATTRIBUTES</strong> <em>(tree <var>type1</var>, tree <var>type2</var>)</em></dt>
<dd><p>Define this target hook if the merging of type attributes needs special
handling.  If defined, the result is a list of the combined
<code>TYPE_ATTRIBUTES</code> of <var>type1</var> and <var>type2</var>.  It is assumed
that <code>comptypes</code> has already been called and returned 1.  This
function may call <code>merge_attributes</code> to handle machine-independent
merging.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMERGE_005fDECL_005fATTRIBUTES"></a>Target Hook: <em>tree</em> <strong>TARGET_MERGE_DECL_ATTRIBUTES</strong> <em>(tree <var>olddecl</var>, tree <var>newdecl</var>)</em></dt>
<dd><p>Define this target hook if the merging of decl attributes needs special
handling.  If defined, the result is a list of the combined
<code>DECL_ATTRIBUTES</code> of <var>olddecl</var> and <var>newdecl</var>.
<var>newdecl</var> is a duplicate declaration of <var>olddecl</var>.  Examples of
when this is needed are when one attribute overrides another, or when an
attribute is nullified by a subsequent definition.  This function may
call <code>merge_attributes</code> to handle machine-independent merging.
</p>
<a name="index-TARGET_005fDLLIMPORT_005fDECL_005fATTRIBUTES"></a>
<p>If the only target-specific handling you require is &lsquo;<samp>dllimport</samp>&rsquo;
for Microsoft Windows targets, you should define the macro
<code>TARGET_DLLIMPORT_DECL_ATTRIBUTES</code> to <code>1</code>.  The compiler
will then define a function called
<code>merge_dllimport_decl_attributes</code> which can then be defined as
the expansion of <code>TARGET_MERGE_DECL_ATTRIBUTES</code>.  You can also
add <code>handle_dll_attribute</code> in the attribute table for your port
to perform initial processing of the &lsquo;<samp>dllimport</samp>&rsquo; and
&lsquo;<samp>dllexport</samp>&rsquo; attributes.  This is done in <samp>i386/cygwin.h</samp> and
<samp>i386/i386.c</samp>, for example.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fVALID_005fDLLIMPORT_005fATTRIBUTE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_VALID_DLLIMPORT_ATTRIBUTE_P</strong> <em>(const_tree <var>decl</var>)</em></dt>
<dd><p><var>decl</var> is a variable or function with <code>__attribute__((dllimport))</code> specified.  Use this hook if the target needs to add extra validation checks to <code>handle_dll_attribute</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDECLSPEC"></a>Macro: <strong>TARGET_DECLSPEC</strong></dt>
<dd><p>Define this macro to a nonzero value if you want to treat
<code>__declspec(X)</code> as equivalent to <code>__attribute((X))</code>.  By
default, this behavior is enabled only for targets that define
<code>TARGET_DLLIMPORT_DECL_ATTRIBUTES</code>.  The current implementation
of <code>__declspec</code> is via a built-in macro, but you should not rely
on this implementation detail.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINSERT_005fATTRIBUTES"></a>Target Hook: <em>void</em> <strong>TARGET_INSERT_ATTRIBUTES</strong> <em>(tree <var>node</var>, tree *<var>attr_ptr</var>)</em></dt>
<dd><p>Define this target hook if you want to be able to add attributes to a decl
when it is being created.  This is normally useful for back ends which
wish to implement a pragma by using the attributes which correspond to
the pragma&rsquo;s effect.  The <var>node</var> argument is the decl which is being
created.  The <var>attr_ptr</var> argument is a pointer to the attribute list
for this decl.  The list itself should not be modified, since it may be
shared with other decls, but attributes may be chained on the head of
the list and <code>*<var>attr_ptr</var></code> modified to point to the new
attributes, or a copy of the list may be made if further changes are
needed.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFUNCTION_005fATTRIBUTE_005fINLINABLE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_FUNCTION_ATTRIBUTE_INLINABLE_P</strong> <em>(const_tree <var>fndecl</var>)</em></dt>
<dd><a name="index-inlining"></a>
<p>This target hook returns <code>true</code> if it is ok to inline <var>fndecl</var>
into the current function, despite its having target-specific
attributes, <code>false</code> otherwise.  By default, if a function has a
target specific attribute attached to it, it will not be inlined.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fVALID_005fATTRIBUTE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_OPTION_VALID_ATTRIBUTE_P</strong> <em>(tree <var>fndecl</var>, tree <var>name</var>, tree <var>args</var>, int <var>flags</var>)</em></dt>
<dd><p>This hook is called to parse <code>attribute(target(&quot;...&quot;))</code>, which
allows setting target-specific options on individual functions.
These function-specific options may differ
from the options specified on the command line.  The hook should return
<code>true</code> if the options are valid.
</p>
<p>The hook should set the <code>DECL_FUNCTION_SPECIFIC_TARGET</code> field in
the function declaration to hold a pointer to a target-specific
<code>struct cl_target_option</code> structure.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fSAVE"></a>Target Hook: <em>void</em> <strong>TARGET_OPTION_SAVE</strong> <em>(struct cl_target_option *<var>ptr</var>)</em></dt>
<dd><p>This hook is called to save any additional target-specific information
in the <code>struct cl_target_option</code> structure for function-specific
options.
See <a href="#Option-file-format">Option file format</a>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fRESTORE"></a>Target Hook: <em>void</em> <strong>TARGET_OPTION_RESTORE</strong> <em>(struct cl_target_option *<var>ptr</var>)</em></dt>
<dd><p>This hook is called to restore any additional target-specific
information in the <code>struct cl_target_option</code> structure for
function-specific options.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fPRINT"></a>Target Hook: <em>void</em> <strong>TARGET_OPTION_PRINT</strong> <em>(FILE *<var>file</var>, int <var>indent</var>, struct cl_target_option *<var>ptr</var>)</em></dt>
<dd><p>This hook is called to print any additional target-specific
information in the <code>struct cl_target_option</code> structure for
function-specific options.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fPRAGMA_005fPARSE"></a>Target Hook: <em>bool</em> <strong>TARGET_OPTION_PRAGMA_PARSE</strong> <em>(tree <var>args</var>, tree <var>pop_target</var>)</em></dt>
<dd><p>This target hook parses the options for <code>#pragma GCC target</code>, which
sets the target-specific options for functions that occur later in the
input stream.  The options accepted should be the same as those handled by the
<code>TARGET_OPTION_VALID_ATTRIBUTE_P</code> hook.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fOVERRIDE"></a>Target Hook: <em>void</em> <strong>TARGET_OPTION_OVERRIDE</strong> <em>(void)</em></dt>
<dd><p>Sometimes certain combinations of command options do not make sense on
a particular target machine.  You can override the hook
<code>TARGET_OPTION_OVERRIDE</code> to take account of this.  This hooks is called
once just after all the command options have been parsed.
</p>
<p>Don&rsquo;t use this hook to turn on various extra optimizations for
<samp>-O</samp>.  That is what <code>TARGET_OPTION_OPTIMIZATION</code> is for.
</p>
<p>If you need to do something whenever the optimization level is
changed via the optimize attribute or pragma, see
<code>TARGET_OVERRIDE_OPTIONS_AFTER_CHANGE</code>
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTION_005fFUNCTION_005fVERSIONS"></a>Target Hook: <em>bool</em> <strong>TARGET_OPTION_FUNCTION_VERSIONS</strong> <em>(tree <var>decl1</var>, tree <var>decl2</var>)</em></dt>
<dd><p>This target hook returns <code>true</code> if <var>DECL1</var> and <var>DECL2</var> are
versions of the same function.  <var>DECL1</var> and <var>DECL2</var> are function
versions if and only if they have the same function signature and
different target specific attributes, that is, they are compiled for
different target machines.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCAN_005fINLINE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_CAN_INLINE_P</strong> <em>(tree <var>caller</var>, tree <var>callee</var>)</em></dt>
<dd><p>This target hook returns <code>false</code> if the <var>caller</var> function
cannot inline <var>callee</var>, based on target specific information.  By
default, inlining is not allowed if the callee function has function
specific target options and the caller does not use the same options.
</p></dd></dl>

<hr>
<a name="Emulated-TLS"></a>
<div class="header">
<p>
Next: <a href="#MIPS-Coprocessors" accesskey="n" rel="next">MIPS Coprocessors</a>, Previous: <a href="#Target-Attributes" accesskey="p" rel="prev">Target Attributes</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Emulating-TLS"></a>
<h3 class="section">17.26 Emulating TLS</h3>
<a name="index-Emulated-TLS"></a>

<p>For targets whose psABI does not provide Thread Local Storage via
specific relocations and instruction sequences, an emulation layer is
used.  A set of target hooks allows this emulation layer to be
configured for the requirements of a particular target.  For instance
the psABI may in fact specify TLS support in terms of an emulation
layer.
</p>
<p>The emulation layer works by creating a control object for every TLS
object.  To access the TLS object, a lookup function is provided
which, when given the address of the control object, will return the
address of the current thread&rsquo;s instance of the TLS object.
</p>
<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fGET_005fADDRESS"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_GET_ADDRESS</strong></dt>
<dd><p>Contains the name of the helper function that uses a TLS control
object to locate a TLS instance.  The default causes libgcc&rsquo;s
emulated TLS helper function to be used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fREGISTER_005fCOMMON"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_REGISTER_COMMON</strong></dt>
<dd><p>Contains the name of the helper function that should be used at
program startup to register TLS objects that are implicitly
initialized to zero.  If this is <code>NULL</code>, all TLS objects will
have explicit initializers.  The default causes libgcc&rsquo;s emulated TLS
registration function to be used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fVAR_005fSECTION"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_VAR_SECTION</strong></dt>
<dd><p>Contains the name of the section in which TLS control variables should
be placed.  The default of <code>NULL</code> allows these to be placed in
any section.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fTMPL_005fSECTION"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_TMPL_SECTION</strong></dt>
<dd><p>Contains the name of the section in which TLS initializers should be
placed.  The default of <code>NULL</code> allows these to be placed in any
section.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fVAR_005fPREFIX"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_VAR_PREFIX</strong></dt>
<dd><p>Contains the prefix to be prepended to TLS control variable names.
The default of <code>NULL</code> uses a target-specific prefix.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fTMPL_005fPREFIX"></a>Target Hook: <em>const char *</em> <strong>TARGET_EMUTLS_TMPL_PREFIX</strong></dt>
<dd><p>Contains the prefix to be prepended to TLS initializer objects.  The
default of <code>NULL</code> uses a target-specific prefix.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fVAR_005fFIELDS"></a>Target Hook: <em>tree</em> <strong>TARGET_EMUTLS_VAR_FIELDS</strong> <em>(tree <var>type</var>, tree *<var>name</var>)</em></dt>
<dd><p>Specifies a function that generates the FIELD_DECLs for a TLS control
object type.  <var>type</var> is the RECORD_TYPE the fields are for and
<var>name</var> should be filled with the structure tag, if the default of
<code>__emutls_object</code> is unsuitable.  The default creates a type suitable
for libgcc&rsquo;s emulated TLS function.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fVAR_005fINIT"></a>Target Hook: <em>tree</em> <strong>TARGET_EMUTLS_VAR_INIT</strong> <em>(tree <var>var</var>, tree <var>decl</var>, tree <var>tmpl_addr</var>)</em></dt>
<dd><p>Specifies a function that generates the CONSTRUCTOR to initialize a
TLS control object.  <var>var</var> is the TLS control object, <var>decl</var>
is the TLS object and <var>tmpl_addr</var> is the address of the
initializer.  The default initializes libgcc&rsquo;s emulated TLS control object.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fVAR_005fALIGN_005fFIXED"></a>Target Hook: <em>bool</em> <strong>TARGET_EMUTLS_VAR_ALIGN_FIXED</strong></dt>
<dd><p>Specifies whether the alignment of TLS control variable objects is
fixed and should not be increased as some backends may do to optimize
single objects.  The default is false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEMUTLS_005fDEBUG_005fFORM_005fTLS_005fADDRESS"></a>Target Hook: <em>bool</em> <strong>TARGET_EMUTLS_DEBUG_FORM_TLS_ADDRESS</strong></dt>
<dd><p>Specifies whether a DWARF <code>DW_OP_form_tls_address</code> location descriptor
may be used to describe emulated TLS control objects.
</p></dd></dl>

<hr>
<a name="MIPS-Coprocessors"></a>
<div class="header">
<p>
Next: <a href="#PCH-Target" accesskey="n" rel="next">PCH Target</a>, Previous: <a href="#Emulated-TLS" accesskey="p" rel="prev">Emulated TLS</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Defining-coprocessor-specifics-for-MIPS-targets_002e"></a>
<h3 class="section">17.27 Defining coprocessor specifics for MIPS targets.</h3>
<a name="index-MIPS-coprocessor_002ddefinition-macros"></a>

<p>The MIPS specification allows MIPS implementations to have as many as 4
coprocessors, each with as many as 32 private registers.  GCC supports
accessing these registers and transferring values between the registers
and memory using asm-ized variables.  For example:
</p>
<div class="smallexample">
<pre class="smallexample">  register unsigned int cp0count asm (&quot;c0r1&quot;);
  unsigned int d;

  d = cp0count + 3;
</pre></div>

<p>(&ldquo;c0r1&rdquo; is the default name of register 1 in coprocessor 0; alternate
names may be added as described below, or the default names may be
overridden entirely in <code>SUBTARGET_CONDITIONAL_REGISTER_USAGE</code>.)
</p>
<p>Coprocessor registers are assumed to be epilogue-used; sets to them will
be preserved even if it does not appear that the register is used again
later in the function.
</p>
<p>Another note: according to the MIPS spec, coprocessor 1 (if present) is
the FPU.  One accesses COP1 registers through standard mips
floating-point support; they are not included in this mechanism.
</p>
<p>There is one macro used in defining the MIPS coprocessor interface which
you may want to override in subtargets; it is described below.
</p>
<hr>
<a name="PCH-Target"></a>
<div class="header">
<p>
Next: <a href="#C_002b_002b-ABI" accesskey="n" rel="next">C++ ABI</a>, Previous: <a href="#MIPS-Coprocessors" accesskey="p" rel="prev">MIPS Coprocessors</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Parameters-for-Precompiled-Header-Validity-Checking"></a>
<h3 class="section">17.28 Parameters for Precompiled Header Validity Checking</h3>
<a name="index-parameters_002c-precompiled-headers"></a>

<dl>
<dt><a name="index-TARGET_005fGET_005fPCH_005fVALIDITY"></a>Target Hook: <em>void *</em> <strong>TARGET_GET_PCH_VALIDITY</strong> <em>(size_t *<var>sz</var>)</em></dt>
<dd><p>This hook returns a pointer to the data needed by
<code>TARGET_PCH_VALID_P</code> and sets
&lsquo;<samp>*<var>sz</var></samp>&rsquo; to the size of the data in bytes.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPCH_005fVALID_005fP"></a>Target Hook: <em>const char *</em> <strong>TARGET_PCH_VALID_P</strong> <em>(const void *<var>data</var>, size_t <var>sz</var>)</em></dt>
<dd><p>This hook checks whether the options used to create a PCH file are
compatible with the current settings.  It returns <code>NULL</code>
if so and a suitable error message if not.  Error messages will
be presented to the user and must be localized using &lsquo;<samp>_(<var>msg</var>)</samp>&rsquo;.
</p>
<p><var>data</var> is the data that was returned by <code>TARGET_GET_PCH_VALIDITY</code>
when the PCH file was created and <var>sz</var> is the size of that data in bytes.
It&rsquo;s safe to assume that the data was created by the same version of the
compiler, so no format checking is needed.
</p>
<p>The default definition of <code>default_pch_valid_p</code> should be
suitable for most targets.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCHECK_005fPCH_005fTARGET_005fFLAGS"></a>Target Hook: <em>const char *</em> <strong>TARGET_CHECK_PCH_TARGET_FLAGS</strong> <em>(int <var>pch_flags</var>)</em></dt>
<dd><p>If this hook is nonnull, the default implementation of
<code>TARGET_PCH_VALID_P</code> will use it to check for compatible values
of <code>target_flags</code>.  <var>pch_flags</var> specifies the value that
<code>target_flags</code> had when the PCH file was created.  The return
value is the same as for <code>TARGET_PCH_VALID_P</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPREPARE_005fPCH_005fSAVE"></a>Target Hook: <em>void</em> <strong>TARGET_PREPARE_PCH_SAVE</strong> <em>(void)</em></dt>
<dd><p>Called before writing out a PCH file.  If the target has some
garbage-collected data that needs to be in a particular state on PCH loads,
it can use this hook to enforce that state.  Very few targets need
to do anything here.
</p></dd></dl>

<hr>
<a name="C_002b_002b-ABI"></a>
<div class="header">
<p>
Next: <a href="#Named-Address-Spaces" accesskey="n" rel="next">Named Address Spaces</a>, Previous: <a href="#PCH-Target" accesskey="p" rel="prev">PCH Target</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="C_002b_002b-ABI-parameters"></a>
<h3 class="section">17.29 C++ ABI parameters</h3>
<a name="index-parameters_002c-c_002b_002b-abi"></a>

<dl>
<dt><a name="index-TARGET_005fCXX_005fGUARD_005fTYPE"></a>Target Hook: <em>tree</em> <strong>TARGET_CXX_GUARD_TYPE</strong> <em>(void)</em></dt>
<dd><p>Define this hook to override the integer type used for guard variables.
These are used to implement one-time construction of static objects.  The
default is long_long_integer_type_node.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fGUARD_005fMASK_005fBIT"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_GUARD_MASK_BIT</strong> <em>(void)</em></dt>
<dd><p>This hook determines how guard variables are used.  It should return
<code>false</code> (the default) if the first byte should be used.  A return value of
<code>true</code> indicates that only the least significant bit should be used.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fGET_005fCOOKIE_005fSIZE"></a>Target Hook: <em>tree</em> <strong>TARGET_CXX_GET_COOKIE_SIZE</strong> <em>(tree <var>type</var>)</em></dt>
<dd><p>This hook returns the size of the cookie to use when allocating an array
whose elements have the indicated <var>type</var>.  Assumes that it is already
known that a cookie is needed.  The default is
<code>max(sizeof (size_t), alignof(type))</code>, as defined in section 2.7 of the
IA64/Generic C++ ABI.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fCOOKIE_005fHAS_005fSIZE"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_COOKIE_HAS_SIZE</strong> <em>(void)</em></dt>
<dd><p>This hook should return <code>true</code> if the element size should be stored in
array cookies.  The default is to return <code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fIMPORT_005fEXPORT_005fCLASS"></a>Target Hook: <em>int</em> <strong>TARGET_CXX_IMPORT_EXPORT_CLASS</strong> <em>(tree <var>type</var>, int <var>import_export</var>)</em></dt>
<dd><p>If defined by a backend this hook allows the decision made to export
class <var>type</var> to be overruled.  Upon entry <var>import_export</var>
will contain 1 if the class is going to be exported, -1 if it is going
to be imported and 0 otherwise.  This function should return the
modified value and perform any other actions necessary to support the
backend&rsquo;s targeted operating system.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fCDTOR_005fRETURNS_005fTHIS"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_CDTOR_RETURNS_THIS</strong> <em>(void)</em></dt>
<dd><p>This hook should return <code>true</code> if constructors and destructors return
the address of the object created/destroyed.  The default is to return
<code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fKEY_005fMETHOD_005fMAY_005fBE_005fINLINE"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_KEY_METHOD_MAY_BE_INLINE</strong> <em>(void)</em></dt>
<dd><p>This hook returns true if the key method for a class (i.e., the method
which, if defined in the current translation unit, causes the virtual
table to be emitted) may be an inline function.  Under the standard
Itanium C++ ABI the key method may be an inline function so long as
the function is not declared inline in the class definition.  Under
some variants of the ABI, an inline function can never be the key
method.  The default is to return <code>true</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fDETERMINE_005fCLASS_005fDATA_005fVISIBILITY"></a>Target Hook: <em>void</em> <strong>TARGET_CXX_DETERMINE_CLASS_DATA_VISIBILITY</strong> <em>(tree <var>decl</var>)</em></dt>
<dd><p><var>decl</var> is a virtual table, virtual table table, typeinfo object, or other similar implicit class data object that will be emitted with external linkage in this translation unit.  No ELF visibility has been explicitly specified.  If the target needs to specify a visibility other than that of the containing class, use this hook to set <code>DECL_VISIBILITY</code> and <code>DECL_VISIBILITY_SPECIFIED</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fCLASS_005fDATA_005fALWAYS_005fCOMDAT"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_CLASS_DATA_ALWAYS_COMDAT</strong> <em>(void)</em></dt>
<dd><p>This hook returns true (the default) if virtual tables and other
similar implicit class data objects are always COMDAT if they have
external linkage.  If this hook returns false, then class data for
classes whose virtual table will be emitted in only one translation
unit will not be COMDAT.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fLIBRARY_005fRTTI_005fCOMDAT"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_LIBRARY_RTTI_COMDAT</strong> <em>(void)</em></dt>
<dd><p>This hook returns true (the default) if the RTTI information for
the basic types which is defined in the C++ runtime should always
be COMDAT, false if it should not be COMDAT.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fUSE_005fAEABI_005fATEXIT"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_USE_AEABI_ATEXIT</strong> <em>(void)</em></dt>
<dd><p>This hook returns true if <code>__aeabi_atexit</code> (as defined by the ARM EABI)
should be used to register static destructors when <samp>-fuse-cxa-atexit</samp>
is in effect.  The default is to return false to use <code>__cxa_atexit</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fUSE_005fATEXIT_005fFOR_005fCXA_005fATEXIT"></a>Target Hook: <em>bool</em> <strong>TARGET_CXX_USE_ATEXIT_FOR_CXA_ATEXIT</strong> <em>(void)</em></dt>
<dd><p>This hook returns true if the target <code>atexit</code> function can be used
in the same manner as <code>__cxa_atexit</code> to register C++ static
destructors. This requires that <code>atexit</code>-registered functions in
shared libraries are run in the correct order when the libraries are
unloaded. The default is to return false.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fADJUST_005fCLASS_005fAT_005fDEFINITION"></a>Target Hook: <em>void</em> <strong>TARGET_CXX_ADJUST_CLASS_AT_DEFINITION</strong> <em>(tree <var>type</var>)</em></dt>
<dd><p><var>type</var> is a C++ class (i.e., RECORD_TYPE or UNION_TYPE) that has just been defined.  Use this hook to make adjustments to the class (eg, tweak visibility or perform any other required target modifications).
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCXX_005fDECL_005fMANGLING_005fCONTEXT"></a>Target Hook: <em>tree</em> <strong>TARGET_CXX_DECL_MANGLING_CONTEXT</strong> <em>(const_tree <var>decl</var>)</em></dt>
<dd><p>Return target-specific mangling context of <var>decl</var> or <code>NULL_TREE</code>.
</p></dd></dl>

<hr>
<a name="Named-Address-Spaces"></a>
<div class="header">
<p>
Next: <a href="#Misc" accesskey="n" rel="next">Misc</a>, Previous: <a href="#C_002b_002b-ABI" accesskey="p" rel="prev">C++ ABI</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Adding-support-for-named-address-spaces"></a>
<h3 class="section">17.30 Adding support for named address spaces</h3>
<a name="index-named-address-spaces"></a>

<p>The draft technical report of the ISO/IEC JTC1 S22 WG14 N1275
standards committee, <cite>Programming Languages - C - Extensions to
support embedded processors</cite>, specifies a syntax for embedded
processors to specify alternate address spaces.  You can configure a
GCC port to support section 5.1 of the draft report to add support for
address spaces other than the default address space.  These address
spaces are new keywords that are similar to the <code>volatile</code> and
<code>const</code> type attributes.
</p>
<p>Pointers to named address spaces can have a different size than
pointers to the generic address space.
</p>
<p>For example, the SPU port uses the <code>__ea</code> address space to refer
to memory in the host processor, rather than memory local to the SPU
processor.  Access to memory in the <code>__ea</code> address space involves
issuing DMA operations to move data between the host processor and the
local processor memory address space.  Pointers in the <code>__ea</code>
address space are either 32 bits or 64 bits based on the
<samp>-mea32</samp> or <samp>-mea64</samp> switches (native SPU pointers are
always 32 bits).
</p>
<p>Internally, address spaces are represented as a small integer in the
range 0 to 15 with address space 0 being reserved for the generic
address space.
</p>
<p>To register a named address space qualifier keyword with the C front end,
the target may call the <code>c_register_addr_space</code> routine.  For example,
the SPU port uses the following to declare <code>__ea</code> as the keyword for
named address space #1:
</p><div class="smallexample">
<pre class="smallexample">#define ADDR_SPACE_EA 1
c_register_addr_space (&quot;__ea&quot;, ADDR_SPACE_EA);
</pre></div>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fPOINTER_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_ADDR_SPACE_POINTER_MODE</strong> <em>(addr_space_t <var>address_space</var>)</em></dt>
<dd><p>Define this to return the machine mode to use for pointers to
<var>address_space</var> if the target supports named address spaces.
The default version of this hook returns <code>ptr_mode</code> for the
generic address space only.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fADDRESS_005fMODE"></a>Target Hook: <em>enum machine_mode</em> <strong>TARGET_ADDR_SPACE_ADDRESS_MODE</strong> <em>(addr_space_t <var>address_space</var>)</em></dt>
<dd><p>Define this to return the machine mode to use for addresses in
<var>address_space</var> if the target supports named address spaces.
The default version of this hook returns <code>Pmode</code> for the
generic address space only.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fVALID_005fPOINTER_005fMODE"></a>Target Hook: <em>bool</em> <strong>TARGET_ADDR_SPACE_VALID_POINTER_MODE</strong> <em>(enum machine_mode <var>mode</var>, addr_space_t <var>as</var>)</em></dt>
<dd><p>Define this to return nonzero if the port can handle pointers
with machine mode <var>mode</var> to address space <var>as</var>.  This target
hook is the same as the <code>TARGET_VALID_POINTER_MODE</code> target hook,
except that it includes explicit named address space support.  The default
version of this hook returns true for the modes returned by either the
<code>TARGET_ADDR_SPACE_POINTER_MODE</code> or <code>TARGET_ADDR_SPACE_ADDRESS_MODE</code>
target hooks for the given address space.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fLEGITIMATE_005fADDRESS_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_ADDR_SPACE_LEGITIMATE_ADDRESS_P</strong> <em>(enum machine_mode <var>mode</var>, rtx <var>exp</var>, bool <var>strict</var>, addr_space_t <var>as</var>)</em></dt>
<dd><p>Define this to return true if <var>exp</var> is a valid address for mode
<var>mode</var> in the named address space <var>as</var>.  The <var>strict</var>
parameter says whether strict addressing is in effect after reload has
finished.  This target hook is the same as the
<code>TARGET_LEGITIMATE_ADDRESS_P</code> target hook, except that it includes
explicit named address space support.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fLEGITIMIZE_005fADDRESS"></a>Target Hook: <em>rtx</em> <strong>TARGET_ADDR_SPACE_LEGITIMIZE_ADDRESS</strong> <em>(rtx <var>x</var>, rtx <var>oldx</var>, enum machine_mode <var>mode</var>, addr_space_t <var>as</var>)</em></dt>
<dd><p>Define this to modify an invalid address <var>x</var> to be a valid address
with mode <var>mode</var> in the named address space <var>as</var>.  This target
hook is the same as the <code>TARGET_LEGITIMIZE_ADDRESS</code> target hook,
except that it includes explicit named address space support.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fSUBSET_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_ADDR_SPACE_SUBSET_P</strong> <em>(addr_space_t <var>subset</var>, addr_space_t <var>superset</var>)</em></dt>
<dd><p>Define this to return whether the <var>subset</var> named address space is
contained within the <var>superset</var> named address space.  Pointers to
a named address space that is a subset of another named address space
will be converted automatically without a cast if used together in
arithmetic operations.  Pointers to a superset address space can be
converted to pointers to a subset address space via explicit casts.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fADDR_005fSPACE_005fCONVERT"></a>Target Hook: <em>rtx</em> <strong>TARGET_ADDR_SPACE_CONVERT</strong> <em>(rtx <var>op</var>, tree <var>from_type</var>, tree <var>to_type</var>)</em></dt>
<dd><p>Define this to convert the pointer expression represented by the RTL
<var>op</var> with type <var>from_type</var> that points to a named address
space to a new pointer expression with type <var>to_type</var> that points
to a different named address space.  When this hook it called, it is
guaranteed that one of the two address spaces is a subset of the other,
as determined by the <code>TARGET_ADDR_SPACE_SUBSET_P</code> target hook.
</p></dd></dl>

<hr>
<a name="Misc"></a>
<div class="header">
<p>
Previous: <a href="#Named-Address-Spaces" accesskey="p" rel="prev">Named Address Spaces</a>, Up: <a href="#Target-Macros" accesskey="u" rel="up">Target Macros</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Miscellaneous-Parameters"></a>
<h3 class="section">17.31 Miscellaneous Parameters</h3>
<a name="index-parameters_002c-miscellaneous"></a>

<p>Here are several miscellaneous parameters.
</p>
<dl>
<dt><a name="index-HAS_005fLONG_005fCOND_005fBRANCH"></a>Macro: <strong>HAS_LONG_COND_BRANCH</strong></dt>
<dd><p>Define this boolean macro to indicate whether or not your architecture
has conditional branches that can span all of memory.  It is used in
conjunction with an optimization that partitions hot and cold basic
blocks into separate sections of the executable.  If this macro is
set to false, gcc will convert any conditional branches that attempt
to cross between sections into unconditional branches or indirect jumps.
</p></dd></dl>

<dl>
<dt><a name="index-HAS_005fLONG_005fUNCOND_005fBRANCH"></a>Macro: <strong>HAS_LONG_UNCOND_BRANCH</strong></dt>
<dd><p>Define this boolean macro to indicate whether or not your architecture
has unconditional branches that can span all of memory.  It is used in
conjunction with an optimization that partitions hot and cold basic
blocks into separate sections of the executable.  If this macro is
set to false, gcc will convert any unconditional branches that attempt
to cross between sections into indirect jumps.
</p></dd></dl>

<dl>
<dt><a name="index-CASE_005fVECTOR_005fMODE"></a>Macro: <strong>CASE_VECTOR_MODE</strong></dt>
<dd><p>An alias for a machine mode name.  This is the machine mode that
elements of a jump-table should have.
</p></dd></dl>

<dl>
<dt><a name="index-CASE_005fVECTOR_005fSHORTEN_005fMODE"></a>Macro: <strong>CASE_VECTOR_SHORTEN_MODE</strong> <em>(<var>min_offset</var>, <var>max_offset</var>, <var>body</var>)</em></dt>
<dd><p>Optional: return the preferred mode for an <code>addr_diff_vec</code>
when the minimum and maximum offset are known.  If you define this,
it enables extra code in branch shortening to deal with <code>addr_diff_vec</code>.
To make this work, you also have to define <code>INSN_ALIGN</code> and
make the alignment for <code>addr_diff_vec</code> explicit.
The <var>body</var> argument is provided so that the offset_unsigned and scale
flags can be updated.
</p></dd></dl>

<dl>
<dt><a name="index-CASE_005fVECTOR_005fPC_005fRELATIVE"></a>Macro: <strong>CASE_VECTOR_PC_RELATIVE</strong></dt>
<dd><p>Define this macro to be a C expression to indicate when jump-tables
should contain relative addresses.  You need not define this macro if
jump-tables never contain relative addresses, or jump-tables should
contain relative addresses only when <samp>-fPIC</samp> or <samp>-fPIC</samp>
is in effect.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCASE_005fVALUES_005fTHRESHOLD"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_CASE_VALUES_THRESHOLD</strong> <em>(void)</em></dt>
<dd><p>This function return the smallest number of different values for which it
is best to use a jump-table instead of a tree of conditional branches.
The default is four for machines with a <code>casesi</code> instruction and
five otherwise.  This is best for most machines.
</p></dd></dl>

<dl>
<dt><a name="index-WORD_005fREGISTER_005fOPERATIONS"></a>Macro: <strong>WORD_REGISTER_OPERATIONS</strong></dt>
<dd><p>Define this macro if operations between registers with integral mode
smaller than a word are always performed on the entire register.
Most RISC machines have this property and most CISC machines do not.
</p></dd></dl>

<dl>
<dt><a name="index-LOAD_005fEXTEND_005fOP"></a>Macro: <strong>LOAD_EXTEND_OP</strong> <em>(<var>mem_mode</var>)</em></dt>
<dd><p>Define this macro to be a C expression indicating when insns that read
memory in <var>mem_mode</var>, an integral mode narrower than a word, set the
bits outside of <var>mem_mode</var> to be either the sign-extension or the
zero-extension of the data read.  Return <code>SIGN_EXTEND</code> for values
of <var>mem_mode</var> for which the
insn sign-extends, <code>ZERO_EXTEND</code> for which it zero-extends, and
<code>UNKNOWN</code> for other modes.
</p>
<p>This macro is not called with <var>mem_mode</var> non-integral or with a width
greater than or equal to <code>BITS_PER_WORD</code>, so you may return any
value in this case.  Do not define this macro if it would always return
<code>UNKNOWN</code>.  On machines where this macro is defined, you will normally
define it as the constant <code>SIGN_EXTEND</code> or <code>ZERO_EXTEND</code>.
</p>
<p>You may return a non-<code>UNKNOWN</code> value even if for some hard registers
the sign extension is not performed, if for the <code>REGNO_REG_CLASS</code>
of these hard registers <code>CANNOT_CHANGE_MODE_CLASS</code> returns nonzero
when the <var>from</var> mode is <var>mem_mode</var> and the <var>to</var> mode is any
integral mode larger than this but not larger than <code>word_mode</code>.
</p>
<p>You must return <code>UNKNOWN</code> if for some hard registers that allow this
mode, <code>CANNOT_CHANGE_MODE_CLASS</code> says that they cannot change to
<code>word_mode</code>, but that they can change to another integral mode that
is larger then <var>mem_mode</var> but still smaller than <code>word_mode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-SHORT_005fIMMEDIATES_005fSIGN_005fEXTEND"></a>Macro: <strong>SHORT_IMMEDIATES_SIGN_EXTEND</strong></dt>
<dd><p>Define this macro if loading short immediate values into registers sign
extends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMIN_005fDIVISIONS_005fFOR_005fRECIP_005fMUL"></a>Target Hook: <em>unsigned int</em> <strong>TARGET_MIN_DIVISIONS_FOR_RECIP_MUL</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>When <samp>-ffast-math</samp> is in effect, GCC tries to optimize
divisions by the same divisor, by turning them into multiplications by
the reciprocal.  This target hook specifies the minimum number of divisions
that should be there for GCC to perform the optimization for a variable
of mode <var>mode</var>.  The default implementation returns 3 if the machine
has an instruction for the division, and 2 if it does not.
</p></dd></dl>

<dl>
<dt><a name="index-MOVE_005fMAX"></a>Macro: <strong>MOVE_MAX</strong></dt>
<dd><p>The maximum number of bytes that a single instruction can move quickly
between memory and registers or between two memory locations.
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fMOVE_005fMAX"></a>Macro: <strong>MAX_MOVE_MAX</strong></dt>
<dd><p>The maximum number of bytes that a single instruction can move quickly
between memory and registers or between two memory locations.  If this
is undefined, the default is <code>MOVE_MAX</code>.  Otherwise, it is the
constant value that is the largest value that <code>MOVE_MAX</code> can have
at run-time.
</p></dd></dl>

<dl>
<dt><a name="index-SHIFT_005fCOUNT_005fTRUNCATED"></a>Macro: <strong>SHIFT_COUNT_TRUNCATED</strong></dt>
<dd><p>A C expression that is nonzero if on this machine the number of bits
actually used for the count of a shift operation is equal to the number
of bits needed to represent the size of the object being shifted.  When
this macro is nonzero, the compiler will assume that it is safe to omit
a sign-extend, zero-extend, and certain bitwise &lsquo;and&rsquo; instructions that
truncates the count of a shift operation.  On machines that have
instructions that act on bit-fields at variable positions, which may
include &lsquo;bit test&rsquo; instructions, a nonzero <code>SHIFT_COUNT_TRUNCATED</code>
also enables deletion of truncations of the values that serve as
arguments to bit-field instructions.
</p>
<p>If both types of instructions truncate the count (for shifts) and
position (for bit-field operations), or if no variable-position bit-field
instructions exist, you should define this macro.
</p>
<p>However, on some machines, such as the 80386 and the 680x0, truncation
only applies to shift operations and not the (real or pretended)
bit-field operations.  Define <code>SHIFT_COUNT_TRUNCATED</code> to be zero on
such machines.  Instead, add patterns to the <samp>md</samp> file that include
the implied truncation of the shift instructions.
</p>
<p>You need not define this macro if it would always have the value of zero.
</p></dd></dl>

<a name="TARGET_005fSHIFT_005fTRUNCATION_005fMASK"></a><dl>
<dt><a name="index-TARGET_005fSHIFT_005fTRUNCATION_005fMASK"></a>Target Hook: <em>unsigned HOST_WIDE_INT</em> <strong>TARGET_SHIFT_TRUNCATION_MASK</strong> <em>(enum machine_mode <var>mode</var>)</em></dt>
<dd><p>This function describes how the standard shift patterns for <var>mode</var>
deal with shifts by negative amounts or by more than the width of the mode.
See <a href="#shift-patterns">shift patterns</a>.
</p>
<p>On many machines, the shift patterns will apply a mask <var>m</var> to the
shift count, meaning that a fixed-width shift of <var>x</var> by <var>y</var> is
equivalent to an arbitrary-width shift of <var>x</var> by <var>y &amp; m</var>.  If
this is true for mode <var>mode</var>, the function should return <var>m</var>,
otherwise it should return 0.  A return value of 0 indicates that no
particular behavior is guaranteed.
</p>
<p>Note that, unlike <code>SHIFT_COUNT_TRUNCATED</code>, this function does
<em>not</em> apply to general shift rtxes; it applies only to instructions
that are generated by the named shift patterns.
</p>
<p>The default implementation of this function returns
<code>GET_MODE_BITSIZE (<var>mode</var>) - 1</code> if <code>SHIFT_COUNT_TRUNCATED</code>
and 0 otherwise.  This definition is always safe, but if
<code>SHIFT_COUNT_TRUNCATED</code> is false, and some shift patterns
nevertheless truncate the shift count, you may get better code
by overriding it.
</p></dd></dl>

<dl>
<dt><a name="index-TRULY_005fNOOP_005fTRUNCATION"></a>Macro: <strong>TRULY_NOOP_TRUNCATION</strong> <em>(<var>outprec</var>, <var>inprec</var>)</em></dt>
<dd><p>A C expression which is nonzero if on this machine it is safe to
&ldquo;convert&rdquo; an integer of <var>inprec</var> bits to one of <var>outprec</var>
bits (where <var>outprec</var> is smaller than <var>inprec</var>) by merely
operating on it as if it had only <var>outprec</var> bits.
</p>
<p>On many machines, this expression can be 1.
</p>
<p>When <code>TRULY_NOOP_TRUNCATION</code> returns 1 for a pair of sizes for
modes for which <code>MODES_TIEABLE_P</code> is 0, suboptimal code can result.
If this is the case, making <code>TRULY_NOOP_TRUNCATION</code> return 0 in
such cases may improve things.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMODE_005fREP_005fEXTENDED"></a>Target Hook: <em>int</em> <strong>TARGET_MODE_REP_EXTENDED</strong> <em>(enum machine_mode <var>mode</var>, enum machine_mode <var>rep_mode</var>)</em></dt>
<dd><p>The representation of an integral mode can be such that the values
are always extended to a wider integral mode.  Return
<code>SIGN_EXTEND</code> if values of <var>mode</var> are represented in
sign-extended form to <var>rep_mode</var>.  Return <code>UNKNOWN</code>
otherwise.  (Currently, none of the targets use zero-extended
representation this way so unlike <code>LOAD_EXTEND_OP</code>,
<code>TARGET_MODE_REP_EXTENDED</code> is expected to return either
<code>SIGN_EXTEND</code> or <code>UNKNOWN</code>.  Also no target extends
<var>mode</var> to <var>rep_mode</var> so that <var>rep_mode</var> is not the next
widest integral mode and currently we take advantage of this fact.)
</p>
<p>Similarly to <code>LOAD_EXTEND_OP</code> you may return a non-<code>UNKNOWN</code>
value even if the extension is not performed on certain hard registers
as long as for the <code>REGNO_REG_CLASS</code> of these hard registers
<code>CANNOT_CHANGE_MODE_CLASS</code> returns nonzero.
</p>
<p>Note that <code>TARGET_MODE_REP_EXTENDED</code> and <code>LOAD_EXTEND_OP</code>
describe two related properties.  If you define
<code>TARGET_MODE_REP_EXTENDED (mode, word_mode)</code> you probably also want
to define <code>LOAD_EXTEND_OP (mode)</code> to return the same type of
extension.
</p>
<p>In order to enforce the representation of <code>mode</code>,
<code>TRULY_NOOP_TRUNCATION</code> should return false when truncating to
<code>mode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STORE_005fFLAG_005fVALUE"></a>Macro: <strong>STORE_FLAG_VALUE</strong></dt>
<dd><p>A C expression describing the value returned by a comparison operator
with an integral mode and stored by a store-flag instruction
(&lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo;) when the condition is true.  This description must
apply to <em>all</em> the &lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo; patterns and all the
comparison operators whose results have a <code>MODE_INT</code> mode.
</p>
<p>A value of 1 or -1 means that the instruction implementing the
comparison operator returns exactly 1 or -1 when the comparison is true
and 0 when the comparison is false.  Otherwise, the value indicates
which bits of the result are guaranteed to be 1 when the comparison is
true.  This value is interpreted in the mode of the comparison
operation, which is given by the mode of the first operand in the
&lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo; pattern.  Either the low bit or the sign bit of
<code>STORE_FLAG_VALUE</code> be on.  Presently, only those bits are used by
the compiler.
</p>
<p>If <code>STORE_FLAG_VALUE</code> is neither 1 or -1, the compiler will
generate code that depends only on the specified bits.  It can also
replace comparison operators with equivalent operations if they cause
the required bits to be set, even if the remaining bits are undefined.
For example, on a machine whose comparison operators return an
<code>SImode</code> value and where <code>STORE_FLAG_VALUE</code> is defined as
&lsquo;<samp>0x80000000</samp>&rsquo;, saying that just the sign bit is relevant, the
expression
</p>
<div class="smallexample">
<pre class="smallexample">(ne:SI (and:SI <var>x</var> (const_int <var>power-of-2</var>)) (const_int 0))
</pre></div>

<p>can be converted to
</p>
<div class="smallexample">
<pre class="smallexample">(ashift:SI <var>x</var> (const_int <var>n</var>))
</pre></div>

<p>where <var>n</var> is the appropriate shift count to move the bit being
tested into the sign bit.
</p>
<p>There is no way to describe a machine that always sets the low-order bit
for a true value, but does not guarantee the value of any other bits,
but we do not know of any machine that has such an instruction.  If you
are trying to port GCC to such a machine, include an instruction to
perform a logical-and of the result with 1 in the pattern for the
comparison operators and let us know at <a href="mailto:gcc@gcc.gnu.org">gcc@gcc.gnu.org</a>.
</p>
<p>Often, a machine will have multiple instructions that obtain a value
from a comparison (or the condition codes).  Here are rules to guide the
choice of value for <code>STORE_FLAG_VALUE</code>, and hence the instructions
to be used:
</p>
<ul>
<li> Use the shortest sequence that yields a valid definition for
<code>STORE_FLAG_VALUE</code>.  It is more efficient for the compiler to
&ldquo;normalize&rdquo; the value (convert it to, e.g., 1 or 0) than for the
comparison operators to do so because there may be opportunities to
combine the normalization with other operations.

</li><li> For equal-length sequences, use a value of 1 or -1, with -1 being
slightly preferred on machines with expensive jumps and 1 preferred on
other machines.

</li><li> As a second choice, choose a value of &lsquo;<samp>0x80000001</samp>&rsquo; if instructions
exist that set both the sign and low-order bits but do not define the
others.

</li><li> Otherwise, use a value of &lsquo;<samp>0x80000000</samp>&rsquo;.
</li></ul>

<p>Many machines can produce both the value chosen for
<code>STORE_FLAG_VALUE</code> and its negation in the same number of
instructions.  On those machines, you should also define a pattern for
those cases, e.g., one matching
</p>
<div class="smallexample">
<pre class="smallexample">(set <var>A</var> (neg:<var>m</var> (ne:<var>m</var> <var>B</var> <var>C</var>)))
</pre></div>

<p>Some machines can also perform <code>and</code> or <code>plus</code> operations on
condition code values with less instructions than the corresponding
&lsquo;<samp>cstore<var>mode</var>4</samp>&rsquo; insn followed by <code>and</code> or <code>plus</code>.  On those
machines, define the appropriate patterns.  Use the names <code>incscc</code>
and <code>decscc</code>, respectively, for the patterns which perform
<code>plus</code> or <code>minus</code> operations on condition code values.  See
<samp>rs6000.md</samp> for some examples.  The GNU Superoptimizer can be used to
find such instruction sequences on other machines.
</p>
<p>If this macro is not defined, the default value, 1, is used.  You need
not define <code>STORE_FLAG_VALUE</code> if the machine has no store-flag
instructions, or if the value generated by these instructions is 1.
</p></dd></dl>

<dl>
<dt><a name="index-FLOAT_005fSTORE_005fFLAG_005fVALUE"></a>Macro: <strong>FLOAT_STORE_FLAG_VALUE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression that gives a nonzero <code>REAL_VALUE_TYPE</code> value that is
returned when comparison operators with floating-point results are true.
Define this macro on machines that have comparison operations that return
floating-point values.  If there are no such operations, do not define
this macro.
</p></dd></dl>

<dl>
<dt><a name="index-VECTOR_005fSTORE_005fFLAG_005fVALUE"></a>Macro: <strong>VECTOR_STORE_FLAG_VALUE</strong> <em>(<var>mode</var>)</em></dt>
<dd><p>A C expression that gives a rtx representing the nonzero true element
for vector comparisons.  The returned rtx should be valid for the inner
mode of <var>mode</var> which is guaranteed to be a vector mode.  Define
this macro on machines that have vector comparison operations that
return a vector result.  If there are no such operations, do not define
this macro.  Typically, this macro is defined as <code>const1_rtx</code> or
<code>constm1_rtx</code>.  This macro may return <code>NULL_RTX</code> to prevent
the compiler optimizing such vector comparison operations for the
given mode.
</p></dd></dl>

<dl>
<dt><a name="index-CLZ_005fDEFINED_005fVALUE_005fAT_005fZERO"></a>Macro: <strong>CLZ_DEFINED_VALUE_AT_ZERO</strong> <em>(<var>mode</var>, <var>value</var>)</em></dt>
<dt><a name="index-CTZ_005fDEFINED_005fVALUE_005fAT_005fZERO"></a>Macro: <strong>CTZ_DEFINED_VALUE_AT_ZERO</strong> <em>(<var>mode</var>, <var>value</var>)</em></dt>
<dd><p>A C expression that indicates whether the architecture defines a value
for <code>clz</code> or <code>ctz</code> with a zero operand.
A result of <code>0</code> indicates the value is undefined.
If the value is defined for only the RTL expression, the macro should
evaluate to <code>1</code>; if the value applies also to the corresponding optab
entry (which is normally the case if it expands directly into
the corresponding RTL), then the macro should evaluate to <code>2</code>.
In the cases where the value is defined, <var>value</var> should be set to
this value.
</p>
<p>If this macro is not defined, the value of <code>clz</code> or
<code>ctz</code> at zero is assumed to be undefined.
</p>
<p>This macro must be defined if the target&rsquo;s expansion for <code>ffs</code>
relies on a particular value to get correct results.  Otherwise it
is not necessary, though it may be used to optimize some corner cases, and
to provide a default expansion for the <code>ffs</code> optab.
</p>
<p>Note that regardless of this macro the &ldquo;definedness&rdquo; of <code>clz</code>
and <code>ctz</code> at zero do <em>not</em> extend to the builtin functions
visible to the user.  Thus one may be free to adjust the value at will
to match the target expansion of these operations without fear of
breaking the API.
</p></dd></dl>

<dl>
<dt><a name="index-Pmode"></a>Macro: <strong>Pmode</strong></dt>
<dd><p>An alias for the machine mode for pointers.  On most machines, define
this to be the integer mode corresponding to the width of a hardware
pointer; <code>SImode</code> on 32-bit machine or <code>DImode</code> on 64-bit machines.
On some machines you must define this to be one of the partial integer
modes, such as <code>PSImode</code>.
</p>
<p>The width of <code>Pmode</code> must be at least as large as the value of
<code>POINTER_SIZE</code>.  If it is not equal, you must define the macro
<code>POINTERS_EXTEND_UNSIGNED</code> to specify how pointers are extended
to <code>Pmode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-FUNCTION_005fMODE"></a>Macro: <strong>FUNCTION_MODE</strong></dt>
<dd><p>An alias for the machine mode used for memory references to functions
being called, in <code>call</code> RTL expressions.  On most CISC machines,
where an instruction can begin at any byte address, this should be
<code>QImode</code>.  On most RISC machines, where all instructions have fixed
size and alignment, this should be a mode with the same size and alignment
as the machine instruction words - typically <code>SImode</code> or <code>HImode</code>.
</p></dd></dl>

<dl>
<dt><a name="index-STDC_005f0_005fIN_005fSYSTEM_005fHEADERS"></a>Macro: <strong>STDC_0_IN_SYSTEM_HEADERS</strong></dt>
<dd><p>In normal operation, the preprocessor expands <code>__STDC__</code> to the
constant 1, to signify that GCC conforms to ISO Standard C.  On some
hosts, like Solaris, the system compiler uses a different convention,
where <code>__STDC__</code> is normally 0, but is 1 if the user specifies
strict conformance to the C Standard.
</p>
<p>Defining <code>STDC_0_IN_SYSTEM_HEADERS</code> makes GNU CPP follows the host
convention when processing system header files, but when processing user
files <code>__STDC__</code> will always expand to 1.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fC_005fPREINCLUDE"></a>C Target Hook: <em>const char *</em> <strong>TARGET_C_PREINCLUDE</strong> <em>(void)</em></dt>
<dd><p>Define this hook to return the name of a header file to be included at the start of all compilations, as if it had been included with <code>#include &lt;<var>file</var>&gt;</code>.  If this hook returns <code>NULL</code>, or is not defined, or the header is not found, or if the user specifies <samp>-ffreestanding</samp> or <samp>-nostdinc</samp>, no header is included.
</p>
<p>This hook can be used together with a header provided by the system C library to implement ISO C requirements for certain macros to be predefined that describe properties of the whole implementation rather than just the compiler.
</p></dd></dl>

<dl>
<dt><a name="index-NO_005fIMPLICIT_005fEXTERN_005fC"></a>Macro: <strong>NO_IMPLICIT_EXTERN_C</strong></dt>
<dd><p>Define this macro if the system header files support C++ as well as C.
This macro inhibits the usual method of using system header files in
C++, which is to pretend that the file&rsquo;s contents are enclosed in
&lsquo;<samp>extern &quot;C&quot; {&hellip;}</samp>&rsquo;.
</p></dd></dl>

<a name="index-_0023pragma"></a>
<a name="index-pragma"></a>
<dl>
<dt><a name="index-REGISTER_005fTARGET_005fPRAGMAS"></a>Macro: <strong>REGISTER_TARGET_PRAGMAS</strong> <em>()</em></dt>
<dd><p>Define this macro if you want to implement any target-specific pragmas.
If defined, it is a C expression which makes a series of calls to
<code>c_register_pragma</code> or <code>c_register_pragma_with_expansion</code>
for each pragma.  The macro may also do any
setup required for the pragmas.
</p>
<p>The primary reason to define this macro is to provide compatibility with
other compilers for the same target.  In general, we discourage
definition of target-specific pragmas for GCC.
</p>
<p>If the pragma can be implemented by attributes then you should consider
defining the target hook &lsquo;<samp>TARGET_INSERT_ATTRIBUTES</samp>&rsquo; as well.
</p>
<p>Preprocessor macros that appear on pragma lines are not expanded.  All
&lsquo;<samp>#pragma</samp>&rsquo; directives that do not match any registered pragma are
silently ignored, unless the user specifies <samp>-Wunknown-pragmas</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-c_005fregister_005fpragma"></a>Function: <em>void</em> <strong>c_register_pragma</strong> <em>(const char *<var>space</var>, const char *<var>name</var>, void (*<var>callback</var>) (struct cpp_reader *))</em></dt>
<dt><a name="index-c_005fregister_005fpragma_005fwith_005fexpansion"></a>Function: <em>void</em> <strong>c_register_pragma_with_expansion</strong> <em>(const char *<var>space</var>, const char *<var>name</var>, void (*<var>callback</var>) (struct cpp_reader *))</em></dt>
<dd>
<p>Each call to <code>c_register_pragma</code> or
<code>c_register_pragma_with_expansion</code> establishes one pragma.  The
<var>callback</var> routine will be called when the preprocessor encounters a
pragma of the form
</p>
<div class="smallexample">
<pre class="smallexample">#pragma [<var>space</var>] <var>name</var> &hellip;
</pre></div>

<p><var>space</var> is the case-sensitive namespace of the pragma, or
<code>NULL</code> to put the pragma in the global namespace.  The callback
routine receives <var>pfile</var> as its first argument, which can be passed
on to cpplib&rsquo;s functions if necessary.  You can lex tokens after the
<var>name</var> by calling <code>pragma_lex</code>.  Tokens that are not read by the
callback will be silently ignored.  The end of the line is indicated by
a token of type <code>CPP_EOF</code>.  Macro expansion occurs on the
arguments of pragmas registered with
<code>c_register_pragma_with_expansion</code> but not on the arguments of
pragmas registered with <code>c_register_pragma</code>.
</p>
<p>Note that the use of <code>pragma_lex</code> is specific to the C and C++
compilers.  It will not work in the Java or Fortran compilers, or any
other language compilers for that matter.  Thus if <code>pragma_lex</code> is going
to be called from target-specific code, it must only be done so when
building the C and C++ compilers.  This can be done by defining the
variables <code>c_target_objs</code> and <code>cxx_target_objs</code> in the
target entry in the <samp>config.gcc</samp> file.  These variables should name
the target-specific, language-specific object file which contains the
code that uses <code>pragma_lex</code>.  Note it will also be necessary to add a
rule to the makefile fragment pointed to by <code>tmake_file</code> that shows
how to build this object file.
</p></dd></dl>

<dl>
<dt><a name="index-HANDLE_005fPRAGMA_005fPACK_005fWITH_005fEXPANSION"></a>Macro: <strong>HANDLE_PRAGMA_PACK_WITH_EXPANSION</strong></dt>
<dd><p>Define this macro if macros should be expanded in the
arguments of &lsquo;<samp>#pragma pack</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fDEFAULT_005fPACK_005fSTRUCT"></a>Macro: <strong>TARGET_DEFAULT_PACK_STRUCT</strong></dt>
<dd><p>If your target requires a structure packing default other than 0 (meaning
the machine default), define this macro to the necessary value (in bytes).
This must be a value that would also be valid to use with
&lsquo;<samp>#pragma pack()</samp>&rsquo; (that is, a small power of two).
</p></dd></dl>

<dl>
<dt><a name="index-DOLLARS_005fIN_005fIDENTIFIERS"></a>Macro: <strong>DOLLARS_IN_IDENTIFIERS</strong></dt>
<dd><p>Define this macro to control use of the character &lsquo;<samp>$</samp>&rsquo; in
identifier names for the C family of languages.  0 means &lsquo;<samp>$</samp>&rsquo; is
not allowed by default; 1 means it is allowed.  1 is the default;
there is no need to define this macro in that case.
</p></dd></dl>

<dl>
<dt><a name="index-INSN_005fSETS_005fARE_005fDELAYED"></a>Macro: <strong>INSN_SETS_ARE_DELAYED</strong> <em>(<var>insn</var>)</em></dt>
<dd><p>Define this macro as a C expression that is nonzero if it is safe for the
delay slot scheduler to place instructions in the delay slot of <var>insn</var>,
even if they appear to use a resource set or clobbered in <var>insn</var>.
<var>insn</var> is always a <code>jump_insn</code> or an <code>insn</code>; GCC knows that
every <code>call_insn</code> has this behavior.  On machines where some <code>insn</code>
or <code>jump_insn</code> is really a function call and hence has this behavior,
you should define this macro.
</p>
<p>You need not define this macro if it would always return zero.
</p></dd></dl>

<dl>
<dt><a name="index-INSN_005fREFERENCES_005fARE_005fDELAYED"></a>Macro: <strong>INSN_REFERENCES_ARE_DELAYED</strong> <em>(<var>insn</var>)</em></dt>
<dd><p>Define this macro as a C expression that is nonzero if it is safe for the
delay slot scheduler to place instructions in the delay slot of <var>insn</var>,
even if they appear to set or clobber a resource referenced in <var>insn</var>.
<var>insn</var> is always a <code>jump_insn</code> or an <code>insn</code>.  On machines where
some <code>insn</code> or <code>jump_insn</code> is really a function call and its operands
are registers whose use is actually in the subroutine it calls, you should
define this macro.  Doing so allows the delay slot scheduler to move
instructions which copy arguments into the argument registers into the delay
slot of <var>insn</var>.
</p>
<p>You need not define this macro if it would always return zero.
</p></dd></dl>

<dl>
<dt><a name="index-MULTIPLE_005fSYMBOL_005fSPACES"></a>Macro: <strong>MULTIPLE_SYMBOL_SPACES</strong></dt>
<dd><p>Define this macro as a C expression that is nonzero if, in some cases,
global symbols from one translation unit may not be bound to undefined
symbols in another translation unit without user intervention.  For
instance, under Microsoft Windows symbols must be explicitly imported
from shared libraries (DLLs).
</p>
<p>You need not define this macro if it would always evaluate to zero.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMD_005fASM_005fCLOBBERS"></a>Target Hook: <em>tree</em> <strong>TARGET_MD_ASM_CLOBBERS</strong> <em>(tree <var>outputs</var>, tree <var>inputs</var>, tree <var>clobbers</var>)</em></dt>
<dd><p>This target hook should add to <var>clobbers</var> <code>STRING_CST</code> trees for
any hard regs the port wishes to automatically clobber for an asm.
It should return the result of the last <code>tree_cons</code> used to add a
clobber.  The <var>outputs</var>, <var>inputs</var> and <var>clobber</var> lists are the
corresponding parameters to the asm and may be inspected to avoid
clobbering a register that is an input or output of the asm.  You can use
<code>tree_overlaps_hard_reg_set</code>, declared in <samp>tree.h</samp>, to test
for overlap with regards to asm-declared registers.
</p></dd></dl>

<dl>
<dt><a name="index-MATH_005fLIBRARY"></a>Macro: <strong>MATH_LIBRARY</strong></dt>
<dd><p>Define this macro as a C string constant for the linker argument to link
in the system math library, minus the initial &lsquo;<samp>&quot;-l&quot;</samp>&rsquo;, or
&lsquo;<samp>&quot;&quot;</samp>&rsquo; if the target does not have a
separate math library.
</p>
<p>You need only define this macro if the default of &lsquo;<samp>&quot;m&quot;</samp>&rsquo; is wrong.
</p></dd></dl>

<dl>
<dt><a name="index-LIBRARY_005fPATH_005fENV"></a>Macro: <strong>LIBRARY_PATH_ENV</strong></dt>
<dd><p>Define this macro as a C string constant for the environment variable that
specifies where the linker should look for libraries.
</p>
<p>You need only define this macro if the default of &lsquo;<samp>&quot;LIBRARY_PATH&quot;</samp>&rsquo;
is wrong.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPOSIX_005fIO"></a>Macro: <strong>TARGET_POSIX_IO</strong></dt>
<dd><p>Define this macro if the target supports the following POSIX&nbsp;file
functions, access, mkdir and  file locking with fcntl / F_SETLKW.
Defining <code>TARGET_POSIX_IO</code> will enable the test coverage code
to use file locking when exiting a program, which avoids race conditions
if the program has forked. It will also create directories at run-time
for cross-profiling.
</p></dd></dl>

<dl>
<dt><a name="index-MAX_005fCONDITIONAL_005fEXECUTE"></a>Macro: <strong>MAX_CONDITIONAL_EXECUTE</strong></dt>
<dd>
<p>A C expression for the maximum number of instructions to execute via
conditional execution instructions instead of a branch.  A value of
<code>BRANCH_COST</code>+1 is the default if the machine does not use cc0, and
1 if it does use cc0.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMODIFY_005fTESTS"></a>Macro: <strong>IFCVT_MODIFY_TESTS</strong> <em>(<var>ce_info</var>, <var>true_expr</var>, <var>false_expr</var>)</em></dt>
<dd><p>Used if the target needs to perform machine-dependent modifications on the
conditionals used for turning basic blocks into conditionally executed code.
<var>ce_info</var> points to a data structure, <code>struct ce_if_block</code>, which
contains information about the currently processed blocks.  <var>true_expr</var>
and <var>false_expr</var> are the tests that are used for converting the
then-block and the else-block, respectively.  Set either <var>true_expr</var> or
<var>false_expr</var> to a null pointer if the tests cannot be converted.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMODIFY_005fMULTIPLE_005fTESTS"></a>Macro: <strong>IFCVT_MODIFY_MULTIPLE_TESTS</strong> <em>(<var>ce_info</var>, <var>bb</var>, <var>true_expr</var>, <var>false_expr</var>)</em></dt>
<dd><p>Like <code>IFCVT_MODIFY_TESTS</code>, but used when converting more complicated
if-statements into conditions combined by <code>and</code> and <code>or</code> operations.
<var>bb</var> contains the basic block that contains the test that is currently
being processed and about to be turned into a condition.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMODIFY_005fINSN"></a>Macro: <strong>IFCVT_MODIFY_INSN</strong> <em>(<var>ce_info</var>, <var>pattern</var>, <var>insn</var>)</em></dt>
<dd><p>A C expression to modify the <var>PATTERN</var> of an <var>INSN</var> that is to
be converted to conditional execution format.  <var>ce_info</var> points to
a data structure, <code>struct ce_if_block</code>, which contains information
about the currently processed blocks.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMODIFY_005fFINAL"></a>Macro: <strong>IFCVT_MODIFY_FINAL</strong> <em>(<var>ce_info</var>)</em></dt>
<dd><p>A C expression to perform any final machine dependent modifications in
converting code to conditional execution.  The involved basic blocks
can be found in the <code>struct ce_if_block</code> structure that is pointed
to by <var>ce_info</var>.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMODIFY_005fCANCEL"></a>Macro: <strong>IFCVT_MODIFY_CANCEL</strong> <em>(<var>ce_info</var>)</em></dt>
<dd><p>A C expression to cancel any machine dependent modifications in
converting code to conditional execution.  The involved basic blocks
can be found in the <code>struct ce_if_block</code> structure that is pointed
to by <var>ce_info</var>.
</p></dd></dl>

<dl>
<dt><a name="index-IFCVT_005fMACHDEP_005fINIT"></a>Macro: <strong>IFCVT_MACHDEP_INIT</strong> <em>(<var>ce_info</var>)</em></dt>
<dd><p>A C expression to initialize any machine specific data for if-conversion
of the if-block in the <code>struct ce_if_block</code> structure that is pointed
to by <var>ce_info</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMACHINE_005fDEPENDENT_005fREORG"></a>Target Hook: <em>void</em> <strong>TARGET_MACHINE_DEPENDENT_REORG</strong> <em>(void)</em></dt>
<dd><p>If non-null, this hook performs a target-specific pass over the
instruction stream.  The compiler will run it at all optimization levels,
just before the point at which it normally does delayed-branch scheduling.
</p>
<p>The exact purpose of the hook varies from target to target.  Some use
it to do transformations that are necessary for correctness, such as
laying out in-function constant pools or avoiding hardware hazards.
Others use it as an opportunity to do some machine-dependent optimizations.
</p>
<p>You need not implement the hook if it has nothing to do.  The default
definition is null.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINIT_005fBUILTINS"></a>Target Hook: <em>void</em> <strong>TARGET_INIT_BUILTINS</strong> <em>(void)</em></dt>
<dd><p>Define this hook if you have any machine-specific built-in functions
that need to be defined.  It should be a function that performs the
necessary setup.
</p>
<p>Machine specific built-in functions can be useful to expand special machine
instructions that would otherwise not normally be generated because
they have no equivalent in the source language (for example, SIMD vector
instructions or prefetch instructions).
</p>
<p>To create a built-in function, call the function
<code>lang_hooks.builtin_function</code>
which is defined by the language front end.  You can use any type nodes set
up by <code>build_common_tree_nodes</code>;
only language front ends that use those two functions will call
&lsquo;<samp>TARGET_INIT_BUILTINS</samp>&rsquo;.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBUILTIN_005fDECL"></a>Target Hook: <em>tree</em> <strong>TARGET_BUILTIN_DECL</strong> <em>(unsigned <var>code</var>, bool <var>initialize_p</var>)</em></dt>
<dd><p>Define this hook if you have any machine-specific built-in functions
that need to be defined.  It should be a function that returns the
builtin function declaration for the builtin function code <var>code</var>.
If there is no such builtin and it cannot be initialized at this time
if <var>initialize_p</var> is true the function should return <code>NULL_TREE</code>.
If <var>code</var> is out of range the function should return
<code>error_mark_node</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXPAND_005fBUILTIN"></a>Target Hook: <em>rtx</em> <strong>TARGET_EXPAND_BUILTIN</strong> <em>(tree <var>exp</var>, rtx <var>target</var>, rtx <var>subtarget</var>, enum machine_mode <var>mode</var>, int <var>ignore</var>)</em></dt>
<dd>
<p>Expand a call to a machine specific built-in function that was set up by
&lsquo;<samp>TARGET_INIT_BUILTINS</samp>&rsquo;.  <var>exp</var> is the expression for the
function call; the result should go to <var>target</var> if that is
convenient, and have mode <var>mode</var> if that is convenient.
<var>subtarget</var> may be used as the target for computing one of
<var>exp</var>&rsquo;s operands.  <var>ignore</var> is nonzero if the value is to be
ignored.  This function should return the result of the call to the
built-in function.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fRESOLVE_005fOVERLOADED_005fBUILTIN"></a>Target Hook: <em>tree</em> <strong>TARGET_RESOLVE_OVERLOADED_BUILTIN</strong> <em>(unsigned int <var>loc</var>, tree <var>fndecl</var>, void *<var>arglist</var>)</em></dt>
<dd><p>Select a replacement for a machine specific built-in function that
was set up by &lsquo;<samp>TARGET_INIT_BUILTINS</samp>&rsquo;.  This is done
<em>before</em> regular type checking, and so allows the target to
implement a crude form of function overloading.  <var>fndecl</var> is the
declaration of the built-in function.  <var>arglist</var> is the list of
arguments passed to the built-in function.  The result is a
complete expression that implements the operation, usually
another <code>CALL_EXPR</code>.
<var>arglist</var> really has type &lsquo;<samp>VEC(tree,gc)*</samp>&rsquo;
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFOLD_005fBUILTIN"></a>Target Hook: <em>tree</em> <strong>TARGET_FOLD_BUILTIN</strong> <em>(tree <var>fndecl</var>, int <var>n_args</var>, tree *<var>argp</var>, bool <var>ignore</var>)</em></dt>
<dd><p>Fold a call to a machine specific built-in function that was set up by
&lsquo;<samp>TARGET_INIT_BUILTINS</samp>&rsquo;.  <var>fndecl</var> is the declaration of the
built-in function.  <var>n_args</var> is the number of arguments passed to
the function; the arguments themselves are pointed to by <var>argp</var>.
The result is another tree containing a simplified expression for the
call&rsquo;s result.  If <var>ignore</var> is true the value will be ignored.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCOMPARE_005fVERSION_005fPRIORITY"></a>Target Hook: <em>int</em> <strong>TARGET_COMPARE_VERSION_PRIORITY</strong> <em>(tree <var>decl1</var>, tree <var>decl2</var>)</em></dt>
<dd><p>This hook is used to compare the target attributes in two functions to
determine which function&rsquo;s features get higher priority.  This is used
during function multi-versioning to figure out the order in which two
versions must be dispatched.  A function version with a higher priority
is checked for dispatching earlier.  <var>decl1</var> and <var>decl2</var> are
 the two function decls that will be compared.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGET_005fFUNCTION_005fVERSIONS_005fDISPATCHER"></a>Target Hook: <em>tree</em> <strong>TARGET_GET_FUNCTION_VERSIONS_DISPATCHER</strong> <em>(void *<var>decl</var>)</em></dt>
<dd><p>This hook is used to get the dispatcher function for a set of function
versions.  The dispatcher function is called to invoke the right function
version at run-time. <var>decl</var> is one version from a set of semantically
identical versions.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGENERATE_005fVERSION_005fDISPATCHER_005fBODY"></a>Target Hook: <em>tree</em> <strong>TARGET_GENERATE_VERSION_DISPATCHER_BODY</strong> <em>(void *<var>arg</var>)</em></dt>
<dd><p>This hook is used to generate the dispatcher logic to invoke the right
function version at run-time for a given set of function versions.
<var>arg</var> points to the callgraph node of the dispatcher function whose
body must be generated.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fWITHIN_005fDOLOOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_WITHIN_DOLOOP</strong> <em>(const_rtx <var>insn</var>)</em></dt>
<dd>
<p>Take an instruction in <var>insn</var> and return NULL if it is valid within a
low-overhead loop, otherwise return a string explaining why doloop
could not be applied.
</p>
<p>Many targets use special registers for low-overhead looping. For any
instruction that clobbers these this function should return a string indicating
the reason why the doloop could not be applied.
By default, the RTL loop optimizer does not use a present doloop pattern for
loops containing function calls or branch on table instructions.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLEGITIMATE_005fCOMBINED_005fINSN"></a>Target Hook: <em>bool</em> <strong>TARGET_LEGITIMATE_COMBINED_INSN</strong> <em>(rtx <var>insn</var>)</em></dt>
<dd><p>Take an instruction in <var>insn</var> and return <code>false</code> if the instruction is not appropriate as a combination of two or more instructions.  The default is to accept all instructions.
</p></dd></dl>

<dl>
<dt><a name="index-MD_005fCAN_005fREDIRECT_005fBRANCH"></a>Macro: <strong>MD_CAN_REDIRECT_BRANCH</strong> <em>(<var>branch1</var>, <var>branch2</var>)</em></dt>
<dd>
<p>Take a branch insn in <var>branch1</var> and another in <var>branch2</var>.
Return true if redirecting <var>branch1</var> to the destination of
<var>branch2</var> is possible.
</p>
<p>On some targets, branches may have a limited range.  Optimizing the
filling of delay slots can result in branches being redirected, and this
may in turn cause a branch offset to overflow.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCAN_005fFOLLOW_005fJUMP"></a>Target Hook: <em>bool</em> <strong>TARGET_CAN_FOLLOW_JUMP</strong> <em>(const_rtx <var>follower</var>, const_rtx <var>followee</var>)</em></dt>
<dd><p>FOLLOWER and FOLLOWEE are JUMP_INSN instructions;  return true if FOLLOWER may be modified to follow FOLLOWEE;  false, if it can&rsquo;t.  For example, on some targets, certain kinds of branches can&rsquo;t be made to  follow through a hot/cold partitioning.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCOMMUTATIVE_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_COMMUTATIVE_P</strong> <em>(const_rtx <var>x</var>, int <var>outer_code</var>)</em></dt>
<dd><p>This target hook returns <code>true</code> if <var>x</var> is considered to be commutative.
Usually, this is just COMMUTATIVE_P (<var>x</var>), but the HP PA doesn&rsquo;t consider
PLUS to be commutative inside a MEM.  <var>outer_code</var> is the rtx code
of the enclosing rtl, if known, otherwise it is UNKNOWN.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fALLOCATE_005fINITIAL_005fVALUE"></a>Target Hook: <em>rtx</em> <strong>TARGET_ALLOCATE_INITIAL_VALUE</strong> <em>(rtx <var>hard_reg</var>)</em></dt>
<dd>
<p>When the initial value of a hard register has been copied in a pseudo
register, it is often not necessary to actually allocate another register
to this pseudo register, because the original hard register or a stack slot
it has been saved into can be used.  <code>TARGET_ALLOCATE_INITIAL_VALUE</code>
is called at the start of register allocation once for each hard register
that had its initial value copied by using
<code>get_func_hard_reg_initial_val</code> or <code>get_hard_reg_initial_val</code>.
Possible values are <code>NULL_RTX</code>, if you don&rsquo;t want
to do any special allocation, a <code>REG</code> rtx&mdash;that would typically be
the hard register itself, if it is known not to be clobbered&mdash;or a
<code>MEM</code>.
If you are returning a <code>MEM</code>, this is only a hint for the allocator;
it might decide to use another register anyways.
You may use <code>current_function_is_leaf</code> or 
<code>REG_N_SETS</code> in the hook to determine if the hard
register in question will not be clobbered.
The default value of this hook is <code>NULL</code>, which disables any special
allocation.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUNSPEC_005fMAY_005fTRAP_005fP"></a>Target Hook: <em>int</em> <strong>TARGET_UNSPEC_MAY_TRAP_P</strong> <em>(const_rtx <var>x</var>, unsigned <var>flags</var>)</em></dt>
<dd><p>This target hook returns nonzero if <var>x</var>, an <code>unspec</code> or
<code>unspec_volatile</code> operation, might cause a trap.  Targets can use
this hook to enhance precision of analysis for <code>unspec</code> and
<code>unspec_volatile</code> operations.  You may call <code>may_trap_p_1</code>
to analyze inner elements of <var>x</var> in which case <var>flags</var> should be
passed along.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fSET_005fCURRENT_005fFUNCTION"></a>Target Hook: <em>void</em> <strong>TARGET_SET_CURRENT_FUNCTION</strong> <em>(tree <var>decl</var>)</em></dt>
<dd><p>The compiler invokes this hook whenever it changes its current function
context (<code>cfun</code>).  You can define this function if
the back end needs to perform any initialization or reset actions on a
per-function basis.  For example, it may be used to implement function
attributes that affect register usage or code generation patterns.
The argument <var>decl</var> is the declaration for the new function context,
and may be null to indicate that the compiler has left a function context
and is returning to processing at the top level.
The default hook function does nothing.
</p>
<p>GCC sets <code>cfun</code> to a dummy function context during initialization of
some parts of the back end.  The hook function is not invoked in this
situation; you need not worry about the hook being invoked recursively,
or when the back end is in a partially-initialized state.
<code>cfun</code> might be <code>NULL</code> to indicate processing at top level,
outside of any function scope.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOBJECT_005fSUFFIX"></a>Macro: <strong>TARGET_OBJECT_SUFFIX</strong></dt>
<dd><p>Define this macro to be a C string representing the suffix for object
files on your target machine.  If you do not define this macro, GCC will
use &lsquo;<samp>.o</samp>&rsquo; as the suffix for object files.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXECUTABLE_005fSUFFIX"></a>Macro: <strong>TARGET_EXECUTABLE_SUFFIX</strong></dt>
<dd><p>Define this macro to be a C string representing the suffix to be
automatically added to executable files on your target machine.  If you
do not define this macro, GCC will use the null string as the suffix for
executable files.
</p></dd></dl>

<dl>
<dt><a name="index-COLLECT_005fEXPORT_005fLIST"></a>Macro: <strong>COLLECT_EXPORT_LIST</strong></dt>
<dd><p>If defined, <code>collect2</code> will scan the individual object files
specified on its command line and create an export list for the linker.
Define this macro for systems like AIX, where the linker discards
object files that are not referenced from <code>main</code> and uses export
lists.
</p></dd></dl>

<dl>
<dt><a name="index-MODIFY_005fJNI_005fMETHOD_005fCALL"></a>Macro: <strong>MODIFY_JNI_METHOD_CALL</strong> <em>(<var>mdecl</var>)</em></dt>
<dd><p>Define this macro to a C expression representing a variant of the
method call <var>mdecl</var>, if Java Native Interface (JNI) methods
must be invoked differently from other methods on your target.
For example, on 32-bit Microsoft Windows, JNI methods must be invoked using
the <code>stdcall</code> calling convention and this macro is then
defined as this expression:
</p>
<div class="smallexample">
<pre class="smallexample">build_type_attribute_variant (<var>mdecl</var>,
                              build_tree_list
                              (get_identifier (&quot;stdcall&quot;),
                               NULL))
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-TARGET_005fCANNOT_005fMODIFY_005fJUMPS_005fP"></a>Target Hook: <em>bool</em> <strong>TARGET_CANNOT_MODIFY_JUMPS_P</strong> <em>(void)</em></dt>
<dd><p>This target hook returns <code>true</code> past the point in which new jump
instructions could be created.  On machines that require a register for
every jump such as the SHmedia ISA of SH5, this point would typically be
reload, so this target hook should be defined to a function such as:
</p>
<div class="smallexample">
<pre class="smallexample">static bool
cannot_modify_jumps_past_reload_p ()
{
  return (reload_completed || reload_in_progress);
}
</pre></div>
</dd></dl>

<dl>
<dt><a name="index-TARGET_005fBRANCH_005fTARGET_005fREGISTER_005fCLASS"></a>Target Hook: <em>reg_class_t</em> <strong>TARGET_BRANCH_TARGET_REGISTER_CLASS</strong> <em>(void)</em></dt>
<dd><p>This target hook returns a register class for which branch target register
optimizations should be applied.  All registers in this class should be
usable interchangeably.  After reload, registers in this class will be
re-allocated and loads will be hoisted out of loops and be subjected
to inter-block scheduling.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fBRANCH_005fTARGET_005fREGISTER_005fCALLEE_005fSAVED"></a>Target Hook: <em>bool</em> <strong>TARGET_BRANCH_TARGET_REGISTER_CALLEE_SAVED</strong> <em>(bool <var>after_prologue_epilogue_gen</var>)</em></dt>
<dd><p>Branch target register optimization will by default exclude callee-saved
registers
that are not already live during the current function; if this target hook
returns true, they will be included.  The target code must than make sure
that all target registers in the class returned by
&lsquo;<samp>TARGET_BRANCH_TARGET_REGISTER_CLASS</samp>&rsquo; that might need saving are
saved.  <var>after_prologue_epilogue_gen</var> indicates if prologues and
epilogues have already been generated.  Note, even if you only return
true when <var>after_prologue_epilogue_gen</var> is false, you still are likely
to have to make special provisions in <code>INITIAL_ELIMINATION_OFFSET</code>
to reserve space for caller-saved target registers.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fHAVE_005fCONDITIONAL_005fEXECUTION"></a>Target Hook: <em>bool</em> <strong>TARGET_HAVE_CONDITIONAL_EXECUTION</strong> <em>(void)</em></dt>
<dd><p>This target hook returns true if the target supports conditional execution.
This target hook is required only when the target has several different
modes and they have different conditional execution capability, such as ARM.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fLOOP_005fUNROLL_005fADJUST"></a>Target Hook: <em>unsigned</em> <strong>TARGET_LOOP_UNROLL_ADJUST</strong> <em>(unsigned <var>nunroll</var>, struct loop *<var>loop</var>)</em></dt>
<dd><p>This target hook returns a new value for the number of times <var>loop</var>
should be unrolled. The parameter <var>nunroll</var> is the number of times
the loop is to be unrolled. The parameter <var>loop</var> is a pointer to
the loop, which is going to be checked for unrolling. This target hook
is required only when the target has special constraints like maximum
number of memory accesses.
</p></dd></dl>

<dl>
<dt><a name="index-POWI_005fMAX_005fMULTS"></a>Macro: <strong>POWI_MAX_MULTS</strong></dt>
<dd><p>If defined, this macro is interpreted as a signed integer C expression
that specifies the maximum number of floating point multiplications
that should be emitted when expanding exponentiation by an integer
constant inline.  When this value is defined, exponentiation requiring
more than this number of multiplications is implemented by calling the
system library&rsquo;s <code>pow</code>, <code>powf</code> or <code>powl</code> routines.
The default value places no upper bound on the multiplication count.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXTRA_005fINCLUDES"></a>Macro: <em>void</em> <strong>TARGET_EXTRA_INCLUDES</strong> <em>(const char *<var>sysroot</var>, const char *<var>iprefix</var>, int <var>stdinc</var>)</em></dt>
<dd><p>This target hook should register any extra include files for the
target.  The parameter <var>stdinc</var> indicates if normal include files
are present.  The parameter <var>sysroot</var> is the system root directory.
The parameter <var>iprefix</var> is the prefix for the gcc directory.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fEXTRA_005fPRE_005fINCLUDES"></a>Macro: <em>void</em> <strong>TARGET_EXTRA_PRE_INCLUDES</strong> <em>(const char *<var>sysroot</var>, const char *<var>iprefix</var>, int <var>stdinc</var>)</em></dt>
<dd><p>This target hook should register any extra include files for the
target before any standard headers.  The parameter <var>stdinc</var>
indicates if normal include files are present.  The parameter
<var>sysroot</var> is the system root directory.  The parameter
<var>iprefix</var> is the prefix for the gcc directory.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOPTF"></a>Macro: <em>void</em> <strong>TARGET_OPTF</strong> <em>(char *<var>path</var>)</em></dt>
<dd><p>This target hook should register special include paths for the target.
The parameter <var>path</var> is the include to register.  On Darwin
systems, this is used for Framework includes, which have semantics
that are different from <samp>-I</samp>.
</p></dd></dl>

<dl>
<dt><a name="index-bool"></a>Macro: <strong>bool</strong> <em>TARGET_USE_LOCAL_THUNK_ALIAS_P (tree <var>fndecl</var>)</em></dt>
<dd><p>This target macro returns <code>true</code> if it is safe to use a local alias
for a virtual function <var>fndecl</var> when constructing thunks,
<code>false</code> otherwise.  By default, the macro returns <code>true</code> for all
functions, if a target supports aliases (i.e. defines
<code>ASM_OUTPUT_DEF</code>), <code>false</code> otherwise,
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fFORMAT_005fTYPES"></a>Macro: <strong>TARGET_FORMAT_TYPES</strong></dt>
<dd><p>If defined, this macro is the name of a global variable containing
target-specific format checking information for the <samp>-Wformat</samp>
option.  The default is to have no target-specific format checks.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fN_005fFORMAT_005fTYPES"></a>Macro: <strong>TARGET_N_FORMAT_TYPES</strong></dt>
<dd><p>If defined, this macro is the number of entries in
<code>TARGET_FORMAT_TYPES</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOVERRIDES_005fFORMAT_005fATTRIBUTES"></a>Macro: <strong>TARGET_OVERRIDES_FORMAT_ATTRIBUTES</strong></dt>
<dd><p>If defined, this macro is the name of a global variable containing
target-specific format overrides for the <samp>-Wformat</samp> option. The
default is to have no target-specific format overrides. If defined,
<code>TARGET_FORMAT_TYPES</code> must be defined, too.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOVERRIDES_005fFORMAT_005fATTRIBUTES_005fCOUNT"></a>Macro: <strong>TARGET_OVERRIDES_FORMAT_ATTRIBUTES_COUNT</strong></dt>
<dd><p>If defined, this macro specifies the number of entries in
<code>TARGET_OVERRIDES_FORMAT_ATTRIBUTES</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fOVERRIDES_005fFORMAT_005fINIT"></a>Macro: <strong>TARGET_OVERRIDES_FORMAT_INIT</strong></dt>
<dd><p>If defined, this macro specifies the optional initialization
routine for target specific customizations of the system printf
and scanf formatter settings.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fRELAXED_005fORDERING"></a>Target Hook: <em>bool</em> <strong>TARGET_RELAXED_ORDERING</strong></dt>
<dd><p>If set to <code>true</code>, means that the target&rsquo;s memory model does not
guarantee that loads which do not depend on one another will access
main memory in the order of the instruction stream; if ordering is
important, an explicit memory barrier must be used.  This is true of
many recent processors which implement a policy of &ldquo;relaxed,&rdquo;
&ldquo;weak,&rdquo; or &ldquo;release&rdquo; memory consistency, such as Alpha, PowerPC,
and ia64.  The default is <code>false</code>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fARG_005fFOR_005fUNPROTOTYPED_005fFN"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_ARG_FOR_UNPROTOTYPED_FN</strong> <em>(const_tree <var>typelist</var>, const_tree <var>funcdecl</var>, const_tree <var>val</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
illegal to pass argument <var>val</var> to function <var>funcdecl</var>
with prototype <var>typelist</var>.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fCONVERSION"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_CONVERSION</strong> <em>(const_tree <var>fromtype</var>, const_tree <var>totype</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
invalid to convert from <var>fromtype</var> to <var>totype</var>, or <code>NULL</code>
if validity should be determined by the front end.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fUNARY_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_UNARY_OP</strong> <em>(int <var>op</var>, const_tree <var>type</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
invalid to apply operation <var>op</var> (where unary plus is denoted by
<code>CONVERT_EXPR</code>) to an operand of type <var>type</var>, or <code>NULL</code>
if validity should be determined by the front end.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fBINARY_005fOP"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_BINARY_OP</strong> <em>(int <var>op</var>, const_tree <var>type1</var>, const_tree <var>type2</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
invalid to apply operation <var>op</var> to operands of types <var>type1</var>
and <var>type2</var>, or <code>NULL</code> if validity should be determined by
the front end.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fPARAMETER_005fTYPE"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_PARAMETER_TYPE</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
invalid for functions to include parameters of type <var>type</var>,
or <code>NULL</code> if validity should be determined by
the front end.  This is currently used only by the C and C++ front ends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fINVALID_005fRETURN_005fTYPE"></a>Target Hook: <em>const char *</em> <strong>TARGET_INVALID_RETURN_TYPE</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>If defined, this macro returns the diagnostic message when it is
invalid for functions to have return type <var>type</var>,
or <code>NULL</code> if validity should be determined by
the front end.  This is currently used only by the C and C++ front ends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fPROMOTED_005fTYPE"></a>Target Hook: <em>tree</em> <strong>TARGET_PROMOTED_TYPE</strong> <em>(const_tree <var>type</var>)</em></dt>
<dd><p>If defined, this target hook returns the type to which values of
<var>type</var> should be promoted when they appear in expressions,
analogous to the integer promotions, or <code>NULL_TREE</code> to use the
front end&rsquo;s normal promotion rules.  This hook is useful when there are
target-specific types with special promotion rules.
This is currently used only by the C and C++ front ends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCONVERT_005fTO_005fTYPE"></a>Target Hook: <em>tree</em> <strong>TARGET_CONVERT_TO_TYPE</strong> <em>(tree <var>type</var>, tree <var>expr</var>)</em></dt>
<dd><p>If defined, this hook returns the result of converting <var>expr</var> to
<var>type</var>.  It should return the converted expression,
or <code>NULL_TREE</code> to apply the front end&rsquo;s normal conversion rules.
This hook is useful when there are target-specific types with special
conversion rules.
This is currently used only by the C and C++ front ends.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUSE_005fJCR_005fSECTION"></a>Macro: <strong>TARGET_USE_JCR_SECTION</strong></dt>
<dd><p>This macro determines whether to use the JCR section to register Java
classes. By default, TARGET_USE_JCR_SECTION is defined to 1 if both
SUPPORTS_WEAK and TARGET_HAVE_NAMED_SECTIONS are true, else 0.
</p></dd></dl>

<dl>
<dt><a name="index-OBJC_005fJBLEN"></a>Macro: <strong>OBJC_JBLEN</strong></dt>
<dd><p>This macro determines the size of the objective C jump buffer for the
NeXT runtime. By default, OBJC_JBLEN is defined to an innocuous value.
</p></dd></dl>

<dl>
<dt><a name="index-LIBGCC2_005fUNWIND_005fATTRIBUTE"></a>Macro: <strong>LIBGCC2_UNWIND_ATTRIBUTE</strong></dt>
<dd><p>Define this macro if any target-specific attributes need to be attached
to the functions in <samp>libgcc</samp> that provide low-level support for
call stack unwinding.  It is used in declarations in <samp>unwind-generic.h</samp>
and the associated definitions of those functions.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fUPDATE_005fSTACK_005fBOUNDARY"></a>Target Hook: <em>void</em> <strong>TARGET_UPDATE_STACK_BOUNDARY</strong> <em>(void)</em></dt>
<dd><p>Define this macro to update the current function stack boundary if
necessary.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fGET_005fDRAP_005fRTX"></a>Target Hook: <em>rtx</em> <strong>TARGET_GET_DRAP_RTX</strong> <em>(void)</em></dt>
<dd><p>This hook should return an rtx for Dynamic Realign Argument Pointer (DRAP) if a
different argument pointer register is needed to access the function&rsquo;s
argument list due to stack realignment.  Return <code>NULL</code> if no DRAP
is needed.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fALLOCATE_005fSTACK_005fSLOTS_005fFOR_005fARGS"></a>Target Hook: <em>bool</em> <strong>TARGET_ALLOCATE_STACK_SLOTS_FOR_ARGS</strong> <em>(void)</em></dt>
<dd><p>When optimization is disabled, this hook indicates whether or not
arguments should be allocated to stack slots.  Normally, GCC allocates
stacks slots for arguments when not optimizing in order to make
debugging easier.  However, when a function is declared with
<code>__attribute__((naked))</code>, there is no stack frame, and the compiler
cannot safely move arguments from the registers in which they are passed
to the stack.  Therefore, this hook should return true in general, but
false for naked functions.  The default implementation always returns true.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fCONST_005fANCHOR"></a>Target Hook: <em>unsigned HOST_WIDE_INT</em> <strong>TARGET_CONST_ANCHOR</strong></dt>
<dd><p>On some architectures it can take multiple instructions to synthesize
a constant.  If there is another constant already in a register that
is close enough in value then it is preferable that the new constant
is computed from this register using immediate addition or
subtraction.  We accomplish this through CSE.  Besides the value of
the constant we also add a lower and an upper constant anchor to the
available expressions.  These are then queried when encountering new
constants.  The anchors are computed by rounding the constant up and
down to a multiple of the value of <code>TARGET_CONST_ANCHOR</code>.
<code>TARGET_CONST_ANCHOR</code> should be the maximum positive value
accepted by immediate-add plus one.  We currently assume that the
value of <code>TARGET_CONST_ANCHOR</code> is a power of 2.  For example, on
MIPS, where add-immediate takes a 16-bit signed value,
<code>TARGET_CONST_ANCHOR</code> is set to &lsquo;<samp>0x8000</samp>&rsquo;.  The default value
is zero, which disables this optimization.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fASAN_005fSHADOW_005fOFFSET"></a>Target Hook: <em>unsigned HOST_WIDE_INT</em> <strong>TARGET_ASAN_SHADOW_OFFSET</strong> <em>(void)</em></dt>
<dd><p>Return the offset bitwise ored into shifted address to get corresponding
Address Sanitizer shadow memory address.  NULL if Address Sanitizer is not
supported by the target.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fMEMMODEL_005fCHECK"></a>Target Hook: <em>unsigned HOST_WIDE_INT</em> <strong>TARGET_MEMMODEL_CHECK</strong> <em>(unsigned HOST_WIDE_INT <var>val</var>)</em></dt>
<dd><p>Validate target specific memory model mask bits. When NULL no target specific
memory model bits are allowed.
</p></dd></dl>

<dl>
<dt><a name="index-TARGET_005fATOMIC_005fTEST_005fAND_005fSET_005fTRUEVAL"></a>Target Hook: <em>unsigned char</em> <strong>TARGET_ATOMIC_TEST_AND_SET_TRUEVAL</strong></dt>
<dd><p>This value should be set if the result written by <code>atomic_test_and_set</code> is not exactly 1, i.e. the <code>bool</code> <code>true</code>.
</p></dd></dl>

<hr>
<a name="Host-Config"></a>
<div class="header">
<p>
Next: <a href="#Fragments" accesskey="n" rel="next">Fragments</a>, Previous: <a href="#Target-Macros" accesskey="p" rel="prev">Target Macros</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Host-Configuration"></a>
<h2 class="chapter">18 Host Configuration</h2>
<a name="index-host-configuration"></a>

<p>Most details about the machine and system on which the compiler is
actually running are detected by the <code>configure</code> script.  Some
things are impossible for <code>configure</code> to detect; these are
described in two ways, either by macros defined in a file named
<samp>xm-<var>machine</var>.h</samp> or by hook functions in the file specified
by the <var>out_host_hook_obj</var> variable in <samp>config.gcc</samp>.  (The
intention is that very few hosts will need a header file but nearly
every fully supported host will need to override some hooks.)
</p>
<p>If you need to define only a few macros, and they have simple
definitions, consider using the <code>xm_defines</code> variable in your
<samp>config.gcc</samp> entry instead of creating a host configuration
header.  See <a href="#System-Config">System Config</a>.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Host-Common" accesskey="1">Host Common</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Things every host probably needs implemented.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Filesystem" accesskey="2">Filesystem</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Your host can&rsquo;t have the letter &lsquo;a&rsquo; in filenames?
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Host-Misc" accesskey="3">Host Misc</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Rare configuration options for hosts.
</td></tr>
</table>

<hr>
<a name="Host-Common"></a>
<div class="header">
<p>
Next: <a href="#Filesystem" accesskey="n" rel="next">Filesystem</a>, Up: <a href="#Host-Config" accesskey="u" rel="up">Host Config</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Host-Common-1"></a>
<h3 class="section">18.1 Host Common</h3>
<a name="index-host-hooks"></a>
<a name="index-host-functions"></a>

<p>Some things are just not portable, even between similar operating systems,
and are too difficult for autoconf to detect.  They get implemented using
hook functions in the file specified by the <var>host_hook_obj</var>
variable in <samp>config.gcc</samp>.
</p>
<dl>
<dt><a name="index-HOST_005fHOOKS_005fEXTRA_005fSIGNALS"></a>Host Hook: <em>void</em> <strong>HOST_HOOKS_EXTRA_SIGNALS</strong> <em>(void)</em></dt>
<dd><p>This host hook is used to set up handling for extra signals.  The most
common thing to do in this hook is to detect stack overflow.
</p></dd></dl>

<dl>
<dt><a name="index-HOST_005fHOOKS_005fGT_005fPCH_005fGET_005fADDRESS"></a>Host Hook: <em>void *</em> <strong>HOST_HOOKS_GT_PCH_GET_ADDRESS</strong> <em>(size_t   <var>size</var>, int <var>fd</var>)</em></dt>
<dd><p>This host hook returns the address of some space that is likely to be
free in some subsequent invocation of the compiler.  We intend to load
the PCH data at this address such that the data need not be relocated.
The area should be able to hold <var>size</var> bytes.  If the host uses
<code>mmap</code>, <var>fd</var> is an open file descriptor that can be used for
probing.
</p></dd></dl>

<dl>
<dt><a name="index-HOST_005fHOOKS_005fGT_005fPCH_005fUSE_005fADDRESS"></a>Host Hook: <em>int</em> <strong>HOST_HOOKS_GT_PCH_USE_ADDRESS</strong> <em>(void * <var>address</var>,   size_t <var>size</var>, int <var>fd</var>, size_t <var>offset</var>)</em></dt>
<dd><p>This host hook is called when a PCH file is about to be loaded.
We want to load <var>size</var> bytes from <var>fd</var> at <var>offset</var>
into memory at <var>address</var>.  The given address will be the result of
a previous invocation of <code>HOST_HOOKS_GT_PCH_GET_ADDRESS</code>.
Return -1 if we couldn&rsquo;t allocate <var>size</var> bytes at <var>address</var>.
Return 0 if the memory is allocated but the data is not loaded.  Return 1
if the hook has performed everything.
</p>
<p>If the implementation uses reserved address space, free any reserved
space beyond <var>size</var>, regardless of the return value.  If no PCH will
be loaded, this hook may be called with <var>size</var> zero, in which case
all reserved address space should be freed.
</p>
<p>Do not try to handle values of <var>address</var> that could not have been
returned by this executable; just return -1.  Such values usually
indicate an out-of-date PCH file (built by some other GCC executable),
and such a PCH file won&rsquo;t work.
</p></dd></dl>

<dl>
<dt><a name="index-HOST_005fHOOKS_005fGT_005fPCH_005fALLOC_005fGRANULARITY"></a>Host Hook: <em>size_t</em> <strong>HOST_HOOKS_GT_PCH_ALLOC_GRANULARITY</strong> <em>(void);</em></dt>
<dd><p>This host hook returns the alignment required for allocating virtual
memory.  Usually this is the same as getpagesize, but on some hosts the
alignment for reserving memory differs from the pagesize for committing
memory.
</p></dd></dl>

<hr>
<a name="Filesystem"></a>
<div class="header">
<p>
Next: <a href="#Host-Misc" accesskey="n" rel="next">Host Misc</a>, Previous: <a href="#Host-Common" accesskey="p" rel="prev">Host Common</a>, Up: <a href="#Host-Config" accesskey="u" rel="up">Host Config</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Host-Filesystem"></a>
<h3 class="section">18.2 Host Filesystem</h3>
<a name="index-configuration-file"></a>
<a name="index-xm_002dmachine_002eh"></a>

<p>GCC needs to know a number of things about the semantics of the host
machine&rsquo;s filesystem.  Filesystems with Unix and MS-DOS semantics are
automatically detected.  For other systems, you can define the
following macros in <samp>xm-<var>machine</var>.h</samp>.
</p>
<dl compact="compact">
<dt><code>HAVE_DOS_BASED_FILE_SYSTEM</code>
<a name="index-HAVE_005fDOS_005fBASED_005fFILE_005fSYSTEM"></a>
</dt>
<dd><p>This macro is automatically defined by <samp>system.h</samp> if the host
file system obeys the semantics defined by MS-DOS instead of Unix.
DOS file systems are case insensitive, file specifications may begin
with a drive letter, and both forward slash and backslash (&lsquo;<samp>/</samp>&rsquo;
and &lsquo;<samp>\</samp>&rsquo;) are directory separators.
</p>
</dd>
<dt><code>DIR_SEPARATOR</code>
<a name="index-DIR_005fSEPARATOR"></a>
</dt>
<dt><code>DIR_SEPARATOR_2</code>
<a name="index-DIR_005fSEPARATOR_005f2"></a>
</dt>
<dd><p>If defined, these macros expand to character constants specifying
separators for directory names within a file specification.
<samp>system.h</samp> will automatically give them appropriate values on
Unix and MS-DOS file systems.  If your file system is neither of
these, define one or both appropriately in <samp>xm-<var>machine</var>.h</samp>.
</p>
<p>However, operating systems like VMS, where constructing a pathname is
more complicated than just stringing together directory names
separated by a special character, should not define either of these
macros.
</p>
</dd>
<dt><code>PATH_SEPARATOR</code>
<a name="index-PATH_005fSEPARATOR"></a>
</dt>
<dd><p>If defined, this macro should expand to a character constant
specifying the separator for elements of search paths.  The default
value is a colon (&lsquo;<samp>:</samp>&rsquo;).  DOS-based systems usually, but not
always, use semicolon (&lsquo;<samp>;</samp>&rsquo;).
</p>
</dd>
<dt><code>VMS</code>
<a name="index-VMS"></a>
</dt>
<dd><p>Define this macro if the host system is VMS.
</p>
</dd>
<dt><code>HOST_OBJECT_SUFFIX</code>
<a name="index-HOST_005fOBJECT_005fSUFFIX"></a>
</dt>
<dd><p>Define this macro to be a C string representing the suffix for object
files on your host machine.  If you do not define this macro, GCC will
use &lsquo;<samp>.o</samp>&rsquo; as the suffix for object files.
</p>
</dd>
<dt><code>HOST_EXECUTABLE_SUFFIX</code>
<a name="index-HOST_005fEXECUTABLE_005fSUFFIX"></a>
</dt>
<dd><p>Define this macro to be a C string representing the suffix for
executable files on your host machine.  If you do not define this macro,
GCC will use the null string as the suffix for executable files.
</p>
</dd>
<dt><code>HOST_BIT_BUCKET</code>
<a name="index-HOST_005fBIT_005fBUCKET"></a>
</dt>
<dd><p>A pathname defined by the host operating system, which can be opened as
a file and written to, but all the information written is discarded.
This is commonly known as a <em>bit bucket</em> or <em>null device</em>.  If
you do not define this macro, GCC will use &lsquo;<samp>/dev/null</samp>&rsquo; as the bit
bucket.  If the host does not support a bit bucket, define this macro to
an invalid filename.
</p>
</dd>
<dt><code>UPDATE_PATH_HOST_CANONICALIZE (<var>path</var>)</code>
<a name="index-UPDATE_005fPATH_005fHOST_005fCANONICALIZE-_0028path_0029"></a>
</dt>
<dd><p>If defined, a C statement (sans semicolon) that performs host-dependent
canonicalization when a path used in a compilation driver or
preprocessor is canonicalized.  <var>path</var> is a malloc-ed path to be
canonicalized.  If the C statement does canonicalize <var>path</var> into a
different buffer, the old path should be freed and the new buffer should
have been allocated with malloc.
</p>
</dd>
<dt><code>DUMPFILE_FORMAT</code>
<a name="index-DUMPFILE_005fFORMAT"></a>
</dt>
<dd><p>Define this macro to be a C string representing the format to use for
constructing the index part of debugging dump file names.  The resultant
string must fit in fifteen bytes.  The full filename will be the
concatenation of: the prefix of the assembler file name, the string
resulting from applying this format to an index number, and a string
unique to each dump file kind, e.g. &lsquo;<samp>rtl</samp>&rsquo;.
</p>
<p>If you do not define this macro, GCC will use &lsquo;<samp>.%02d.</samp>&rsquo;.  You should
define this macro if using the default will create an invalid file name.
</p>
</dd>
<dt><code>DELETE_IF_ORDINARY</code>
<a name="index-DELETE_005fIF_005fORDINARY"></a>
</dt>
<dd><p>Define this macro to be a C statement (sans semicolon) that performs
host-dependent removal of ordinary temp files in the compilation driver.
</p>
<p>If you do not define this macro, GCC will use the default version.  You
should define this macro if the default version does not reliably remove
the temp file as, for example, on VMS which allows multiple versions
of a file.
</p>
</dd>
<dt><code>HOST_LACKS_INODE_NUMBERS</code>
<a name="index-HOST_005fLACKS_005fINODE_005fNUMBERS"></a>
</dt>
<dd><p>Define this macro if the host filesystem does not report meaningful inode
numbers in struct stat.
</p></dd>
</dl>

<hr>
<a name="Host-Misc"></a>
<div class="header">
<p>
Previous: <a href="#Filesystem" accesskey="p" rel="prev">Filesystem</a>, Up: <a href="#Host-Config" accesskey="u" rel="up">Host Config</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Host-Misc-1"></a>
<h3 class="section">18.3 Host Misc</h3>
<a name="index-configuration-file-1"></a>
<a name="index-xm_002dmachine_002eh-1"></a>

<dl compact="compact">
<dt><code>FATAL_EXIT_CODE</code>
<a name="index-FATAL_005fEXIT_005fCODE"></a>
</dt>
<dd><p>A C expression for the status code to be returned when the compiler
exits after serious errors.  The default is the system-provided macro
&lsquo;<samp>EXIT_FAILURE</samp>&rsquo;, or &lsquo;<samp>1</samp>&rsquo; if the system doesn&rsquo;t define that
macro.  Define this macro only if these defaults are incorrect.
</p>
</dd>
<dt><code>SUCCESS_EXIT_CODE</code>
<a name="index-SUCCESS_005fEXIT_005fCODE"></a>
</dt>
<dd><p>A C expression for the status code to be returned when the compiler
exits without serious errors.  (Warnings are not serious errors.)  The
default is the system-provided macro &lsquo;<samp>EXIT_SUCCESS</samp>&rsquo;, or &lsquo;<samp>0</samp>&rsquo; if
the system doesn&rsquo;t define that macro.  Define this macro only if these
defaults are incorrect.
</p>
</dd>
<dt><code>USE_C_ALLOCA</code>
<a name="index-USE_005fC_005fALLOCA"></a>
</dt>
<dd><p>Define this macro if GCC should use the C implementation of <code>alloca</code>
provided by <samp>libiberty.a</samp>.  This only affects how some parts of the
compiler itself allocate memory.  It does not change code generation.
</p>
<p>When GCC is built with a compiler other than itself, the C <code>alloca</code>
is always used.  This is because most other implementations have serious
bugs.  You should define this macro only on a system where no
stack-based <code>alloca</code> can possibly work.  For instance, if a system
has a small limit on the size of the stack, GCC&rsquo;s builtin <code>alloca</code>
will not work reliably.
</p>
</dd>
<dt><code>COLLECT2_HOST_INITIALIZATION</code>
<a name="index-COLLECT2_005fHOST_005fINITIALIZATION"></a>
</dt>
<dd><p>If defined, a C statement (sans semicolon) that performs host-dependent
initialization when <code>collect2</code> is being initialized.
</p>
</dd>
<dt><code>GCC_DRIVER_HOST_INITIALIZATION</code>
<a name="index-GCC_005fDRIVER_005fHOST_005fINITIALIZATION"></a>
</dt>
<dd><p>If defined, a C statement (sans semicolon) that performs host-dependent
initialization when a compilation driver is being initialized.
</p>
</dd>
<dt><code>HOST_LONG_LONG_FORMAT</code>
<a name="index-HOST_005fLONG_005fLONG_005fFORMAT"></a>
</dt>
<dd><p>If defined, the string used to indicate an argument of type <code>long
long</code> to functions like <code>printf</code>.  The default value is
<code>&quot;ll&quot;</code>.
</p>
</dd>
<dt><code>HOST_LONG_FORMAT</code>
<a name="index-HOST_005fLONG_005fFORMAT"></a>
</dt>
<dd><p>If defined, the string used to indicate an argument of type <code>long</code>
to functions like <code>printf</code>.  The default value is <code>&quot;l&quot;</code>.
</p>
</dd>
<dt><code>HOST_PTR_PRINTF</code>
<a name="index-HOST_005fPTR_005fPRINTF"></a>
</dt>
<dd><p>If defined, the string used to indicate an argument of type <code>void *</code>
to functions like <code>printf</code>.  The default value is <code>&quot;%p&quot;</code>.
</p></dd>
</dl>

<p>In addition, if <code>configure</code> generates an incorrect definition of
any of the macros in <samp>auto-host.h</samp>, you can override that
definition in a host configuration header.  If you need to do this,
first see if it is possible to fix <code>configure</code>.
</p>
<hr>
<a name="Fragments"></a>
<div class="header">
<p>
Next: <a href="#Collect2" accesskey="n" rel="next">Collect2</a>, Previous: <a href="#Host-Config" accesskey="p" rel="prev">Host Config</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Makefile-Fragments"></a>
<h2 class="chapter">19 Makefile Fragments</h2>
<a name="index-makefile-fragment"></a>

<p>When you configure GCC using the <samp>configure</samp> script, it will
construct the file <samp>Makefile</samp> from the template file
<samp>Makefile.in</samp>.  When it does this, it can incorporate makefile
fragments from the <samp>config</samp> directory.  These are used to set
Makefile parameters that are not amenable to being calculated by
autoconf.  The list of fragments to incorporate is set by
<samp>config.gcc</samp> (and occasionally <samp>config.build</samp>
and <samp>config.host</samp>); See <a href="#System-Config">System Config</a>.
</p>
<p>Fragments are named either <samp>t-<var>target</var></samp> or <samp>x-<var>host</var></samp>,
depending on whether they are relevant to configuring GCC to produce
code for a particular target, or to configuring GCC to run on a
particular host.  Here <var>target</var> and <var>host</var> are mnemonics
which usually have some relationship to the canonical system name, but
no formal connection.
</p>
<p>If these files do not exist, it means nothing needs to be added for a
given target or host.  Most targets need a few <samp>t-<var>target</var></samp>
fragments, but needing <samp>x-<var>host</var></samp> fragments is rare.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Target-Fragment" accesskey="1">Target Fragment</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Writing <samp>t-<var>target</var></samp> files.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Host-Fragment" accesskey="2">Host Fragment</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Writing <samp>x-<var>host</var></samp> files.
</td></tr>
</table>

<hr>
<a name="Target-Fragment"></a>
<div class="header">
<p>
Next: <a href="#Host-Fragment" accesskey="n" rel="next">Host Fragment</a>, Up: <a href="#Fragments" accesskey="u" rel="up">Fragments</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Target-Makefile-Fragments"></a>
<h3 class="section">19.1 Target Makefile Fragments</h3>
<a name="index-target-makefile-fragment"></a>
<a name="index-t_002dtarget"></a>

<p>Target makefile fragments can set these Makefile variables.
</p>
<dl compact="compact">
<dd><a name="index-LIBGCC2_005fCFLAGS"></a>
</dd>
<dt><code>LIBGCC2_CFLAGS</code></dt>
<dd><p>Compiler flags to use when compiling <samp>libgcc2.c</samp>.
</p>
<a name="index-LIB2FUNCS_005fEXTRA"></a>
</dd>
<dt><code>LIB2FUNCS_EXTRA</code></dt>
<dd><p>A list of source file names to be compiled or assembled and inserted
into <samp>libgcc.a</samp>.
</p>
<a name="index-CRTSTUFF_005fT_005fCFLAGS"></a>
</dd>
<dt><code>CRTSTUFF_T_CFLAGS</code></dt>
<dd><p>Special flags used when compiling <samp>crtstuff.c</samp>.
See <a href="#Initialization">Initialization</a>.
</p>
<a name="index-CRTSTUFF_005fT_005fCFLAGS_005fS"></a>
</dd>
<dt><code>CRTSTUFF_T_CFLAGS_S</code></dt>
<dd><p>Special flags used when compiling <samp>crtstuff.c</samp> for shared
linking.  Used if you use <samp>crtbeginS.o</samp> and <samp>crtendS.o</samp>
in <code>EXTRA-PARTS</code>.
See <a href="#Initialization">Initialization</a>.
</p>
<a name="index-MULTILIB_005fOPTIONS"></a>
</dd>
<dt><code>MULTILIB_OPTIONS</code></dt>
<dd><p>For some targets, invoking GCC in different ways produces objects
that can not be linked together.  For example, for some targets GCC
produces both big and little endian code.  For these targets, you must
arrange for multiple versions of <samp>libgcc.a</samp> to be compiled, one for
each set of incompatible options.  When GCC invokes the linker, it
arranges to link in the right version of <samp>libgcc.a</samp>, based on
the command line options used.
</p>
<p>The <code>MULTILIB_OPTIONS</code> macro lists the set of options for which
special versions of <samp>libgcc.a</samp> must be built.  Write options that
are mutually incompatible side by side, separated by a slash.  Write
options that may be used together separated by a space.  The build
procedure will build all combinations of compatible options.
</p>
<p>For example, if you set <code>MULTILIB_OPTIONS</code> to &lsquo;<samp>m68000/m68020
msoft-float</samp>&rsquo;, <samp>Makefile</samp> will build special versions of
<samp>libgcc.a</samp> using the following sets of options:  <samp>-m68000</samp>,
<samp>-m68020</samp>, <samp>-msoft-float</samp>, &lsquo;<samp>-m68000 -msoft-float</samp>&rsquo;, and
&lsquo;<samp>-m68020 -msoft-float</samp>&rsquo;.
</p>
<a name="index-MULTILIB_005fDIRNAMES"></a>
</dd>
<dt><code>MULTILIB_DIRNAMES</code></dt>
<dd><p>If <code>MULTILIB_OPTIONS</code> is used, this variable specifies the
directory names that should be used to hold the various libraries.
Write one element in <code>MULTILIB_DIRNAMES</code> for each element in
<code>MULTILIB_OPTIONS</code>.  If <code>MULTILIB_DIRNAMES</code> is not used, the
default value will be <code>MULTILIB_OPTIONS</code>, with all slashes treated
as spaces.
</p>
<p><code>MULTILIB_DIRNAMES</code> describes the multilib directories using GCC
conventions and is applied to directories that are part of the GCC
installation.  When multilib-enabled, the compiler will add a
subdirectory of the form <var>prefix</var>/<var>multilib</var> before each
directory in the search path for libraries and crt files.
</p>
<p>For example, if <code>MULTILIB_OPTIONS</code> is set to &lsquo;<samp>m68000/m68020
msoft-float</samp>&rsquo;, then the default value of <code>MULTILIB_DIRNAMES</code> is
&lsquo;<samp>m68000 m68020 msoft-float</samp>&rsquo;.  You may specify a different value if
you desire a different set of directory names.
</p>
<a name="index-MULTILIB_005fMATCHES"></a>
</dd>
<dt><code>MULTILIB_MATCHES</code></dt>
<dd><p>Sometimes the same option may be written in two different ways.  If an
option is listed in <code>MULTILIB_OPTIONS</code>, GCC needs to know about
any synonyms.  In that case, set <code>MULTILIB_MATCHES</code> to a list of
items of the form &lsquo;<samp>option=option</samp>&rsquo; to describe all relevant
synonyms.  For example, &lsquo;<samp>m68000=mc68000 m68020=mc68020</samp>&rsquo;.
</p>
<a name="index-MULTILIB_005fEXCEPTIONS"></a>
</dd>
<dt><code>MULTILIB_EXCEPTIONS</code></dt>
<dd><p>Sometimes when there are multiple sets of <code>MULTILIB_OPTIONS</code> being
specified, there are combinations that should not be built.  In that
case, set <code>MULTILIB_EXCEPTIONS</code> to be all of the switch exceptions
in shell case syntax that should not be built.
</p>
<p>For example the ARM processor cannot execute both hardware floating
point instructions and the reduced size THUMB instructions at the same
time, so there is no need to build libraries with both of these
options enabled.  Therefore <code>MULTILIB_EXCEPTIONS</code> is set to:
</p><div class="smallexample">
<pre class="smallexample">*mthumb/*mhard-float*
</pre></div>

<a name="index-MULTILIB_005fREQUIRED"></a>
</dd>
<dt><code>MULTILIB_REQUIRED</code></dt>
<dd><p>Sometimes when there are only a few combinations are required, it would
be a big effort to come up with a <code>MULTILIB_EXCEPTIONS</code> list to
cover all undesired ones.  In such a case, just listing all the required
combinations in <code>MULTILIB_REQUIRED</code> would be more straightforward.
</p>
<p>The way to specify the entries in <code>MULTILIB_REQUIRED</code> is same with
the way used for <code>MULTILIB_EXCEPTIONS</code>, only this time what are
required will be specified.  Suppose there are multiple sets of
<code>MULTILIB_OPTIONS</code> and only two combinations are required, one
for ARMv7-M and one for ARMv7-R with hard floating-point ABI and FPU, the
<code>MULTILIB_REQUIRED</code> can be set to:
</p><div class="smallexample">
<pre class="smallexample"><code>MULTILIB_REQUIRED</code> =  mthumb/march=armv7-m
<code>MULTILIB_REQUIRED</code> += march=armv7-r/mfloat-abi=hard/mfpu=vfpv3-d16
</pre></div>

<p>The <code>MULTILIB_REQUIRED</code> can be used together with
<code>MULTILIB_EXCEPTIONS</code>.  The option combinations generated from
<code>MULTILIB_OPTIONS</code> will be filtered by <code>MULTILIB_EXCEPTIONS</code>
and then by <code>MULTILIB_REQUIRED</code>.
</p>
<a name="index-MULTILIB_005fREUSE"></a>
</dd>
<dt><code>MULTILIB_REUSE</code></dt>
<dd><p>Sometimes it is desirable to reuse one existing multilib for different
sets of options.  Such kind of reuse can minimize the number of multilib
variants.  And for some targets it is better to reuse an existing multilib
than to fall back to default multilib when there is no corresponding multilib.
This can be done by adding reuse rules to <code>MULTILIB_REUSE</code>.
</p>
<p>A reuse rule is comprised of two parts connected by equality sign.  The left part
is option set used to build multilib and the right part is option set that will
reuse this multilib.  The order of options in the left part matters and should be
same with those specified in <code>MULTILIB_REQUIRED</code> or aligned with order in
<code>MULTILIB_OPTIONS</code>.  There is no such limitation for options in right part
as we don&rsquo;t build multilib from them.  But the equality sign in both parts should
be replaced with period.
</p>
<p>The <code>MULTILIB_REUSE</code> is different from <code>MULTILIB_MATCHES</code> in that it
sets up relations between two option sets rather than two options.  Here is an
example to demo how we reuse libraries built in Thumb mode for applications built
in ARM mode:
</p><div class="smallexample">
<pre class="smallexample"><code>MULTILIB_REUSE</code> = mthumb/march.armv7-r=marm/march.armv7-r
</pre></div>

<p>Before the advent of <code>MULTILIB_REUSE</code>, GCC select multilib by comparing command
line options with options used to build multilib.  The <code>MULTILIB_REUSE</code> is
complementary to that way.  Only when the original comparison matches nothing it will
work to see if it is OK to reuse some existing multilib.
</p>
<a name="index-MULTILIB_005fEXTRA_005fOPTS"></a>
</dd>
<dt><code>MULTILIB_EXTRA_OPTS</code></dt>
<dd><p>Sometimes it is desirable that when building multiple versions of
<samp>libgcc.a</samp> certain options should always be passed on to the
compiler.  In that case, set <code>MULTILIB_EXTRA_OPTS</code> to be the list
of options to be used for all builds.  If you set this, you should
probably set <code>CRTSTUFF_T_CFLAGS</code> to a dash followed by it.
</p>
<a name="index-MULTILIB_005fOSDIRNAMES"></a>
</dd>
<dt><code>MULTILIB_OSDIRNAMES</code></dt>
<dd><p>If <code>MULTILIB_OPTIONS</code> is used, this variable specifies 
a list of subdirectory names, that are used to modify the search
path depending on the chosen multilib.  Unlike <code>MULTILIB_DIRNAMES</code>,
<code>MULTILIB_OSDIRNAMES</code> describes the multilib directories using
operating systems conventions, and is applied to the directories such as
<code>lib</code> or those in the <code>LIBRARY_PATH</code> environment variable.
The format is either the same as of
<code>MULTILIB_DIRNAMES</code>, or a set of mappings.  When it is the same
as <code>MULTILIB_DIRNAMES</code>, it describes the multilib directories
using operating system conventions, rather than GCC conventions.  When it is a set
of mappings of the form <var>gccdir</var>=<var>osdir</var>, the left side gives
the GCC convention and the right gives the equivalent OS defined
location.  If the <var>osdir</var> part begins with a &lsquo;<samp>!</samp>&rsquo;,
GCC will not search in the non-multilib directory and use
exclusively the multilib directory.  Otherwise, the compiler will
examine the search path for libraries and crt files twice; the first
time it will add <var>multilib</var> to each directory in the search path,
the second it will not.
</p>
<p>For configurations that support both multilib and multiarch,
<code>MULTILIB_OSDIRNAMES</code> also encodes the multiarch name, thus
subsuming <code>MULTIARCH_DIRNAME</code>.  The multiarch name is appended to
each directory name, separated by a colon (e.g.
&lsquo;<samp>../lib32:i386-linux-gnu</samp>&rsquo;).
</p>
<p>Each multiarch subdirectory will be searched before the corresponding OS
multilib directory, for example &lsquo;<samp>/lib/i386-linux-gnu</samp>&rsquo; before
&lsquo;<samp>/lib/../lib32</samp>&rsquo;.  The multiarch name will also be used to modify the
system header search path, as explained for <code>MULTIARCH_DIRNAME</code>.
</p>
<a name="index-MULTIARCH_005fDIRNAME"></a>
</dd>
<dt><code>MULTIARCH_DIRNAME</code></dt>
<dd><p>This variable specifies the multiarch name for configurations that are
multiarch-enabled but not multilibbed configurations.
</p>
<p>The multiarch name is used to augment the search path for libraries, crt
files and system header files with additional locations.  The compiler
will add a multiarch subdirectory of the form
<var>prefix</var>/<var>multiarch</var> before each directory in the library and
crt search path.  It will also add two directories
<code>LOCAL_INCLUDE_DIR</code>/<var>multiarch</var> and
<code>NATIVE_SYSTEM_HEADER_DIR</code>/<var>multiarch</var>) to the system header
search path, respectively before <code>LOCAL_INCLUDE_DIR</code> and
<code>NATIVE_SYSTEM_HEADER_DIR</code>.
</p>
<p><code>MULTIARCH_DIRNAME</code> is not used for configurations that support
both multilib and multiarch.  In that case, multiarch names are encoded
in <code>MULTILIB_OSDIRNAMES</code> instead.
</p>
<p>More documentation about multiarch can be found at
<a href="http://wiki.debian.org/Multiarch">http://wiki.debian.org/Multiarch</a>.
</p>
<a name="index-SPECS"></a>
</dd>
<dt><code>SPECS</code></dt>
<dd><p>Unfortunately, setting <code>MULTILIB_EXTRA_OPTS</code> is not enough, since
it does not affect the build of target libraries, at least not the
build of the default multilib.  One possible work-around is to use
<code>DRIVER_SELF_SPECS</code> to bring options from the <samp>specs</samp> file
as if they had been passed in the compiler driver command line.
However, you don&rsquo;t want to be adding these options after the toolchain
is installed, so you can instead tweak the <samp>specs</samp> file that will
be used during the toolchain build, while you still install the
original, built-in <samp>specs</samp>.  The trick is to set <code>SPECS</code> to
some other filename (say <samp>specs.install</samp>), that will then be
created out of the built-in specs, and introduce a <samp>Makefile</samp>
rule to generate the <samp>specs</samp> file that&rsquo;s going to be used at
build time out of your <samp>specs.install</samp>.
</p>
</dd>
<dt><code>T_CFLAGS</code></dt>
<dd><p>These are extra flags to pass to the C compiler.  They are used both
when building GCC, and when compiling things with the just-built GCC.
This variable is deprecated and should not be used.
</p></dd>
</dl>

<hr>
<a name="Host-Fragment"></a>
<div class="header">
<p>
Previous: <a href="#Target-Fragment" accesskey="p" rel="prev">Target Fragment</a>, Up: <a href="#Fragments" accesskey="u" rel="up">Fragments</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Host-Makefile-Fragments"></a>
<h3 class="section">19.2 Host Makefile Fragments</h3>
<a name="index-host-makefile-fragment"></a>
<a name="index-x_002dhost"></a>

<p>The use of <samp>x-<var>host</var></samp> fragments is discouraged.  You should only
use it for makefile dependencies.
</p>
<hr>
<a name="Collect2"></a>
<div class="header">
<p>
Next: <a href="#Header-Dirs" accesskey="n" rel="next">Header Dirs</a>, Previous: <a href="#Fragments" accesskey="p" rel="prev">Fragments</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="collect2"></a>
<h2 class="chapter">20 <code>collect2</code></h2>

<p>GCC uses a utility called <code>collect2</code> on nearly all systems to arrange
to call various initialization functions at start time.
</p>
<p>The program <code>collect2</code> works by linking the program once and
looking through the linker output file for symbols with particular names
indicating they are constructor functions.  If it finds any, it
creates a new temporary &lsquo;<samp>.c</samp>&rsquo; file containing a table of them,
compiles it, and links the program a second time including that file.
</p>
<a name="index-_005f_005fmain"></a>
<a name="index-constructors_002c-automatic-calls"></a>
<p>The actual calls to the constructors are carried out by a subroutine
called <code>__main</code>, which is called (automatically) at the beginning
of the body of <code>main</code> (provided <code>main</code> was compiled with GNU
CC).  Calling <code>__main</code> is necessary, even when compiling C code, to
allow linking C and C++ object code together.  (If you use
<samp>-nostdlib</samp>, you get an unresolved reference to <code>__main</code>,
since it&rsquo;s defined in the standard GCC library.  Include <samp>-lgcc</samp> at
the end of your compiler command line to resolve this reference.)
</p>
<p>The program <code>collect2</code> is installed as <code>ld</code> in the directory
where the passes of the compiler are installed.  When <code>collect2</code>
needs to find the <em>real</em> <code>ld</code>, it tries the following file
names:
</p>
<ul>
<li> a hard coded linker file name, if GCC was configured with the
<samp>--with-ld</samp> option.

</li><li> <samp>real-ld</samp> in the directories listed in the compiler&rsquo;s search
directories.

</li><li> <samp>real-ld</samp> in the directories listed in the environment variable
<code>PATH</code>.

</li><li> The file specified in the <code>REAL_LD_FILE_NAME</code> configuration macro,
if specified.

</li><li> <samp>ld</samp> in the compiler&rsquo;s search directories, except that
<code>collect2</code> will not execute itself recursively.

</li><li> <samp>ld</samp> in <code>PATH</code>.
</li></ul>

<p>&ldquo;The compiler&rsquo;s search directories&rdquo; means all the directories where
<code>gcc</code> searches for passes of the compiler.  This includes
directories that you specify with <samp>-B</samp>.
</p>
<p>Cross-compilers search a little differently:
</p>
<ul>
<li> <samp>real-ld</samp> in the compiler&rsquo;s search directories.

</li><li> <samp><var>target</var>-real-ld</samp> in <code>PATH</code>.

</li><li> The file specified in the <code>REAL_LD_FILE_NAME</code> configuration macro,
if specified.

</li><li> <samp>ld</samp> in the compiler&rsquo;s search directories.

</li><li> <samp><var>target</var>-ld</samp> in <code>PATH</code>.
</li></ul>

<p><code>collect2</code> explicitly avoids running <code>ld</code> using the file name
under which <code>collect2</code> itself was invoked.  In fact, it remembers
up a list of such names&mdash;in case one copy of <code>collect2</code> finds
another copy (or version) of <code>collect2</code> installed as <code>ld</code> in a
second place in the search path.
</p>
<p><code>collect2</code> searches for the utilities <code>nm</code> and <code>strip</code>
using the same algorithm as above for <code>ld</code>.
</p>
<hr>
<a name="Header-Dirs"></a>
<div class="header">
<p>
Next: <a href="#Type-Information" accesskey="n" rel="next">Type Information</a>, Previous: <a href="#Collect2" accesskey="p" rel="prev">Collect2</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Standard-Header-File-Directories"></a>
<h2 class="chapter">21 Standard Header File Directories</h2>

<p><code>GCC_INCLUDE_DIR</code> means the same thing for native and cross.  It is
where GCC stores its private include files, and also where GCC
stores the fixed include files.  A cross compiled GCC runs
<code>fixincludes</code> on the header files in <samp>$(tooldir)/include</samp>.
(If the cross compilation header files need to be fixed, they must be
installed before GCC is built.  If the cross compilation header files
are already suitable for GCC, nothing special need be done).
</p>
<p><code>GPLUSPLUS_INCLUDE_DIR</code> means the same thing for native and cross.  It
is where <code>g++</code> looks first for header files.  The C++ library
installs only target independent header files in that directory.
</p>
<p><code>LOCAL_INCLUDE_DIR</code> is used only by native compilers.  GCC
doesn&rsquo;t install anything there.  It is normally
<samp>/usr/local/include</samp>.  This is where local additions to a packaged
system should place header files.
</p>
<p><code>CROSS_INCLUDE_DIR</code> is used only by cross compilers.  GCC
doesn&rsquo;t install anything there.
</p>
<p><code>TOOL_INCLUDE_DIR</code> is used for both native and cross compilers.  It
is the place for other packages to install header files that GCC will
use.  For a cross-compiler, this is the equivalent of
<samp>/usr/include</samp>.  When you build a cross-compiler,
<code>fixincludes</code> processes any header files in this directory.
</p>
<hr>
<a name="Type-Information"></a>
<div class="header">
<p>
Next: <a href="#Plugins" accesskey="n" rel="next">Plugins</a>, Previous: <a href="#Header-Dirs" accesskey="p" rel="prev">Header Dirs</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Memory-Management-and-Type-Information"></a>
<h2 class="chapter">22 Memory Management and Type Information</h2>
<a name="index-GGC"></a>
<a name="index-GTY"></a>

<p>GCC uses some fairly sophisticated memory management techniques, which
involve determining information about GCC&rsquo;s data structures from GCC&rsquo;s
source code and using this information to perform garbage collection and
implement precompiled headers.
</p>
<p>A full C++ parser would be too complicated for this task, so a limited
subset of C++ is interpreted and special markers are used to determine
what parts of the source to look at.  All <code>struct</code>, <code>union</code>
and <code>template</code> structure declarations that define data structures
that are allocated under control of the garbage collector must be
marked.  All global variables that hold pointers to garbage-collected
memory must also be marked.  Finally, all global variables that need
to be saved and restored by a precompiled header must be marked.  (The
precompiled header mechanism can only save static variables if they&rsquo;re
scalar. Complex data structures must be allocated in garbage-collected
memory to be saved in a precompiled header.)
</p>
<p>The full format of a marker is
</p><div class="smallexample">
<pre class="smallexample">GTY (([<var>option</var>] [(<var>param</var>)], [<var>option</var>] [(<var>param</var>)] &hellip;))
</pre></div>
<p>but in most cases no options are needed.  The outer double parentheses
are still necessary, though: <code>GTY(())</code>.  Markers can appear:
</p>
<ul>
<li> In a structure definition, before the open brace;
</li><li> In a global variable declaration, after the keyword <code>static</code> or
<code>extern</code>; and
</li><li> In a structure field definition, before the name of the field.
</li></ul>

<p>Here are some examples of marking simple data structures and globals.
</p>
<div class="smallexample">
<pre class="smallexample">struct GTY(()) <var>tag</var>
{
  <var>fields</var>&hellip;
};

typedef struct GTY(()) <var>tag</var>
{
  <var>fields</var>&hellip;
} *<var>typename</var>;

static GTY(()) struct <var>tag</var> *<var>list</var>;   /* <span class="roman">points to GC memory</span> */
static GTY(()) int <var>counter</var>;        /* <span class="roman">save counter in a PCH</span> */
</pre></div>

<p>The parser understands simple typedefs such as
<code>typedef struct <var>tag</var> *<var>name</var>;</code> and
<code>typedef int <var>name</var>;</code>.
These don&rsquo;t need to be marked.
</p>
<p>Since <code>gengtype</code>&rsquo;s understanding of C++ is limited, there are
several constructs and declarations that are not supported inside
classes/structures marked for automatic GC code generation.  The
following C++ constructs produce a <code>gengtype</code> error on
structures/classes marked for automatic GC code generation:
</p>
<ul>
<li> Type definitions inside classes/structures are not supported.
</li><li> Enumerations inside classes/structures are not supported.
</li></ul>

<p>If you have a class or structure using any of the above constructs,
you need to mark that class as <code>GTY ((user))</code> and provide your
own marking routines (see section <a href="#User-GC">User GC</a> for details).
</p>
<p>It is always valid to include function definitions inside classes.
Those are always ignored by <code>gengtype</code>, as it only cares about
data members.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#GTY-Options" accesskey="1">GTY Options</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">What goes inside a <code>GTY(())</code>.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#User-GC" accesskey="2">User GC</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Adding user-provided GC marking routines.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#GGC-Roots" accesskey="3">GGC Roots</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Making global variables GGC roots.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Files" accesskey="4">Files</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How the generated files work.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Invoking-the-garbage-collector" accesskey="5">Invoking the garbage collector</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How to invoke the garbage collector.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Troubleshooting" accesskey="6">Troubleshooting</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">When something does not work as expected.
</td></tr>
</table>

<hr>
<a name="GTY-Options"></a>
<div class="header">
<p>
Next: <a href="#User-GC" accesskey="n" rel="next">User GC</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-Inside-of-a-GTY_0028_0028_0029_0029"></a>
<h3 class="section">22.1 The Inside of a <code>GTY(())</code></h3>

<p>Sometimes the C code is not enough to fully describe the type
structure.  Extra information can be provided with <code>GTY</code> options
and additional markers.  Some options take a parameter, which may be
either a string or a type name, depending on the parameter.  If an
option takes no parameter, it is acceptable either to omit the
parameter entirely, or to provide an empty string as a parameter.  For
example, <code>GTY&nbsp;((skip))<!-- /@w --></code> and <code>GTY&nbsp;((skip&nbsp;(&quot;&quot;)))<!-- /@w --></code> are
equivalent.
</p>
<p>When the parameter is a string, often it is a fragment of C code.  Four
special escapes may be used in these strings, to refer to pieces of
the data structure being marked:
</p>
<a name="index-_0025-in-GTY-option"></a>
<dl compact="compact">
<dt><code>%h</code></dt>
<dd><p>The current structure.
</p></dd>
<dt><code>%1</code></dt>
<dd><p>The structure that immediately contains the current structure.
</p></dd>
<dt><code>%0</code></dt>
<dd><p>The outermost structure that contains the current structure.
</p></dd>
<dt><code>%a</code></dt>
<dd><p>A partial expression of the form <code>[i1][i2]&hellip;</code> that indexes
the array item currently being marked.
</p></dd>
</dl>

<p>For instance, suppose that you have a structure of the form
</p><div class="smallexample">
<pre class="smallexample">struct A {
  &hellip;
};
struct B {
  struct A foo[12];
};
</pre></div>
<p>and <code>b</code> is a variable of type <code>struct B</code>.  When marking
&lsquo;<samp>b.foo[11]</samp>&rsquo;, <code>%h</code> would expand to &lsquo;<samp>b.foo[11]</samp>&rsquo;,
<code>%0</code> and <code>%1</code> would both expand to &lsquo;<samp>b</samp>&rsquo;, and <code>%a</code>
would expand to &lsquo;<samp>[11]</samp>&rsquo;.
</p>
<p>As in ordinary C, adjacent strings will be concatenated; this is
helpful when you have a complicated expression.
</p><div class="smallexample">
<pre class="smallexample">GTY ((chain_next (&quot;TREE_CODE (&amp;%h.generic) == INTEGER_TYPE&quot;
                  &quot; ? TYPE_NEXT_VARIANT (&amp;%h.generic)&quot;
                  &quot; : TREE_CHAIN (&amp;%h.generic)&quot;)))
</pre></div>

<p>The available options are:
</p>
<dl compact="compact">
<dd><a name="index-length"></a>
</dd>
<dt><code>length (&quot;<var>expression</var>&quot;)</code></dt>
<dd>
<p>There are two places the type machinery will need to be explicitly told
the length of an array of non-atomic objects.  The first case is when a
structure ends in a variable-length array, like this:
</p><div class="smallexample">
<pre class="smallexample">struct GTY(()) rtvec_def {
  int num_elem;         /* <span class="roman">number of elements</span> */
  rtx GTY ((length (&quot;%h.num_elem&quot;))) elem[1];
};
</pre></div>

<p>In this case, the <code>length</code> option is used to override the specified
array length (which should usually be <code>1</code>).  The parameter of the
option is a fragment of C code that calculates the length.
</p>
<p>The second case is when a structure or a global variable contains a
pointer to an array, like this:
</p><div class="smallexample">
<pre class="smallexample">struct gimple_omp_for_iter * GTY((length (&quot;%h.collapse&quot;))) iter;
</pre></div>
<p>In this case, <code>iter</code> has been allocated by writing something like
</p><div class="smallexample">
<pre class="smallexample">  x-&gt;iter = ggc_alloc_cleared_vec_gimple_omp_for_iter (collapse);
</pre></div>
<p>and the <code>collapse</code> provides the length of the field.
</p>
<p>This second use of <code>length</code> also works on global variables, like:
</p><pre class="verbatim">static GTY((length(&quot;reg_known_value_size&quot;))) rtx *reg_known_value;
</pre>
<p>Note that the <code>length</code> option is only meant for use with arrays of
non-atomic objects, that is, objects that contain pointers pointing to
other GTY-managed objects.  For other GC-allocated arrays and strings
you should use <code>atomic</code>.
</p>
<a name="index-skip"></a>
</dd>
<dt><code>skip</code></dt>
<dd>
<p>If <code>skip</code> is applied to a field, the type machinery will ignore it.
This is somewhat dangerous; the only safe use is in a union when one
field really isn&rsquo;t ever used.
</p>
<a name="index-desc"></a>
<a name="index-tag"></a>
<a name="index-default"></a>
</dd>
<dt><code>desc (&quot;<var>expression</var>&quot;)</code></dt>
<dt><code>tag (&quot;<var>constant</var>&quot;)</code></dt>
<dt><code>default</code></dt>
<dd>
<p>The type machinery needs to be told which field of a <code>union</code> is
currently active.  This is done by giving each field a constant
<code>tag</code> value, and then specifying a discriminator using <code>desc</code>.
The value of the expression given by <code>desc</code> is compared against
each <code>tag</code> value, each of which should be different.  If no
<code>tag</code> is matched, the field marked with <code>default</code> is used if
there is one, otherwise no field in the union will be marked.
</p>
<p>In the <code>desc</code> option, the &ldquo;current structure&rdquo; is the union that
it discriminates.  Use <code>%1</code> to mean the structure containing it.
There are no escapes available to the <code>tag</code> option, since it is a
constant.
</p>
<p>For example,
</p><div class="smallexample">
<pre class="smallexample">struct GTY(()) tree_binding
{
  struct tree_common common;
  union tree_binding_u {
    tree GTY ((tag (&quot;0&quot;))) scope;
    struct cp_binding_level * GTY ((tag (&quot;1&quot;))) level;
  } GTY ((desc (&quot;BINDING_HAS_LEVEL_P ((tree)&amp;%0)&quot;))) xscope;
  tree value;
};
</pre></div>

<p>In this example, the value of BINDING_HAS_LEVEL_P when applied to a
<code>struct tree_binding *</code> is presumed to be 0 or 1.  If 1, the type
mechanism will treat the field <code>level</code> as being present and if 0,
will treat the field <code>scope</code> as being present.
</p>
<a name="index-param_005fis"></a>
<a name="index-use_005fparam"></a>
</dd>
<dt><code>param_is (<var>type</var>)</code></dt>
<dt><code>use_param</code></dt>
<dd>
<p>Sometimes it&rsquo;s convenient to define some data structure to work on
generic pointers (that is, <code>PTR</code>) and then use it with a specific
type.  <code>param_is</code> specifies the real type pointed to, and
<code>use_param</code> says where in the generic data structure that type
should be put.
</p>
<p>For instance, to have a <code>htab_t</code> that points to trees, one would
write the definition of <code>htab_t</code> like this:
</p><div class="smallexample">
<pre class="smallexample">typedef struct GTY(()) {
  &hellip;
  void ** GTY ((use_param, &hellip;)) entries;
  &hellip;
} htab_t;
</pre></div>
<p>and then declare variables like this:
</p><div class="smallexample">
<pre class="smallexample">  static htab_t GTY ((param_is (union tree_node))) ict;
</pre></div>

<a name="index-paramn_005fis"></a>
<a name="index-use_005fparamn"></a>
</dd>
<dt><code>param<var>n</var>_is (<var>type</var>)</code></dt>
<dt><code>use_param<var>n</var></code></dt>
<dd>
<p>In more complicated cases, the data structure might need to work on
several different types, which might not necessarily all be pointers.
For this, <code>param1_is</code> through <code>param9_is</code> may be used to
specify the real type of a field identified by <code>use_param1</code> through
<code>use_param9</code>.
</p>
<a name="index-use_005fparams"></a>
</dd>
<dt><code>use_params</code></dt>
<dd>
<p>When a structure contains another structure that is parameterized,
there&rsquo;s no need to do anything special, the inner structure inherits the
parameters of the outer one.  When a structure contains a pointer to a
parameterized structure, the type machinery won&rsquo;t automatically detect
this (it could, it just doesn&rsquo;t yet), so it&rsquo;s necessary to tell it that
the pointed-to structure should use the same parameters as the outer
structure.  This is done by marking the pointer with the
<code>use_params</code> option.
</p>
<a name="index-deletable"></a>
</dd>
<dt><code>deletable</code></dt>
<dd>
<p><code>deletable</code>, when applied to a global variable, indicates that when
garbage collection runs, there&rsquo;s no need to mark anything pointed to
by this variable, it can just be set to <code>NULL</code> instead.  This is used
to keep a list of free structures around for re-use.
</p>
<a name="index-if_005fmarked"></a>
</dd>
<dt><code>if_marked (&quot;<var>expression</var>&quot;)</code></dt>
<dd>
<p>Suppose you want some kinds of object to be unique, and so you put them
in a hash table.  If garbage collection marks the hash table, these
objects will never be freed, even if the last other reference to them
goes away.  GGC has special handling to deal with this: if you use the
<code>if_marked</code> option on a global hash table, GGC will call the
routine whose name is the parameter to the option on each hash table
entry.  If the routine returns nonzero, the hash table entry will
be marked as usual.  If the routine returns zero, the hash table entry
will be deleted.
</p>
<p>The routine <code>ggc_marked_p</code> can be used to determine if an element
has been marked already; in fact, the usual case is to use
<code>if_marked (&quot;ggc_marked_p&quot;)</code>.
</p>
<a name="index-mark_005fhook"></a>
</dd>
<dt><code>mark_hook (&quot;<var>hook-routine-name</var>&quot;)</code></dt>
<dd>
<p>If provided for a structure or union type, the given
<var>hook-routine-name</var> (between double-quotes) is the name of a
routine called when the garbage collector has just marked the data as
reachable. This routine should not change the data, or call any ggc
routine. Its only argument is a pointer to the just marked (const)
structure or union.
</p>
<a name="index-maybe_005fundef"></a>
</dd>
<dt><code>maybe_undef</code></dt>
<dd>
<p>When applied to a field, <code>maybe_undef</code> indicates that it&rsquo;s OK if
the structure that this fields points to is never defined, so long as
this field is always <code>NULL</code>.  This is used to avoid requiring
backends to define certain optional structures.  It doesn&rsquo;t work with
language frontends.
</p>
<a name="index-nested_005fptr"></a>
</dd>
<dt><code>nested_ptr (<var>type</var>, &quot;<var>to expression</var>&quot;, &quot;<var>from expression</var>&quot;)</code></dt>
<dd>
<p>The type machinery expects all pointers to point to the start of an
object.  Sometimes for abstraction purposes it&rsquo;s convenient to have
a pointer which points inside an object.  So long as it&rsquo;s possible to
convert the original object to and from the pointer, such pointers
can still be used.  <var>type</var> is the type of the original object,
the <var>to expression</var> returns the pointer given the original object,
and the <var>from expression</var> returns the original object given
the pointer.  The pointer will be available using the <code>%h</code>
escape.
</p>
<a name="index-chain_005fnext"></a>
<a name="index-chain_005fprev"></a>
<a name="index-chain_005fcircular"></a>
</dd>
<dt><code>chain_next (&quot;<var>expression</var>&quot;)</code></dt>
<dt><code>chain_prev (&quot;<var>expression</var>&quot;)</code></dt>
<dt><code>chain_circular (&quot;<var>expression</var>&quot;)</code></dt>
<dd>
<p>It&rsquo;s helpful for the type machinery to know if objects are often
chained together in long lists; this lets it generate code that uses
less stack space by iterating along the list instead of recursing down
it.  <code>chain_next</code> is an expression for the next item in the list,
<code>chain_prev</code> is an expression for the previous item.  For singly
linked lists, use only <code>chain_next</code>; for doubly linked lists, use
both.  The machinery requires that taking the next item of the
previous item gives the original item.  <code>chain_circular</code> is similar
to <code>chain_next</code>, but can be used for circular single linked lists.
</p>
<a name="index-reorder"></a>
</dd>
<dt><code>reorder (&quot;<var>function name</var>&quot;)</code></dt>
<dd>
<p>Some data structures depend on the relative ordering of pointers.  If
the precompiled header machinery needs to change that ordering, it
will call the function referenced by the <code>reorder</code> option, before
changing the pointers in the object that&rsquo;s pointed to by the field the
option applies to.  The function must take four arguments, with the
signature &lsquo;<samp>void&nbsp;*,&nbsp;void&nbsp;*,&nbsp;<span class="nolinebreak">gt_pointer_operator,</span>&nbsp;void&nbsp;*<!-- /@w --></samp>&rsquo;.
The first parameter is a pointer to the structure that contains the
object being updated, or the object itself if there is no containing
structure.  The second parameter is a cookie that should be ignored.
The third parameter is a routine that, given a pointer, will update it
to its correct new value.  The fourth parameter is a cookie that must
be passed to the second parameter.
</p>
<p>PCH cannot handle data structures that depend on the absolute values
of pointers.  <code>reorder</code> functions can be expensive.  When
possible, it is better to depend on properties of the data, like an ID
number or the hash of a string instead.
</p>
<a name="index-variable_005fsize"></a>
</dd>
<dt><code>variable_size</code></dt>
<dd>
<p>The type machinery expects the types to be of constant size.  When this
is not true, for example, with structs that have array fields or unions,
the type machinery cannot tell how many bytes need to be allocated at
each allocation.  The <code>variable_size</code> is used to mark such types.
The type machinery then provides allocators that take a parameter
indicating an exact size of object being allocated.  Note that the size
must be provided in bytes whereas the <code>length</code> option works with
array lengths in number of elements.
</p>
<p>For example,
</p><div class="smallexample">
<pre class="smallexample">struct GTY((variable_size)) sorted_fields_type {
  int len;
  tree GTY((length (&quot;%h.len&quot;))) elts[1];
};
</pre></div>

<p>Then the objects of <code>struct sorted_fields_type</code> are allocated in GC
memory as follows:
</p><div class="smallexample">
<pre class="smallexample">  field_vec = ggc_alloc_sorted_fields_type (size);
</pre></div>

<p>If <var>field_vec-&gt;elts</var> stores <var>n</var> elements, then <var>size</var>
could be calculated as follows:
</p><div class="smallexample">
<pre class="smallexample">  size_t size = sizeof (struct sorted_fields_type) + n * sizeof (tree);
</pre></div>

<a name="index-atomic"></a>
</dd>
<dt><code>atomic</code></dt>
<dd>
<p>The <code>atomic</code> option can only be used with pointers.  It informs
the GC machinery that the memory that the pointer points to does not
contain any pointers, and hence it should be treated by the GC and PCH
machinery as an &ldquo;atomic&rdquo; block of memory that does not need to be
examined when scanning memory for pointers.  In particular, the
machinery will not scan that memory for pointers to mark them as
reachable (when marking pointers for GC) or to relocate them (when
writing a PCH file).
</p>
<p>The <code>atomic</code> option differs from the <code>skip</code> option.
<code>atomic</code> keeps the memory under Garbage Collection, but makes the
GC ignore the contents of the memory.  <code>skip</code> is more drastic in
that it causes the pointer and the memory to be completely ignored by
the Garbage Collector.  So, memory marked as <code>atomic</code> is
automatically freed when no longer reachable, while memory marked as
<code>skip</code> is not.
</p>
<p>The <code>atomic</code> option must be used with great care, because all
sorts of problem can occur if used incorrectly, that is, if the memory
the pointer points to does actually contain a pointer.
</p>
<p>Here is an example of how to use it:
</p><div class="smallexample">
<pre class="smallexample">struct GTY(()) my_struct {
  int number_of_elements;
  unsigned int * GTY ((atomic)) elements;
};
</pre></div>
<p>In this case, <code>elements</code> is a pointer under GC, and the memory it
points to needs to be allocated using the Garbage Collector, and will
be freed automatically by the Garbage Collector when it is no longer
referenced.  But the memory that the pointer points to is an array of
<code>unsigned int</code> elements, and the GC must not try to scan it to
find pointers to mark or relocate, which is why it is marked with the
<code>atomic</code> option.
</p>
<p>Note that, currently, global variables can not be marked with
<code>atomic</code>; only fields of a struct can.  This is a known
limitation.  It would be useful to be able to mark global pointers
with <code>atomic</code> to make the PCH machinery aware of them so that
they are saved and restored correctly to PCH files.
</p>
<a name="index-special"></a>
</dd>
<dt><code>special (&quot;<var>name</var>&quot;)</code></dt>
<dd>
<p>The <code>special</code> option is used to mark types that have to be dealt
with by special case machinery.  The parameter is the name of the
special case.  See <samp>gengtype.c</samp> for further details.  Avoid
adding new special cases unless there is no other alternative.
</p>
<a name="index-user"></a>
</dd>
<dt><code>user</code></dt>
<dd>
<p>The <code>user</code> option indicates that the code to mark structure
fields is completely handled by user-provided routines.  See section
<a href="#User-GC">User GC</a> for details on what functions need to be provided.
</p></dd>
</dl>

<hr>
<a name="User-GC"></a>
<div class="header">
<p>
Next: <a href="#GGC-Roots" accesskey="n" rel="next">GGC Roots</a>, Previous: <a href="#GTY-Options" accesskey="p" rel="prev">GTY Options</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Support-for-user_002dprovided-GC-marking-routines"></a>
<h3 class="section">22.2 Support for user-provided GC marking routines</h3>
<a name="index-user-gc"></a>
<p>The garbage collector supports types for which no automatic marking
code is generated.  For these types, the user is required to provide
three functions: one to act as a marker for garbage collection, and
two functions to act as marker and pointer walker for pre-compiled
headers.
</p>
<p>Given a structure <code>struct GTY((user)) my_struct</code>, the following functions
should be defined to mark <code>my_struct</code>:
</p>
<div class="smallexample">
<pre class="smallexample">void gt_ggc_mx (my_struct *p)
{
  /* This marks field 'fld'.  */
  gt_ggc_mx (p-&gt;fld);
}

void gt_pch_nx (my_struct *p)
{
  /* This marks field 'fld'.  */
  gt_pch_nx (tp-&gt;fld);
}

void gt_pch_nx (my_struct *p, gt_pointer_operator op, void *cookie)
{
  /* For every field 'fld', call the given pointer operator.  */
  op (&amp;(tp-&gt;fld), cookie);
}
</pre></div>

<p>In general, each marker <code>M</code> should call <code>M</code> for every
pointer field in the structure.  Fields that are not allocated in GC
or are not pointers must be ignored.
</p>
<p>For embedded lists (e.g., structures with a <code>next</code> or <code>prev</code>
pointer), the marker must follow the chain and mark every element in
it.
</p>
<p>Note that the rules for the pointer walker <code>gt_pch_nx (my_struct
*, gt_pointer_operator, void *)</code> are slightly different.  In this
case, the operation <code>op</code> must be applied to the <em>address</em> of
every pointer field.
</p>
<a name="User_002dprovided-marking-routines-for-template-types"></a>
<h4 class="subsection">22.2.1 User-provided marking routines for template types</h4>
<p>When a template type <code>TP</code> is marked with <code>GTY</code>, all
instances of that type are considered user-provided types.  This means
that the individual instances of <code>TP</code> do not need to be marked
with <code>GTY</code>.  The user needs to provide template functions to mark
all the fields of the type.
</p>
<p>The following code snippets represent all the functions that need to
be provided. Note that type <code>TP</code> may reference to more than one
type. In these snippets, there is only one type <code>T</code>, but there
could be more.
</p>
<div class="smallexample">
<pre class="smallexample">template&lt;typename T&gt;
void gt_ggc_mx (TP&lt;T&gt; *tp)
{
  extern void gt_ggc_mx (T&amp;);

  /* This marks field 'fld' of type 'T'.  */
  gt_ggc_mx (tp-&gt;fld);
}

template&lt;typename T&gt;
void gt_pch_nx (TP&lt;T&gt; *tp)
{
  extern void gt_pch_nx (T&amp;);

  /* This marks field 'fld' of type 'T'.  */
  gt_pch_nx (tp-&gt;fld);
}

template&lt;typename T&gt;
void gt_pch_nx (TP&lt;T *&gt; *tp, gt_pointer_operator op, void *cookie)
{
  /* For every field 'fld' of 'tp' with type 'T *', call the given
     pointer operator.  */
  op (&amp;(tp-&gt;fld), cookie);
}

template&lt;typename T&gt;
void gt_pch_nx (TP&lt;T&gt; *tp, gt_pointer_operator, void *cookie)
{
  extern void gt_pch_nx (T *, gt_pointer_operator, void *);

  /* For every field 'fld' of 'tp' with type 'T', call the pointer
     walker for all the fields of T.  */
  gt_pch_nx (&amp;(tp-&gt;fld), op, cookie);
}
</pre></div>

<p>Support for user-defined types is currently limited. The following
restrictions apply:
</p>
<ol>
<li> Type <code>TP</code> and all the argument types <code>T</code> must be
marked with <code>GTY</code>.

</li><li> Type <code>TP</code> can only have type names in its argument list.

</li><li> The pointer walker functions are different for <code>TP&lt;T&gt;</code> and
<code>TP&lt;T *&gt;</code>. In the case of <code>TP&lt;T&gt;</code>, references to
<code>T</code> must be handled by calling <code>gt_pch_nx</code> (which
will, in turn, walk all the pointers inside fields of <code>T</code>).
In the case of <code>TP&lt;T *&gt;</code>, references to <code>T *</code> must be
handled by calling the <code>op</code> function on the address of the
pointer (see the code snippets above).
</li></ol>

<hr>
<a name="GGC-Roots"></a>
<div class="header">
<p>
Next: <a href="#Files" accesskey="n" rel="next">Files</a>, Previous: <a href="#User-GC" accesskey="p" rel="prev">User GC</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Marking-Roots-for-the-Garbage-Collector"></a>
<h3 class="section">22.3 Marking Roots for the Garbage Collector</h3>
<a name="index-roots_002c-marking"></a>
<a name="index-marking-roots"></a>

<p>In addition to keeping track of types, the type machinery also locates
the global variables (<em>roots</em>) that the garbage collector starts
at.  Roots must be declared using one of the following syntaxes:
</p>
<ul>
<li> <code>extern GTY(([<var>options</var>])) <var>type</var> <var>name</var>;</code>
</li><li> <code>static GTY(([<var>options</var>])) <var>type</var> <var>name</var>;</code>
</li></ul>
<p>The syntax
</p><ul>
<li> <code>GTY(([<var>options</var>])) <var>type</var> <var>name</var>;</code>
</li></ul>
<p>is <em>not</em> accepted.  There should be an <code>extern</code> declaration
of such a variable in a header somewhere&mdash;mark that, not the
definition.  Or, if the variable is only used in one file, make it
<code>static</code>.
</p>
<hr>
<a name="Files"></a>
<div class="header">
<p>
Next: <a href="#Invoking-the-garbage-collector" accesskey="n" rel="next">Invoking the garbage collector</a>, Previous: <a href="#GGC-Roots" accesskey="p" rel="prev">GGC Roots</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Source-Files-Containing-Type-Information"></a>
<h3 class="section">22.4 Source Files Containing Type Information</h3>
<a name="index-generated-files"></a>
<a name="index-files_002c-generated"></a>

<p>Whenever you add <code>GTY</code> markers to a source file that previously
had none, or create a new source file containing <code>GTY</code> markers,
there are three things you need to do:
</p>
<ol>
<li> You need to add the file to the list of source files the type
machinery scans.  There are four cases:

<ol>
<li> For a back-end file, this is usually done
automatically; if not, you should add it to <code>target_gtfiles</code> in
the appropriate port&rsquo;s entries in <samp>config.gcc</samp>.

</li><li> For files shared by all front ends, add the filename to the
<code>GTFILES</code> variable in <samp>Makefile.in</samp>.

</li><li> For files that are part of one front end, add the filename to the
<code>gtfiles</code> variable defined in the appropriate
<samp>config-lang.in</samp>.
Headers should appear before non-headers in this list.

</li><li> For files that are part of some but not all front ends, add the
filename to the <code>gtfiles</code> variable of <em>all</em> the front ends
that use it.
</li></ol>

</li><li> If the file was a header file, you&rsquo;ll need to check that it&rsquo;s included
in the right place to be visible to the generated files.  For a back-end
header file, this should be done automatically.  For a front-end header
file, it needs to be included by the same file that includes
<samp>gtype-<var>lang</var>.h</samp>.  For other header files, it needs to be
included in <samp>gtype-desc.c</samp>, which is a generated file, so add it to
<code>ifiles</code> in <code>open_base_file</code> in <samp>gengtype.c</samp>.

<p>For source files that aren&rsquo;t header files, the machinery will generate a
header file that should be included in the source file you just changed.
The file will be called <samp>gt-<var>path</var>.h</samp> where <var>path</var> is the
pathname relative to the <samp>gcc</samp> directory with slashes replaced by
<tt>-</tt>, so for example the header file to be included in
<samp>cp/parser.c</samp> is called <samp>gt-cp-parser.c</samp>.  The
generated header file should be included after everything else in the
source file.  Don&rsquo;t forget to mention this file as a dependency in the
<samp>Makefile</samp>!
</p>
</li></ol>

<p>For language frontends, there is another file that needs to be included
somewhere.  It will be called <samp>gtype-<var>lang</var>.h</samp>, where
<var>lang</var> is the name of the subdirectory the language is contained in.
</p>
<p>Plugins can add additional root tables.  Run the <code>gengtype</code>
utility in plugin mode as <code>gengtype -P pluginout.h <var>source-dir</var>
<var>file-list</var> <var>plugin*.c</var></code> with your plugin files
<var>plugin*.c</var> using <code>GTY</code> to generate the <var>pluginout.h</var> file.
The GCC build tree is needed to be present in that mode.
</p>

<hr>
<a name="Invoking-the-garbage-collector"></a>
<div class="header">
<p>
Next: <a href="#Troubleshooting" accesskey="n" rel="next">Troubleshooting</a>, Previous: <a href="#Files" accesskey="p" rel="prev">Files</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="How-to-invoke-the-garbage-collector"></a>
<h3 class="section">22.5 How to invoke the garbage collector</h3>
<a name="index-garbage-collector_002c-invocation"></a>
<a name="index-ggc_005fcollect"></a>

<p>The GCC garbage collector GGC is only invoked explicitly. In contrast
with many other garbage collectors, it is not implicitly invoked by
allocation routines when a lot of memory has been consumed. So the
only way to have GGC reclaim storage is to call the <code>ggc_collect</code>
function explicitly.  This call is an expensive operation, as it may
have to scan the entire heap.  Beware that local variables (on the GCC
call stack) are not followed by such an invocation (as many other
garbage collectors do): you should reference all your data from static
or external <code>GTY</code>-ed variables, and it is advised to call
<code>ggc_collect</code> with a shallow call stack.  The GGC is an exact mark
and sweep garbage collector (so it does not scan the call stack for
pointers).  In practice GCC passes don&rsquo;t often call <code>ggc_collect</code>
themselves, because it is called by the pass manager between passes.
</p>
<p>At the time of the <code>ggc_collect</code> call all pointers in the GC-marked
structures must be valid or <code>NULL</code>.  In practice this means that
there should not be uninitialized pointer fields in the structures even
if your code never reads or writes those fields at a particular
instance.  One way to ensure this is to use cleared versions of
allocators unless all the fields are initialized manually immediately
after allocation.
</p>
<hr>
<a name="Troubleshooting"></a>
<div class="header">
<p>
Previous: <a href="#Invoking-the-garbage-collector" accesskey="p" rel="prev">Invoking the garbage collector</a>, Up: <a href="#Type-Information" accesskey="u" rel="up">Type Information</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Troubleshooting-the-garbage-collector"></a>
<h3 class="section">22.6 Troubleshooting the garbage collector</h3>
<a name="index-garbage-collector_002c-troubleshooting"></a>

<p>With the current garbage collector implementation, most issues should
show up as GCC compilation errors.  Some of the most commonly
encountered issues are described below.
</p>
<ul>
<li> Gengtype does not produce allocators for a <code>GTY</code>-marked type.
Gengtype checks if there is at least one possible path from GC roots to
at least one instance of each type before outputting allocators.  If
there is no such path, the <code>GTY</code> markers will be ignored and no
allocators will be output.  Solve this by making sure that there exists
at least one such path.  If creating it is unfeasible or raises a &ldquo;code
smell&rdquo;, consider if you really must use GC for allocating such type.

</li><li> Link-time errors about undefined <code>gt_ggc_r_foo_bar</code> and
similarly-named symbols.  Check if your <samp>foo_bar</samp> source file has
<code>#include &quot;gt-foo_bar.h&quot;</code> as its very last line.

</li></ul>

<hr>
<a name="Plugins"></a>
<div class="header">
<p>
Next: <a href="#LTO" accesskey="n" rel="next">LTO</a>, Previous: <a href="#Type-Information" accesskey="p" rel="prev">Type Information</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Plugins-1"></a>
<h2 class="chapter">23 Plugins</h2>
<a name="index-Plugins"></a>

<p>GCC plugin is a loadable module that provides extra
features to the compiler, which they can further pass
around as a shareable module.
</p>
<p>GCC plugins provide developers with a rich subset of
the GCC API to allow them to extend GCC as they see fit.
Whether it is writing an additional optimization pass,
transforming code, or analyzing information, plugins
can be quite useful.
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#Plugins-loading" accesskey="1">Plugins loading</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How can we load plugins.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugin-API" accesskey="2">Plugin API</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">The APIs for plugins.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-pass" accesskey="3">Plugins pass</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How a plugin interact with the pass manager.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-GC" accesskey="4">Plugins GC</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How a plugin Interact with GCC Garbage Collector.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-description" accesskey="5">Plugins description</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Giving information about a plugin itself.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-attr" accesskey="6">Plugins attr</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Registering custom attributes or pragmas.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-recording" accesskey="7">Plugins recording</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Recording information about pass execution.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-gate" accesskey="8">Plugins gate</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Controlling which passes are being run.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-tracking" accesskey="9">Plugins tracking</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Keeping track of available passes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Plugins-building">Plugins building</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">How can we build a plugin.
</td></tr>
</table>

<hr>
<a name="Plugins-loading"></a>
<div class="header">
<p>
Next: <a href="#Plugin-API" accesskey="n" rel="next">Plugin API</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Loading-Plugins"></a>
<h3 class="section">23.1 Loading Plugins</h3>

<p>Plugins are supported on platforms that support <samp>-ldl
-rdynamic</samp>.  They are loaded by the compiler using <code>dlopen</code>
and invoked at pre-determined locations in the compilation
process.
</p>
<p>Plugins are loaded with
</p>
<p><samp>-fplugin=/path/to/<var>name</var>.so</samp> <samp>-fplugin-arg-<var>name</var>-<var>key1</var>[=<var>value1</var>]</samp>
</p>
<p>The plugin arguments are parsed by GCC and passed to respective
plugins as key-value pairs. Multiple plugins can be invoked by
specifying multiple <samp>-fplugin</samp> arguments.
</p>
<p>A plugin can be simply given by its short name (no dots or
slashes). When simply passing <samp>-fplugin=<var>name</var></samp>, the plugin is
loaded from the <samp>plugin</samp> directory, so <samp>-fplugin=<var>name</var></samp> is
the same as <samp>-fplugin=`gcc -print-file-name=plugin`/<var>name</var>.so</samp>,
using backquote shell syntax to query the <samp>plugin</samp> directory.
</p>
<hr>
<a name="Plugin-API"></a>
<div class="header">
<p>
Next: <a href="#Plugins-pass" accesskey="n" rel="next">Plugins pass</a>, Previous: <a href="#Plugins-loading" accesskey="p" rel="prev">Plugins loading</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Plugin-API-1"></a>
<h3 class="section">23.2 Plugin API</h3>

<p>Plugins are activated by the compiler at specific events as defined in
<samp>gcc-plugin.h</samp>.  For each event of interest, the plugin should
call <code>register_callback</code> specifying the name of the event and
address of the callback function that will handle that event.
</p>
<p>The header <samp>gcc-plugin.h</samp> must be the first gcc header to be included.
</p>
<a name="Plugin-license-check"></a>
<h4 class="subsection">23.2.1 Plugin license check</h4>

<p>Every plugin should define the global symbol <code>plugin_is_GPL_compatible</code>
to assert that it has been licensed under a GPL-compatible license.
If this symbol does not exist, the compiler will emit a fatal error
and exit with the error message:
</p>
<div class="smallexample">
<pre class="smallexample">fatal error: plugin <var>name</var> is not licensed under a GPL-compatible license
<var>name</var>: undefined symbol: plugin_is_GPL_compatible
compilation terminated
</pre></div>

<p>The declared type of the symbol should be int, to match a forward declaration
in <samp>gcc-plugin.h</samp> that suppresses C++ mangling.  It does not need to be in
any allocated section, though.  The compiler merely asserts that
the symbol exists in the global scope.  Something like this is enough:
</p>
<div class="smallexample">
<pre class="smallexample">int plugin_is_GPL_compatible;
</pre></div>

<a name="Plugin-initialization"></a>
<h4 class="subsection">23.2.2 Plugin initialization</h4>

<p>Every plugin should export a function called <code>plugin_init</code> that
is called right after the plugin is loaded. This function is
responsible for registering all the callbacks required by the plugin
and do any other required initialization.
</p>
<p>This function is called from <code>compile_file</code> right before invoking
the parser.  The arguments to <code>plugin_init</code> are:
</p>
<ul>
<li> <code>plugin_info</code>: Plugin invocation information.
</li><li> <code>version</code>: GCC version.
</li></ul>

<p>The <code>plugin_info</code> struct is defined as follows:
</p>
<div class="smallexample">
<pre class="smallexample">struct plugin_name_args
{
  char *base_name;              /* Short name of the plugin
                                   (filename without .so suffix). */
  const char *full_name;        /* Path to the plugin as specified with
                                   -fplugin=. */
  int argc;                     /* Number of arguments specified with
                                   -fplugin-arg-.... */
  struct plugin_argument *argv; /* Array of ARGC key-value pairs. */
  const char *version;          /* Version string provided by plugin. */
  const char *help;             /* Help string provided by plugin. */
}
</pre></div>

<p>If initialization fails, <code>plugin_init</code> must return a non-zero
value.  Otherwise, it should return 0.
</p>
<p>The version of the GCC compiler loading the plugin is described by the
following structure:
</p>
<div class="smallexample">
<pre class="smallexample">struct plugin_gcc_version
{
  const char *basever;
  const char *datestamp;
  const char *devphase;
  const char *revision;
  const char *configuration_arguments;
};
</pre></div>

<p>The function <code>plugin_default_version_check</code> takes two pointers to
such structure and compare them field by field. It can be used by the
plugin&rsquo;s <code>plugin_init</code> function.
</p>
<p>The version of GCC used to compile the plugin can be found in the symbol
<code>gcc_version</code> defined in the header <samp>plugin-version.h</samp>. The
recommended version check to perform looks like
</p>
<div class="smallexample">
<pre class="smallexample">#include &quot;plugin-version.h&quot;
...

int
plugin_init (struct plugin_name_args *plugin_info,
             struct plugin_gcc_version *version)
{
  if (!plugin_default_version_check (version, &amp;gcc_version))
    return 1;

}
</pre></div>

<p>but you can also check the individual fields if you want a less strict check.
</p>
<a name="Plugin-callbacks"></a>
<h4 class="subsection">23.2.3 Plugin callbacks</h4>

<p>Callback functions have the following prototype:
</p>
<div class="smallexample">
<pre class="smallexample">/* The prototype for a plugin callback function.
     gcc_data  - event-specific data provided by GCC
     user_data - plugin-specific data provided by the plug-in.  */
typedef void (*plugin_callback_func)(void *gcc_data, void *user_data);
</pre></div>

<p>Callbacks can be invoked at the following pre-determined events:
</p>

<div class="smallexample">
<pre class="smallexample">enum plugin_event
{
  PLUGIN_PASS_MANAGER_SETUP,    /* To hook into pass manager.  */
  PLUGIN_FINISH_TYPE,           /* After finishing parsing a type.  */
  PLUGIN_FINISH_DECL,           /* After finishing parsing a declaration. */
  PLUGIN_FINISH_UNIT,           /* Useful for summary processing.  */
  PLUGIN_PRE_GENERICIZE,        /* Allows to see low level AST in C and C++ frontends.  */
  PLUGIN_FINISH,                /* Called before GCC exits.  */
  PLUGIN_INFO,                  /* Information about the plugin. */
  PLUGIN_GGC_START,             /* Called at start of GCC Garbage Collection. */
  PLUGIN_GGC_MARKING,           /* Extend the GGC marking. */
  PLUGIN_GGC_END,               /* Called at end of GGC. */
  PLUGIN_REGISTER_GGC_ROOTS,    /* Register an extra GGC root table. */
  PLUGIN_REGISTER_GGC_CACHES,   /* Register an extra GGC cache table. */
  PLUGIN_ATTRIBUTES,            /* Called during attribute registration */
  PLUGIN_START_UNIT,            /* Called before processing a translation unit.  */
  PLUGIN_PRAGMAS,               /* Called during pragma registration. */
  /* Called before first pass from all_passes.  */
  PLUGIN_ALL_PASSES_START,
  /* Called after last pass from all_passes.  */
  PLUGIN_ALL_PASSES_END,
  /* Called before first ipa pass.  */
  PLUGIN_ALL_IPA_PASSES_START,
  /* Called after last ipa pass.  */
  PLUGIN_ALL_IPA_PASSES_END,
  /* Allows to override pass gate decision for current_pass.  */
  PLUGIN_OVERRIDE_GATE,
  /* Called before executing a pass.  */
  PLUGIN_PASS_EXECUTION,
  /* Called before executing subpasses of a GIMPLE_PASS in
     execute_ipa_pass_list.  */
  PLUGIN_EARLY_GIMPLE_PASSES_START,
  /* Called after executing subpasses of a GIMPLE_PASS in
     execute_ipa_pass_list.  */
  PLUGIN_EARLY_GIMPLE_PASSES_END,
  /* Called when a pass is first instantiated.  */
  PLUGIN_NEW_PASS,

  PLUGIN_EVENT_FIRST_DYNAMIC    /* Dummy event used for indexing callback
                                   array.  */
};
</pre></div>

<p>In addition, plugins can also look up the enumerator of a named event,
and / or generate new events dynamically, by calling the function
<code>get_named_event_id</code>.
</p>
<p>To register a callback, the plugin calls <code>register_callback</code> with
the arguments:
</p>
<ul>
<li> <code>char *name</code>: Plugin name.
</li><li> <code>int event</code>: The event code.
</li><li> <code>plugin_callback_func callback</code>: The function that handles <code>event</code>.
</li><li> <code>void *user_data</code>: Pointer to plugin-specific data.
</li></ul>

<p>For the PLUGIN_PASS_MANAGER_SETUP, PLUGIN_INFO, PLUGIN_REGISTER_GGC_ROOTS
and PLUGIN_REGISTER_GGC_CACHES pseudo-events the <code>callback</code> should be
null, and the <code>user_data</code> is specific.
</p>
<p>When the PLUGIN_PRAGMAS event is triggered (with a null
pointer as data from GCC), plugins may register their own pragmas
using functions like <code>c_register_pragma</code> or
<code>c_register_pragma_with_expansion</code>.
</p>
<hr>
<a name="Plugins-pass"></a>
<div class="header">
<p>
Next: <a href="#Plugins-GC" accesskey="n" rel="next">Plugins GC</a>, Previous: <a href="#Plugin-API" accesskey="p" rel="prev">Plugin API</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Interacting-with-the-pass-manager"></a>
<h3 class="section">23.3 Interacting with the pass manager</h3>

<p>There needs to be a way to add/reorder/remove passes dynamically. This
is useful for both analysis plugins (plugging in after a certain pass
such as CFG or an IPA pass) and optimization plugins.
</p>
<p>Basic support for inserting new passes or replacing existing passes is
provided. A plugin registers a new pass with GCC by calling
<code>register_callback</code> with the <code>PLUGIN_PASS_MANAGER_SETUP</code>
event and a pointer to a <code>struct register_pass_info</code> object defined as follows
</p>
<div class="smallexample">
<pre class="smallexample">enum pass_positioning_ops
{
  PASS_POS_INSERT_AFTER,  // Insert after the reference pass.
  PASS_POS_INSERT_BEFORE, // Insert before the reference pass.
  PASS_POS_REPLACE        // Replace the reference pass.
};

struct register_pass_info
{
  struct opt_pass *pass;            /* New pass provided by the plugin.  */
  const char *reference_pass_name;  /* Name of the reference pass for hooking
                                       up the new pass.  */
  int ref_pass_instance_number;     /* Insert the pass at the specified
                                       instance number of the reference pass.  */
                                    /* Do it for every instance if it is 0.  */
  enum pass_positioning_ops pos_op; /* how to insert the new pass.  */
};


/* Sample plugin code that registers a new pass.  */
int
plugin_init (struct plugin_name_args *plugin_info,
             struct plugin_gcc_version *version)
{
  struct register_pass_info pass_info;

  ...

  /* Code to fill in the pass_info object with new pass information.  */

  ...

  /* Register the new pass.  */
  register_callback (plugin_info-&gt;base_name, PLUGIN_PASS_MANAGER_SETUP, NULL, &amp;pass_info);

  ...
}
</pre></div>


<hr>
<a name="Plugins-GC"></a>
<div class="header">
<p>
Next: <a href="#Plugins-description" accesskey="n" rel="next">Plugins description</a>, Previous: <a href="#Plugins-pass" accesskey="p" rel="prev">Plugins pass</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Interacting-with-the-GCC-Garbage-Collector"></a>
<h3 class="section">23.4 Interacting with the GCC Garbage Collector</h3>

<p>Some plugins may want to be informed when GGC (the GCC Garbage
Collector) is running. They can register callbacks for the
<code>PLUGIN_GGC_START</code> and <code>PLUGIN_GGC_END</code> events (for which
the callback is called with a null <code>gcc_data</code>) to be notified of
the start or end of the GCC garbage collection.
</p>
<p>Some plugins may need to have GGC mark additional data. This can be
done by registering a callback (called with a null <code>gcc_data</code>)
for the <code>PLUGIN_GGC_MARKING</code> event. Such callbacks can call the
<code>ggc_set_mark</code> routine, preferably through the <code>ggc_mark</code> macro
(and conversely, these routines should usually not be used in plugins
outside of the <code>PLUGIN_GGC_MARKING</code> event).
</p>
<p>Some plugins may need to add extra GGC root tables, e.g. to handle their own
<code>GTY</code>-ed data. This can be done with the <code>PLUGIN_REGISTER_GGC_ROOTS</code>
pseudo-event with a null callback and the extra root table (of type <code>struct
ggc_root_tab*</code>) as <code>user_data</code>.  Plugins that want to use the
<code>if_marked</code> hash table option can add the extra GGC cache tables generated
by <code>gengtype</code> using the <code>PLUGIN_REGISTER_GGC_CACHES</code> pseudo-event with
a null callback and the extra cache table (of type <code>struct ggc_cache_tab*</code>)
as <code>user_data</code>.  Running the <code>gengtype -p <var>source-dir</var>
<var>file-list</var> <var>plugin*.c</var> ...</code> utility generates these extra root tables.
</p>
<p>You should understand the details of memory management inside GCC
before using <code>PLUGIN_GGC_MARKING</code>, <code>PLUGIN_REGISTER_GGC_ROOTS</code>
or <code>PLUGIN_REGISTER_GGC_CACHES</code>.
</p>

<hr>
<a name="Plugins-description"></a>
<div class="header">
<p>
Next: <a href="#Plugins-attr" accesskey="n" rel="next">Plugins attr</a>, Previous: <a href="#Plugins-GC" accesskey="p" rel="prev">Plugins GC</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Giving-information-about-a-plugin"></a>
<h3 class="section">23.5 Giving information about a plugin</h3>

<p>A plugin should give some information to the user about itself. This
uses the following structure:
</p>
<div class="smallexample">
<pre class="smallexample">struct plugin_info
{
  const char *version;
  const char *help;
};
</pre></div>

<p>Such a structure is passed as the <code>user_data</code> by the plugin&rsquo;s
init routine using <code>register_callback</code> with the
<code>PLUGIN_INFO</code> pseudo-event and a null callback.
</p>
<hr>
<a name="Plugins-attr"></a>
<div class="header">
<p>
Next: <a href="#Plugins-recording" accesskey="n" rel="next">Plugins recording</a>, Previous: <a href="#Plugins-description" accesskey="p" rel="prev">Plugins description</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Registering-custom-attributes-or-pragmas"></a>
<h3 class="section">23.6 Registering custom attributes or pragmas</h3>

<p>For analysis (or other) purposes it is useful to be able to add custom
attributes or pragmas.
</p>
<p>The <code>PLUGIN_ATTRIBUTES</code> callback is called during attribute
registration. Use the <code>register_attribute</code> function to register
custom attributes.
</p>
<div class="smallexample">
<pre class="smallexample">/* Attribute handler callback */
static tree
handle_user_attribute (tree *node, tree name, tree args,
                       int flags, bool *no_add_attrs)
{
  return NULL_TREE;
}

/* Attribute definition */
static struct attribute_spec user_attr =
  { &quot;user&quot;, 1, 1, false,  false, false, handle_user_attribute, false };

/* Plugin callback called during attribute registration.
Registered with register_callback (plugin_name, PLUGIN_ATTRIBUTES, register_attributes, NULL)
*/
static void
register_attributes (void *event_data, void *data)
{
  warning (0, G_(&quot;Callback to register attributes&quot;));
  register_attribute (&amp;user_attr);
}

</pre></div>


<p>The <code>PLUGIN_PRAGMAS</code> callback is called during pragmas
registration. Use the <code>c_register_pragma</code> or
<code>c_register_pragma_with_expansion</code> functions to register custom
pragmas.
</p>
<div class="smallexample">
<pre class="smallexample">/* Plugin callback called during pragmas registration. Registered with
     register_callback (plugin_name, PLUGIN_PRAGMAS,
                        register_my_pragma, NULL);
*/
static void
register_my_pragma (void *event_data, void *data)
{
  warning (0, G_(&quot;Callback to register pragmas&quot;));
  c_register_pragma (&quot;GCCPLUGIN&quot;, &quot;sayhello&quot;, handle_pragma_sayhello);
}
</pre></div>

<p>It is suggested to pass <code>&quot;GCCPLUGIN&quot;</code> (or a short name identifying
your plugin) as the &ldquo;space&rdquo; argument of your pragma.
</p>

<hr>
<a name="Plugins-recording"></a>
<div class="header">
<p>
Next: <a href="#Plugins-gate" accesskey="n" rel="next">Plugins gate</a>, Previous: <a href="#Plugins-attr" accesskey="p" rel="prev">Plugins attr</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Recording-information-about-pass-execution"></a>
<h3 class="section">23.7 Recording information about pass execution</h3>

<p>The event PLUGIN_PASS_EXECUTION passes the pointer to the executed pass
(the same as current_pass) as <code>gcc_data</code> to the callback.  You can also
inspect cfun to find out about which function this pass is executed for.
Note that this event will only be invoked if the gate check (if
applicable, modified by PLUGIN_OVERRIDE_GATE) succeeds.
You can use other hooks, like <code>PLUGIN_ALL_PASSES_START</code>,
<code>PLUGIN_ALL_PASSES_END</code>, <code>PLUGIN_ALL_IPA_PASSES_START</code>,
<code>PLUGIN_ALL_IPA_PASSES_END</code>, <code>PLUGIN_EARLY_GIMPLE_PASSES_START</code>,
and/or <code>PLUGIN_EARLY_GIMPLE_PASSES_END</code> to manipulate global state
in your plugin(s) in order to get context for the pass execution.
</p>

<hr>
<a name="Plugins-gate"></a>
<div class="header">
<p>
Next: <a href="#Plugins-tracking" accesskey="n" rel="next">Plugins tracking</a>, Previous: <a href="#Plugins-recording" accesskey="p" rel="prev">Plugins recording</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Controlling-which-passes-are-being-run"></a>
<h3 class="section">23.8 Controlling which passes are being run</h3>

<p>After the original gate function for a pass is called, its result
- the gate status - is stored as an integer.
Then the event <code>PLUGIN_OVERRIDE_GATE</code> is invoked, with a pointer
to the gate status in the <code>gcc_data</code> parameter to the callback function.
A nonzero value of the gate status means that the pass is to be executed.
You can both read and write the gate status via the passed pointer.
</p>

<hr>
<a name="Plugins-tracking"></a>
<div class="header">
<p>
Next: <a href="#Plugins-building" accesskey="n" rel="next">Plugins building</a>, Previous: <a href="#Plugins-gate" accesskey="p" rel="prev">Plugins gate</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Keeping-track-of-available-passes"></a>
<h3 class="section">23.9 Keeping track of available passes</h3>

<p>When your plugin is loaded, you can inspect the various
pass lists to determine what passes are available.  However, other
plugins might add new passes.  Also, future changes to GCC might cause
generic passes to be added after plugin loading.
When a pass is first added to one of the pass lists, the event
<code>PLUGIN_NEW_PASS</code> is invoked, with the callback parameter
<code>gcc_data</code> pointing to the new pass.
</p>

<hr>
<a name="Plugins-building"></a>
<div class="header">
<p>
Previous: <a href="#Plugins-tracking" accesskey="p" rel="prev">Plugins tracking</a>, Up: <a href="#Plugins" accesskey="u" rel="up">Plugins</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Building-GCC-plugins"></a>
<h3 class="section">23.10 Building GCC plugins</h3>

<p>If plugins are enabled, GCC installs the headers needed to build a
plugin (somewhere in the installation tree, e.g. under
<samp>/usr/local</samp>).  In particular a <samp>plugin/include</samp> directory
is installed, containing all the header files needed to build plugins.
</p>
<p>On most systems, you can query this <code>plugin</code> directory by
invoking <code>gcc -print-file-name=plugin</code> (replace if needed
<code>gcc</code> with the appropriate program path).
</p>
<p>Inside plugins, this <code>plugin</code> directory name can be queried by
calling <code>default_plugin_dir_name ()</code>.
</p>
<p>Plugins may know, when they are compiled, the GCC version for which
<samp>plugin-version.h</samp> is provided.  The constant macros
<code>GCCPLUGIN_VERSION_MAJOR</code>, <code>GCCPLUGIN_VERSION_MINOR</code>,
<code>GCCPLUGIN_VERSION_PATCHLEVEL</code>, <code>GCCPLUGIN_VERSION</code> are
integer numbers, so a plugin could ensure it is built for GCC 4.7 with 
</p><div class="smallexample">
<pre class="smallexample">#if GCCPLUGIN_VERSION != 4007
#error this GCC plugin is for GCC 4.7
#endif
</pre></div>

<p>The following GNU Makefile excerpt shows how to build a simple plugin:
</p>
<div class="smallexample">
<pre class="smallexample">GCC=gcc
PLUGIN_SOURCE_FILES= plugin1.c plugin2.c
PLUGIN_OBJECT_FILES= $(patsubst %.c,%.o,$(PLUGIN_SOURCE_FILES))
GCCPLUGINS_DIR:= $(shell $(GCC) -print-file-name=plugin)
CFLAGS+= -I$(GCCPLUGINS_DIR)/include -fPIC -O2

plugin.so: $(PLUGIN_OBJECT_FILES)
   $(GCC) -shared $^ -o $@
</pre></div>

<p>A single source file plugin may be built with <code>gcc -I`gcc
-print-file-name=plugin`/include -fPIC -shared -O2 plugin.c -o
plugin.so</code>, using backquote shell syntax to query the <samp>plugin</samp>
directory.
</p>
<p>When a plugin needs to use <code>gengtype</code>, be sure that both
<samp>gengtype</samp> and <samp>gtype.state</samp> have the same version as the
GCC for which the plugin is built.
</p>
<hr>
<a name="LTO"></a>
<div class="header">
<p>
Next: <a href="#Funding" accesskey="n" rel="next">Funding</a>, Previous: <a href="#Plugins" accesskey="p" rel="prev">Plugins</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Link-Time-Optimization"></a>
<h2 class="chapter">24 Link Time Optimization</h2>
<a name="index-lto"></a>
<a name="index-whopr"></a>
<a name="index-wpa"></a>
<a name="index-ltrans"></a>

<p>Link Time Optimization (LTO) gives GCC the capability of
dumping its internal representation (GIMPLE) to disk,
so that all the different compilation units that make up
a single executable can be optimized as a single module.
This expands the scope of inter-procedural optimizations
to encompass the whole program (or, rather, everything
that is visible at link time).
</p>
<table class="menu" border="0" cellspacing="0">
<tr><td align="left" valign="top">&bull; <a href="#LTO-Overview" accesskey="1">LTO Overview</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Overview of LTO.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#LTO-object-file-layout" accesskey="2">LTO object file layout</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">LTO file sections in ELF.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#IPA" accesskey="3">IPA</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Using summary information in IPA passes.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#WHOPR" accesskey="4">WHOPR</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Whole program assumptions,
                            linker plugin and symbol visibilities.
</td></tr>
<tr><td align="left" valign="top">&bull; <a href="#Internal-flags" accesskey="5">Internal flags</a>:</td><td>&nbsp;&nbsp;</td><td align="left" valign="top">Internal flags controlling <code>lto1</code>.
</td></tr>
</table>

<hr>
<a name="LTO-Overview"></a>
<div class="header">
<p>
Next: <a href="#LTO-object-file-layout" accesskey="n" rel="next">LTO object file layout</a>, Up: <a href="#LTO" accesskey="u" rel="up">LTO</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Design-Overview"></a>
<h3 class="section">24.1 Design Overview</h3>

<p>Link time optimization is implemented as a GCC front end for a
bytecode representation of GIMPLE that is emitted in special sections
of <code>.o</code> files.  Currently, LTO support is enabled in most
ELF-based systems, as well as darwin, cygwin and mingw systems.
</p>
<p>Since GIMPLE bytecode is saved alongside final object code, object
files generated with LTO support are larger than regular object files.
This &ldquo;fat&rdquo; object format makes it easy to integrate LTO into
existing build systems, as one can, for instance, produce archives of
the files.  Additionally, one might be able to ship one set of fat
objects which could be used both for development and the production of
optimized builds.  A, perhaps surprising, side effect of this feature
is that any mistake in the toolchain that leads to LTO information not
being used (e.g. an older <code>libtool</code> calling <code>ld</code> directly).
This is both an advantage, as the system is more robust, and a
disadvantage, as the user is not informed that the optimization has
been disabled.
</p>
<p>The current implementation only produces &ldquo;fat&rdquo; objects, effectively
doubling compilation time and increasing file sizes up to 5x the
original size.  This hides the problem that some tools, such as
<code>ar</code> and <code>nm</code>, need to understand symbol tables of LTO
sections.  These tools were extended to use the plugin infrastructure,
and with these problems solved, GCC will also support &ldquo;slim&rdquo; objects
consisting of the intermediate code alone.
</p>
<p>At the highest level, LTO splits the compiler in two.  The first half
(the &ldquo;writer&rdquo;) produces a streaming representation of all the
internal data structures needed to optimize and generate code.  This
includes declarations, types, the callgraph and the GIMPLE representation
of function bodies.
</p>
<p>When <samp>-flto</samp> is given during compilation of a source file, the
pass manager executes all the passes in <code>all_lto_gen_passes</code>.
Currently, this phase is composed of two IPA passes:
</p>
<ul>
<li> <code>pass_ipa_lto_gimple_out</code>
This pass executes the function <code>lto_output</code> in
<samp>lto-streamer-out.c</samp>, which traverses the call graph encoding
every reachable declaration, type and function.  This generates a
memory representation of all the file sections described below.

</li><li> <code>pass_ipa_lto_finish_out</code>
This pass executes the function <code>produce_asm_for_decls</code> in
<samp>lto-streamer-out.c</samp>, which takes the memory image built in the
previous pass and encodes it in the corresponding ELF file sections.
</li></ul>

<p>The second half of LTO support is the &ldquo;reader&rdquo;.  This is implemented
as the GCC front end <samp>lto1</samp> in <samp>lto/lto.c</samp>.  When
<samp>collect2</samp> detects a link set of <code>.o</code>/<code>.a</code> files with
LTO information and the <samp>-flto</samp> is enabled, it invokes
<samp>lto1</samp> which reads the set of files and aggregates them into a
single translation unit for optimization.  The main entry point for
the reader is <samp>lto/lto.c</samp>:<code>lto_main</code>.
</p>
<a name="LTO-modes-of-operation"></a>
<h4 class="subsection">24.1.1 LTO modes of operation</h4>

<p>One of the main goals of the GCC link-time infrastructure was to allow
effective compilation of large programs.  For this reason GCC implements two
link-time compilation modes.
</p>
<ol>
<li> <em>LTO mode</em>, in which the whole program is read into the
compiler at link-time and optimized in a similar way as if it
were a single source-level compilation unit.

</li><li> <em>WHOPR or partitioned mode</em>, designed to utilize multiple
CPUs and/or a distributed compilation environment to quickly link
large applications.  WHOPR stands for WHOle Program optimizeR (not to
be confused with the semantics of <samp>-fwhole-program</samp>).  It
partitions the aggregated callgraph from many different <code>.o</code>
files and distributes the compilation of the sub-graphs to different
CPUs.

<p>Note that distributed compilation is not implemented yet, but since
the parallelism is facilitated via generating a <code>Makefile</code>, it
would be easy to implement.
</p></li></ol>

<p>WHOPR splits LTO into three main stages:
</p><ol>
<li> Local generation (LGEN)
This stage executes in parallel.  Every file in the program is compiled
into the intermediate language and packaged together with the local
call-graph and summary information.  This stage is the same for both
the LTO and WHOPR compilation mode.

</li><li> Whole Program Analysis (WPA)
WPA is performed sequentially.  The global call-graph is generated, and
a global analysis procedure makes transformation decisions.  The global
call-graph is partitioned to facilitate parallel optimization during
phase 3.  The results of the WPA stage are stored into new object files
which contain the partitions of program expressed in the intermediate
language and the optimization decisions.

</li><li> Local transformations (LTRANS)
This stage executes in parallel.  All the decisions made during phase 2
are implemented locally in each partitioned object file, and the final
object code is generated.  Optimizations which cannot be decided
efficiently during the phase 2 may be performed on the local
call-graph partitions.
</li></ol>

<p>WHOPR can be seen as an extension of the usual LTO mode of
compilation.  In LTO, WPA and LTRANS are executed within a single
execution of the compiler, after the whole program has been read into
memory.
</p>
<p>When compiling in WHOPR mode, the callgraph is partitioned during
the WPA stage.  The whole program is split into a given number of
partitions of roughly the same size.  The compiler tries to
minimize the number of references which cross partition boundaries.
The main advantage of WHOPR is to allow the parallel execution of
LTRANS stages, which are the most time-consuming part of the
compilation process.  Additionally, it avoids the need to load the
whole program into memory.
</p>

<hr>
<a name="LTO-object-file-layout"></a>
<div class="header">
<p>
Next: <a href="#IPA" accesskey="n" rel="next">IPA</a>, Previous: <a href="#LTO-Overview" accesskey="p" rel="prev">LTO Overview</a>, Up: <a href="#LTO" accesskey="u" rel="up">LTO</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="LTO-file-sections"></a>
<h3 class="section">24.2 LTO file sections</h3>

<p>LTO information is stored in several ELF sections inside object files.
Data structures and enum codes for sections are defined in
<samp>lto-streamer.h</samp>.
</p>
<p>These sections are emitted from <samp>lto-streamer-out.c</samp> and mapped
in all at once from <samp>lto/lto.c</samp>:<code>lto_file_read</code>.  The
individual functions dealing with the reading/writing of each section
are described below.
</p>
<ul>
<li> Command line options (<code>.gnu.lto_.opts</code>)

<p>This section contains the command line options used to generate the
object files.  This is used at link time to determine the optimization
level and other settings when they are not explicitly specified at the
linker command line.
</p>
<p>Currently, GCC does not support combining LTO object files compiled
with different set of the command line options into a single binary.
At link time, the options given on the command line and the options
saved on all the files in a link-time set are applied globally.  No
attempt is made at validating the combination of flags (other than the
usual validation done by option processing).  This is implemented in
<samp>lto/lto.c</samp>:<code>lto_read_all_file_options</code>.
</p>

</li><li> Symbol table (<code>.gnu.lto_.symtab</code>)

<p>This table replaces the ELF symbol table for functions and variables
represented in the LTO IL.  Symbols used and exported by the optimized
assembly code of &ldquo;fat&rdquo; objects might not match the ones used and
exported by the intermediate code.  This table is necessary because
the intermediate code is less optimized and thus requires a separate
symbol table.
</p>
<p>Additionally, the binary code in the &ldquo;fat&rdquo; object will lack a call
to a function, since the call was optimized out at compilation time
after the intermediate language was streamed out.  In some special
cases, the same optimization may not happen during link-time
optimization.  This would lead to an undefined symbol if only one
symbol table was used.
</p>
<p>The symbol table is emitted in
<samp>lto-streamer-out.c</samp>:<code>produce_symtab</code>.
</p>

</li><li> Global declarations and types (<code>.gnu.lto_.decls</code>)

<p>This section contains an intermediate language dump of all
declarations and types required to represent the callgraph, static
variables and top-level debug info.
</p>
<p>The contents of this section are emitted in
<samp>lto-streamer-out.c</samp>:<code>produce_asm_for_decls</code>.  Types and
symbols are emitted in a topological order that preserves the sharing
of pointers when the file is read back in
(<samp>lto.c</samp>:<code>read_cgraph_and_symbols</code>).
</p>

</li><li> The callgraph (<code>.gnu.lto_.cgraph</code>)

<p>This section contains the basic data structure used by the GCC
inter-procedural optimization infrastructure.  This section stores an
annotated multi-graph which represents the functions and call sites as
well as the variables, aliases and top-level <code>asm</code> statements.
</p>
<p>This section is emitted in
<samp>lto-streamer-out.c</samp>:<code>output_cgraph</code> and read in
<samp>lto-cgraph.c</samp>:<code>input_cgraph</code>.
</p>

</li><li> IPA references (<code>.gnu.lto_.refs</code>)

<p>This section contains references between function and static
variables.  It is emitted by <samp>lto-cgraph.c</samp>:<code>output_refs</code>
and read by <samp>lto-cgraph.c</samp>:<code>input_refs</code>.
</p>

</li><li> Function bodies (<code>.gnu.lto_.function_body.&lt;name&gt;</code>)

<p>This section contains function bodies in the intermediate language
representation.  Every function body is in a separate section to allow
copying of the section independently to different object files or
reading the function on demand.
</p>
<p>Functions are emitted in
<samp>lto-streamer-out.c</samp>:<code>output_function</code> and read in
<samp>lto-streamer-in.c</samp>:<code>input_function</code>.
</p>

</li><li> Static variable initializers (<code>.gnu.lto_.vars</code>)

<p>This section contains all the symbols in the global variable pool.  It
is emitted by <samp>lto-cgraph.c</samp>:<code>output_varpool</code> and read in
<samp>lto-cgraph.c</samp>:<code>input_cgraph</code>.
</p>
</li><li> Summaries and optimization summaries used by IPA passes
(<code>.gnu.lto_.&lt;xxx&gt;</code>, where <code>&lt;xxx&gt;</code> is one of <code>jmpfuncs</code>,
<code>pureconst</code> or <code>reference</code>)

<p>These sections are used by IPA passes that need to emit summary
information during LTO generation to be read and aggregated at
link time.  Each pass is responsible for implementing two pass manager
hooks: one for writing the summary and another for reading it in.  The
format of these sections is entirely up to each individual pass.  The
only requirement is that the writer and reader hooks agree on the
format.
</p></li></ul>


<hr>
<a name="IPA"></a>
<div class="header">
<p>
Next: <a href="#WHOPR" accesskey="n" rel="next">WHOPR</a>, Previous: <a href="#LTO-object-file-layout" accesskey="p" rel="prev">LTO object file layout</a>, Up: <a href="#LTO" accesskey="u" rel="up">LTO</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Using-summary-information-in-IPA-passes"></a>
<h3 class="section">24.3 Using summary information in IPA passes</h3>

<p>Programs are represented internally as a <em>callgraph</em> (a
multi-graph where nodes are functions and edges are call sites)
and a <em>varpool</em> (a list of static and external variables in
the program).
</p>
<p>The inter-procedural optimization is organized as a sequence of
individual passes, which operate on the callgraph and the
varpool.  To make the implementation of WHOPR possible, every
inter-procedural optimization pass is split into several stages
that are executed at different times during WHOPR compilation:
</p>
<ul>
<li> LGEN time
<ol>
<li> <em>Generate summary</em> (<code>generate_summary</code> in
<code>struct ipa_opt_pass_d</code>).  This stage analyzes every function
body and variable initializer is examined and stores relevant
information into a pass-specific data structure.

</li><li> <em>Write summary</em> (<code>write_summary</code> in
<code>struct ipa_opt_pass_d</code>).  This stage writes all the
pass-specific information generated by <code>generate_summary</code>.
Summaries go into their own <code>LTO_section_*</code> sections that
have to be declared in <samp>lto-streamer.h</samp>:<code>enum
lto_section_type</code>.  A new section is created by calling
<code>create_output_block</code> and data can be written using the
<code>lto_output_*</code> routines.
</li></ol>

</li><li> WPA time
<ol>
<li> <em>Read summary</em> (<code>read_summary</code> in
<code>struct ipa_opt_pass_d</code>).  This stage reads all the
pass-specific information in exactly the same order that it was
written by <code>write_summary</code>.

</li><li> <em>Execute</em> (<code>execute</code> in <code>struct
opt_pass</code>).  This performs inter-procedural propagation.  This
must be done without actual access to the individual function
bodies or variable initializers.  Typically, this results in a
transitive closure operation over the summary information of all
the nodes in the callgraph.

</li><li> <em>Write optimization summary</em>
(<code>write_optimization_summary</code> in <code>struct
ipa_opt_pass_d</code>).  This writes the result of the inter-procedural
propagation into the object file.  This can use the same data
structures and helper routines used in <code>write_summary</code>.
</li></ol>

</li><li> LTRANS time
<ol>
<li> <em>Read optimization summary</em>
(<code>read_optimization_summary</code> in <code>struct
ipa_opt_pass_d</code>).  The counterpart to
<code>write_optimization_summary</code>.  This reads the interprocedural
optimization decisions in exactly the same format emitted by
<code>write_optimization_summary</code>.

</li><li> <em>Transform</em> (<code>function_transform</code> and
<code>variable_transform</code> in <code>struct ipa_opt_pass_d</code>).
The actual function bodies and variable initializers are updated
based on the information passed down from the <em>Execute</em> stage.
</li></ol>
</li></ul>

<p>The implementation of the inter-procedural passes are shared
between LTO, WHOPR and classic non-LTO compilation.
</p>
<ul>
<li> During the traditional file-by-file mode every pass executes its
own <em>Generate summary</em>, <em>Execute</em>, and <em>Transform</em>
stages within the single execution context of the compiler.

</li><li> In LTO compilation mode, every pass uses <em>Generate
summary</em> and <em>Write summary</em> stages at compilation time,
while the <em>Read summary</em>, <em>Execute</em>, and
<em>Transform</em> stages are executed at link time.

</li><li> In WHOPR mode all stages are used.
</li></ul>

<p>To simplify development, the GCC pass manager differentiates
between normal inter-procedural passes and small inter-procedural
passes.  A <em>small inter-procedural pass</em>
(<code>SIMPLE_IPA_PASS</code>) is a pass that does
everything at once and thus it can not be executed during WPA in
WHOPR mode.  It defines only the <em>Execute</em> stage and during
this stage it accesses and modifies the function bodies.  Such
passes are useful for optimization at LGEN or LTRANS time and are
used, for example, to implement early optimization before writing
object files.  The simple inter-procedural passes can also be used
for easier prototyping and development of a new inter-procedural
pass.
</p>

<a name="Virtual-clones"></a>
<h4 class="subsection">24.3.1 Virtual clones</h4>

<p>One of the main challenges of introducing the WHOPR compilation
mode was addressing the interactions between optimization passes.
In LTO compilation mode, the passes are executed in a sequence,
each of which consists of analysis (or <em>Generate summary</em>),
propagation (or <em>Execute</em>) and <em>Transform</em> stages.
Once the work of one pass is finished, the next pass sees the
updated program representation and can execute.  This makes the
individual passes dependent on each other.
</p>
<p>In WHOPR mode all passes first execute their <em>Generate
summary</em> stage.  Then summary writing marks the end of the LGEN
stage.  At WPA time,
the summaries are read back into memory and all passes run the
<em>Execute</em> stage.  Optimization summaries are streamed and
sent to LTRANS, where all the passes execute the <em>Transform</em>
stage.
</p>
<p>Most optimization passes split naturally into analysis,
propagation and transformation stages.  But some do not.  The
main problem arises when one pass performs changes and the
following pass gets confused by seeing different callgraphs
between the <em>Transform</em> stage and the <em>Generate summary</em>
or <em>Execute</em> stage.  This means that the passes are required
to communicate their decisions with each other.
</p>
<p>To facilitate this communication, the GCC callgraph
infrastructure implements <em>virtual clones</em>, a method of
representing the changes performed by the optimization passes in
the callgraph without needing to update function bodies.
</p>
<p>A <em>virtual clone</em> in the callgraph is a function that has no
associated body, just a description of how to create its body based
on a different function (which itself may be a virtual clone).
</p>
<p>The description of function modifications includes adjustments to
the function&rsquo;s signature (which allows, for example, removing or
adding function arguments), substitutions to perform on the
function body, and, for inlined functions, a pointer to the
function that it will be inlined into.
</p>
<p>It is also possible to redirect any edge of the callgraph from a
function to its virtual clone.  This implies updating of the call
site to adjust for the new function signature.
</p>
<p>Most of the transformations performed by inter-procedural
optimizations can be represented via virtual clones.  For
instance, a constant propagation pass can produce a virtual clone
of the function which replaces one of its arguments by a
constant.  The inliner can represent its decisions by producing a
clone of a function whose body will be later integrated into
a given function.
</p>
<p>Using <em>virtual clones</em>, the program can be easily updated
during the <em>Execute</em> stage, solving most of pass interactions
problems that would otherwise occur during <em>Transform</em>.
</p>
<p>Virtual clones are later materialized in the LTRANS stage and
turned into real functions.  Passes executed after the virtual
clone were introduced also perform their <em>Transform</em> stage
on new functions, so for a pass there is no significant
difference between operating on a real function or a virtual
clone introduced before its <em>Execute</em> stage.
</p>
<p>Optimization passes then work on virtual clones introduced before
their <em>Execute</em> stage as if they were real functions.  The
only difference is that clones are not visible during the
<em>Generate Summary</em> stage.
</p>
<p>To keep function summaries updated, the callgraph interface
allows an optimizer to register a callback that is called every
time a new clone is introduced as well as when the actual
function or variable is generated or when a function or variable
is removed.  These hooks are registered in the <em>Generate
summary</em> stage and allow the pass to keep its information intact
until the <em>Execute</em> stage.  The same hooks can also be
registered during the <em>Execute</em> stage to keep the
optimization summaries updated for the <em>Transform</em> stage.
</p>
<a name="IPA-references"></a>
<h4 class="subsection">24.3.2 IPA references</h4>

<p>GCC represents IPA references in the callgraph.  For a function
or variable <code>A</code>, the <em>IPA reference</em> is a list of all
locations where the address of <code>A</code> is taken and, when
<code>A</code> is a variable, a list of all direct stores and reads
to/from <code>A</code>.  References represent an oriented multi-graph on
the union of nodes of the callgraph and the varpool.  See
<samp>ipa-reference.c</samp>:<code>ipa_reference_write_optimization_summary</code>
and
<samp>ipa-reference.c</samp>:<code>ipa_reference_read_optimization_summary</code>
for details.
</p>
<a name="Jump-functions"></a>
<h4 class="subsection">24.3.3 Jump functions</h4>
<p>Suppose that an optimization pass sees a function <code>A</code> and it
knows the values of (some of) its arguments.  The <em>jump
function</em> describes the value of a parameter of a given function
call in function <code>A</code> based on this knowledge.
</p>
<p>Jump functions are used by several optimizations, such as the
inter-procedural constant propagation pass and the
devirtualization pass.  The inliner also uses jump functions to
perform inlining of callbacks.
</p>
<hr>
<a name="WHOPR"></a>
<div class="header">
<p>
Next: <a href="#Internal-flags" accesskey="n" rel="next">Internal flags</a>, Previous: <a href="#IPA" accesskey="p" rel="prev">IPA</a>, Up: <a href="#LTO" accesskey="u" rel="up">LTO</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Whole-program-assumptions_002c-linker-plugin-and-symbol-visibilities"></a>
<h3 class="section">24.4 Whole program assumptions, linker plugin and symbol visibilities</h3>

<p>Link-time optimization gives relatively minor benefits when used
alone.  The problem is that propagation of inter-procedural
information does not work well across functions and variables
that are called or referenced by other compilation units (such as
from a dynamically linked library).  We say that such functions
and variables are <em>externally visible</em>.
</p>
<p>To make the situation even more difficult, many applications
organize themselves as a set of shared libraries, and the default
ELF visibility rules allow one to overwrite any externally
visible symbol with a different symbol at runtime.  This
basically disables any optimizations across such functions and
variables, because the compiler cannot be sure that the function
body it is seeing is the same function body that will be used at
runtime.  Any function or variable not declared <code>static</code> in
the sources degrades the quality of inter-procedural
optimization.
</p>
<p>To avoid this problem the compiler must assume that it sees the
whole program when doing link-time optimization.  Strictly
speaking, the whole program is rarely visible even at link-time.
Standard system libraries are usually linked dynamically or not
provided with the link-time information.  In GCC, the whole
program option (<samp>-fwhole-program</samp>) asserts that every
function and variable defined in the current compilation
unit is static, except for function <code>main</code> (note: at
link time, the current unit is the union of all objects compiled
with LTO).  Since some functions and variables need to
be referenced externally, for example by another DSO or from an
assembler file, GCC also provides the function and variable
attribute <code>externally_visible</code> which can be used to disable
the effect of <samp>-fwhole-program</samp> on a specific symbol.
</p>
<p>The whole program mode assumptions are slightly more complex in
C++, where inline functions in headers are put into <em>COMDAT</em>
sections.  COMDAT function and variables can be defined by
multiple object files and their bodies are unified at link-time
and dynamic link-time.  COMDAT functions are changed to local only
when their address is not taken and thus un-sharing them with a
library is not harmful.  COMDAT variables always remain externally
visible, however for readonly variables it is assumed that their
initializers cannot be overwritten by a different value.
</p>
<p>GCC provides the function and variable attribute
<code>visibility</code> that can be used to specify the visibility of
externally visible symbols (or alternatively an
<samp>-fdefault-visibility</samp> command line option).  ELF defines
the <code>default</code>, <code>protected</code>, <code>hidden</code> and
<code>internal</code> visibilities.
</p>
<p>The most commonly used is visibility is <code>hidden</code>.  It
specifies that the symbol cannot be referenced from outside of
the current shared library.  Unfortunately, this information
cannot be used directly by the link-time optimization in the
compiler since the whole shared library also might contain
non-LTO objects and those are not visible to the compiler.
</p>
<p>GCC solves this problem using linker plugins.  A <em>linker
plugin</em> is an interface to the linker that allows an external
program to claim the ownership of a given object file.  The linker
then performs the linking procedure by querying the plugin about
the symbol table of the claimed objects and once the linking
decisions are complete, the plugin is allowed to provide the
final object file before the actual linking is made.  The linker
plugin obtains the symbol resolution information which specifies
which symbols provided by the claimed objects are bound from the
rest of a binary being linked.
</p>
<p>Currently, the linker plugin  works only in combination
with the Gold linker, but a GNU ld implementation is under
development.
</p>
<p>GCC is designed to be independent of the rest of the toolchain
and aims to support linkers without plugin support.  For this
reason it does not use the linker plugin by default.  Instead,
the object files are examined by <code>collect2</code> before being
passed to the linker and objects found to have LTO sections are
passed to <code>lto1</code> first.  This mode does not work for
library archives.  The decision on what object files from the
archive are needed depends on the actual linking and thus GCC
would have to implement the linker itself.  The resolution
information is missing too and thus GCC needs to make an educated
guess based on <samp>-fwhole-program</samp>.  Without the linker
plugin GCC also assumes that symbols are declared <code>hidden</code>
and not referred by non-LTO code by default.
</p>
<hr>
<a name="Internal-flags"></a>
<div class="header">
<p>
Previous: <a href="#WHOPR" accesskey="p" rel="prev">WHOPR</a>, Up: <a href="#LTO" accesskey="u" rel="up">LTO</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Internal-flags-controlling-lto1"></a>
<h3 class="section">24.5 Internal flags controlling <code>lto1</code></h3>

<p>The following flags are passed into <code>lto1</code> and are not
meant to be used directly from the command line.
</p>
<ul>
<li> -fwpa
<a name="index-fwpa"></a>
This option runs the serial part of the link-time optimizer
performing the inter-procedural propagation (WPA mode).  The
compiler reads in summary information from all inputs and
performs an analysis based on summary information only.  It
generates object files for subsequent runs of the link-time
optimizer where individual object files are optimized using both
summary information from the WPA mode and the actual function
bodies.  It then drives the LTRANS phase.

</li><li> -fltrans
<a name="index-fltrans"></a>
This option runs the link-time optimizer in the
local-transformation (LTRANS) mode, which reads in output from a
previous run of the LTO in WPA mode.  In the LTRANS mode, LTO
optimizes an object and produces the final assembly.

</li><li> -fltrans-output-list=<var>file</var>
<a name="index-fltrans_002doutput_002dlist"></a>
This option specifies a file to which the names of LTRANS output
files are written.  This option is only meaningful in conjunction
with <samp>-fwpa</samp>.
</li></ul>

<hr>
<a name="Funding"></a>
<div class="header">
<p>
Next: <a href="#GNU-Project" accesskey="n" rel="next">GNU Project</a>, Previous: <a href="#LTO" accesskey="p" rel="prev">LTO</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Funding-Free-Software"></a>
<h2 class="unnumbered">Funding Free Software</h2>

<p>If you want to have more free software a few years from now, it makes
sense for you to help encourage people to contribute funds for its
development.  The most effective approach known is to encourage
commercial redistributors to donate.
</p>
<p>Users of free software systems can boost the pace of development by
encouraging for-a-fee distributors to donate part of their selling price
to free software developers&mdash;the Free Software Foundation, and others.
</p>
<p>The way to convince distributors to do this is to demand it and expect
it from them.  So when you compare distributors, judge them partly by
how much they give to free software development.  Show distributors
they must compete to be the one who gives the most.
</p>
<p>To make this approach work, you must insist on numbers that you can
compare, such as, &ldquo;We will donate ten dollars to the Frobnitz project
for each disk sold.&rdquo;  Don&rsquo;t be satisfied with a vague promise, such as
&ldquo;A portion of the profits are donated,&rdquo; since it doesn&rsquo;t give a basis
for comparison.
</p>
<p>Even a precise fraction &ldquo;of the profits from this disk&rdquo; is not very
meaningful, since creative accounting and unrelated business decisions
can greatly alter what fraction of the sales price counts as profit.
If the price you pay is $50, ten percent of the profit is probably
less than a dollar; it might be a few cents, or nothing at all.
</p>
<p>Some redistributors do development work themselves.  This is useful too;
but to keep everyone honest, you need to inquire how much they do, and
what kind.  Some kinds of development make much more long-term
difference than others.  For example, maintaining a separate version of
a program contributes very little; maintaining the standard version of a
program for the whole community contributes much.  Easy new ports
contribute little, since someone else would surely do them; difficult
ports such as adding a new CPU to the GNU Compiler Collection contribute more;
major new features or packages contribute the most.
</p>
<p>By establishing the idea that supporting further development is &ldquo;the
proper thing to do&rdquo; when distributing free software for a fee, we can
assure a steady flow of resources into making more free software.
</p>
<div class="display">
<pre class="display">Copyright &copy; 1994 Free Software Foundation, Inc.
Verbatim copying and redistribution of this section is permitted
without royalty; alteration is not permitted.
</pre></div>

<hr>
<a name="GNU-Project"></a>
<div class="header">
<p>
Next: <a href="#Copying" accesskey="n" rel="next">Copying</a>, Previous: <a href="#Funding" accesskey="p" rel="prev">Funding</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="The-GNU-Project-and-GNU_002fLinux"></a>
<h2 class="unnumbered">The GNU Project and GNU/Linux</h2>

<p>The GNU Project was launched in 1984 to develop a complete Unix-like
operating system which is free software: the GNU system.  (GNU is a
recursive acronym for &ldquo;GNU&rsquo;s Not Unix&rdquo;; it is pronounced
&ldquo;guh-NEW&rdquo;.)  Variants of the GNU operating system, which use the
kernel Linux, are now widely used; though these systems are often
referred to as &ldquo;Linux&rdquo;, they are more accurately called GNU/Linux
systems.
</p>
<p>For more information, see:
</p><div class="smallexample">
<pre class="smallexample"><a href="http://www.gnu.org/">http://www.gnu.org/</a>
<a href="http://www.gnu.org/gnu/linux-and-gnu.html">http://www.gnu.org/gnu/linux-and-gnu.html</a>
</pre></div>
<hr>
<a name="Copying"></a>
<div class="header">
<p>
Next: <a href="#GNU-Free-Documentation-License" accesskey="n" rel="next">GNU Free Documentation License</a>, Previous: <a href="#GNU-Project" accesskey="p" rel="prev">GNU Project</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GNU-General-Public-License"></a>
<h2 class="unnumbered">GNU General Public License</h2>
<div align="center">Version 3, 29 June 2007
</div>

<div class="display">
<pre class="display">Copyright &copy; 2007 Free Software Foundation, Inc. <a href="http://fsf.org/">http://fsf.org/</a>

Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.
</pre></div>

<a name="Preamble"></a>
<h3 class="heading">Preamble</h3>

<p>The GNU General Public License is a free, copyleft license for
software and other kinds of works.
</p>
<p>The licenses for most software and other practical works are designed
to take away your freedom to share and change the works.  By contrast,
the GNU General Public License is intended to guarantee your freedom
to share and change all versions of a program&ndash;to make sure it remains
free software for all its users.  We, the Free Software Foundation,
use the GNU General Public License for most of our software; it
applies also to any other work released this way by its authors.  You
can apply it to your programs, too.
</p>
<p>When we speak of free software, we are referring to freedom, not
price.  Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
them if you wish), that you receive source code or can get it if you
want it, that you can change the software or use pieces of it in new
free programs, and that you know you can do these things.
</p>
<p>To protect your rights, we need to prevent others from denying you
these rights or asking you to surrender the rights.  Therefore, you
have certain responsibilities if you distribute copies of the
software, or if you modify it: responsibilities to respect the freedom
of others.
</p>
<p>For example, if you distribute copies of such a program, whether
gratis or for a fee, you must pass on to the recipients the same
freedoms that you received.  You must make sure that they, too,
receive or can get the source code.  And you must show them these
terms so they know their rights.
</p>
<p>Developers that use the GNU GPL protect your rights with two steps:
(1) assert copyright on the software, and (2) offer you this License
giving you legal permission to copy, distribute and/or modify it.
</p>
<p>For the developers&rsquo; and authors&rsquo; protection, the GPL clearly explains
that there is no warranty for this free software.  For both users&rsquo; and
authors&rsquo; sake, the GPL requires that modified versions be marked as
changed, so that their problems will not be attributed erroneously to
authors of previous versions.
</p>
<p>Some devices are designed to deny users access to install or run
modified versions of the software inside them, although the
manufacturer can do so.  This is fundamentally incompatible with the
aim of protecting users&rsquo; freedom to change the software.  The
systematic pattern of such abuse occurs in the area of products for
individuals to use, which is precisely where it is most unacceptable.
Therefore, we have designed this version of the GPL to prohibit the
practice for those products.  If such problems arise substantially in
other domains, we stand ready to extend this provision to those
domains in future versions of the GPL, as needed to protect the
freedom of users.
</p>
<p>Finally, every program is threatened constantly by software patents.
States should not allow patents to restrict development and use of
software on general-purpose computers, but in those that do, we wish
to avoid the special danger that patents applied to a free program
could make it effectively proprietary.  To prevent this, the GPL
assures that patents cannot be used to render the program non-free.
</p>
<p>The precise terms and conditions for copying, distribution and
modification follow.
</p>
<a name="TERMS-AND-CONDITIONS"></a>
<h3 class="heading">TERMS AND CONDITIONS</h3>

<ol>
<li> Definitions.

<p>&ldquo;This License&rdquo; refers to version 3 of the GNU General Public License.
</p>
<p>&ldquo;Copyright&rdquo; also means copyright-like laws that apply to other kinds
of works, such as semiconductor masks.
</p>
<p>&ldquo;The Program&rdquo; refers to any copyrightable work licensed under this
License.  Each licensee is addressed as &ldquo;you&rdquo;.  &ldquo;Licensees&rdquo; and
&ldquo;recipients&rdquo; may be individuals or organizations.
</p>
<p>To &ldquo;modify&rdquo; a work means to copy from or adapt all or part of the work
in a fashion requiring copyright permission, other than the making of
an exact copy.  The resulting work is called a &ldquo;modified version&rdquo; of
the earlier work or a work &ldquo;based on&rdquo; the earlier work.
</p>
<p>A &ldquo;covered work&rdquo; means either the unmodified Program or a work based
on the Program.
</p>
<p>To &ldquo;propagate&rdquo; a work means to do anything with it that, without
permission, would make you directly or secondarily liable for
infringement under applicable copyright law, except executing it on a
computer or modifying a private copy.  Propagation includes copying,
distribution (with or without modification), making available to the
public, and in some countries other activities as well.
</p>
<p>To &ldquo;convey&rdquo; a work means any kind of propagation that enables other
parties to make or receive copies.  Mere interaction with a user
through a computer network, with no transfer of a copy, is not
conveying.
</p>
<p>An interactive user interface displays &ldquo;Appropriate Legal Notices&rdquo; to
the extent that it includes a convenient and prominently visible
feature that (1) displays an appropriate copyright notice, and (2)
tells the user that there is no warranty for the work (except to the
extent that warranties are provided), that licensees may convey the
work under this License, and how to view a copy of this License.  If
the interface presents a list of user commands or options, such as a
menu, a prominent item in the list meets this criterion.
</p>
</li><li> Source Code.

<p>The &ldquo;source code&rdquo; for a work means the preferred form of the work for
making modifications to it.  &ldquo;Object code&rdquo; means any non-source form
of a work.
</p>
<p>A &ldquo;Standard Interface&rdquo; means an interface that either is an official
standard defined by a recognized standards body, or, in the case of
interfaces specified for a particular programming language, one that
is widely used among developers working in that language.
</p>
<p>The &ldquo;System Libraries&rdquo; of an executable work include anything, other
than the work as a whole, that (a) is included in the normal form of
packaging a Major Component, but which is not part of that Major
Component, and (b) serves only to enable use of the work with that
Major Component, or to implement a Standard Interface for which an
implementation is available to the public in source code form.  A
&ldquo;Major Component&rdquo;, in this context, means a major essential component
(kernel, window system, and so on) of the specific operating system
(if any) on which the executable work runs, or a compiler used to
produce the work, or an object code interpreter used to run it.
</p>
<p>The &ldquo;Corresponding Source&rdquo; for a work in object code form means all
the source code needed to generate, install, and (for an executable
work) run the object code and to modify the work, including scripts to
control those activities.  However, it does not include the work&rsquo;s
System Libraries, or general-purpose tools or generally available free
programs which are used unmodified in performing those activities but
which are not part of the work.  For example, Corresponding Source
includes interface definition files associated with source files for
the work, and the source code for shared libraries and dynamically
linked subprograms that the work is specifically designed to require,
such as by intimate data communication or control flow between those
subprograms and other parts of the work.
</p>
<p>The Corresponding Source need not include anything that users can
regenerate automatically from other parts of the Corresponding Source.
</p>
<p>The Corresponding Source for a work in source code form is that same
work.
</p>
</li><li> Basic Permissions.

<p>All rights granted under this License are granted for the term of
copyright on the Program, and are irrevocable provided the stated
conditions are met.  This License explicitly affirms your unlimited
permission to run the unmodified Program.  The output from running a
covered work is covered by this License only if the output, given its
content, constitutes a covered work.  This License acknowledges your
rights of fair use or other equivalent, as provided by copyright law.
</p>
<p>You may make, run and propagate covered works that you do not convey,
without conditions so long as your license otherwise remains in force.
You may convey covered works to others for the sole purpose of having
them make modifications exclusively for you, or provide you with
facilities for running those works, provided that you comply with the
terms of this License in conveying all material for which you do not
control copyright.  Those thus making or running the covered works for
you must do so exclusively on your behalf, under your direction and
control, on terms that prohibit them from making any copies of your
copyrighted material outside their relationship with you.
</p>
<p>Conveying under any other circumstances is permitted solely under the
conditions stated below.  Sublicensing is not allowed; section 10
makes it unnecessary.
</p>
</li><li> Protecting Users&rsquo; Legal Rights From Anti-Circumvention Law.

<p>No covered work shall be deemed part of an effective technological
measure under any applicable law fulfilling obligations under article
11 of the WIPO copyright treaty adopted on 20 December 1996, or
similar laws prohibiting or restricting circumvention of such
measures.
</p>
<p>When you convey a covered work, you waive any legal power to forbid
circumvention of technological measures to the extent such
circumvention is effected by exercising rights under this License with
respect to the covered work, and you disclaim any intention to limit
operation or modification of the work as a means of enforcing, against
the work&rsquo;s users, your or third parties&rsquo; legal rights to forbid
circumvention of technological measures.
</p>
</li><li> Conveying Verbatim Copies.

<p>You may convey verbatim copies of the Program&rsquo;s source code as you
receive it, in any medium, provided that you conspicuously and
appropriately publish on each copy an appropriate copyright notice;
keep intact all notices stating that this License and any
non-permissive terms added in accord with section 7 apply to the code;
keep intact all notices of the absence of any warranty; and give all
recipients a copy of this License along with the Program.
</p>
<p>You may charge any price or no price for each copy that you convey,
and you may offer support or warranty protection for a fee.
</p>
</li><li> Conveying Modified Source Versions.

<p>You may convey a work based on the Program, or the modifications to
produce it from the Program, in the form of source code under the
terms of section 4, provided that you also meet all of these
conditions:
</p>
<ol>
<li> The work must carry prominent notices stating that you modified it,
and giving a relevant date.

</li><li> The work must carry prominent notices stating that it is released
under this License and any conditions added under section 7.  This
requirement modifies the requirement in section 4 to &ldquo;keep intact all
notices&rdquo;.

</li><li> You must license the entire work, as a whole, under this License to
anyone who comes into possession of a copy.  This License will
therefore apply, along with any applicable section 7 additional terms,
to the whole of the work, and all its parts, regardless of how they
are packaged.  This License gives no permission to license the work in
any other way, but it does not invalidate such permission if you have
separately received it.

</li><li> If the work has interactive user interfaces, each must display
Appropriate Legal Notices; however, if the Program has interactive
interfaces that do not display Appropriate Legal Notices, your work
need not make them do so.
</li></ol>

<p>A compilation of a covered work with other separate and independent
works, which are not by their nature extensions of the covered work,
and which are not combined with it such as to form a larger program,
in or on a volume of a storage or distribution medium, is called an
&ldquo;aggregate&rdquo; if the compilation and its resulting copyright are not
used to limit the access or legal rights of the compilation&rsquo;s users
beyond what the individual works permit.  Inclusion of a covered work
in an aggregate does not cause this License to apply to the other
parts of the aggregate.
</p>
</li><li> Conveying Non-Source Forms.

<p>You may convey a covered work in object code form under the terms of
sections 4 and 5, provided that you also convey the machine-readable
Corresponding Source under the terms of this License, in one of these
ways:
</p>
<ol>
<li> Convey the object code in, or embodied in, a physical product
(including a physical distribution medium), accompanied by the
Corresponding Source fixed on a durable physical medium customarily
used for software interchange.

</li><li> Convey the object code in, or embodied in, a physical product
(including a physical distribution medium), accompanied by a written
offer, valid for at least three years and valid for as long as you
offer spare parts or customer support for that product model, to give
anyone who possesses the object code either (1) a copy of the
Corresponding Source for all the software in the product that is
covered by this License, on a durable physical medium customarily used
for software interchange, for a price no more than your reasonable
cost of physically performing this conveying of source, or (2) access
to copy the Corresponding Source from a network server at no charge.

</li><li> Convey individual copies of the object code with a copy of the written
offer to provide the Corresponding Source.  This alternative is
allowed only occasionally and noncommercially, and only if you
received the object code with such an offer, in accord with subsection
6b.

</li><li> Convey the object code by offering access from a designated place
(gratis or for a charge), and offer equivalent access to the
Corresponding Source in the same way through the same place at no
further charge.  You need not require recipients to copy the
Corresponding Source along with the object code.  If the place to copy
the object code is a network server, the Corresponding Source may be
on a different server (operated by you or a third party) that supports
equivalent copying facilities, provided you maintain clear directions
next to the object code saying where to find the Corresponding Source.
Regardless of what server hosts the Corresponding Source, you remain
obligated to ensure that it is available for as long as needed to
satisfy these requirements.

</li><li> Convey the object code using peer-to-peer transmission, provided you
inform other peers where the object code and Corresponding Source of
the work are being offered to the general public at no charge under
subsection 6d.

</li></ol>

<p>A separable portion of the object code, whose source code is excluded
from the Corresponding Source as a System Library, need not be
included in conveying the object code work.
</p>
<p>A &ldquo;User Product&rdquo; is either (1) a &ldquo;consumer product&rdquo;, which means any
tangible personal property which is normally used for personal,
family, or household purposes, or (2) anything designed or sold for
incorporation into a dwelling.  In determining whether a product is a
consumer product, doubtful cases shall be resolved in favor of
coverage.  For a particular product received by a particular user,
&ldquo;normally used&rdquo; refers to a typical or common use of that class of
product, regardless of the status of the particular user or of the way
in which the particular user actually uses, or expects or is expected
to use, the product.  A product is a consumer product regardless of
whether the product has substantial commercial, industrial or
non-consumer uses, unless such uses represent the only significant
mode of use of the product.
</p>
<p>&ldquo;Installation Information&rdquo; for a User Product means any methods,
procedures, authorization keys, or other information required to
install and execute modified versions of a covered work in that User
Product from a modified version of its Corresponding Source.  The
information must suffice to ensure that the continued functioning of
the modified object code is in no case prevented or interfered with
solely because modification has been made.
</p>
<p>If you convey an object code work under this section in, or with, or
specifically for use in, a User Product, and the conveying occurs as
part of a transaction in which the right of possession and use of the
User Product is transferred to the recipient in perpetuity or for a
fixed term (regardless of how the transaction is characterized), the
Corresponding Source conveyed under this section must be accompanied
by the Installation Information.  But this requirement does not apply
if neither you nor any third party retains the ability to install
modified object code on the User Product (for example, the work has
been installed in ROM).
</p>
<p>The requirement to provide Installation Information does not include a
requirement to continue to provide support service, warranty, or
updates for a work that has been modified or installed by the
recipient, or for the User Product in which it has been modified or
installed.  Access to a network may be denied when the modification
itself materially and adversely affects the operation of the network
or violates the rules and protocols for communication across the
network.
</p>
<p>Corresponding Source conveyed, and Installation Information provided,
in accord with this section must be in a format that is publicly
documented (and with an implementation available to the public in
source code form), and must require no special password or key for
unpacking, reading or copying.
</p>
</li><li> Additional Terms.

<p>&ldquo;Additional permissions&rdquo; are terms that supplement the terms of this
License by making exceptions from one or more of its conditions.
Additional permissions that are applicable to the entire Program shall
be treated as though they were included in this License, to the extent
that they are valid under applicable law.  If additional permissions
apply only to part of the Program, that part may be used separately
under those permissions, but the entire Program remains governed by
this License without regard to the additional permissions.
</p>
<p>When you convey a copy of a covered work, you may at your option
remove any additional permissions from that copy, or from any part of
it.  (Additional permissions may be written to require their own
removal in certain cases when you modify the work.)  You may place
additional permissions on material, added by you to a covered work,
for which you have or can give appropriate copyright permission.
</p>
<p>Notwithstanding any other provision of this License, for material you
add to a covered work, you may (if authorized by the copyright holders
of that material) supplement the terms of this License with terms:
</p>
<ol>
<li> Disclaiming warranty or limiting liability differently from the terms
of sections 15 and 16 of this License; or

</li><li> Requiring preservation of specified reasonable legal notices or author
attributions in that material or in the Appropriate Legal Notices
displayed by works containing it; or

</li><li> Prohibiting misrepresentation of the origin of that material, or
requiring that modified versions of such material be marked in
reasonable ways as different from the original version; or

</li><li> Limiting the use for publicity purposes of names of licensors or
authors of the material; or

</li><li> Declining to grant rights under trademark law for use of some trade
names, trademarks, or service marks; or

</li><li> Requiring indemnification of licensors and authors of that material by
anyone who conveys the material (or modified versions of it) with
contractual assumptions of liability to the recipient, for any
liability that these contractual assumptions directly impose on those
licensors and authors.
</li></ol>

<p>All other non-permissive additional terms are considered &ldquo;further
restrictions&rdquo; within the meaning of section 10.  If the Program as you
received it, or any part of it, contains a notice stating that it is
governed by this License along with a term that is a further
restriction, you may remove that term.  If a license document contains
a further restriction but permits relicensing or conveying under this
License, you may add to a covered work material governed by the terms
of that license document, provided that the further restriction does
not survive such relicensing or conveying.
</p>
<p>If you add terms to a covered work in accord with this section, you
must place, in the relevant source files, a statement of the
additional terms that apply to those files, or a notice indicating
where to find the applicable terms.
</p>
<p>Additional terms, permissive or non-permissive, may be stated in the
form of a separately written license, or stated as exceptions; the
above requirements apply either way.
</p>
</li><li> Termination.

<p>You may not propagate or modify a covered work except as expressly
provided under this License.  Any attempt otherwise to propagate or
modify it is void, and will automatically terminate your rights under
this License (including any patent licenses granted under the third
paragraph of section 11).
</p>
<p>However, if you cease all violation of this License, then your license
from a particular copyright holder is reinstated (a) provisionally,
unless and until the copyright holder explicitly and finally
terminates your license, and (b) permanently, if the copyright holder
fails to notify you of the violation by some reasonable means prior to
60 days after the cessation.
</p>
<p>Moreover, your license from a particular copyright holder is
reinstated permanently if the copyright holder notifies you of the
violation by some reasonable means, this is the first time you have
received notice of violation of this License (for any work) from that
copyright holder, and you cure the violation prior to 30 days after
your receipt of the notice.
</p>
<p>Termination of your rights under this section does not terminate the
licenses of parties who have received copies or rights from you under
this License.  If your rights have been terminated and not permanently
reinstated, you do not qualify to receive new licenses for the same
material under section 10.
</p>
</li><li> Acceptance Not Required for Having Copies.

<p>You are not required to accept this License in order to receive or run
a copy of the Program.  Ancillary propagation of a covered work
occurring solely as a consequence of using peer-to-peer transmission
to receive a copy likewise does not require acceptance.  However,
nothing other than this License grants you permission to propagate or
modify any covered work.  These actions infringe copyright if you do
not accept this License.  Therefore, by modifying or propagating a
covered work, you indicate your acceptance of this License to do so.
</p>
</li><li> Automatic Licensing of Downstream Recipients.

<p>Each time you convey a covered work, the recipient automatically
receives a license from the original licensors, to run, modify and
propagate that work, subject to this License.  You are not responsible
for enforcing compliance by third parties with this License.
</p>
<p>An &ldquo;entity transaction&rdquo; is a transaction transferring control of an
organization, or substantially all assets of one, or subdividing an
organization, or merging organizations.  If propagation of a covered
work results from an entity transaction, each party to that
transaction who receives a copy of the work also receives whatever
licenses to the work the party&rsquo;s predecessor in interest had or could
give under the previous paragraph, plus a right to possession of the
Corresponding Source of the work from the predecessor in interest, if
the predecessor has it or can get it with reasonable efforts.
</p>
<p>You may not impose any further restrictions on the exercise of the
rights granted or affirmed under this License.  For example, you may
not impose a license fee, royalty, or other charge for exercise of
rights granted under this License, and you may not initiate litigation
(including a cross-claim or counterclaim in a lawsuit) alleging that
any patent claim is infringed by making, using, selling, offering for
sale, or importing the Program or any portion of it.
</p>
</li><li> Patents.

<p>A &ldquo;contributor&rdquo; is a copyright holder who authorizes use under this
License of the Program or a work on which the Program is based.  The
work thus licensed is called the contributor&rsquo;s &ldquo;contributor version&rdquo;.
</p>
<p>A contributor&rsquo;s &ldquo;essential patent claims&rdquo; are all patent claims owned
or controlled by the contributor, whether already acquired or
hereafter acquired, that would be infringed by some manner, permitted
by this License, of making, using, or selling its contributor version,
but do not include claims that would be infringed only as a
consequence of further modification of the contributor version.  For
purposes of this definition, &ldquo;control&rdquo; includes the right to grant
patent sublicenses in a manner consistent with the requirements of
this License.
</p>
<p>Each contributor grants you a non-exclusive, worldwide, royalty-free
patent license under the contributor&rsquo;s essential patent claims, to
make, use, sell, offer for sale, import and otherwise run, modify and
propagate the contents of its contributor version.
</p>
<p>In the following three paragraphs, a &ldquo;patent license&rdquo; is any express
agreement or commitment, however denominated, not to enforce a patent
(such as an express permission to practice a patent or covenant not to
sue for patent infringement).  To &ldquo;grant&rdquo; such a patent license to a
party means to make such an agreement or commitment not to enforce a
patent against the party.
</p>
<p>If you convey a covered work, knowingly relying on a patent license,
and the Corresponding Source of the work is not available for anyone
to copy, free of charge and under the terms of this License, through a
publicly available network server or other readily accessible means,
then you must either (1) cause the Corresponding Source to be so
available, or (2) arrange to deprive yourself of the benefit of the
patent license for this particular work, or (3) arrange, in a manner
consistent with the requirements of this License, to extend the patent
license to downstream recipients.  &ldquo;Knowingly relying&rdquo; means you have
actual knowledge that, but for the patent license, your conveying the
covered work in a country, or your recipient&rsquo;s use of the covered work
in a country, would infringe one or more identifiable patents in that
country that you have reason to believe are valid.
</p>
<p>If, pursuant to or in connection with a single transaction or
arrangement, you convey, or propagate by procuring conveyance of, a
covered work, and grant a patent license to some of the parties
receiving the covered work authorizing them to use, propagate, modify
or convey a specific copy of the covered work, then the patent license
you grant is automatically extended to all recipients of the covered
work and works based on it.
</p>
<p>A patent license is &ldquo;discriminatory&rdquo; if it does not include within the
scope of its coverage, prohibits the exercise of, or is conditioned on
the non-exercise of one or more of the rights that are specifically
granted under this License.  You may not convey a covered work if you
are a party to an arrangement with a third party that is in the
business of distributing software, under which you make payment to the
third party based on the extent of your activity of conveying the
work, and under which the third party grants, to any of the parties
who would receive the covered work from you, a discriminatory patent
license (a) in connection with copies of the covered work conveyed by
you (or copies made from those copies), or (b) primarily for and in
connection with specific products or compilations that contain the
covered work, unless you entered into that arrangement, or that patent
license was granted, prior to 28 March 2007.
</p>
<p>Nothing in this License shall be construed as excluding or limiting
any implied license or other defenses to infringement that may
otherwise be available to you under applicable patent law.
</p>
</li><li> No Surrender of Others&rsquo; Freedom.

<p>If conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License.  If you cannot convey
a covered work so as to satisfy simultaneously your obligations under
this License and any other pertinent obligations, then as a
consequence you may not convey it at all.  For example, if you agree
to terms that obligate you to collect a royalty for further conveying
from those to whom you convey the Program, the only way you could
satisfy both those terms and this License would be to refrain entirely
from conveying the Program.
</p>
</li><li> Use with the GNU Affero General Public License.

<p>Notwithstanding any other provision of this License, you have
permission to link or combine any covered work with a work licensed
under version 3 of the GNU Affero General Public License into a single
combined work, and to convey the resulting work.  The terms of this
License will continue to apply to the part which is the covered work,
but the special requirements of the GNU Affero General Public License,
section 13, concerning interaction through a network will apply to the
combination as such.
</p>
</li><li> Revised Versions of this License.

<p>The Free Software Foundation may publish revised and/or new versions
of the GNU General Public License from time to time.  Such new
versions will be similar in spirit to the present version, but may
differ in detail to address new problems or concerns.
</p>
<p>Each version is given a distinguishing version number.  If the Program
specifies that a certain numbered version of the GNU General Public
License &ldquo;or any later version&rdquo; applies to it, you have the option of
following the terms and conditions either of that numbered version or
of any later version published by the Free Software Foundation.  If
the Program does not specify a version number of the GNU General
Public License, you may choose any version ever published by the Free
Software Foundation.
</p>
<p>If the Program specifies that a proxy can decide which future versions
of the GNU General Public License can be used, that proxy&rsquo;s public
statement of acceptance of a version permanently authorizes you to
choose that version for the Program.
</p>
<p>Later license versions may give you additional or different
permissions.  However, no additional obligations are imposed on any
author or copyright holder as a result of your choosing to follow a
later version.
</p>
</li><li> Disclaimer of Warranty.

<p>THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM &ldquo;AS IS&rdquo; WITHOUT
WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND
PERFORMANCE OF THE PROGRAM IS WITH YOU.  SHOULD THE PROGRAM PROVE
DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR
CORRECTION.
</p>
</li><li> Limitation of Liability.

<p>IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR
CONVEYS THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES
ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT
NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR
LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM
TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER
PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
</p>
</li><li> Interpretation of Sections 15 and 16.

<p>If the disclaimer of warranty and limitation of liability provided
above cannot be given local legal effect according to their terms,
reviewing courts shall apply local law that most closely approximates
an absolute waiver of all civil liability in connection with the
Program, unless a warranty or assumption of liability accompanies a
copy of the Program in return for a fee.
</p>
</li></ol>

<a name="END-OF-TERMS-AND-CONDITIONS"></a>
<h3 class="heading">END OF TERMS AND CONDITIONS</h3>

<a name="How-to-Apply-These-Terms-to-Your-New-Programs"></a>
<h3 class="heading">How to Apply These Terms to Your New Programs</h3>

<p>If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these
terms.
</p>
<p>To do so, attach the following notices to the program.  It is safest
to attach them to the start of each source file to most effectively
state the exclusion of warranty; and each file should have at least
the &ldquo;copyright&rdquo; line and a pointer to where the full notice is found.
</p>
<div class="smallexample">
<pre class="smallexample"><var>one line to give the program's name and a brief idea of what it does.</var>  
Copyright (C) <var>year</var> <var>name of author</var>

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or (at
your option) any later version.

This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <a href="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</a>.
</pre></div>

<p>Also add information on how to contact you by electronic and paper mail.
</p>
<p>If the program does terminal interaction, make it output a short
notice like this when it starts in an interactive mode:
</p>
<div class="smallexample">
<pre class="smallexample"><var>program</var> Copyright (C) <var>year</var> <var>name of author</var> 
This program comes with ABSOLUTELY NO WARRANTY; for details type &lsquo;<samp>show w</samp>&rsquo;.
This is free software, and you are welcome to redistribute it
under certain conditions; type &lsquo;<samp>show c</samp>&rsquo; for details.
</pre></div>

<p>The hypothetical commands &lsquo;<samp>show w</samp>&rsquo; and &lsquo;<samp>show c</samp>&rsquo; should show
the appropriate parts of the General Public License.  Of course, your
program&rsquo;s commands might be different; for a GUI interface, you would
use an &ldquo;about box&rdquo;.
</p>
<p>You should also get your employer (if you work as a programmer) or school,
if any, to sign a &ldquo;copyright disclaimer&rdquo; for the program, if necessary.
For more information on this, and how to apply and follow the GNU GPL, see
<a href="http://www.gnu.org/licenses/">http://www.gnu.org/licenses/</a>.
</p>
<p>The GNU General Public License does not permit incorporating your
program into proprietary programs.  If your program is a subroutine
library, you may consider it more useful to permit linking proprietary
applications with the library.  If this is what you want to do, use
the GNU Lesser General Public License instead of this License.  But
first, please read <a href="http://www.gnu.org/philosophy/why-not-lgpl.html">http://www.gnu.org/philosophy/why-not-lgpl.html</a>.
</p>

<hr>
<a name="GNU-Free-Documentation-License"></a>
<div class="header">
<p>
Next: <a href="#Contributors" accesskey="n" rel="next">Contributors</a>, Previous: <a href="#Copying" accesskey="p" rel="prev">Copying</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="GNU-Free-Documentation-License-1"></a>
<h2 class="unnumbered">GNU Free Documentation License</h2>

<a name="index-FDL_002c-GNU-Free-Documentation-License"></a>
<div align="center">Version 1.3, 3 November 2008
</div>
<div class="display">
<pre class="display">Copyright &copy; 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.
<a href="http://fsf.org/">http://fsf.org/</a>

Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
</pre></div>

<ol>
<li> PREAMBLE

<p>The purpose of this License is to make a manual, textbook, or other
functional and useful document <em>free</em> in the sense of freedom: to
assure everyone the effective freedom to copy and redistribute it,
with or without modifying it, either commercially or noncommercially.
Secondarily, this License preserves for the author and publisher a way
to get credit for their work, while not being considered responsible
for modifications made by others.
</p>
<p>This License is a kind of &ldquo;copyleft&rdquo;, which means that derivative
works of the document must themselves be free in the same sense.  It
complements the GNU General Public License, which is a copyleft
license designed for free software.
</p>
<p>We have designed this License in order to use it for manuals for free
software, because free software needs free documentation: a free
program should come with manuals providing the same freedoms that the
software does.  But this License is not limited to software manuals;
it can be used for any textual work, regardless of subject matter or
whether it is published as a printed book.  We recommend this License
principally for works whose purpose is instruction or reference.
</p>
</li><li> APPLICABILITY AND DEFINITIONS

<p>This License applies to any manual or other work, in any medium, that
contains a notice placed by the copyright holder saying it can be
distributed under the terms of this License.  Such a notice grants a
world-wide, royalty-free license, unlimited in duration, to use that
work under the conditions stated herein.  The &ldquo;Document&rdquo;, below,
refers to any such manual or work.  Any member of the public is a
licensee, and is addressed as &ldquo;you&rdquo;.  You accept the license if you
copy, modify or distribute the work in a way requiring permission
under copyright law.
</p>
<p>A &ldquo;Modified Version&rdquo; of the Document means any work containing the
Document or a portion of it, either copied verbatim, or with
modifications and/or translated into another language.
</p>
<p>A &ldquo;Secondary Section&rdquo; is a named appendix or a front-matter section
of the Document that deals exclusively with the relationship of the
publishers or authors of the Document to the Document&rsquo;s overall
subject (or to related matters) and contains nothing that could fall
directly within that overall subject.  (Thus, if the Document is in
part a textbook of mathematics, a Secondary Section may not explain
any mathematics.)  The relationship could be a matter of historical
connection with the subject or with related matters, or of legal,
commercial, philosophical, ethical or political position regarding
them.
</p>
<p>The &ldquo;Invariant Sections&rdquo; are certain Secondary Sections whose titles
are designated, as being those of Invariant Sections, in the notice
that says that the Document is released under this License.  If a
section does not fit the above definition of Secondary then it is not
allowed to be designated as Invariant.  The Document may contain zero
Invariant Sections.  If the Document does not identify any Invariant
Sections then there are none.
</p>
<p>The &ldquo;Cover Texts&rdquo; are certain short passages of text that are listed,
as Front-Cover Texts or Back-Cover Texts, in the notice that says that
the Document is released under this License.  A Front-Cover Text may
be at most 5 words, and a Back-Cover Text may be at most 25 words.
</p>
<p>A &ldquo;Transparent&rdquo; copy of the Document means a machine-readable copy,
represented in a format whose specification is available to the
general public, that is suitable for revising the document
straightforwardly with generic text editors or (for images composed of
pixels) generic paint programs or (for drawings) some widely available
drawing editor, and that is suitable for input to text formatters or
for automatic translation to a variety of formats suitable for input
to text formatters.  A copy made in an otherwise Transparent file
format whose markup, or absence of markup, has been arranged to thwart
or discourage subsequent modification by readers is not Transparent.
An image format is not Transparent if used for any substantial amount
of text.  A copy that is not &ldquo;Transparent&rdquo; is called &ldquo;Opaque&rdquo;.
</p>
<p>Examples of suitable formats for Transparent copies include plain
<small>ASCII</small> without markup, Texinfo input format, LaTeX input
format, <acronym>SGML</acronym> or <acronym>XML</acronym> using a publicly available
<acronym>DTD</acronym>, and standard-conforming simple <acronym>HTML</acronym>,
PostScript or <acronym>PDF</acronym> designed for human modification.  Examples
of transparent image formats include <acronym>PNG</acronym>, <acronym>XCF</acronym> and
<acronym>JPG</acronym>.  Opaque formats include proprietary formats that can be
read and edited only by proprietary word processors, <acronym>SGML</acronym> or
<acronym>XML</acronym> for which the <acronym>DTD</acronym> and/or processing tools are
not generally available, and the machine-generated <acronym>HTML</acronym>,
PostScript or <acronym>PDF</acronym> produced by some word processors for
output purposes only.
</p>
<p>The &ldquo;Title Page&rdquo; means, for a printed book, the title page itself,
plus such following pages as are needed to hold, legibly, the material
this License requires to appear in the title page.  For works in
formats which do not have any title page as such, &ldquo;Title Page&rdquo; means
the text near the most prominent appearance of the work&rsquo;s title,
preceding the beginning of the body of the text.
</p>
<p>The &ldquo;publisher&rdquo; means any person or entity that distributes copies
of the Document to the public.
</p>
<p>A section &ldquo;Entitled XYZ&rdquo; means a named subunit of the Document whose
title either is precisely XYZ or contains XYZ in parentheses following
text that translates XYZ in another language.  (Here XYZ stands for a
specific section name mentioned below, such as &ldquo;Acknowledgements&rdquo;,
&ldquo;Dedications&rdquo;, &ldquo;Endorsements&rdquo;, or &ldquo;History&rdquo;.)  To &ldquo;Preserve the Title&rdquo;
of such a section when you modify the Document means that it remains a
section &ldquo;Entitled XYZ&rdquo; according to this definition.
</p>
<p>The Document may include Warranty Disclaimers next to the notice which
states that this License applies to the Document.  These Warranty
Disclaimers are considered to be included by reference in this
License, but only as regards disclaiming warranties: any other
implication that these Warranty Disclaimers may have is void and has
no effect on the meaning of this License.
</p>
</li><li> VERBATIM COPYING

<p>You may copy and distribute the Document in any medium, either
commercially or noncommercially, provided that this License, the
copyright notices, and the license notice saying this License applies
to the Document are reproduced in all copies, and that you add no other
conditions whatsoever to those of this License.  You may not use
technical measures to obstruct or control the reading or further
copying of the copies you make or distribute.  However, you may accept
compensation in exchange for copies.  If you distribute a large enough
number of copies you must also follow the conditions in section 3.
</p>
<p>You may also lend copies, under the same conditions stated above, and
you may publicly display copies.
</p>
</li><li> COPYING IN QUANTITY

<p>If you publish printed copies (or copies in media that commonly have
printed covers) of the Document, numbering more than 100, and the
Document&rsquo;s license notice requires Cover Texts, you must enclose the
copies in covers that carry, clearly and legibly, all these Cover
Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
the back cover.  Both covers must also clearly and legibly identify
you as the publisher of these copies.  The front cover must present
the full title with all words of the title equally prominent and
visible.  You may add other material on the covers in addition.
Copying with changes limited to the covers, as long as they preserve
the title of the Document and satisfy these conditions, can be treated
as verbatim copying in other respects.
</p>
<p>If the required texts for either cover are too voluminous to fit
legibly, you should put the first ones listed (as many as fit
reasonably) on the actual cover, and continue the rest onto adjacent
pages.
</p>
<p>If you publish or distribute Opaque copies of the Document numbering
more than 100, you must either include a machine-readable Transparent
copy along with each Opaque copy, or state in or with each Opaque copy
a computer-network location from which the general network-using
public has access to download using public-standard network protocols
a complete Transparent copy of the Document, free of added material.
If you use the latter option, you must take reasonably prudent steps,
when you begin distribution of Opaque copies in quantity, to ensure
that this Transparent copy will remain thus accessible at the stated
location until at least one year after the last time you distribute an
Opaque copy (directly or through your agents or retailers) of that
edition to the public.
</p>
<p>It is requested, but not required, that you contact the authors of the
Document well before redistributing any large number of copies, to give
them a chance to provide you with an updated version of the Document.
</p>
</li><li> MODIFICATIONS

<p>You may copy and distribute a Modified Version of the Document under
the conditions of sections 2 and 3 above, provided that you release
the Modified Version under precisely this License, with the Modified
Version filling the role of the Document, thus licensing distribution
and modification of the Modified Version to whoever possesses a copy
of it.  In addition, you must do these things in the Modified Version:
</p>
<ol>
<li> Use in the Title Page (and on the covers, if any) a title distinct
from that of the Document, and from those of previous versions
(which should, if there were any, be listed in the History section
of the Document).  You may use the same title as a previous version
if the original publisher of that version gives permission.

</li><li> List on the Title Page, as authors, one or more persons or entities
responsible for authorship of the modifications in the Modified
Version, together with at least five of the principal authors of the
Document (all of its principal authors, if it has fewer than five),
unless they release you from this requirement.

</li><li> State on the Title page the name of the publisher of the
Modified Version, as the publisher.

</li><li> Preserve all the copyright notices of the Document.

</li><li> Add an appropriate copyright notice for your modifications
adjacent to the other copyright notices.

</li><li> Include, immediately after the copyright notices, a license notice
giving the public permission to use the Modified Version under the
terms of this License, in the form shown in the Addendum below.

</li><li> Preserve in that license notice the full lists of Invariant Sections
and required Cover Texts given in the Document&rsquo;s license notice.

</li><li> Include an unaltered copy of this License.

</li><li> Preserve the section Entitled &ldquo;History&rdquo;, Preserve its Title, and add
to it an item stating at least the title, year, new authors, and
publisher of the Modified Version as given on the Title Page.  If
there is no section Entitled &ldquo;History&rdquo; in the Document, create one
stating the title, year, authors, and publisher of the Document as
given on its Title Page, then add an item describing the Modified
Version as stated in the previous sentence.

</li><li> Preserve the network location, if any, given in the Document for
public access to a Transparent copy of the Document, and likewise
the network locations given in the Document for previous versions
it was based on.  These may be placed in the &ldquo;History&rdquo; section.
You may omit a network location for a work that was published at
least four years before the Document itself, or if the original
publisher of the version it refers to gives permission.

</li><li> For any section Entitled &ldquo;Acknowledgements&rdquo; or &ldquo;Dedications&rdquo;, Preserve
the Title of the section, and preserve in the section all the
substance and tone of each of the contributor acknowledgements and/or
dedications given therein.

</li><li> Preserve all the Invariant Sections of the Document,
unaltered in their text and in their titles.  Section numbers
or the equivalent are not considered part of the section titles.

</li><li> Delete any section Entitled &ldquo;Endorsements&rdquo;.  Such a section
may not be included in the Modified Version.

</li><li> Do not retitle any existing section to be Entitled &ldquo;Endorsements&rdquo; or
to conflict in title with any Invariant Section.

</li><li> Preserve any Warranty Disclaimers.
</li></ol>

<p>If the Modified Version includes new front-matter sections or
appendices that qualify as Secondary Sections and contain no material
copied from the Document, you may at your option designate some or all
of these sections as invariant.  To do this, add their titles to the
list of Invariant Sections in the Modified Version&rsquo;s license notice.
These titles must be distinct from any other section titles.
</p>
<p>You may add a section Entitled &ldquo;Endorsements&rdquo;, provided it contains
nothing but endorsements of your Modified Version by various
parties&mdash;for example, statements of peer review or that the text has
been approved by an organization as the authoritative definition of a
standard.
</p>
<p>You may add a passage of up to five words as a Front-Cover Text, and a
passage of up to 25 words as a Back-Cover Text, to the end of the list
of Cover Texts in the Modified Version.  Only one passage of
Front-Cover Text and one of Back-Cover Text may be added by (or
through arrangements made by) any one entity.  If the Document already
includes a cover text for the same cover, previously added by you or
by arrangement made by the same entity you are acting on behalf of,
you may not add another; but you may replace the old one, on explicit
permission from the previous publisher that added the old one.
</p>
<p>The author(s) and publisher(s) of the Document do not by this License
give permission to use their names for publicity for or to assert or
imply endorsement of any Modified Version.
</p>
</li><li> COMBINING DOCUMENTS

<p>You may combine the Document with other documents released under this
License, under the terms defined in section 4 above for modified
versions, provided that you include in the combination all of the
Invariant Sections of all of the original documents, unmodified, and
list them all as Invariant Sections of your combined work in its
license notice, and that you preserve all their Warranty Disclaimers.
</p>
<p>The combined work need only contain one copy of this License, and
multiple identical Invariant Sections may be replaced with a single
copy.  If there are multiple Invariant Sections with the same name but
different contents, make the title of each such section unique by
adding at the end of it, in parentheses, the name of the original
author or publisher of that section if known, or else a unique number.
Make the same adjustment to the section titles in the list of
Invariant Sections in the license notice of the combined work.
</p>
<p>In the combination, you must combine any sections Entitled &ldquo;History&rdquo;
in the various original documents, forming one section Entitled
&ldquo;History&rdquo;; likewise combine any sections Entitled &ldquo;Acknowledgements&rdquo;,
and any sections Entitled &ldquo;Dedications&rdquo;.  You must delete all
sections Entitled &ldquo;Endorsements.&rdquo;
</p>
</li><li> COLLECTIONS OF DOCUMENTS

<p>You may make a collection consisting of the Document and other documents
released under this License, and replace the individual copies of this
License in the various documents with a single copy that is included in
the collection, provided that you follow the rules of this License for
verbatim copying of each of the documents in all other respects.
</p>
<p>You may extract a single document from such a collection, and distribute
it individually under this License, provided you insert a copy of this
License into the extracted document, and follow this License in all
other respects regarding verbatim copying of that document.
</p>
</li><li> AGGREGATION WITH INDEPENDENT WORKS

<p>A compilation of the Document or its derivatives with other separate
and independent documents or works, in or on a volume of a storage or
distribution medium, is called an &ldquo;aggregate&rdquo; if the copyright
resulting from the compilation is not used to limit the legal rights
of the compilation&rsquo;s users beyond what the individual works permit.
When the Document is included in an aggregate, this License does not
apply to the other works in the aggregate which are not themselves
derivative works of the Document.
</p>
<p>If the Cover Text requirement of section 3 is applicable to these
copies of the Document, then if the Document is less than one half of
the entire aggregate, the Document&rsquo;s Cover Texts may be placed on
covers that bracket the Document within the aggregate, or the
electronic equivalent of covers if the Document is in electronic form.
Otherwise they must appear on printed covers that bracket the whole
aggregate.
</p>
</li><li> TRANSLATION

<p>Translation is considered a kind of modification, so you may
distribute translations of the Document under the terms of section 4.
Replacing Invariant Sections with translations requires special
permission from their copyright holders, but you may include
translations of some or all Invariant Sections in addition to the
original versions of these Invariant Sections.  You may include a
translation of this License, and all the license notices in the
Document, and any Warranty Disclaimers, provided that you also include
the original English version of this License and the original versions
of those notices and disclaimers.  In case of a disagreement between
the translation and the original version of this License or a notice
or disclaimer, the original version will prevail.
</p>
<p>If a section in the Document is Entitled &ldquo;Acknowledgements&rdquo;,
&ldquo;Dedications&rdquo;, or &ldquo;History&rdquo;, the requirement (section 4) to Preserve
its Title (section 1) will typically require changing the actual
title.
</p>
</li><li> TERMINATION

<p>You may not copy, modify, sublicense, or distribute the Document
except as expressly provided under this License.  Any attempt
otherwise to copy, modify, sublicense, or distribute it is void, and
will automatically terminate your rights under this License.
</p>
<p>However, if you cease all violation of this License, then your license
from a particular copyright holder is reinstated (a) provisionally,
unless and until the copyright holder explicitly and finally
terminates your license, and (b) permanently, if the copyright holder
fails to notify you of the violation by some reasonable means prior to
60 days after the cessation.
</p>
<p>Moreover, your license from a particular copyright holder is
reinstated permanently if the copyright holder notifies you of the
violation by some reasonable means, this is the first time you have
received notice of violation of this License (for any work) from that
copyright holder, and you cure the violation prior to 30 days after
your receipt of the notice.
</p>
<p>Termination of your rights under this section does not terminate the
licenses of parties who have received copies or rights from you under
this License.  If your rights have been terminated and not permanently
reinstated, receipt of a copy of some or all of the same material does
not give you any rights to use it.
</p>
</li><li> FUTURE REVISIONS OF THIS LICENSE

<p>The Free Software Foundation may publish new, revised versions
of the GNU Free Documentation License from time to time.  Such new
versions will be similar in spirit to the present version, but may
differ in detail to address new problems or concerns.  See
<a href="http://www.gnu.org/copyleft/">http://www.gnu.org/copyleft/</a>.
</p>
<p>Each version of the License is given a distinguishing version number.
If the Document specifies that a particular numbered version of this
License &ldquo;or any later version&rdquo; applies to it, you have the option of
following the terms and conditions either of that specified version or
of any later version that has been published (not as a draft) by the
Free Software Foundation.  If the Document does not specify a version
number of this License, you may choose any version ever published (not
as a draft) by the Free Software Foundation.  If the Document
specifies that a proxy can decide which future versions of this
License can be used, that proxy&rsquo;s public statement of acceptance of a
version permanently authorizes you to choose that version for the
Document.
</p>
</li><li> RELICENSING

<p>&ldquo;Massive Multiauthor Collaboration Site&rdquo; (or &ldquo;MMC Site&rdquo;) means any
World Wide Web server that publishes copyrightable works and also
provides prominent facilities for anybody to edit those works.  A
public wiki that anybody can edit is an example of such a server.  A
&ldquo;Massive Multiauthor Collaboration&rdquo; (or &ldquo;MMC&rdquo;) contained in the
site means any set of copyrightable works thus published on the MMC
site.
</p>
<p>&ldquo;CC-BY-SA&rdquo; means the Creative Commons Attribution-Share Alike 3.0
license published by Creative Commons Corporation, a not-for-profit
corporation with a principal place of business in San Francisco,
California, as well as future copyleft versions of that license
published by that same organization.
</p>
<p>&ldquo;Incorporate&rdquo; means to publish or republish a Document, in whole or
in part, as part of another Document.
</p>
<p>An MMC is &ldquo;eligible for relicensing&rdquo; if it is licensed under this
License, and if all works that were first published under this License
somewhere other than this MMC, and subsequently incorporated in whole
or in part into the MMC, (1) had no cover texts or invariant sections,
and (2) were thus incorporated prior to November 1, 2008.
</p>
<p>The operator of an MMC Site may republish an MMC contained in the site
under CC-BY-SA on the same site at any time before August 1, 2009,
provided the MMC is eligible for relicensing.
</p>
</li></ol>

<a name="ADDENDUM_003a-How-to-use-this-License-for-your-documents"></a>
<h3 class="unnumberedsec">ADDENDUM: How to use this License for your documents</h3>

<p>To use this License in a document you have written, include a copy of
the License in the document and put the following copyright and
license notices just after the title page:
</p>
<div class="smallexample">
<pre class="smallexample">  Copyright (C)  <var>year</var>  <var>your name</var>.
  Permission is granted to copy, distribute and/or modify this document
  under the terms of the GNU Free Documentation License, Version 1.3
  or any later version published by the Free Software Foundation;
  with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
  Texts.  A copy of the license is included in the section entitled ``GNU
  Free Documentation License''.
</pre></div>

<p>If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts,
replace the &ldquo;with...Texts.&rdquo; line with this:
</p>
<div class="smallexample">
<pre class="smallexample">    with the Invariant Sections being <var>list their titles</var>, with
    the Front-Cover Texts being <var>list</var>, and with the Back-Cover Texts
    being <var>list</var>.
</pre></div>

<p>If you have Invariant Sections without Cover Texts, or some other
combination of the three, merge those two alternatives to suit the
situation.
</p>
<p>If your document contains nontrivial examples of program code, we
recommend releasing these examples in parallel under your choice of
free software license, such as the GNU General Public License,
to permit their use in free software.
</p>



<hr>
<a name="Contributors"></a>
<div class="header">
<p>
Next: <a href="#Option-Index" accesskey="n" rel="next">Option Index</a>, Previous: <a href="#GNU-Free-Documentation-License" accesskey="p" rel="prev">GNU Free Documentation License</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Contributors-to-GCC"></a>
<h2 class="unnumbered">Contributors to GCC</h2>
<a name="index-contributors"></a>

<p>The GCC project would like to thank its many contributors.  Without them the
project would not have been nearly as successful as it has been.  Any omissions
in this list are accidental.  Feel free to contact
<a href="mailto:law@redhat.com">law@redhat.com</a> or <a href="mailto:gerald@pfeifer.com">gerald@pfeifer.com</a> if you have been left
out or some of your contributions are not listed.  Please keep this list in
alphabetical order.
</p>
<ul>
<li> Analog Devices helped implement the support for complex data types
and iterators.

</li><li> John David Anglin for threading-related fixes and improvements to
libstdc++-v3, and the HP-UX port.

</li><li> James van Artsdalen wrote the code that makes efficient use of
the Intel 80387 register stack.

</li><li> Abramo and Roberto Bagnara for the SysV68 Motorola 3300 Delta Series
port.

</li><li> Alasdair Baird for various bug fixes.

</li><li> Giovanni Bajo for analyzing lots of complicated C++ problem reports.

</li><li> Peter Barada for his work to improve code generation for new
ColdFire cores.

</li><li> Gerald Baumgartner added the signature extension to the C++ front end.

</li><li> Godmar Back for his Java improvements and encouragement.

</li><li> Scott Bambrough for help porting the Java compiler.

</li><li> Wolfgang Bangerth for processing tons of bug reports.

</li><li> Jon Beniston for his Microsoft Windows port of Java and port to Lattice Mico32.

</li><li> Daniel Berlin for better DWARF2 support, faster/better optimizations,
improved alias analysis, plus migrating GCC to Bugzilla.

</li><li> Geoff Berry for his Java object serialization work and various patches.

</li><li> David Binderman tests weekly snapshots of GCC trunk against Fedora Rawhide
for several architectures.

</li><li> Uros Bizjak for the implementation of x87 math built-in functions and
for various middle end and i386 back end improvements and bug fixes.

</li><li> Eric Blake for helping to make GCJ and libgcj conform to the
specifications.

</li><li> Janne Blomqvist for contributions to GNU Fortran.

</li><li> Segher Boessenkool for various fixes.

</li><li> Hans-J. Boehm for his <a href="http://www.hpl.hp.com/personal/Hans_Boehm/gc/">garbage collector</a>, IA-64 libffi port, and other Java work.

</li><li> Neil Booth for work on cpplib, lang hooks, debug hooks and other
miscellaneous clean-ups.

</li><li> Steven Bosscher for integrating the GNU Fortran front end into GCC and for
contributing to the tree-ssa branch.

</li><li> Eric Botcazou for fixing middle- and backend bugs left and right.

</li><li> Per Bothner for his direction via the steering committee and various
improvements to the infrastructure for supporting new languages.  Chill
front end implementation.  Initial implementations of
cpplib, fix-header, config.guess, libio, and past C++ library (libg++)
maintainer.  Dreaming up, designing and implementing much of GCJ.

</li><li> Devon Bowen helped port GCC to the Tahoe.

</li><li> Don Bowman for mips-vxworks contributions.

</li><li> Dave Brolley for work on cpplib and Chill.

</li><li> Paul Brook for work on the ARM architecture and maintaining GNU Fortran.

</li><li> Robert Brown implemented the support for Encore 32000 systems.

</li><li> Christian Bruel for improvements to local store elimination.

</li><li> Herman A.J. ten Brugge for various fixes.

</li><li> Joerg Brunsmann for Java compiler hacking and help with the GCJ FAQ.

</li><li> Joe Buck for his direction via the steering committee.

</li><li> Craig Burley for leadership of the G77 Fortran effort.

</li><li> Stephan Buys for contributing Doxygen notes for libstdc++.

</li><li> Paolo Carlini for libstdc++ work: lots of efficiency improvements to
the C++ strings, streambufs and formatted I/O, hard detective work on
the frustrating localization issues, and keeping up with the problem reports.

</li><li> John Carr for his alias work, SPARC hacking, infrastructure improvements,
previous contributions to the steering committee, loop optimizations, etc.

</li><li> Stephane Carrez for 68HC11 and 68HC12 ports.

</li><li> Steve Chamberlain for support for the Renesas SH and H8 processors
and the PicoJava processor, and for GCJ config fixes.

</li><li> Glenn Chambers for help with the GCJ FAQ.

</li><li> John-Marc Chandonia for various libgcj patches.

</li><li> Denis Chertykov for contributing and maintaining the AVR port, the first GCC port
for an 8-bit architecture.

</li><li> Scott Christley for his Objective-C contributions.

</li><li> Eric Christopher for his Java porting help and clean-ups.

</li><li> Branko Cibej for more warning contributions.

</li><li> The <a href="http://www.gnu.org/software/classpath/">GNU Classpath project</a>
for all of their merged runtime code.

</li><li> Nick Clifton for arm, mcore, fr30, v850, m32r, rx work,
<samp>--help</samp>, and other random hacking.

</li><li> Michael Cook for libstdc++ cleanup patches to reduce warnings.

</li><li> R. Kelley Cook for making GCC buildable from a read-only directory as
well as other miscellaneous build process and documentation clean-ups.

</li><li> Ralf Corsepius for SH testing and minor bug fixing.

</li><li> Stan Cox for care and feeding of the x86 port and lots of behind
the scenes hacking.

</li><li> Alex Crain provided changes for the 3b1.

</li><li> Ian Dall for major improvements to the NS32k port.

</li><li> Paul Dale for his work to add uClinux platform support to the
m68k backend.

</li><li> Dario Dariol contributed the four varieties of sample programs
that print a copy of their source.

</li><li> Russell Davidson for fstream and stringstream fixes in libstdc++.

</li><li> Bud Davis for work on the G77 and GNU Fortran compilers.

</li><li> Mo DeJong for GCJ and libgcj bug fixes.

</li><li> DJ Delorie for the DJGPP port, build and libiberty maintenance,
various bug fixes, and the M32C, MeP, and RL78 ports.

</li><li> Arnaud Desitter for helping to debug GNU Fortran.

</li><li> Gabriel Dos Reis for contributions to G++, contributions and
maintenance of GCC diagnostics infrastructure, libstdc++-v3,
including <code>valarray&lt;&gt;</code>, <code>complex&lt;&gt;</code>, maintaining the numerics library
(including that pesky <code>&lt;limits&gt;</code> :-) and keeping up-to-date anything
to do with numbers.

</li><li> Ulrich Drepper for his work on glibc, testing of GCC using glibc, ISO C99
support, CFG dumping support, etc., plus support of the C++ runtime
libraries including for all kinds of C interface issues, contributing and
maintaining <code>complex&lt;&gt;</code>, sanity checking and disbursement, configuration
architecture, libio maintenance, and early math work.

</li><li> Zdenek Dvorak for a new loop unroller and various fixes.

</li><li> Michael Eager for his work on the Xilinx MicroBlaze port.

</li><li> Richard Earnshaw for his ongoing work with the ARM.

</li><li> David Edelsohn for his direction via the steering committee, ongoing work
with the RS6000/PowerPC port, help cleaning up Haifa loop changes,
doing the entire AIX port of libstdc++ with his bare hands, and for
ensuring GCC properly keeps working on AIX.

</li><li> Kevin Ediger for the floating point formatting of num_put::do_put in
libstdc++.

</li><li> Phil Edwards for libstdc++ work including configuration hackery,
documentation maintainer, chief breaker of the web pages, the occasional
iostream bug fix, and work on shared library symbol versioning.

</li><li> Paul Eggert for random hacking all over GCC.

</li><li> Mark Elbrecht for various DJGPP improvements, and for libstdc++
configuration support for locales and fstream-related fixes.

</li><li> Vadim Egorov for libstdc++ fixes in strings, streambufs, and iostreams.

</li><li> Christian Ehrhardt for dealing with bug reports.

</li><li> Ben Elliston for his work to move the Objective-C runtime into its
own subdirectory and for his work on autoconf.

</li><li> Revital Eres for work on the PowerPC 750CL port.

</li><li> Marc Espie for OpenBSD support.

</li><li> Doug Evans for much of the global optimization framework, arc, m32r,
and SPARC work.

</li><li> Christopher Faylor for his work on the Cygwin port and for caring and
feeding the gcc.gnu.org box and saving its users tons of spam.

</li><li> Fred Fish for BeOS support and Ada fixes.

</li><li> Ivan Fontes Garcia for the Portuguese translation of the GCJ FAQ.

</li><li> Peter Gerwinski for various bug fixes and the Pascal front end.

</li><li> Kaveh R. Ghazi for his direction via the steering committee, amazing
work to make &lsquo;<samp>-W -Wall -W* -Werror</samp>&rsquo; useful, and 
testing GCC on a plethora of platforms.  Kaveh extends his gratitude to
the CAIP Center at Rutgers University for providing him with computing
resources to work on Free Software from the late 1980s to 2010.

</li><li> John Gilmore for a donation to the FSF earmarked improving GNU Java.

</li><li> Judy Goldberg for c++ contributions.

</li><li> Torbjorn Granlund for various fixes and the c-torture testsuite,
multiply- and divide-by-constant optimization, improved long long
support, improved leaf function register allocation, and his direction
via the steering committee.

</li><li> Anthony Green for his <samp>-Os</samp> contributions, the moxie port, and
Java front end work.

</li><li> Stu Grossman for gdb hacking, allowing GCJ developers to debug Java code.

</li><li> Michael K. Gschwind contributed the port to the PDP-11.

</li><li> Richard Guenther for his ongoing middle-end contributions and bug fixes
and for release management.

</li><li> Ron Guilmette implemented the <code>protoize</code> and <code>unprotoize</code>
tools, the support for Dwarf symbolic debugging information, and much of
the support for System V Release 4.  He has also worked heavily on the
Intel 386 and 860 support.

</li><li> Sumanth Gundapaneni for contributing the CR16 port.

</li><li> Mostafa Hagog for Swing Modulo Scheduling (SMS) and post reload GCSE.

</li><li> Bruno Haible for improvements in the runtime overhead for EH, new
warnings and assorted bug fixes.

</li><li> Andrew Haley for his amazing Java compiler and library efforts.

</li><li> Chris Hanson assisted in making GCC work on HP-UX for the 9000 series 300.

</li><li> Michael Hayes for various thankless work he&rsquo;s done trying to get
the c30/c40 ports functional.  Lots of loop and unroll improvements and
fixes.

</li><li> Dara Hazeghi for wading through myriads of target-specific bug reports.

</li><li> Kate Hedstrom for staking the G77 folks with an initial testsuite.

</li><li> Richard Henderson for his ongoing SPARC, alpha, ia32, and ia64 work, loop
opts, and generally fixing lots of old problems we&rsquo;ve ignored for
years, flow rewrite and lots of further stuff, including reviewing
tons of patches.

</li><li> Aldy Hernandez for working on the PowerPC port, SIMD support, and
various fixes.

</li><li> Nobuyuki Hikichi of Software Research Associates, Tokyo, contributed
the support for the Sony NEWS machine.

</li><li> Kazu Hirata for caring and feeding the Renesas H8/300 port and various fixes.

</li><li> Katherine Holcomb for work on GNU Fortran.

</li><li> Manfred Hollstein for his ongoing work to keep the m88k alive, lots
of testing and bug fixing, particularly of GCC configury code.

</li><li> Steve Holmgren for MachTen patches.

</li><li> Mat Hostetter for work on the TILE-Gx and TILEPro ports.

</li><li> Jan Hubicka for his x86 port improvements.

</li><li> Falk Hueffner for working on C and optimization bug reports.

</li><li> Bernardo Innocenti for his m68k work, including merging of
ColdFire improvements and uClinux support.

</li><li> Christian Iseli for various bug fixes.

</li><li> Kamil Iskra for general m68k hacking.

</li><li> Lee Iverson for random fixes and MIPS testing.

</li><li> Andreas Jaeger for testing and benchmarking of GCC and various bug fixes.

</li><li> Jakub Jelinek for his SPARC work and sibling call optimizations as well
as lots of bug fixes and test cases, and for improving the Java build
system.

</li><li> Janis Johnson for ia64 testing and fixes, her quality improvement
sidetracks, and web page maintenance.

</li><li> Kean Johnston for SCO OpenServer support and various fixes.

</li><li> Tim Josling for the sample language treelang based originally on Richard
Kenner&rsquo;s &ldquo;toy&rdquo; language.

</li><li> Nicolai Josuttis for additional libstdc++ documentation.

</li><li> Klaus Kaempf for his ongoing work to make alpha-vms a viable target.

</li><li> Steven G. Kargl for work on GNU Fortran.

</li><li> David Kashtan of SRI adapted GCC to VMS.

</li><li> Ryszard Kabatek for many, many libstdc++ bug fixes and optimizations of
strings, especially member functions, and for auto_ptr fixes.

</li><li> Geoffrey Keating for his ongoing work to make the PPC work for GNU/Linux
and his automatic regression tester.

</li><li> Brendan Kehoe for his ongoing work with G++ and for a lot of early work
in just about every part of libstdc++.

</li><li> Oliver M. Kellogg of Deutsche Aerospace contributed the port to the
MIL-STD-1750A.

</li><li> Richard Kenner of the New York University Ultracomputer Research
Laboratory wrote the machine descriptions for the AMD 29000, the DEC
Alpha, the IBM RT PC, and the IBM RS/6000 as well as the support for
instruction attributes.  He also made changes to better support RISC
processors including changes to common subexpression elimination,
strength reduction, function calling sequence handling, and condition
code support, in addition to generalizing the code for frame pointer
elimination and delay slot scheduling.  Richard Kenner was also the
head maintainer of GCC for several years.

</li><li> Mumit Khan for various contributions to the Cygwin and Mingw32 ports and
maintaining binary releases for Microsoft Windows hosts, and for massive libstdc++
porting work to Cygwin/Mingw32.

</li><li> Robin Kirkham for cpu32 support.

</li><li> Mark Klein for PA improvements.

</li><li> Thomas Koenig for various bug fixes.

</li><li> Bruce Korb for the new and improved fixincludes code.

</li><li> Benjamin Kosnik for his G++ work and for leading the libstdc++-v3 effort.

</li><li> Charles LaBrec contributed the support for the Integrated Solutions
68020 system.

</li><li> Asher Langton and Mike Kumbera for contributing Cray pointer support
to GNU Fortran, and for other GNU Fortran improvements.

</li><li> Jeff Law for his direction via the steering committee, coordinating the
entire egcs project and GCC 2.95, rolling out snapshots and releases,
handling merges from GCC2, reviewing tons of patches that might have
fallen through the cracks else, and random but extensive hacking.

</li><li> Walter Lee for work on the TILE-Gx and TILEPro ports.

</li><li> Marc Lehmann for his direction via the steering committee and helping
with analysis and improvements of x86 performance.

</li><li> Victor Leikehman for work on GNU Fortran.

</li><li> Ted Lemon wrote parts of the RTL reader and printer.

</li><li> Kriang Lerdsuwanakij for C++ improvements including template as template
parameter support, and many C++ fixes.

</li><li> Warren Levy for tremendous work on libgcj (Java Runtime Library) and
random work on the Java front end.

</li><li> Alain Lichnewsky ported GCC to the MIPS CPU.

</li><li> Oskar Liljeblad for hacking on AWT and his many Java bug reports and
patches.

</li><li> Robert Lipe for OpenServer support, new testsuites, testing, etc.

</li><li> Chen Liqin for various S+core related fixes/improvement, and for
maintaining the S+core port.

</li><li> Weiwen Liu for testing and various bug fixes.

</li><li> Manuel L&oacute;pez-Ib&aacute;&ntilde;ez for improving <samp>-Wconversion</samp> and
many other diagnostics fixes and improvements.

</li><li> Dave Love for his ongoing work with the Fortran front end and
runtime libraries.

</li><li> Martin von L&ouml;wis for internal consistency checking infrastructure,
various C++ improvements including namespace support, and tons of
assistance with libstdc++/compiler merges.

</li><li> H.J. Lu for his previous contributions to the steering committee, many x86
bug reports, prototype patches, and keeping the GNU/Linux ports working.

</li><li> Greg McGary for random fixes and (someday) bounded pointers.

</li><li> Andrew MacLeod for his ongoing work in building a real EH system,
various code generation improvements, work on the global optimizer, etc.

</li><li> Vladimir Makarov for hacking some ugly i960 problems, PowerPC hacking
improvements to compile-time performance, overall knowledge and
direction in the area of instruction scheduling, and design and
implementation of the automaton based instruction scheduler.

</li><li> Bob Manson for his behind the scenes work on dejagnu.

</li><li> Philip Martin for lots of libstdc++ string and vector iterator fixes and
improvements, and string clean up and testsuites.

</li><li> All of the Mauve project
<a href="http://sourceware.org/cgi-bin/cvsweb.cgi/~checkout~/mauve/THANKS?rev=1.2&amp;cvsroot=mauve&amp;only_with_tag=HEAD">contributors</a>,
for Java test code.

</li><li> Bryce McKinlay for numerous GCJ and libgcj fixes and improvements.

</li><li> Adam Megacz for his work on the Microsoft Windows port of GCJ.

</li><li> Michael Meissner for LRS framework, ia32, m32r, v850, m88k, MIPS,
powerpc, haifa, ECOFF debug support, and other assorted hacking.

</li><li> Jason Merrill for his direction via the steering committee and leading
the G++ effort.

</li><li> Martin Michlmayr for testing GCC on several architectures using the
entire Debian archive.

</li><li> David Miller for his direction via the steering committee, lots of
SPARC work, improvements in jump.c and interfacing with the Linux kernel
developers.

</li><li> Gary Miller ported GCC to Charles River Data Systems machines.

</li><li> Alfred Minarik for libstdc++ string and ios bug fixes, and turning the
entire libstdc++ testsuite namespace-compatible.

</li><li> Mark Mitchell for his direction via the steering committee, mountains of
C++ work, load/store hoisting out of loops, alias analysis improvements,
ISO C <code>restrict</code> support, and serving as release manager from 2000
to 2011.

</li><li> Alan Modra for various GNU/Linux bits and testing.

</li><li> Toon Moene for his direction via the steering committee, Fortran
maintenance, and his ongoing work to make us make Fortran run fast.

</li><li> Jason Molenda for major help in the care and feeding of all the services
on the gcc.gnu.org (formerly egcs.cygnus.com) machine&mdash;mail, web
services, ftp services, etc etc.  Doing all this work on scrap paper and
the backs of envelopes would have been&hellip; difficult.

</li><li> Catherine Moore for fixing various ugly problems we have sent her
way, including the haifa bug which was killing the Alpha &amp; PowerPC
Linux kernels.

</li><li> Mike Moreton for his various Java patches.

</li><li> David Mosberger-Tang for various Alpha improvements, and for the initial
IA-64 port.

</li><li> Stephen Moshier contributed the floating point emulator that assists in
cross-compilation and permits support for floating point numbers wider
than 64 bits and for ISO C99 support.

</li><li> Bill Moyer for his behind the scenes work on various issues.

</li><li> Philippe De Muyter for his work on the m68k port.

</li><li> Joseph S. Myers for his work on the PDP-11 port, format checking and ISO
C99 support, and continuous emphasis on (and contributions to) documentation.

</li><li> Nathan Myers for his work on libstdc++-v3: architecture and authorship
through the first three snapshots, including implementation of locale
infrastructure, string, shadow C headers, and the initial project
documentation (DESIGN, CHECKLIST, and so forth).  Later, more work on
MT-safe string and shadow headers.

</li><li> Felix Natter for documentation on porting libstdc++.

</li><li> Nathanael Nerode for cleaning up the configuration/build process.

</li><li> NeXT, Inc. donated the front end that supports the Objective-C
language.

</li><li> Hans-Peter Nilsson for the CRIS and MMIX ports, improvements to the search
engine setup, various documentation fixes and other small fixes.

</li><li> Geoff Noer for his work on getting cygwin native builds working.

</li><li> Diego Novillo for his work on Tree SSA, OpenMP, SPEC performance
tracking web pages, GIMPLE tuples, and assorted fixes.

</li><li> David O&rsquo;Brien for the FreeBSD/alpha, FreeBSD/AMD x86-64, FreeBSD/ARM,
FreeBSD/PowerPC, and FreeBSD/SPARC64 ports and related infrastructure
improvements.

</li><li> Alexandre Oliva for various build infrastructure improvements, scripts and
amazing testing work, including keeping libtool issues sane and happy.

</li><li> Stefan Olsson for work on mt_alloc.

</li><li> Melissa O&rsquo;Neill for various NeXT fixes.

</li><li> Rainer Orth for random MIPS work, including improvements to GCC&rsquo;s o32
ABI support, improvements to dejagnu&rsquo;s MIPS support, Java configuration
clean-ups and porting work, and maintaining the IRIX, Solaris 2, and
Tru64 UNIX ports.

</li><li> Hartmut Penner for work on the s390 port.

</li><li> Paul Petersen wrote the machine description for the Alliant FX/8.

</li><li> Alexandre Petit-Bianco for implementing much of the Java compiler and
continued Java maintainership.

</li><li> Matthias Pfaller for major improvements to the NS32k port.

</li><li> Gerald Pfeifer for his direction via the steering committee, pointing
out lots of problems we need to solve, maintenance of the web pages, and
taking care of documentation maintenance in general.

</li><li> Andrew Pinski for processing bug reports by the dozen.

</li><li> Ovidiu Predescu for his work on the Objective-C front end and runtime
libraries.

</li><li> Jerry Quinn for major performance improvements in C++ formatted I/O.

</li><li> Ken Raeburn for various improvements to checker, MIPS ports and various
cleanups in the compiler.

</li><li> Rolf W. Rasmussen for hacking on AWT.

</li><li> David Reese of Sun Microsystems contributed to the Solaris on PowerPC
port.

</li><li> Volker Reichelt for keeping up with the problem reports.

</li><li> Joern Rennecke for maintaining the sh port, loop, regmove &amp; reload
hacking and developing and maintaining the Epiphany port.

</li><li> Loren J. Rittle for improvements to libstdc++-v3 including the FreeBSD
port, threading fixes, thread-related configury changes, critical
threading documentation, and solutions to really tricky I/O problems,
as well as keeping GCC properly working on FreeBSD and continuous testing.

</li><li> Craig Rodrigues for processing tons of bug reports.

</li><li> Ola R&ouml;nnerup for work on mt_alloc.

</li><li> Gavin Romig-Koch for lots of behind the scenes MIPS work.

</li><li> David Ronis inspired and encouraged Craig to rewrite the G77
documentation in texinfo format by contributing a first pass at a
translation of the old <samp>g77-0.5.16/f/DOC</samp> file.

</li><li> Ken Rose for fixes to GCC&rsquo;s delay slot filling code.

</li><li> Paul Rubin wrote most of the preprocessor.

</li><li> P&eacute;tur Run&oacute;lfsson for major performance improvements in C++ formatted I/O and
large file support in C++ filebuf.

</li><li> Chip Salzenberg for libstdc++ patches and improvements to locales, traits,
Makefiles, libio, libtool hackery, and &ldquo;long long&rdquo; support.

</li><li> Juha Sarlin for improvements to the H8 code generator.

</li><li> Greg Satz assisted in making GCC work on HP-UX for the 9000 series 300.

</li><li> Roger Sayle for improvements to constant folding and GCC&rsquo;s RTL optimizers
as well as for fixing numerous bugs.

</li><li> Bradley Schatz for his work on the GCJ FAQ.

</li><li> Peter Schauer wrote the code to allow debugging to work on the Alpha.

</li><li> William Schelter did most of the work on the Intel 80386 support.

</li><li> Tobias Schl&uuml;ter for work on GNU Fortran.

</li><li> Bernd Schmidt for various code generation improvements and major
work in the reload pass, serving as release manager for
GCC 2.95.3, and work on the Blackfin and C6X ports.

</li><li> Peter Schmid for constant testing of libstdc++&mdash;especially application
testing, going above and beyond what was requested for the release
criteria&mdash;and libstdc++ header file tweaks.

</li><li> Jason Schroeder for jcf-dump patches.

</li><li> Andreas Schwab for his work on the m68k port.

</li><li> Lars Segerlund for work on GNU Fortran.

</li><li> Dodji Seketeli for numerous C++ bug fixes and debug info improvements.

</li><li> Joel Sherrill for his direction via the steering committee, RTEMS
contributions and RTEMS testing.

</li><li> Nathan Sidwell for many C++ fixes/improvements.

</li><li> Jeffrey Siegal for helping RMS with the original design of GCC, some
code which handles the parse tree and RTL data structures, constant
folding and help with the original VAX &amp; m68k ports.

</li><li> Kenny Simpson for prompting libstdc++ fixes due to defect reports from
the LWG (thereby keeping GCC in line with updates from the ISO).

</li><li> Franz Sirl for his ongoing work with making the PPC port stable
for GNU/Linux.

</li><li> Andrey Slepuhin for assorted AIX hacking.

</li><li> Trevor Smigiel for contributing the SPU port.

</li><li> Christopher Smith did the port for Convex machines.

</li><li> Danny Smith for his major efforts on the Mingw (and Cygwin) ports.

</li><li> Randy Smith finished the Sun FPA support.

</li><li> Scott Snyder for queue, iterator, istream, and string fixes and libstdc++
testsuite entries.  Also for providing the patch to G77 to add
rudimentary support for <code>INTEGER*1</code>, <code>INTEGER*2</code>, and
<code>LOGICAL*1</code>.

</li><li> Zdenek Sojka for running automated regression testing of GCC and reporting
numerous bugs.

</li><li> Jayant Sonar for contributing the CR16 port.

</li><li> Brad Spencer for contributions to the GLIBCPP_FORCE_NEW technique.

</li><li> Richard Stallman, for writing the original GCC and launching the GNU project.

</li><li> Jan Stein of the Chalmers Computer Society provided support for
Genix, as well as part of the 32000 machine description.

</li><li> Nigel Stephens for various mips16 related fixes/improvements.

</li><li> Jonathan Stone wrote the machine description for the Pyramid computer.

</li><li> Graham Stott for various infrastructure improvements.

</li><li> John Stracke for his Java HTTP protocol fixes.

</li><li> Mike Stump for his Elxsi port, G++ contributions over the years and more
recently his vxworks contributions

</li><li> Jeff Sturm for Java porting help, bug fixes, and encouragement.

</li><li> Shigeya Suzuki for this fixes for the bsdi platforms.

</li><li> Ian Lance Taylor for the Go frontend, the initial mips16 and mips64
support, general configury hacking, fixincludes, etc.

</li><li> Holger Teutsch provided the support for the Clipper CPU.

</li><li> Gary Thomas for his ongoing work to make the PPC work for GNU/Linux.

</li><li> Philipp Thomas for random bug fixes throughout the compiler

</li><li> Jason Thorpe for thread support in libstdc++ on NetBSD.

</li><li> Kresten Krab Thorup wrote the run time support for the Objective-C
language and the fantastic Java bytecode interpreter.

</li><li> Michael Tiemann for random bug fixes, the first instruction scheduler,
initial C++ support, function integration, NS32k, SPARC and M88k
machine description work, delay slot scheduling.

</li><li> Andreas Tobler for his work porting libgcj to Darwin.

</li><li> Teemu Torma for thread safe exception handling support.

</li><li> Leonard Tower wrote parts of the parser, RTL generator, and RTL
definitions, and of the VAX machine description.

</li><li> Daniel Towner and Hariharan Sandanagobalane contributed and
maintain the picoChip port.

</li><li> Tom Tromey for internationalization support and for his many Java
contributions and libgcj maintainership.

</li><li> Lassi Tuura for improvements to config.guess to determine HP processor
types.

</li><li> Petter Urkedal for libstdc++ CXXFLAGS, math, and algorithms fixes.

</li><li> Andy Vaught for the design and initial implementation of the GNU Fortran
front end.

</li><li> Brent Verner for work with the libstdc++ cshadow files and their
associated configure steps.

</li><li> Todd Vierling for contributions for NetBSD ports.

</li><li> Jonathan Wakely for contributing libstdc++ Doxygen notes and XHTML
guidance.

</li><li> Dean Wakerley for converting the install documentation from HTML to texinfo
in time for GCC 3.0.

</li><li> Krister Walfridsson for random bug fixes.

</li><li> Feng Wang for contributions to GNU Fortran.

</li><li> Stephen M. Webb for time and effort on making libstdc++ shadow files
work with the tricky Solaris 8+ headers, and for pushing the build-time
header tree.

</li><li> John Wehle for various improvements for the x86 code generator,
related infrastructure improvements to help x86 code generation,
value range propagation and other work, WE32k port.

</li><li> Ulrich Weigand for work on the s390 port.

</li><li> Zack Weinberg for major work on cpplib and various other bug fixes.

</li><li> Matt Welsh for help with Linux Threads support in GCJ.

</li><li> Urban Widmark for help fixing java.io.

</li><li> Mark Wielaard for new Java library code and his work integrating with
Classpath.

</li><li> Dale Wiles helped port GCC to the Tahoe.

</li><li> Bob Wilson from Tensilica, Inc. for the Xtensa port.

</li><li> Jim Wilson for his direction via the steering committee, tackling hard
problems in various places that nobody else wanted to work on, strength
reduction and other loop optimizations.

</li><li> Paul Woegerer and Tal Agmon for the CRX port.

</li><li> Carlo Wood for various fixes.

</li><li> Tom Wood for work on the m88k port.

</li><li> Canqun Yang for work on GNU Fortran.

</li><li> Masanobu Yuhara of Fujitsu Laboratories implemented the machine
description for the Tron architecture (specifically, the Gmicro).

</li><li> Kevin Zachmann helped port GCC to the Tahoe.

</li><li> Ayal Zaks for Swing Modulo Scheduling (SMS).

</li><li> Xiaoqiang Zhang for work on GNU Fortran.

</li><li> Gilles Zunino for help porting Java to Irix.

</li></ul>

<p>The following people are recognized for their contributions to GNAT,
the Ada front end of GCC:
</p><ul>
<li> Bernard Banner

</li><li> Romain Berrendonner

</li><li> Geert Bosch

</li><li> Emmanuel Briot

</li><li> Joel Brobecker

</li><li> Ben Brosgol

</li><li> Vincent Celier

</li><li> Arnaud Charlet

</li><li> Chien Chieng

</li><li> Cyrille Comar

</li><li> Cyrille Crozes

</li><li> Robert Dewar

</li><li> Gary Dismukes

</li><li> Robert Duff

</li><li> Ed Falis

</li><li> Ramon Fernandez

</li><li> Sam Figueroa

</li><li> Vasiliy Fofanov

</li><li> Michael Friess

</li><li> Franco Gasperoni

</li><li> Ted Giering

</li><li> Matthew Gingell

</li><li> Laurent Guerby

</li><li> Jerome Guitton

</li><li> Olivier Hainque

</li><li> Jerome Hugues

</li><li> Hristian Kirtchev

</li><li> Jerome Lambourg

</li><li> Bruno Leclerc

</li><li> Albert Lee

</li><li> Sean McNeil

</li><li> Javier Miranda

</li><li> Laurent Nana

</li><li> Pascal Obry

</li><li> Dong-Ik Oh

</li><li> Laurent Pautet

</li><li> Brett Porter

</li><li> Thomas Quinot

</li><li> Nicolas Roche

</li><li> Pat Rogers

</li><li> Jose Ruiz

</li><li> Douglas Rupp

</li><li> Sergey Rybin

</li><li> Gail Schenker

</li><li> Ed Schonberg

</li><li> Nicolas Setton

</li><li> Samuel Tardieu

</li></ul>


<p>The following people are recognized for their contributions of new
features, bug reports, testing and integration of classpath/libgcj for
GCC version 4.1:
</p><ul>
<li> Lillian Angel for <code>JTree</code> implementation and lots Free Swing
additions and bug fixes.

</li><li> Wolfgang Baer for <code>GapContent</code> bug fixes.

</li><li> Anthony Balkissoon for <code>JList</code>, Free Swing 1.5 updates and mouse event
fixes, lots of Free Swing work including <code>JTable</code> editing.

</li><li> Stuart Ballard for RMI constant fixes.

</li><li> Goffredo Baroncelli for <code>HTTPURLConnection</code> fixes.

</li><li> Gary Benson for <code>MessageFormat</code> fixes.

</li><li> Daniel Bonniot for <code>Serialization</code> fixes.

</li><li> Chris Burdess for lots of gnu.xml and http protocol fixes, <code>StAX</code>
and <code>DOM xml:id</code> support.

</li><li> Ka-Hing Cheung for <code>TreePath</code> and <code>TreeSelection</code> fixes.

</li><li> Archie Cobbs for build fixes, VM interface updates,
<code>URLClassLoader</code> updates.

</li><li> Kelley Cook for build fixes.

</li><li> Martin Cordova for Suggestions for better <code>SocketTimeoutException</code>.

</li><li> David Daney for <code>BitSet</code> bug fixes, <code>HttpURLConnection</code>
rewrite and improvements.

</li><li> Thomas Fitzsimmons for lots of upgrades to the gtk+ AWT and Cairo 2D
support. Lots of imageio framework additions, lots of AWT and Free
Swing bug fixes.

</li><li> Jeroen Frijters for <code>ClassLoader</code> and nio cleanups, serialization fixes,
better <code>Proxy</code> support, bug fixes and IKVM integration.

</li><li> Santiago Gala for <code>AccessControlContext</code> fixes.

</li><li> Nicolas Geoffray for <code>VMClassLoader</code> and <code>AccessController</code>
improvements.

</li><li> David Gilbert for <code>basic</code> and <code>metal</code> icon and plaf support
and lots of documenting, Lots of Free Swing and metal theme
additions. <code>MetalIconFactory</code> implementation.

</li><li> Anthony Green for <code>MIDI</code> framework, <code>ALSA</code> and <code>DSSI</code>
providers.

</li><li> Andrew Haley for <code>Serialization</code> and <code>URLClassLoader</code> fixes,
gcj build speedups.

</li><li> Kim Ho for <code>JFileChooser</code> implementation.

</li><li> Andrew John Hughes for <code>Locale</code> and net fixes, URI RFC2986
updates, <code>Serialization</code> fixes, <code>Properties</code> XML support and
generic branch work, VMIntegration guide update.

</li><li> Bastiaan Huisman for <code>TimeZone</code> bug fixing.

</li><li> Andreas Jaeger for mprec updates.

</li><li> Paul Jenner for better <samp>-Werror</samp> support.

</li><li> Ito Kazumitsu for <code>NetworkInterface</code> implementation and updates.

</li><li> Roman Kennke for <code>BoxLayout</code>, <code>GrayFilter</code> and
<code>SplitPane</code>, plus bug fixes all over. Lots of Free Swing work
including styled text.

</li><li> Simon Kitching for <code>String</code> cleanups and optimization suggestions.

</li><li> Michael Koch for configuration fixes, <code>Locale</code> updates, bug and
build fixes.

</li><li> Guilhem Lavaux for configuration, thread and channel fixes and Kaffe
integration. JCL native <code>Pointer</code> updates. Logger bug fixes.

</li><li> David Lichteblau for JCL support library global/local reference
cleanups.

</li><li> Aaron Luchko for JDWP updates and documentation fixes.

</li><li> Ziga Mahkovec for <code>Graphics2D</code> upgraded to Cairo 0.5 and new regex
features.

</li><li> Sven de Marothy for BMP imageio support, CSS and <code>TextLayout</code>
fixes. <code>GtkImage</code> rewrite, 2D, awt, free swing and date/time fixes and
implementing the Qt4 peers.

</li><li> Casey Marshall for crypto algorithm fixes, <code>FileChannel</code> lock,
<code>SystemLogger</code> and <code>FileHandler</code> rotate implementations, NIO
<code>FileChannel.map</code> support, security and policy updates.

</li><li> Bryce McKinlay for RMI work.

</li><li> Audrius Meskauskas for lots of Free Corba, RMI and HTML work plus
testing and documenting.

</li><li> Kalle Olavi Niemitalo for build fixes.

</li><li> Rainer Orth for build fixes.

</li><li> Andrew Overholt for <code>File</code> locking fixes.

</li><li> Ingo Proetel for <code>Image</code>, <code>Logger</code> and <code>URLClassLoader</code>
updates.

</li><li> Olga Rodimina for <code>MenuSelectionManager</code> implementation.

</li><li> Jan Roehrich for <code>BasicTreeUI</code> and <code>JTree</code> fixes.

</li><li> Julian Scheid for documentation updates and gjdoc support.

</li><li> Christian Schlichtherle for zip fixes and cleanups.

</li><li> Robert Schuster for documentation updates and beans fixes,
<code>TreeNode</code> enumerations and <code>ActionCommand</code> and various
fixes, XML and URL, AWT and Free Swing bug fixes.

</li><li> Keith Seitz for lots of JDWP work.

</li><li> Christian Thalinger for 64-bit cleanups, Configuration and VM
interface fixes and <code>CACAO</code> integration, <code>fdlibm</code> updates.

</li><li> Gael Thomas for <code>VMClassLoader</code> boot packages support suggestions.

</li><li> Andreas Tobler for Darwin and Solaris testing and fixing, <code>Qt4</code>
support for Darwin/OS X, <code>Graphics2D</code> support, <code>gtk+</code>
updates.

</li><li> Dalibor Topic for better <code>DEBUG</code> support, build cleanups and
Kaffe integration. <code>Qt4</code> build infrastructure, <code>SHA1PRNG</code>
and <code>GdkPixbugDecoder</code> updates.

</li><li> Tom Tromey for Eclipse integration, generics work, lots of bug fixes
and gcj integration including coordinating The Big Merge.

</li><li> Mark Wielaard for bug fixes, packaging and release management,
<code>Clipboard</code> implementation, system call interrupts and network
timeouts and <code>GdkPixpufDecoder</code> fixes.

</li></ul>


<p>In addition to the above, all of which also contributed time and energy in
testing GCC, we would like to thank the following for their contributions
to testing:
</p>
<ul>
<li> Michael Abd-El-Malek

</li><li> Thomas Arend

</li><li> Bonzo Armstrong

</li><li> Steven Ashe

</li><li> Chris Baldwin

</li><li> David Billinghurst

</li><li> Jim Blandy

</li><li> Stephane Bortzmeyer

</li><li> Horst von Brand

</li><li> Frank Braun

</li><li> Rodney Brown

</li><li> Sidney Cadot

</li><li> Bradford Castalia

</li><li> Robert Clark

</li><li> Jonathan Corbet

</li><li> Ralph Doncaster

</li><li> Richard Emberson

</li><li> Levente Farkas

</li><li> Graham Fawcett

</li><li> Mark Fernyhough

</li><li> Robert A. French

</li><li> J&ouml;rgen Freyh

</li><li> Mark K. Gardner

</li><li> Charles-Antoine Gauthier

</li><li> Yung Shing Gene

</li><li> David Gilbert

</li><li> Simon Gornall

</li><li> Fred Gray

</li><li> John Griffin

</li><li> Patrik Hagglund

</li><li> Phil Hargett

</li><li> Amancio Hasty

</li><li> Takafumi Hayashi

</li><li> Bryan W. Headley

</li><li> Kevin B. Hendricks

</li><li> Joep Jansen

</li><li> Christian Joensson

</li><li> Michel Kern

</li><li> David Kidd

</li><li> Tobias Kuipers

</li><li> Anand Krishnaswamy

</li><li> A. O. V. Le Blanc

</li><li> llewelly

</li><li> Damon Love

</li><li> Brad Lucier

</li><li> Matthias Klose

</li><li> Martin Knoblauch

</li><li> Rick Lutowski

</li><li> Jesse Macnish

</li><li> Stefan Morrell

</li><li> Anon A. Mous

</li><li> Matthias Mueller

</li><li> Pekka Nikander

</li><li> Rick Niles

</li><li> Jon Olson

</li><li> Magnus Persson

</li><li> Chris Pollard

</li><li> Richard Polton

</li><li> Derk Reefman

</li><li> David Rees

</li><li> Paul Reilly

</li><li> Tom Reilly

</li><li> Torsten Rueger

</li><li> Danny Sadinoff

</li><li> Marc Schifer

</li><li> Erik Schnetter

</li><li> Wayne K. Schroll

</li><li> David Schuler

</li><li> Vin Shelton

</li><li> Tim Souder

</li><li> Adam Sulmicki

</li><li> Bill Thorson

</li><li> George Talbot

</li><li> Pedro A. M. Vazquez

</li><li> Gregory Warnes

</li><li> Ian Watson

</li><li> David E. Young

</li><li> And many others
</li></ul>

<p>And finally we&rsquo;d like to thank everyone who uses the compiler, provides
feedback and generally reminds us why we&rsquo;re doing this work in the first
place.
</p>

<hr>
<a name="Option-Index"></a>
<div class="header">
<p>
Next: <a href="#Concept-Index" accesskey="n" rel="next">Concept Index</a>, Previous: <a href="#Contributors" accesskey="p" rel="prev">Contributors</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Option-Index-1"></a>
<h2 class="unnumbered">Option Index</h2>

<p>GCC&rsquo;s command line options are indexed here without any initial &lsquo;<samp>-</samp>&rsquo;
or &lsquo;<samp>--</samp>&rsquo;.  Where an option has both positive and negative forms
(such as <samp>-f<var>option</var></samp> and <samp>-fno-<var>option</var></samp>),
relevant entries in the manual are indexed under the most appropriate
form; it may sometimes be useful to look up both forms.
</p>
<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Option-Index_op_letter-F"><b>F</b></a>
 &nbsp; 
<a class="summary-letter" href="#Option-Index_op_letter-M"><b>M</b></a>
 &nbsp; 
</td></tr></table>
<table class="index-op" border="0">
<tr><td></td><th align="left">Index Entry</th><td>&nbsp;</td><th align="left"> Section</th></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Option-Index_op_letter-F">F</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-fltrans"><code>fltrans</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Internal-flags">Internal flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fltrans_002doutput_002dlist"><code>fltrans-output-list</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Internal-flags">Internal flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fwpa"><code>fwpa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Internal-flags">Internal flags</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Option-Index_op_letter-M">M</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-msoft_002dfloat"><code>msoft-float</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
</table>
<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Option-Index_op_letter-F"><b>F</b></a>
 &nbsp; 
<a class="summary-letter" href="#Option-Index_op_letter-M"><b>M</b></a>
 &nbsp; 
</td></tr></table>

<hr>
<a name="Concept-Index"></a>
<div class="header">
<p>
Previous: <a href="#Option-Index" accesskey="p" rel="prev">Option Index</a>, Up: <a href="#Top" accesskey="u" rel="up">Top</a> &nbsp; [<a href="#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="#Option-Index" title="Index" rel="index">Index</a>]</p>
</div>
<a name="Concept-Index-1"></a>
<h2 class="unnumbered">Concept Index</h2>

<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Concept-Index_cp_symbol-1"><b>!</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-2"><b>#</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-3"><b>%</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-4"><b>&amp;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-5"><b>(</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-6"><b>*</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-7"><b>+</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-8"><b>-</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-9"><b>/</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-10"><b>0</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-11"><b>&lt;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-12"><b>=</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-13"><b>&gt;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-14"><b>?</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-15"><b>\</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-16"><b>_</b></a>
 &nbsp; 
<br>
<a class="summary-letter" href="#Concept-Index_cp_letter-A"><b>A</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-B"><b>B</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-C"><b>C</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-D"><b>D</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-E"><b>E</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-F"><b>F</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-G"><b>G</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-H"><b>H</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-I"><b>I</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-J"><b>J</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-L"><b>L</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-M"><b>M</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-N"><b>N</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-O"><b>O</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-P"><b>P</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-Q"><b>Q</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-R"><b>R</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-S"><b>S</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-T"><b>T</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-U"><b>U</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-V"><b>V</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-W"><b>W</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-X"><b>X</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-Z"><b>Z</b></a>
 &nbsp; 
</td></tr></table>
<table class="index-cp" border="0">
<tr><td></td><th align="left">Index Entry</th><td>&nbsp;</td><th align="left"> Section</th></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-1">!</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0021-in-constraint">&lsquo;<samp>!</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Multi_002dAlternative">Multi-Alternative</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-2">#</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0023-in-constraint">&lsquo;<samp>#</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0023-in-template"><code>#</code> in template</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0023pragma"><code>#pragma</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-3">%</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0025-in-constraint">&lsquo;<samp>%</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0025-in-GTY-option">% in GTY option</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0025-in-template">&lsquo;<samp>%</samp>&rsquo; in template</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-4">&amp;</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0026-in-constraint">&lsquo;<samp>&amp;</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-5">(</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_0028nil_0029">(nil)</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-6">*</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002a-in-constraint">&lsquo;<samp>*</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002a-in-template"><code>*</code> in template</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-7">+</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002b-in-constraint">&lsquo;<samp>+</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-8">-</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002dfsection_002danchors"><samp>-fsection-anchors</samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002dfsection_002danchors-1"><samp>-fsection-anchors</samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-9">/</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fc-in-RTL-dump">&lsquo;<samp>/c</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002ff-in-RTL-dump">&lsquo;<samp>/f</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fi-in-RTL-dump">&lsquo;<samp>/i</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fj-in-RTL-dump">&lsquo;<samp>/j</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fs-in-RTL-dump">&lsquo;<samp>/s</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fu-in-RTL-dump">&lsquo;<samp>/u</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_002fv-in-RTL-dump">&lsquo;<samp>/v</samp>&rsquo; in RTL dump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-10">0</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-0-in-constraint">&lsquo;<samp>0</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-11">&lt;</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_003c-in-constraint">&lsquo;<samp>&lt;</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-12">=</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_003d-in-constraint">&lsquo;<samp>=</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-13">&gt;</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_003e-in-constraint">&lsquo;<samp>&gt;</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-14">?</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_003f-in-constraint">&lsquo;<samp>?</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Multi_002dAlternative">Multi-Alternative</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-15">\</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005c">\</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_symbol-16">_</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fabsvdi2"><code>__absvdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fabsvsi2"><code>__absvsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddda3"><code>__addda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadddf3"><code>__adddf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadddq3"><code>__adddq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddha3"><code>__addha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddhq3"><code>__addhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddqq3"><code>__addqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddsa3"><code>__addsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddsf3"><code>__addsf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddsq3"><code>__addsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddta3"><code>__addta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddtf3"><code>__addtf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadduda3"><code>__adduda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddudq3"><code>__addudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadduha3"><code>__adduha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadduhq3"><code>__adduhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadduqq3"><code>__adduqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddusa3"><code>__addusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddusq3"><code>__addusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fadduta3"><code>__adduta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddvdi3"><code>__addvdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddvsi3"><code>__addvsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005faddxf3"><code>__addxf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlda3"><code>__ashlda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashldi3"><code>__ashldi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashldq3"><code>__ashldq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlha3"><code>__ashlha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlhq3"><code>__ashlhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlqq3"><code>__ashlqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlsa3"><code>__ashlsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlsi3"><code>__ashlsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlsq3"><code>__ashlsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlta3"><code>__ashlta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlti3"><code>__ashlti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashluda3"><code>__ashluda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashludq3"><code>__ashludq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashluha3"><code>__ashluha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashluhq3"><code>__ashluhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashluqq3"><code>__ashluqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlusa3"><code>__ashlusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashlusq3"><code>__ashlusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashluta3"><code>__ashluta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrda3"><code>__ashrda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrdi3"><code>__ashrdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrdq3"><code>__ashrdq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrha3"><code>__ashrha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrhq3"><code>__ashrhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrqq3"><code>__ashrqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrsa3"><code>__ashrsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrsi3"><code>__ashrsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrsq3"><code>__ashrsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrta3"><code>__ashrta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fashrti3"><code>__ashrti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fadddd3"><code>__bid_adddd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005faddsd3"><code>__bid_addsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005faddtd3"><code>__bid_addtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fdivdd3"><code>__bid_divdd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fdivsd3"><code>__bid_divsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fdivtd3"><code>__bid_divtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005feqdd2"><code>__bid_eqdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005feqsd2"><code>__bid_eqsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005feqtd2"><code>__bid_eqtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendddtd2"><code>__bid_extendddtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendddtf"><code>__bid_extendddtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendddxf"><code>__bid_extendddxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextenddfdd"><code>__bid_extenddfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextenddftd"><code>__bid_extenddftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsddd2"><code>__bid_extendsddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsddf"><code>__bid_extendsddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsdtd2"><code>__bid_extendsdtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsdtf"><code>__bid_extendsdtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsdxf"><code>__bid_extendsdxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsfdd"><code>__bid_extendsfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsfsd"><code>__bid_extendsfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendsftd"><code>__bid_extendsftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendtftd"><code>__bid_extendtftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fextendxftd"><code>__bid_extendxftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixdddi"><code>__bid_fixdddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixddsi"><code>__bid_fixddsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixsddi"><code>__bid_fixsddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixsdsi"><code>__bid_fixsdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixtddi"><code>__bid_fixtddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixtdsi"><code>__bid_fixtdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunsdddi"><code>__bid_fixunsdddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunsddsi"><code>__bid_fixunsddsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunssddi"><code>__bid_fixunssddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunssdsi"><code>__bid_fixunssdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunstddi"><code>__bid_fixunstddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffixunstdsi"><code>__bid_fixunstdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatdidd"><code>__bid_floatdidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatdisd"><code>__bid_floatdisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatditd"><code>__bid_floatditd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatsidd"><code>__bid_floatsidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatsisd"><code>__bid_floatsisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatsitd"><code>__bid_floatsitd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunsdidd"><code>__bid_floatunsdidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunsdisd"><code>__bid_floatunsdisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunsditd"><code>__bid_floatunsditd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunssidd"><code>__bid_floatunssidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunssisd"><code>__bid_floatunssisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ffloatunssitd"><code>__bid_floatunssitd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgedd2"><code>__bid_gedd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgesd2"><code>__bid_gesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgetd2"><code>__bid_getd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgtdd2"><code>__bid_gtdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgtsd2"><code>__bid_gtsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fgttd2"><code>__bid_gttd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fledd2"><code>__bid_ledd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005flesd2"><code>__bid_lesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fletd2"><code>__bid_letd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fltdd2"><code>__bid_ltdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fltsd2"><code>__bid_ltsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005flttd2"><code>__bid_lttd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fmuldd3"><code>__bid_muldd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fmulsd3"><code>__bid_mulsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fmultd3"><code>__bid_multd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnedd2"><code>__bid_nedd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnegdd2"><code>__bid_negdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnegsd2"><code>__bid_negsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnegtd2"><code>__bid_negtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnesd2"><code>__bid_nesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fnetd2"><code>__bid_netd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fsubdd3"><code>__bid_subdd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fsubsd3"><code>__bid_subsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005fsubtd3"><code>__bid_subtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncdddf"><code>__bid_truncdddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncddsd2"><code>__bid_truncddsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncddsf"><code>__bid_truncddsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncdfsd"><code>__bid_truncdfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncsdsf"><code>__bid_truncsdsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctddd2"><code>__bid_trunctddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctddf"><code>__bid_trunctddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctdsd2"><code>__bid_trunctdsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctdsf"><code>__bid_trunctdsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctdtf"><code>__bid_trunctdtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctdxf"><code>__bid_trunctdxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctfdd"><code>__bid_trunctfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftrunctfsd"><code>__bid_trunctfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncxfdd"><code>__bid_truncxfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005ftruncxfsd"><code>__bid_truncxfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005funorddd2"><code>__bid_unorddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005funordsd2"><code>__bid_unordsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbid_005funordtd2"><code>__bid_unordtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbswapdi2"><code>__bswapdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbswapsi2"><code>__bswapsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbuiltin_005fclassify_005ftype"><code>__builtin_classify_type</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbuiltin_005fnext_005farg"><code>__builtin_next_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fbuiltin_005fsaveregs"><code>__builtin_saveregs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fclear_005fcache"><code>__clear_cache</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Miscellaneous-routines">Miscellaneous routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fclzdi2"><code>__clzdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fclzsi2"><code>__clzsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fclzti2"><code>__clzti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpda2"><code>__cmpda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpdf2"><code>__cmpdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpdi2"><code>__cmpdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpdq2"><code>__cmpdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpha2"><code>__cmpha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmphq2"><code>__cmphq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpqq2"><code>__cmpqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpsa2"><code>__cmpsa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpsf2"><code>__cmpsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpsq2"><code>__cmpsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpta2"><code>__cmpta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmptf2"><code>__cmptf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpti2"><code>__cmpti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpuda2"><code>__cmpuda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpudq2"><code>__cmpudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpuha2"><code>__cmpuha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpuhq2"><code>__cmpuhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpuqq2"><code>__cmpuqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpusa2"><code>__cmpusa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmpusq2"><code>__cmpusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fcmputa2"><code>__cmputa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fCTOR_005fLIST_005f_005f"><code>__CTOR_LIST__</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fctzdi2"><code>__ctzdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fctzsi2"><code>__ctzsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fctzti2"><code>__ctzti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivda3"><code>__divda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivdc3"><code>__divdc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivdf3"><code>__divdf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivdi3"><code>__divdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivdq3"><code>__divdq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivha3"><code>__divha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivhq3"><code>__divhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivqq3"><code>__divqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivsa3"><code>__divsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivsc3"><code>__divsc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivsf3"><code>__divsf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivsi3"><code>__divsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivsq3"><code>__divsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivta3"><code>__divta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivtc3"><code>__divtc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivtf3"><code>__divtf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivti3"><code>__divti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivxc3"><code>__divxc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdivxf3"><code>__divxf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fadddd3"><code>__dpd_adddd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005faddsd3"><code>__dpd_addsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005faddtd3"><code>__dpd_addtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fdivdd3"><code>__dpd_divdd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fdivsd3"><code>__dpd_divsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fdivtd3"><code>__dpd_divtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005feqdd2"><code>__dpd_eqdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005feqsd2"><code>__dpd_eqsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005feqtd2"><code>__dpd_eqtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendddtd2"><code>__dpd_extendddtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendddtf"><code>__dpd_extendddtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendddxf"><code>__dpd_extendddxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextenddfdd"><code>__dpd_extenddfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextenddftd"><code>__dpd_extenddftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsddd2"><code>__dpd_extendsddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsddf"><code>__dpd_extendsddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsdtd2"><code>__dpd_extendsdtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsdtf"><code>__dpd_extendsdtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsdxf"><code>__dpd_extendsdxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsfdd"><code>__dpd_extendsfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsfsd"><code>__dpd_extendsfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendsftd"><code>__dpd_extendsftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendtftd"><code>__dpd_extendtftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fextendxftd"><code>__dpd_extendxftd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixdddi"><code>__dpd_fixdddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixddsi"><code>__dpd_fixddsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixsddi"><code>__dpd_fixsddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixsdsi"><code>__dpd_fixsdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixtddi"><code>__dpd_fixtddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixtdsi"><code>__dpd_fixtdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunsdddi"><code>__dpd_fixunsdddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunsddsi"><code>__dpd_fixunsddsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunssddi"><code>__dpd_fixunssddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunssdsi"><code>__dpd_fixunssdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunstddi"><code>__dpd_fixunstddi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffixunstdsi"><code>__dpd_fixunstdsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatdidd"><code>__dpd_floatdidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatdisd"><code>__dpd_floatdisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatditd"><code>__dpd_floatditd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatsidd"><code>__dpd_floatsidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatsisd"><code>__dpd_floatsisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatsitd"><code>__dpd_floatsitd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunsdidd"><code>__dpd_floatunsdidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunsdisd"><code>__dpd_floatunsdisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunsditd"><code>__dpd_floatunsditd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunssidd"><code>__dpd_floatunssidd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunssisd"><code>__dpd_floatunssisd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ffloatunssitd"><code>__dpd_floatunssitd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgedd2"><code>__dpd_gedd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgesd2"><code>__dpd_gesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgetd2"><code>__dpd_getd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgtdd2"><code>__dpd_gtdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgtsd2"><code>__dpd_gtsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fgttd2"><code>__dpd_gttd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fledd2"><code>__dpd_ledd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005flesd2"><code>__dpd_lesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fletd2"><code>__dpd_letd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fltdd2"><code>__dpd_ltdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fltsd2"><code>__dpd_ltsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005flttd2"><code>__dpd_lttd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fmuldd3"><code>__dpd_muldd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fmulsd3"><code>__dpd_mulsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fmultd3"><code>__dpd_multd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnedd2"><code>__dpd_nedd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnegdd2"><code>__dpd_negdd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnegsd2"><code>__dpd_negsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnegtd2"><code>__dpd_negtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnesd2"><code>__dpd_nesd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fnetd2"><code>__dpd_netd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fsubdd3"><code>__dpd_subdd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fsubsd3"><code>__dpd_subsd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005fsubtd3"><code>__dpd_subtd3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncdddf"><code>__dpd_truncdddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncddsd2"><code>__dpd_truncddsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncddsf"><code>__dpd_truncddsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncdfsd"><code>__dpd_truncdfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncsdsf"><code>__dpd_truncsdsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctddd2"><code>__dpd_trunctddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctddf"><code>__dpd_trunctddf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctdsd2"><code>__dpd_trunctdsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctdsf"><code>__dpd_trunctdsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctdtf"><code>__dpd_trunctdtf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctdxf"><code>__dpd_trunctdxf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctfdd"><code>__dpd_trunctfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftrunctfsd"><code>__dpd_trunctfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncxfdd"><code>__dpd_truncxfdd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005ftruncxfsd"><code>__dpd_truncxfsd</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005funorddd2"><code>__dpd_unorddd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005funordsd2"><code>__dpd_unordsd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fdpd_005funordtd2"><code>__dpd_unordtd2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fDTOR_005fLIST_005f_005f"><code>__DTOR_LIST__</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005feqdf2"><code>__eqdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005feqsf2"><code>__eqsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005feqtf2"><code>__eqtf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fextenddftf2"><code>__extenddftf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fextenddfxf2"><code>__extenddfxf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fextendsfdf2"><code>__extendsfdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fextendsftf2"><code>__extendsftf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fextendsfxf2"><code>__extendsfxf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fffsdi2"><code>__ffsdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fffsti2"><code>__ffsti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixdfdi"><code>__fixdfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixdfsi"><code>__fixdfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixdfti"><code>__fixdfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixsfdi"><code>__fixsfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixsfsi"><code>__fixsfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixsfti"><code>__fixsfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixtfdi"><code>__fixtfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixtfsi"><code>__fixtfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixtfti"><code>__fixtfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsdfdi"><code>__fixunsdfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsdfsi"><code>__fixunsdfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsdfti"><code>__fixunsdfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunssfdi"><code>__fixunssfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunssfsi"><code>__fixunssfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunssfti"><code>__fixunssfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunstfdi"><code>__fixunstfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunstfsi"><code>__fixunstfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunstfti"><code>__fixunstfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsxfdi"><code>__fixunsxfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsxfsi"><code>__fixunsxfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixunsxfti"><code>__fixunsxfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixxfdi"><code>__fixxfdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixxfsi"><code>__fixxfsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffixxfti"><code>__fixxfti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatdidf"><code>__floatdidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatdisf"><code>__floatdisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatditf"><code>__floatditf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatdixf"><code>__floatdixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatsidf"><code>__floatsidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatsisf"><code>__floatsisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatsitf"><code>__floatsitf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatsixf"><code>__floatsixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloattidf"><code>__floattidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloattisf"><code>__floattisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloattitf"><code>__floattitf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloattixf"><code>__floattixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatundidf"><code>__floatundidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatundisf"><code>__floatundisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatunditf"><code>__floatunditf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatundixf"><code>__floatundixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatunsidf"><code>__floatunsidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatunsisf"><code>__floatunsisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatunsitf"><code>__floatunsitf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatunsixf"><code>__floatunsixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatuntidf"><code>__floatuntidf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatuntisf"><code>__floatuntisf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatuntitf"><code>__floatuntitf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffloatuntixf"><code>__floatuntixf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdadf"><code>__fractdadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdadi"><code>__fractdadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdadq"><code>__fractdadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdaha2"><code>__fractdaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdahi"><code>__fractdahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdahq"><code>__fractdahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdaqi"><code>__fractdaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdaqq"><code>__fractdaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdasa2"><code>__fractdasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdasf"><code>__fractdasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdasi"><code>__fractdasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdasq"><code>__fractdasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdata2"><code>__fractdata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdati"><code>__fractdati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdauda"><code>__fractdauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdaudq"><code>__fractdaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdauha"><code>__fractdauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdauhq"><code>__fractdauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdauqq"><code>__fractdauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdausa"><code>__fractdausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdausq"><code>__fractdausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdauta"><code>__fractdauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfda"><code>__fractdfda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfdq"><code>__fractdfdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfha"><code>__fractdfha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfhq"><code>__fractdfhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfqq"><code>__fractdfqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfsa"><code>__fractdfsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfsq"><code>__fractdfsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfta"><code>__fractdfta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfuda"><code>__fractdfuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfudq"><code>__fractdfudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfuha"><code>__fractdfuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfuhq"><code>__fractdfuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfuqq"><code>__fractdfuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfusa"><code>__fractdfusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfusq"><code>__fractdfusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdfuta"><code>__fractdfuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdida"><code>__fractdida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdidq"><code>__fractdidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiha"><code>__fractdiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdihq"><code>__fractdihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiqq"><code>__fractdiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdisa"><code>__fractdisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdisq"><code>__fractdisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdita"><code>__fractdita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiuda"><code>__fractdiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiudq"><code>__fractdiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiuha"><code>__fractdiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiuhq"><code>__fractdiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiuqq"><code>__fractdiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiusa"><code>__fractdiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiusq"><code>__fractdiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdiuta"><code>__fractdiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqda"><code>__fractdqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqdf"><code>__fractdqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqdi"><code>__fractdqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqha"><code>__fractdqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqhi"><code>__fractdqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqhq2"><code>__fractdqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqqi"><code>__fractdqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqqq2"><code>__fractdqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqsa"><code>__fractdqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqsf"><code>__fractdqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqsi"><code>__fractdqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqsq2"><code>__fractdqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqta"><code>__fractdqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqti"><code>__fractdqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdquda"><code>__fractdquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqudq"><code>__fractdqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdquha"><code>__fractdquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdquhq"><code>__fractdquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdquqq"><code>__fractdquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqusa"><code>__fractdqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdqusq"><code>__fractdqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractdquta"><code>__fractdquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthada2"><code>__fracthada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthadf"><code>__fracthadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthadi"><code>__fracthadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthadq"><code>__fracthadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthahi"><code>__fracthahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthahq"><code>__fracthahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthaqi"><code>__fracthaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthaqq"><code>__fracthaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthasa2"><code>__fracthasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthasf"><code>__fracthasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthasi"><code>__fracthasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthasq"><code>__fracthasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthata2"><code>__fracthata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthati"><code>__fracthati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthauda"><code>__fracthauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthaudq"><code>__fracthaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthauha"><code>__fracthauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthauhq"><code>__fracthauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthauqq"><code>__fracthauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthausa"><code>__fracthausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthausq"><code>__fracthausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthauta"><code>__fracthauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthida"><code>__fracthida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthidq"><code>__fracthidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiha"><code>__fracthiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthihq"><code>__fracthihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiqq"><code>__fracthiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthisa"><code>__fracthisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthisq"><code>__fracthisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthita"><code>__fracthita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiuda"><code>__fracthiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiudq"><code>__fracthiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiuha"><code>__fracthiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiuhq"><code>__fracthiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiuqq"><code>__fracthiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiusa"><code>__fracthiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiusq"><code>__fracthiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthiuta"><code>__fracthiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqda"><code>__fracthqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqdf"><code>__fracthqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqdi"><code>__fracthqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqdq2"><code>__fracthqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqha"><code>__fracthqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqhi"><code>__fracthqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqqi"><code>__fracthqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqqq2"><code>__fracthqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqsa"><code>__fracthqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqsf"><code>__fracthqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqsi"><code>__fracthqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqsq2"><code>__fracthqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqta"><code>__fracthqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqti"><code>__fracthqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthquda"><code>__fracthquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqudq"><code>__fracthqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthquha"><code>__fracthquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthquhq"><code>__fracthquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthquqq"><code>__fracthquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqusa"><code>__fracthqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthqusq"><code>__fracthqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracthquta"><code>__fracthquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqida"><code>__fractqida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqidq"><code>__fractqidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiha"><code>__fractqiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqihq"><code>__fractqihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiqq"><code>__fractqiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqisa"><code>__fractqisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqisq"><code>__fractqisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqita"><code>__fractqita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiuda"><code>__fractqiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiudq"><code>__fractqiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiuha"><code>__fractqiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiuhq"><code>__fractqiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiuqq"><code>__fractqiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiusa"><code>__fractqiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiusq"><code>__fractqiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqiuta"><code>__fractqiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqda"><code>__fractqqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqdf"><code>__fractqqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqdi"><code>__fractqqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqdq2"><code>__fractqqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqha"><code>__fractqqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqhi"><code>__fractqqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqhq2"><code>__fractqqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqqi"><code>__fractqqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqsa"><code>__fractqqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqsf"><code>__fractqqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqsi"><code>__fractqqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqsq2"><code>__fractqqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqta"><code>__fractqqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqti"><code>__fractqqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqquda"><code>__fractqquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqudq"><code>__fractqqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqquha"><code>__fractqquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqquhq"><code>__fractqquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqquqq"><code>__fractqquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqusa"><code>__fractqqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqqusq"><code>__fractqqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractqquta"><code>__fractqquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsada2"><code>__fractsada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsadf"><code>__fractsadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsadi"><code>__fractsadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsadq"><code>__fractsadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsaha2"><code>__fractsaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsahi"><code>__fractsahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsahq"><code>__fractsahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsaqi"><code>__fractsaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsaqq"><code>__fractsaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsasf"><code>__fractsasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsasi"><code>__fractsasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsasq"><code>__fractsasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsata2"><code>__fractsata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsati"><code>__fractsati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsauda"><code>__fractsauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsaudq"><code>__fractsaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsauha"><code>__fractsauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsauhq"><code>__fractsauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsauqq"><code>__fractsauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsausa"><code>__fractsausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsausq"><code>__fractsausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsauta"><code>__fractsauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfda"><code>__fractsfda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfdq"><code>__fractsfdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfha"><code>__fractsfha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfhq"><code>__fractsfhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfqq"><code>__fractsfqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfsa"><code>__fractsfsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfsq"><code>__fractsfsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfta"><code>__fractsfta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfuda"><code>__fractsfuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfudq"><code>__fractsfudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfuha"><code>__fractsfuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfuhq"><code>__fractsfuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfuqq"><code>__fractsfuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfusa"><code>__fractsfusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfusq"><code>__fractsfusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsfuta"><code>__fractsfuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsida"><code>__fractsida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsidq"><code>__fractsidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiha"><code>__fractsiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsihq"><code>__fractsihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiqq"><code>__fractsiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsisa"><code>__fractsisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsisq"><code>__fractsisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsita"><code>__fractsita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiuda"><code>__fractsiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiudq"><code>__fractsiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiuha"><code>__fractsiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiuhq"><code>__fractsiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiuqq"><code>__fractsiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiusa"><code>__fractsiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiusq"><code>__fractsiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsiuta"><code>__fractsiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqda"><code>__fractsqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqdf"><code>__fractsqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqdi"><code>__fractsqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqdq2"><code>__fractsqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqha"><code>__fractsqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqhi"><code>__fractsqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqhq2"><code>__fractsqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqqi"><code>__fractsqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqqq2"><code>__fractsqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqsa"><code>__fractsqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqsf"><code>__fractsqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqsi"><code>__fractsqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqta"><code>__fractsqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqti"><code>__fractsqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsquda"><code>__fractsquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqudq"><code>__fractsqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsquha"><code>__fractsquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsquhq"><code>__fractsquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsquqq"><code>__fractsquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqusa"><code>__fractsqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsqusq"><code>__fractsqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractsquta"><code>__fractsquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttada2"><code>__fracttada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttadf"><code>__fracttadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttadi"><code>__fracttadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttadq"><code>__fracttadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttaha2"><code>__fracttaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttahi"><code>__fracttahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttahq"><code>__fracttahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttaqi"><code>__fracttaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttaqq"><code>__fracttaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttasa2"><code>__fracttasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttasf"><code>__fracttasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttasi"><code>__fracttasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttasq"><code>__fracttasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttati"><code>__fracttati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttauda"><code>__fracttauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttaudq"><code>__fracttaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttauha"><code>__fracttauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttauhq"><code>__fracttauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttauqq"><code>__fracttauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttausa"><code>__fracttausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttausq"><code>__fracttausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttauta"><code>__fracttauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttida"><code>__fracttida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttidq"><code>__fracttidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiha"><code>__fracttiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttihq"><code>__fracttihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiqq"><code>__fracttiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttisa"><code>__fracttisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttisq"><code>__fracttisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttita"><code>__fracttita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiuda"><code>__fracttiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiudq"><code>__fracttiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiuha"><code>__fracttiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiuhq"><code>__fracttiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiuqq"><code>__fracttiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiusa"><code>__fracttiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiusq"><code>__fracttiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffracttiuta"><code>__fracttiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudada"><code>__fractudada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudadf"><code>__fractudadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudadi"><code>__fractudadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudadq"><code>__fractudadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudaha"><code>__fractudaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudahi"><code>__fractudahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudahq"><code>__fractudahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudaqi"><code>__fractudaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudaqq"><code>__fractudaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudasa"><code>__fractudasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudasf"><code>__fractudasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudasi"><code>__fractudasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudasq"><code>__fractudasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudata"><code>__fractudata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudati"><code>__fractudati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudaudq"><code>__fractudaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudauha2"><code>__fractudauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudauhq"><code>__fractudauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudauqq"><code>__fractudauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudausa2"><code>__fractudausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudausq"><code>__fractudausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudauta2"><code>__fractudauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqda"><code>__fractudqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqdf"><code>__fractudqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqdi"><code>__fractudqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqdq"><code>__fractudqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqha"><code>__fractudqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqhi"><code>__fractudqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqhq"><code>__fractudqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqqi"><code>__fractudqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqqq"><code>__fractudqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqsa"><code>__fractudqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqsf"><code>__fractudqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqsi"><code>__fractudqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqsq"><code>__fractudqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqta"><code>__fractudqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqti"><code>__fractudqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudquda"><code>__fractudquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudquha"><code>__fractudquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudquhq2"><code>__fractudquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudquqq2"><code>__fractudquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqusa"><code>__fractudqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudqusq2"><code>__fractudqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractudquta"><code>__fractudquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhada"><code>__fractuhada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhadf"><code>__fractuhadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhadi"><code>__fractuhadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhadq"><code>__fractuhadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhaha"><code>__fractuhaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhahi"><code>__fractuhahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhahq"><code>__fractuhahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhaqi"><code>__fractuhaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhaqq"><code>__fractuhaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhasa"><code>__fractuhasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhasf"><code>__fractuhasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhasi"><code>__fractuhasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhasq"><code>__fractuhasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhata"><code>__fractuhata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhati"><code>__fractuhati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhauda2"><code>__fractuhauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhaudq"><code>__fractuhaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhauhq"><code>__fractuhauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhauqq"><code>__fractuhauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhausa2"><code>__fractuhausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhausq"><code>__fractuhausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhauta2"><code>__fractuhauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqda"><code>__fractuhqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqdf"><code>__fractuhqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqdi"><code>__fractuhqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqdq"><code>__fractuhqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqha"><code>__fractuhqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqhi"><code>__fractuhqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqhq"><code>__fractuhqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqqi"><code>__fractuhqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqqq"><code>__fractuhqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqsa"><code>__fractuhqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqsf"><code>__fractuhqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqsi"><code>__fractuhqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqsq"><code>__fractuhqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqta"><code>__fractuhqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqti"><code>__fractuhqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhquda"><code>__fractuhquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqudq2"><code>__fractuhqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhquha"><code>__fractuhquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhquqq2"><code>__fractuhquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqusa"><code>__fractuhqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhqusq2"><code>__fractuhqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuhquta"><code>__fractuhquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdadi"><code>__fractunsdadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdahi"><code>__fractunsdahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdaqi"><code>__fractunsdaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdasi"><code>__fractunsdasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdati"><code>__fractunsdati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdida"><code>__fractunsdida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdidq"><code>__fractunsdidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiha"><code>__fractunsdiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdihq"><code>__fractunsdihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiqq"><code>__fractunsdiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdisa"><code>__fractunsdisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdisq"><code>__fractunsdisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdita"><code>__fractunsdita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiuda"><code>__fractunsdiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiudq"><code>__fractunsdiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiuha"><code>__fractunsdiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiuhq"><code>__fractunsdiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiuqq"><code>__fractunsdiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiusa"><code>__fractunsdiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiusq"><code>__fractunsdiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdiuta"><code>__fractunsdiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdqdi"><code>__fractunsdqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdqhi"><code>__fractunsdqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdqqi"><code>__fractunsdqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdqsi"><code>__fractunsdqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsdqti"><code>__fractunsdqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshadi"><code>__fractunshadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshahi"><code>__fractunshahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshaqi"><code>__fractunshaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshasi"><code>__fractunshasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshati"><code>__fractunshati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshida"><code>__fractunshida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshidq"><code>__fractunshidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiha"><code>__fractunshiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshihq"><code>__fractunshihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiqq"><code>__fractunshiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshisa"><code>__fractunshisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshisq"><code>__fractunshisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshita"><code>__fractunshita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiuda"><code>__fractunshiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiudq"><code>__fractunshiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiuha"><code>__fractunshiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiuhq"><code>__fractunshiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiuqq"><code>__fractunshiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiusa"><code>__fractunshiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiusq"><code>__fractunshiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshiuta"><code>__fractunshiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshqdi"><code>__fractunshqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshqhi"><code>__fractunshqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshqqi"><code>__fractunshqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshqsi"><code>__fractunshqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunshqti"><code>__fractunshqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqida"><code>__fractunsqida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqidq"><code>__fractunsqidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiha"><code>__fractunsqiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqihq"><code>__fractunsqihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiqq"><code>__fractunsqiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqisa"><code>__fractunsqisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqisq"><code>__fractunsqisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqita"><code>__fractunsqita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiuda"><code>__fractunsqiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiudq"><code>__fractunsqiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiuha"><code>__fractunsqiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiuhq"><code>__fractunsqiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiuqq"><code>__fractunsqiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiusa"><code>__fractunsqiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiusq"><code>__fractunsqiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqiuta"><code>__fractunsqiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqqdi"><code>__fractunsqqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqqhi"><code>__fractunsqqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqqqi"><code>__fractunsqqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqqsi"><code>__fractunsqqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsqqti"><code>__fractunsqqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssadi"><code>__fractunssadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssahi"><code>__fractunssahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssaqi"><code>__fractunssaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssasi"><code>__fractunssasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssati"><code>__fractunssati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssida"><code>__fractunssida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssidq"><code>__fractunssidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiha"><code>__fractunssiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssihq"><code>__fractunssihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiqq"><code>__fractunssiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssisa"><code>__fractunssisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssisq"><code>__fractunssisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssita"><code>__fractunssita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiuda"><code>__fractunssiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiudq"><code>__fractunssiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiuha"><code>__fractunssiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiuhq"><code>__fractunssiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiuqq"><code>__fractunssiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiusa"><code>__fractunssiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiusq"><code>__fractunssiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssiuta"><code>__fractunssiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssqdi"><code>__fractunssqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssqhi"><code>__fractunssqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssqqi"><code>__fractunssqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssqsi"><code>__fractunssqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunssqti"><code>__fractunssqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstadi"><code>__fractunstadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstahi"><code>__fractunstahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstaqi"><code>__fractunstaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstasi"><code>__fractunstasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstati"><code>__fractunstati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstida"><code>__fractunstida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstidq"><code>__fractunstidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiha"><code>__fractunstiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstihq"><code>__fractunstihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiqq"><code>__fractunstiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstisa"><code>__fractunstisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstisq"><code>__fractunstisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstita"><code>__fractunstita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiuda"><code>__fractunstiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiudq"><code>__fractunstiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiuha"><code>__fractunstiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiuhq"><code>__fractunstiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiuqq"><code>__fractunstiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiusa"><code>__fractunstiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiusq"><code>__fractunstiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunstiuta"><code>__fractunstiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudadi"><code>__fractunsudadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudahi"><code>__fractunsudahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudaqi"><code>__fractunsudaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudasi"><code>__fractunsudasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudati"><code>__fractunsudati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudqdi"><code>__fractunsudqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudqhi"><code>__fractunsudqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudqqi"><code>__fractunsudqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudqsi"><code>__fractunsudqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsudqti"><code>__fractunsudqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhadi"><code>__fractunsuhadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhahi"><code>__fractunsuhahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhaqi"><code>__fractunsuhaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhasi"><code>__fractunsuhasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhati"><code>__fractunsuhati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhqdi"><code>__fractunsuhqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhqhi"><code>__fractunsuhqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhqqi"><code>__fractunsuhqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhqsi"><code>__fractunsuhqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuhqti"><code>__fractunsuhqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuqqdi"><code>__fractunsuqqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuqqhi"><code>__fractunsuqqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuqqqi"><code>__fractunsuqqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuqqsi"><code>__fractunsuqqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsuqqti"><code>__fractunsuqqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusadi"><code>__fractunsusadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusahi"><code>__fractunsusahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusaqi"><code>__fractunsusaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusasi"><code>__fractunsusasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusati"><code>__fractunsusati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusqdi"><code>__fractunsusqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusqhi"><code>__fractunsusqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusqqi"><code>__fractunsusqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusqsi"><code>__fractunsusqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsusqti"><code>__fractunsusqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsutadi"><code>__fractunsutadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsutahi"><code>__fractunsutahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsutaqi"><code>__fractunsutaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsutasi"><code>__fractunsutasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractunsutati"><code>__fractunsutati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqda"><code>__fractuqqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqdf"><code>__fractuqqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqdi"><code>__fractuqqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqdq"><code>__fractuqqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqha"><code>__fractuqqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqhi"><code>__fractuqqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqhq"><code>__fractuqqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqqi"><code>__fractuqqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqqq"><code>__fractuqqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqsa"><code>__fractuqqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqsf"><code>__fractuqqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqsi"><code>__fractuqqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqsq"><code>__fractuqqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqta"><code>__fractuqqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqti"><code>__fractuqqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqquda"><code>__fractuqquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqudq2"><code>__fractuqqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqquha"><code>__fractuqquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqquhq2"><code>__fractuqquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqusa"><code>__fractuqqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqqusq2"><code>__fractuqqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractuqquta"><code>__fractuqquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusada"><code>__fractusada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusadf"><code>__fractusadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusadi"><code>__fractusadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusadq"><code>__fractusadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusaha"><code>__fractusaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusahi"><code>__fractusahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusahq"><code>__fractusahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusaqi"><code>__fractusaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusaqq"><code>__fractusaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusasa"><code>__fractusasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusasf"><code>__fractusasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusasi"><code>__fractusasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusasq"><code>__fractusasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusata"><code>__fractusata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusati"><code>__fractusati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusauda2"><code>__fractusauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusaudq"><code>__fractusaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusauha2"><code>__fractusauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusauhq"><code>__fractusauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusauqq"><code>__fractusauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusausq"><code>__fractusausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusauta2"><code>__fractusauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqda"><code>__fractusqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqdf"><code>__fractusqdf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqdi"><code>__fractusqdi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqdq"><code>__fractusqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqha"><code>__fractusqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqhi"><code>__fractusqhi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqhq"><code>__fractusqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqqi"><code>__fractusqqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqqq"><code>__fractusqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqsa"><code>__fractusqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqsf"><code>__fractusqsf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqsi"><code>__fractusqsi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqsq"><code>__fractusqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqta"><code>__fractusqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqti"><code>__fractusqti</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusquda"><code>__fractusquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqudq2"><code>__fractusqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusquha"><code>__fractusquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusquhq2"><code>__fractusquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusquqq2"><code>__fractusquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusqusa"><code>__fractusqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractusquta"><code>__fractusquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutada"><code>__fractutada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutadf"><code>__fractutadf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutadi"><code>__fractutadi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutadq"><code>__fractutadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutaha"><code>__fractutaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutahi"><code>__fractutahi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutahq"><code>__fractutahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutaqi"><code>__fractutaqi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutaqq"><code>__fractutaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutasa"><code>__fractutasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutasf"><code>__fractutasf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutasi"><code>__fractutasi</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutasq"><code>__fractutasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutata"><code>__fractutata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutati"><code>__fractutati</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutauda2"><code>__fractutauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutaudq"><code>__fractutaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutauha2"><code>__fractutauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutauhq"><code>__fractutauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutauqq"><code>__fractutauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutausa2"><code>__fractutausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ffractutausq"><code>__fractutausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgedf2"><code>__gedf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgesf2"><code>__gesf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgetf2"><code>__getf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgtdf2"><code>__gtdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgtsf2"><code>__gtsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fgttf2"><code>__gttf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fledf2"><code>__ledf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flesf2"><code>__lesf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fletf2"><code>__letf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrdi3"><code>__lshrdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrsi3"><code>__lshrsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrti3"><code>__lshrti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshruda3"><code>__lshruda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrudq3"><code>__lshrudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshruha3"><code>__lshruha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshruhq3"><code>__lshruhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshruqq3"><code>__lshruqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrusa3"><code>__lshrusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshrusq3"><code>__lshrusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flshruta3"><code>__lshruta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fltdf2"><code>__ltdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fltsf2"><code>__ltsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005flttf2"><code>__lttf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmain"><code>__main</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Collect2">Collect2</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmoddi3"><code>__moddi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmodsi3"><code>__modsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmodti3"><code>__modti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmorestack_005fcurrent_005fsegment"><code>__morestack_current_segment</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Miscellaneous-routines">Miscellaneous routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmorestack_005finitial_005fsp"><code>__morestack_initial_sp</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Miscellaneous-routines">Miscellaneous routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmorestack_005fsegments"><code>__morestack_segments</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Miscellaneous-routines">Miscellaneous routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulda3"><code>__mulda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuldc3"><code>__muldc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuldf3"><code>__muldf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuldi3"><code>__muldi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuldq3"><code>__muldq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulha3"><code>__mulha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulhq3"><code>__mulhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulqq3"><code>__mulqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulsa3"><code>__mulsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulsc3"><code>__mulsc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulsf3"><code>__mulsf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulsi3"><code>__mulsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulsq3"><code>__mulsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulta3"><code>__multa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmultc3"><code>__multc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmultf3"><code>__multf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulti3"><code>__multi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuluda3"><code>__muluda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuludq3"><code>__muludq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuluha3"><code>__muluha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuluhq3"><code>__muluhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuluqq3"><code>__muluqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulusa3"><code>__mulusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulusq3"><code>__mulusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmuluta3"><code>__muluta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulvdi3"><code>__mulvdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulvsi3"><code>__mulvsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulxc3"><code>__mulxc3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fmulxf3"><code>__mulxf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnedf2"><code>__nedf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegda2"><code>__negda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegdf2"><code>__negdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegdi2"><code>__negdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegdq2"><code>__negdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegha2"><code>__negha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneghq2"><code>__neghq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegqq2"><code>__negqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegsa2"><code>__negsa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegsf2"><code>__negsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegsq2"><code>__negsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegta2"><code>__negta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegtf2"><code>__negtf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegti2"><code>__negti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneguda2"><code>__neguda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegudq2"><code>__negudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneguha2"><code>__neguha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneguhq2"><code>__neguhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneguqq2"><code>__neguqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegusa2"><code>__negusa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegusq2"><code>__negusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fneguta2"><code>__neguta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegvdi2"><code>__negvdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegvsi2"><code>__negvsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnegxf2"><code>__negxf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnesf2"><code>__nesf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fnetf2"><code>__netf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fparitydi2"><code>__paritydi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fparitysi2"><code>__paritysi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fparityti2"><code>__parityti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpopcountdi2"><code>__popcountdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpopcountsi2"><code>__popcountsi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpopcountti2"><code>__popcountti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpowidf2"><code>__powidf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpowisf2"><code>__powisf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpowitf2"><code>__powitf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fpowixf2"><code>__powixf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdadq"><code>__satfractdadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdaha2"><code>__satfractdaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdahq"><code>__satfractdahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdaqq"><code>__satfractdaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdasa2"><code>__satfractdasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdasq"><code>__satfractdasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdata2"><code>__satfractdata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdauda"><code>__satfractdauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdaudq"><code>__satfractdaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdauha"><code>__satfractdauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdauhq"><code>__satfractdauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdauqq"><code>__satfractdauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdausa"><code>__satfractdausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdausq"><code>__satfractdausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdauta"><code>__satfractdauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfda"><code>__satfractdfda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfdq"><code>__satfractdfdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfha"><code>__satfractdfha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfhq"><code>__satfractdfhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfqq"><code>__satfractdfqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfsa"><code>__satfractdfsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfsq"><code>__satfractdfsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfta"><code>__satfractdfta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfuda"><code>__satfractdfuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfudq"><code>__satfractdfudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfuha"><code>__satfractdfuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfuhq"><code>__satfractdfuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfuqq"><code>__satfractdfuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfusa"><code>__satfractdfusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfusq"><code>__satfractdfusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdfuta"><code>__satfractdfuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdida"><code>__satfractdida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdidq"><code>__satfractdidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiha"><code>__satfractdiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdihq"><code>__satfractdihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiqq"><code>__satfractdiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdisa"><code>__satfractdisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdisq"><code>__satfractdisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdita"><code>__satfractdita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiuda"><code>__satfractdiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiudq"><code>__satfractdiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiuha"><code>__satfractdiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiuhq"><code>__satfractdiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiuqq"><code>__satfractdiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiusa"><code>__satfractdiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiusq"><code>__satfractdiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdiuta"><code>__satfractdiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqda"><code>__satfractdqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqha"><code>__satfractdqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqhq2"><code>__satfractdqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqqq2"><code>__satfractdqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqsa"><code>__satfractdqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqsq2"><code>__satfractdqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqta"><code>__satfractdqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdquda"><code>__satfractdquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqudq"><code>__satfractdqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdquha"><code>__satfractdquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdquhq"><code>__satfractdquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdquqq"><code>__satfractdquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqusa"><code>__satfractdqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdqusq"><code>__satfractdqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractdquta"><code>__satfractdquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthada2"><code>__satfracthada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthadq"><code>__satfracthadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthahq"><code>__satfracthahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthaqq"><code>__satfracthaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthasa2"><code>__satfracthasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthasq"><code>__satfracthasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthata2"><code>__satfracthata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthauda"><code>__satfracthauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthaudq"><code>__satfracthaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthauha"><code>__satfracthauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthauhq"><code>__satfracthauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthauqq"><code>__satfracthauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthausa"><code>__satfracthausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthausq"><code>__satfracthausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthauta"><code>__satfracthauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthida"><code>__satfracthida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthidq"><code>__satfracthidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiha"><code>__satfracthiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthihq"><code>__satfracthihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiqq"><code>__satfracthiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthisa"><code>__satfracthisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthisq"><code>__satfracthisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthita"><code>__satfracthita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiuda"><code>__satfracthiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiudq"><code>__satfracthiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiuha"><code>__satfracthiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiuhq"><code>__satfracthiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiuqq"><code>__satfracthiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiusa"><code>__satfracthiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiusq"><code>__satfracthiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthiuta"><code>__satfracthiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqda"><code>__satfracthqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqdq2"><code>__satfracthqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqha"><code>__satfracthqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqqq2"><code>__satfracthqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqsa"><code>__satfracthqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqsq2"><code>__satfracthqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqta"><code>__satfracthqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthquda"><code>__satfracthquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqudq"><code>__satfracthqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthquha"><code>__satfracthquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthquhq"><code>__satfracthquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthquqq"><code>__satfracthquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqusa"><code>__satfracthqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthqusq"><code>__satfracthqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracthquta"><code>__satfracthquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqida"><code>__satfractqida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqidq"><code>__satfractqidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiha"><code>__satfractqiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqihq"><code>__satfractqihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiqq"><code>__satfractqiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqisa"><code>__satfractqisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqisq"><code>__satfractqisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqita"><code>__satfractqita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiuda"><code>__satfractqiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiudq"><code>__satfractqiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiuha"><code>__satfractqiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiuhq"><code>__satfractqiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiuqq"><code>__satfractqiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiusa"><code>__satfractqiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiusq"><code>__satfractqiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqiuta"><code>__satfractqiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqda"><code>__satfractqqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqdq2"><code>__satfractqqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqha"><code>__satfractqqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqhq2"><code>__satfractqqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqsa"><code>__satfractqqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqsq2"><code>__satfractqqsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqta"><code>__satfractqqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqquda"><code>__satfractqquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqudq"><code>__satfractqqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqquha"><code>__satfractqquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqquhq"><code>__satfractqquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqquqq"><code>__satfractqquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqusa"><code>__satfractqqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqqusq"><code>__satfractqqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractqquta"><code>__satfractqquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsada2"><code>__satfractsada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsadq"><code>__satfractsadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsaha2"><code>__satfractsaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsahq"><code>__satfractsahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsaqq"><code>__satfractsaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsasq"><code>__satfractsasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsata2"><code>__satfractsata2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsauda"><code>__satfractsauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsaudq"><code>__satfractsaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsauha"><code>__satfractsauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsauhq"><code>__satfractsauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsauqq"><code>__satfractsauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsausa"><code>__satfractsausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsausq"><code>__satfractsausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsauta"><code>__satfractsauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfda"><code>__satfractsfda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfdq"><code>__satfractsfdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfha"><code>__satfractsfha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfhq"><code>__satfractsfhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfqq"><code>__satfractsfqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfsa"><code>__satfractsfsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfsq"><code>__satfractsfsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfta"><code>__satfractsfta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfuda"><code>__satfractsfuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfudq"><code>__satfractsfudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfuha"><code>__satfractsfuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfuhq"><code>__satfractsfuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfuqq"><code>__satfractsfuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfusa"><code>__satfractsfusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfusq"><code>__satfractsfusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsfuta"><code>__satfractsfuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsida"><code>__satfractsida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsidq"><code>__satfractsidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiha"><code>__satfractsiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsihq"><code>__satfractsihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiqq"><code>__satfractsiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsisa"><code>__satfractsisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsisq"><code>__satfractsisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsita"><code>__satfractsita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiuda"><code>__satfractsiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiudq"><code>__satfractsiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiuha"><code>__satfractsiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiuhq"><code>__satfractsiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiuqq"><code>__satfractsiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiusa"><code>__satfractsiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiusq"><code>__satfractsiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsiuta"><code>__satfractsiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqda"><code>__satfractsqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqdq2"><code>__satfractsqdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqha"><code>__satfractsqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqhq2"><code>__satfractsqhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqqq2"><code>__satfractsqqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqsa"><code>__satfractsqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqta"><code>__satfractsqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsquda"><code>__satfractsquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqudq"><code>__satfractsqudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsquha"><code>__satfractsquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsquhq"><code>__satfractsquhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsquqq"><code>__satfractsquqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqusa"><code>__satfractsqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsqusq"><code>__satfractsqusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractsquta"><code>__satfractsquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttada2"><code>__satfracttada2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttadq"><code>__satfracttadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttaha2"><code>__satfracttaha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttahq"><code>__satfracttahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttaqq"><code>__satfracttaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttasa2"><code>__satfracttasa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttasq"><code>__satfracttasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttauda"><code>__satfracttauda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttaudq"><code>__satfracttaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttauha"><code>__satfracttauha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttauhq"><code>__satfracttauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttauqq"><code>__satfracttauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttausa"><code>__satfracttausa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttausq"><code>__satfracttausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttauta"><code>__satfracttauta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttida"><code>__satfracttida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttidq"><code>__satfracttidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiha"><code>__satfracttiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttihq"><code>__satfracttihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiqq"><code>__satfracttiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttisa"><code>__satfracttisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttisq"><code>__satfracttisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttita"><code>__satfracttita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiuda"><code>__satfracttiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiudq"><code>__satfracttiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiuha"><code>__satfracttiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiuhq"><code>__satfracttiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiuqq"><code>__satfracttiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiusa"><code>__satfracttiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiusq"><code>__satfracttiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfracttiuta"><code>__satfracttiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudada"><code>__satfractudada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudadq"><code>__satfractudadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudaha"><code>__satfractudaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudahq"><code>__satfractudahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudaqq"><code>__satfractudaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudasa"><code>__satfractudasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudasq"><code>__satfractudasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudata"><code>__satfractudata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudaudq"><code>__satfractudaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudauha2"><code>__satfractudauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudauhq"><code>__satfractudauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudauqq"><code>__satfractudauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudausa2"><code>__satfractudausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudausq"><code>__satfractudausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudauta2"><code>__satfractudauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqda"><code>__satfractudqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqdq"><code>__satfractudqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqha"><code>__satfractudqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqhq"><code>__satfractudqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqqq"><code>__satfractudqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqsa"><code>__satfractudqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqsq"><code>__satfractudqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqta"><code>__satfractudqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudquda"><code>__satfractudquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudquha"><code>__satfractudquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudquhq2"><code>__satfractudquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudquqq2"><code>__satfractudquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqusa"><code>__satfractudqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudqusq2"><code>__satfractudqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractudquta"><code>__satfractudquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhada"><code>__satfractuhada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhadq"><code>__satfractuhadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhaha"><code>__satfractuhaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhahq"><code>__satfractuhahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhaqq"><code>__satfractuhaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhasa"><code>__satfractuhasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhasq"><code>__satfractuhasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhata"><code>__satfractuhata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhauda2"><code>__satfractuhauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhaudq"><code>__satfractuhaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhauhq"><code>__satfractuhauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhauqq"><code>__satfractuhauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhausa2"><code>__satfractuhausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhausq"><code>__satfractuhausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhauta2"><code>__satfractuhauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqda"><code>__satfractuhqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqdq"><code>__satfractuhqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqha"><code>__satfractuhqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqhq"><code>__satfractuhqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqqq"><code>__satfractuhqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqsa"><code>__satfractuhqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqsq"><code>__satfractuhqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqta"><code>__satfractuhqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhquda"><code>__satfractuhquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqudq2"><code>__satfractuhqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhquha"><code>__satfractuhquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhquqq2"><code>__satfractuhquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqusa"><code>__satfractuhqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhqusq2"><code>__satfractuhqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuhquta"><code>__satfractuhquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdida"><code>__satfractunsdida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdidq"><code>__satfractunsdidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiha"><code>__satfractunsdiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdihq"><code>__satfractunsdihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiqq"><code>__satfractunsdiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdisa"><code>__satfractunsdisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdisq"><code>__satfractunsdisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdita"><code>__satfractunsdita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiuda"><code>__satfractunsdiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiudq"><code>__satfractunsdiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiuha"><code>__satfractunsdiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiuhq"><code>__satfractunsdiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiuqq"><code>__satfractunsdiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiusa"><code>__satfractunsdiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiusq"><code>__satfractunsdiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsdiuta"><code>__satfractunsdiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshida"><code>__satfractunshida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshidq"><code>__satfractunshidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiha"><code>__satfractunshiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshihq"><code>__satfractunshihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiqq"><code>__satfractunshiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshisa"><code>__satfractunshisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshisq"><code>__satfractunshisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshita"><code>__satfractunshita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiuda"><code>__satfractunshiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiudq"><code>__satfractunshiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiuha"><code>__satfractunshiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiuhq"><code>__satfractunshiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiuqq"><code>__satfractunshiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiusa"><code>__satfractunshiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiusq"><code>__satfractunshiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunshiuta"><code>__satfractunshiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqida"><code>__satfractunsqida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqidq"><code>__satfractunsqidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiha"><code>__satfractunsqiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqihq"><code>__satfractunsqihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiqq"><code>__satfractunsqiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqisa"><code>__satfractunsqisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqisq"><code>__satfractunsqisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqita"><code>__satfractunsqita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiuda"><code>__satfractunsqiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiudq"><code>__satfractunsqiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiuha"><code>__satfractunsqiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiuhq"><code>__satfractunsqiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiuqq"><code>__satfractunsqiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiusa"><code>__satfractunsqiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiusq"><code>__satfractunsqiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunsqiuta"><code>__satfractunsqiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssida"><code>__satfractunssida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssidq"><code>__satfractunssidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiha"><code>__satfractunssiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssihq"><code>__satfractunssihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiqq"><code>__satfractunssiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssisa"><code>__satfractunssisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssisq"><code>__satfractunssisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssita"><code>__satfractunssita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiuda"><code>__satfractunssiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiudq"><code>__satfractunssiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiuha"><code>__satfractunssiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiuhq"><code>__satfractunssiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiuqq"><code>__satfractunssiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiusa"><code>__satfractunssiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiusq"><code>__satfractunssiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunssiuta"><code>__satfractunssiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstida"><code>__satfractunstida</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstidq"><code>__satfractunstidq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiha"><code>__satfractunstiha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstihq"><code>__satfractunstihq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiqq"><code>__satfractunstiqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstisa"><code>__satfractunstisa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstisq"><code>__satfractunstisq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstita"><code>__satfractunstita</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiuda"><code>__satfractunstiuda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiudq"><code>__satfractunstiudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiuha"><code>__satfractunstiuha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiuhq"><code>__satfractunstiuhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiuqq"><code>__satfractunstiuqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiusa"><code>__satfractunstiusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiusq"><code>__satfractunstiusq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractunstiuta"><code>__satfractunstiuta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqda"><code>__satfractuqqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqdq"><code>__satfractuqqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqha"><code>__satfractuqqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqhq"><code>__satfractuqqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqqq"><code>__satfractuqqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqsa"><code>__satfractuqqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqsq"><code>__satfractuqqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqta"><code>__satfractuqqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqquda"><code>__satfractuqquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqudq2"><code>__satfractuqqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqquha"><code>__satfractuqquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqquhq2"><code>__satfractuqquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqusa"><code>__satfractuqqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqqusq2"><code>__satfractuqqusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractuqquta"><code>__satfractuqquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusada"><code>__satfractusada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusadq"><code>__satfractusadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusaha"><code>__satfractusaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusahq"><code>__satfractusahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusaqq"><code>__satfractusaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusasa"><code>__satfractusasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusasq"><code>__satfractusasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusata"><code>__satfractusata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusauda2"><code>__satfractusauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusaudq"><code>__satfractusaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusauha2"><code>__satfractusauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusauhq"><code>__satfractusauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusauqq"><code>__satfractusauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusausq"><code>__satfractusausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusauta2"><code>__satfractusauta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqda"><code>__satfractusqda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqdq"><code>__satfractusqdq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqha"><code>__satfractusqha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqhq"><code>__satfractusqhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqqq"><code>__satfractusqqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqsa"><code>__satfractusqsa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqsq"><code>__satfractusqsq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqta"><code>__satfractusqta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusquda"><code>__satfractusquda</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqudq2"><code>__satfractusqudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusquha"><code>__satfractusquha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusquhq2"><code>__satfractusquhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusquqq2"><code>__satfractusquqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusqusa"><code>__satfractusqusa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractusquta"><code>__satfractusquta</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutada"><code>__satfractutada</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutadq"><code>__satfractutadq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutaha"><code>__satfractutaha</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutahq"><code>__satfractutahq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutaqq"><code>__satfractutaqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutasa"><code>__satfractutasa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutasq"><code>__satfractutasq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutata"><code>__satfractutata</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutauda2"><code>__satfractutauda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutaudq"><code>__satfractutaudq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutauha2"><code>__satfractutauha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutauhq"><code>__satfractutauhq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutauqq"><code>__satfractutauqq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutausa2"><code>__satfractutausa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsatfractutausq"><code>__satfractutausq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsplitstack_005ffind"><code>__splitstack_find</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Miscellaneous-routines">Miscellaneous routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddda3"><code>__ssaddda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssadddq3"><code>__ssadddq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddha3"><code>__ssaddha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddhq3"><code>__ssaddhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddqq3"><code>__ssaddqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddsa3"><code>__ssaddsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddsq3"><code>__ssaddsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssaddta3"><code>__ssaddta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlda3"><code>__ssashlda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashldq3"><code>__ssashldq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlha3"><code>__ssashlha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlhq3"><code>__ssashlhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlsa3"><code>__ssashlsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlsq3"><code>__ssashlsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssashlta3"><code>__ssashlta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivda3"><code>__ssdivda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivdq3"><code>__ssdivdq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivha3"><code>__ssdivha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivhq3"><code>__ssdivhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivqq3"><code>__ssdivqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivsa3"><code>__ssdivsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivsq3"><code>__ssdivsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssdivta3"><code>__ssdivta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulda3"><code>__ssmulda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmuldq3"><code>__ssmuldq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulha3"><code>__ssmulha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulhq3"><code>__ssmulhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulqq3"><code>__ssmulqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulsa3"><code>__ssmulsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulsq3"><code>__ssmulsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssmulta3"><code>__ssmulta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegda2"><code>__ssnegda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegdq2"><code>__ssnegdq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegha2"><code>__ssnegha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssneghq2"><code>__ssneghq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegqq2"><code>__ssnegqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegsa2"><code>__ssnegsa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegsq2"><code>__ssnegsq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fssnegta2"><code>__ssnegta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubda3"><code>__sssubda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubdq3"><code>__sssubdq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubha3"><code>__sssubha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubhq3"><code>__sssubhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubqq3"><code>__sssubqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubsa3"><code>__sssubsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubsq3"><code>__sssubsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsssubta3"><code>__sssubta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubda3"><code>__subda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubdf3"><code>__subdf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubdq3"><code>__subdq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubha3"><code>__subha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubhq3"><code>__subhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubqq3"><code>__subqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubsa3"><code>__subsa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubsf3"><code>__subsf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubsq3"><code>__subsq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubta3"><code>__subta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubtf3"><code>__subtf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubuda3"><code>__subuda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubudq3"><code>__subudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubuha3"><code>__subuha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubuhq3"><code>__subuhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubuqq3"><code>__subuqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubusa3"><code>__subusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubusq3"><code>__subusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubuta3"><code>__subuta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubvdi3"><code>__subvdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubvsi3"><code>__subvsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fsubxf3"><code>__subxf3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ftruncdfsf2"><code>__truncdfsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ftrunctfdf2"><code>__trunctfdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ftrunctfsf2"><code>__trunctfsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ftruncxfdf2"><code>__truncxfdf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005ftruncxfsf2"><code>__truncxfsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fucmpdi2"><code>__ucmpdi2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fucmpti2"><code>__ucmpti2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivdi3"><code>__udivdi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivmoddi4"><code>__udivmoddi4</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivmodti4"><code>__udivmodti4</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivsi3"><code>__udivsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivti3"><code>__udivti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivuda3"><code>__udivuda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivudq3"><code>__udivudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivuha3"><code>__udivuha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivuhq3"><code>__udivuhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivuqq3"><code>__udivuqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivusa3"><code>__udivusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivusq3"><code>__udivusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fudivuta3"><code>__udivuta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fumoddi3"><code>__umoddi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fumodsi3"><code>__umodsi3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fumodti3"><code>__umodti3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Integer-library-routines">Integer library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005funorddf2"><code>__unorddf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005funordsf2"><code>__unordsf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005funordtf2"><code>__unordtf2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusadduda3"><code>__usadduda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusaddudq3"><code>__usaddudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusadduha3"><code>__usadduha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusadduhq3"><code>__usadduhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusadduqq3"><code>__usadduqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusaddusa3"><code>__usaddusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusaddusq3"><code>__usaddusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusadduta3"><code>__usadduta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashluda3"><code>__usashluda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashludq3"><code>__usashludq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashluha3"><code>__usashluha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashluhq3"><code>__usashluhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashluqq3"><code>__usashluqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashlusa3"><code>__usashlusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashlusq3"><code>__usashlusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusashluta3"><code>__usashluta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivuda3"><code>__usdivuda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivudq3"><code>__usdivudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivuha3"><code>__usdivuha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivuhq3"><code>__usdivuhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivuqq3"><code>__usdivuqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivusa3"><code>__usdivusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivusq3"><code>__usdivusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusdivuta3"><code>__usdivuta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuluda3"><code>__usmuluda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuludq3"><code>__usmuludq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuluha3"><code>__usmuluha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuluhq3"><code>__usmuluhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuluqq3"><code>__usmuluqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmulusa3"><code>__usmulusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmulusq3"><code>__usmulusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusmuluta3"><code>__usmuluta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusneguda2"><code>__usneguda2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusnegudq2"><code>__usnegudq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusneguha2"><code>__usneguha2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusneguhq2"><code>__usneguhq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusneguqq2"><code>__usneguqq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusnegusa2"><code>__usnegusa2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusnegusq2"><code>__usnegusq2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fusneguta2"><code>__usneguta2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubuda3"><code>__ussubuda3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubudq3"><code>__ussubudq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubuha3"><code>__ussubuha3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubuhq3"><code>__ussubuhq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubuqq3"><code>__ussubuqq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubusa3"><code>__ussubusa3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubusq3"><code>__ussubusq3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-_005f_005fussubuta3"><code>__ussubuta3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-A">A</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-abort"><code>abort</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Portability">Portability</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-abs"><code>abs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-abs-and-attributes"><code>abs</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-absence_005fset"><code>absence_set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-absm2-instruction-pattern"><code>abs<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-absolute-value">absolute value</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ABS_005fEXPR"><code>ABS_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-access-to-operands">access to operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-access-to-special-operands">access to special operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-accessors">accessors</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ACCUMULATE_005fOUTGOING_005fARGS"><code>ACCUMULATE_OUTGOING_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ACCUMULATE_005fOUTGOING_005fARGS-and-stack-frames"><code>ACCUMULATE_OUTGOING_ARGS</code> and stack frames</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ACCUM_005fTYPE_005fSIZE"><code>ACCUM_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADA_005fLONG_005fTYPE_005fSIZE"><code>ADA_LONG_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Adding-a-new-GIMPLE-statement-code">Adding a new GIMPLE statement code</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Adding-a-new-GIMPLE-statement-code">Adding a new GIMPLE statement code</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADDITIONAL_005fREGISTER_005fNAMES"><code>ADDITIONAL_REGISTER_NAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addm3-instruction-pattern"><code>add<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addmodecc-instruction-pattern"><code>add<var>mode</var>cc</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-address-constraints">address constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addressing-modes">addressing modes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-address_005foperand"><code>address_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-address_005foperand-1"><code>address_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addr_005fdiff_005fvec"><code>addr_diff_vec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addr_005fdiff_005fvec_002c-length-of"><code>addr_diff_vec</code>, length of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADDR_005fEXPR"><code>ADDR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addr_005fvec"><code>addr_vec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-addr_005fvec_002c-length-of"><code>addr_vec</code>, length of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADJUST_005fFIELD_005fALIGN"><code>ADJUST_FIELD_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADJUST_005fINSN_005fLENGTH"><code>ADJUST_INSN_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ADJUST_005fREG_005fALLOC_005fORDER"><code>ADJUST_REG_ALLOC_ORDER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-aggregates-as-return-values">aggregates as return values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-alias">alias</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alias-analysis">Alias analysis</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-allocate_005fstack-instruction-pattern"><code>allocate_stack</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ALL_005fREGS"><code>ALL_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-alternate-entry-points">alternate entry points</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-anchored-addresses">anchored addresses</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-and"><code>and</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-and-and-attributes"><code>and</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-and_002c-canonicalization-of"><code>and</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-andm3-instruction-pattern"><code>and<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-annotations">annotations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Annotations">Annotations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-APPLY_005fRESULT_005fSIZE"><code>APPLY_RESULT_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARGS_005fGROW_005fDOWNWARD"><code>ARGS_GROW_DOWNWARD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-argument-passing">argument passing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arguments-in-registers">arguments in registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arguments-on-stack">arguments on stack</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARG_005fPOINTER_005fCFA_005fOFFSET"><code>ARG_POINTER_CFA_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARG_005fPOINTER_005fREGNUM"><code>ARG_POINTER_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARG_005fPOINTER_005fREGNUM-and-virtual-registers"><code>ARG_POINTER_REGNUM</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arg_005fpointer_005frtx"><code>arg_pointer_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arithmetic-library">arithmetic library</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arithmetic-shift">arithmetic shift</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arithmetic-shift-with-signed-saturation">arithmetic shift with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arithmetic-shift-with-unsigned-saturation">arithmetic shift with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-arithmetic_002c-in-RTL">arithmetic, in RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARITHMETIC_005fTYPE_005fP"><code>ARITHMETIC_TYPE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-array">array</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARRAY_005fRANGE_005fREF"><code>ARRAY_RANGE_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARRAY_005fREF"><code>ARRAY_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ARRAY_005fTYPE"><code>ARRAY_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashift"><code>ashift</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashift-and-attributes"><code>ashift</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashiftrt"><code>ashiftrt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashiftrt-and-attributes"><code>ashiftrt</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashlm3-instruction-pattern"><code>ashl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ashrm3-instruction-pattern"><code>ashr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fAPP_005fOFF"><code>ASM_APP_OFF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fAPP_005fON"><code>ASM_APP_ON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fCOMMENT_005fSTART"><code>ASM_COMMENT_START</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fDECLARE_005fFUNCTION_005fNAME"><code>ASM_DECLARE_FUNCTION_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fDECLARE_005fFUNCTION_005fSIZE"><code>ASM_DECLARE_FUNCTION_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fDECLARE_005fOBJECT_005fNAME"><code>ASM_DECLARE_OBJECT_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fDECLARE_005fREGISTER_005fGLOBAL"><code>ASM_DECLARE_REGISTER_GLOBAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fFINAL_005fSPEC"><code>ASM_FINAL_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fFINISH_005fDECLARE_005fOBJECT"><code>ASM_FINISH_DECLARE_OBJECT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fFORMAT_005fPRIVATE_005fNAME"><code>ASM_FORMAT_PRIVATE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005ffprintf"><code>asm_fprintf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fFPRINTF_005fEXTENSIONS"><code>ASM_FPRINTF_EXTENSIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fGENERATE_005fINTERNAL_005fLABEL"><code>ASM_GENERATE_INTERNAL_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005finput"><code>asm_input</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005finput-and-_002fv"><code>asm_input</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fMAYBE_005fOUTPUT_005fENCODED_005fADDR_005fRTX"><code>ASM_MAYBE_OUTPUT_ENCODED_ADDR_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005fnoperands"><code>asm_noperands</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fNO_005fSKIP_005fIN_005fTEXT"><code>ASM_NO_SKIP_IN_TEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005foperands-and-_002fv"><code>asm_operands</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005foperands_002c-RTL-sharing"><code>asm_operands</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asm_005foperands_002c-usage"><code>asm_operands</code>, usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Assembler">Assembler</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fADDR_005fDIFF_005fELT"><code>ASM_OUTPUT_ADDR_DIFF_ELT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fADDR_005fVEC_005fELT"><code>ASM_OUTPUT_ADDR_VEC_ELT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGN"><code>ASM_OUTPUT_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGNED_005fBSS"><code>ASM_OUTPUT_ALIGNED_BSS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGNED_005fCOMMON"><code>ASM_OUTPUT_ALIGNED_COMMON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGNED_005fDECL_005fCOMMON"><code>ASM_OUTPUT_ALIGNED_DECL_COMMON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGNED_005fDECL_005fLOCAL"><code>ASM_OUTPUT_ALIGNED_DECL_LOCAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGNED_005fLOCAL"><code>ASM_OUTPUT_ALIGNED_LOCAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fALIGN_005fWITH_005fNOP"><code>ASM_OUTPUT_ALIGN_WITH_NOP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fASCII"><code>ASM_OUTPUT_ASCII</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fCASE_005fEND"><code>ASM_OUTPUT_CASE_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fCASE_005fLABEL"><code>ASM_OUTPUT_CASE_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fCOMMON"><code>ASM_OUTPUT_COMMON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDEBUG_005fLABEL"><code>ASM_OUTPUT_DEBUG_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDEF"><code>ASM_OUTPUT_DEF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDEF_005fFROM_005fDECLS"><code>ASM_OUTPUT_DEF_FROM_DECLS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDWARF_005fDELTA"><code>ASM_OUTPUT_DWARF_DELTA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDWARF_005fOFFSET"><code>ASM_OUTPUT_DWARF_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDWARF_005fPCREL"><code>ASM_OUTPUT_DWARF_PCREL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDWARF_005fTABLE_005fREF"><code>ASM_OUTPUT_DWARF_TABLE_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fDWARF_005fVMS_005fDELTA"><code>ASM_OUTPUT_DWARF_VMS_DELTA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fEXTERNAL"><code>ASM_OUTPUT_EXTERNAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fFDESC"><code>ASM_OUTPUT_FDESC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fFUNCTION_005fLABEL"><code>ASM_OUTPUT_FUNCTION_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fINTERNAL_005fLABEL"><code>ASM_OUTPUT_INTERNAL_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fLABEL"><code>ASM_OUTPUT_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fLABELREF"><code>ASM_OUTPUT_LABELREF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fLABEL_005fREF"><code>ASM_OUTPUT_LABEL_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fLOCAL"><code>ASM_OUTPUT_LOCAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Uninitialized-Data">Uninitialized Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fMAX_005fSKIP_005fALIGN"><code>ASM_OUTPUT_MAX_SKIP_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fMEASURED_005fSIZE"><code>ASM_OUTPUT_MEASURED_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fOPCODE"><code>ASM_OUTPUT_OPCODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fPOOL_005fEPILOGUE"><code>ASM_OUTPUT_POOL_EPILOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fPOOL_005fPROLOGUE"><code>ASM_OUTPUT_POOL_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fREG_005fPOP"><code>ASM_OUTPUT_REG_POP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fREG_005fPUSH"><code>ASM_OUTPUT_REG_PUSH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fSIZE_005fDIRECTIVE"><code>ASM_OUTPUT_SIZE_DIRECTIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fSKIP"><code>ASM_OUTPUT_SKIP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fSOURCE_005fFILENAME"><code>ASM_OUTPUT_SOURCE_FILENAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fSPECIAL_005fPOOL_005fENTRY"><code>ASM_OUTPUT_SPECIAL_POOL_ENTRY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fSYMBOL_005fREF"><code>ASM_OUTPUT_SYMBOL_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fTYPE_005fDIRECTIVE"><code>ASM_OUTPUT_TYPE_DIRECTIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fWEAKREF"><code>ASM_OUTPUT_WEAKREF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fOUTPUT_005fWEAK_005fALIAS"><code>ASM_OUTPUT_WEAK_ALIAS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fPREFERRED_005fEH_005fDATA_005fFORMAT"><code>ASM_PREFERRED_EH_DATA_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fSPEC"><code>ASM_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fSTABD_005fOP"><code>ASM_STABD_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fSTABN_005fOP"><code>ASM_STABN_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fSTABS_005fOP"><code>ASM_STABS_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fWEAKEN_005fDECL"><code>ASM_WEAKEN_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASM_005fWEAKEN_005fLABEL"><code>ASM_WEAKEN_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-assembler-format">assembler format</a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-assembler-instructions-in-RTL">assembler instructions in RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Assembler">Assembler</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASSEMBLER_005fDIALECT"><code>ASSEMBLER_DIALECT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-assemble_005fname"><code>assemble_name</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-assemble_005fname_005fraw"><code>assemble_name_raw</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-assigning-attribute-values-to-insns">assigning attribute values to insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ASSUME_005fEXTENDED_005fUNWIND_005fCONTEXT"><code>ASSUME_EXTENDED_UNWIND_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-asterisk-in-template">asterisk in template</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-AS_005fNEEDS_005fDASH_005fFOR_005fPIPED_005fINPUT"><code>AS_NEEDS_DASH_FOR_PIPED_INPUT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atan2m3-instruction-pattern"><code>atan2<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic"><code>atomic</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005faddmode-instruction-pattern"><code>atomic_add<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fadd_005ffetchmode-instruction-pattern"><code>atomic_add_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fandmode-instruction-pattern"><code>atomic_and<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fand_005ffetchmode-instruction-pattern"><code>atomic_and_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fcompare_005fand_005fswapmode-instruction-pattern"><code>atomic_compare_and_swap<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fexchangemode-instruction-pattern"><code>atomic_exchange<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005faddmode-instruction-pattern"><code>atomic_fetch_add<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005fandmode-instruction-pattern"><code>atomic_fetch_and<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005fnandmode-instruction-pattern"><code>atomic_fetch_nand<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005formode-instruction-pattern"><code>atomic_fetch_or<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005fsubmode-instruction-pattern"><code>atomic_fetch_sub<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ffetch_005fxormode-instruction-pattern"><code>atomic_fetch_xor<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005floadmode-instruction-pattern"><code>atomic_load<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fnandmode-instruction-pattern"><code>atomic_nand<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fnand_005ffetchmode-instruction-pattern"><code>atomic_nand_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005formode-instruction-pattern"><code>atomic_or<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005for_005ffetchmode-instruction-pattern"><code>atomic_or_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fstoremode-instruction-pattern"><code>atomic_store<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fsubmode-instruction-pattern"><code>atomic_sub<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fsub_005ffetchmode-instruction-pattern"><code>atomic_sub_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005ftest_005fand_005fset-instruction-pattern"><code>atomic_test_and_set</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fxormode-instruction-pattern"><code>atomic_xor<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-atomic_005fxor_005ffetchmode-instruction-pattern"><code>atomic_xor_fetch<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attr"><code>attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attr-1"><code>attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attribute-expressions">attribute expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attribute-specifications">attribute specifications</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Attr-Example">Attr Example</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attribute-specifications-example">attribute specifications example</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Attr-Example">Attr Example</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attributes">attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Attributes">Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attributes_002c-defining">attributes, defining</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Attributes">Defining Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attributes_002c-target_002dspecific">attributes, target-specific</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ATTRIBUTE_005fALIGNED_005fVALUE"><code>ATTRIBUTE_ALIGNED_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-attr_005fflag"><code>attr_flag</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-autoincrement-addressing_002c-availability">autoincrement addressing, availability</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Portability">Portability</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-autoincrement_002fdecrement-addressing">autoincrement/decrement addressing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-automata_005foption"><code>automata_option</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-automaton-based-pipeline-description">automaton based pipeline description</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-automaton-based-pipeline-description-1">automaton based pipeline description</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-automaton-based-scheduler">automaton based scheduler</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-AVOID_005fCCMODE_005fCOPIES"><code>AVOID_CCMODE_COPIES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-B">B</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-backslash">backslash</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-barrier"><code>barrier</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-barrier-and-_002ff"><code>barrier</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-barrier-and-_002fv"><code>barrier</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BASE_005fREG_005fCLASS"><code>BASE_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-basic-block">basic block</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Basic-Statements">Basic Statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Statements">Basic Statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-basic_002dblock_002eh"><code>basic-block.h</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Control-Flow">Control Flow</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-basic_005fblock"><code>basic_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BASIC_005fBLOCK"><code>BASIC_BLOCK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BB_005fHEAD_002c-BB_005fEND"><code>BB_HEAD, BB_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bb_005fseq"><code>bb_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIGGEST_005fALIGNMENT"><code>BIGGEST_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIGGEST_005fFIELD_005fALIGNMENT"><code>BIGGEST_FIELD_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BImode"><code>BImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIND_005fEXPR"><code>BIND_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BINFO_005fTYPE"><code>BINFO_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bit_002dfields">bit-fields</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Bit_002dFields">Bit-Fields</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BITFIELD_005fNBYTES_005fLIMITED"><code>BITFIELD_NBYTES_LIMITED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BITS_005fBIG_005fENDIAN"><code>BITS_BIG_ENDIAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BITS_005fBIG_005fENDIAN_002c-effect-on-sign_005fextract"><code>BITS_BIG_ENDIAN</code>, effect on <code>sign_extract</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Bit_002dFields">Bit-Fields</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BITS_005fPER_005fUNIT"><code>BITS_PER_UNIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BITS_005fPER_005fWORD"><code>BITS_PER_WORD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bitwise-complement">bitwise complement</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bitwise-exclusive_002dor">bitwise exclusive-or</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bitwise-inclusive_002dor">bitwise inclusive-or</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bitwise-logical_002dand">bitwise logical-and</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIT_005fAND_005fEXPR"><code>BIT_AND_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIT_005fIOR_005fEXPR"><code>BIT_IOR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIT_005fNOT_005fEXPR"><code>BIT_NOT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BIT_005fXOR_005fEXPR"><code>BIT_XOR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BLKmode"><code>BLKmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BLKmode_002c-and-function-return-values"><code>BLKmode</code>, and function return values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Calls">Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-blockage-instruction-pattern"><code>blockage</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Blocks">Blocks</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Blocks">Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BLOCK_005fFOR_005fINSN_002c-gimple_005fbb"><code>BLOCK_FOR_INSN, gimple_bb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BLOCK_005fREG_005fPADDING"><code>BLOCK_REG_PADDING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bool"><code>bool</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BOOLEAN_005fTYPE"><code>BOOLEAN_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BOOL_005fTYPE_005fSIZE"><code>BOOL_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-branch-prediction">branch prediction</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BRANCH_005fCOST"><code>BRANCH_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-break_005fout_005fmemory_005frefs"><code>break_out_memory_refs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BREAK_005fSTMT"><code>BREAK_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BSS_005fSECTION_005fASM_005fOP"><code>BSS_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bswap"><code>bswap</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-bswapm2-instruction-pattern"><code>bswap<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-btruncm2-instruction-pattern"><code>btrunc<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build0"><code>build0</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build1"><code>build1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build2"><code>build2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build3"><code>build3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build4"><code>build4</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build5"><code>build5</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-build6"><code>build6</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-builtin_005flongjmp-instruction-pattern"><code>builtin_longjmp</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-builtin_005fsetjmp_005freceiver-instruction-pattern"><code>builtin_setjmp_receiver</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-builtin_005fsetjmp_005fsetup-instruction-pattern"><code>builtin_setjmp_setup</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BYTES_005fBIG_005fENDIAN"><code>BYTES_BIG_ENDIAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-BYTES_005fBIG_005fENDIAN_002c-effect-on-subreg"><code>BYTES_BIG_ENDIAN</code>, effect on <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-byte_005fmode"><code>byte_mode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-C">C</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-C-statements-for-assembler-output">C statements for assembler output</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-C99-math-functions_002c-implicit-usage">C99 math functions, implicit usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call"><code>call</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call-1"><code>call</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call-instruction-pattern"><code>call</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call-usage"><code>call</code> usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Calls">Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002c-in-call_005finsn"><code>call</code>, in <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002c-in-mem"><code>call</code>, in <code>mem</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dclobbered-register">call-clobbered register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dclobbered-register-1">call-clobbered register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dclobbered-register-2">call-clobbered register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dsaved-register">call-saved register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dsaved-register-1">call-saved register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dsaved-register-2">call-saved register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dused-register">call-used register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dused-register-1">call-used register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_002dused-register-2">call-used register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALLER_005fSAVE_005fPROFITABLE"><code>CALLER_SAVE_PROFITABLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Caller-Saves">Caller Saves</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-calling-conventions">calling conventions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-and-Calling">Stack and Calling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-calling-functions-in-RTL">calling functions in RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Calls">Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALL_005fEXPR"><code>CALL_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn"><code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fc"><code>call_insn</code> and &lsquo;<samp>/c</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002ff"><code>call_insn</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fi"><code>call_insn</code> and &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fj"><code>call_insn</code> and &lsquo;<samp>/j</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fs"><code>call_insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fs-1"><code>call_insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fu"><code>call_insn</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fu-1"><code>call_insn</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fu-or-_002fi"><code>call_insn</code> and &lsquo;<samp>/u</samp>&rsquo; or &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005finsn-and-_002fv"><code>call_insn</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALL_005fINSN_005fFUNCTION_005fUSAGE"><code>CALL_INSN_FUNCTION_USAGE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005fpop-instruction-pattern"><code>call_pop</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALL_005fPOPS_005fARGS"><code>CALL_POPS_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALL_005fREALLY_005fUSED_005fREGISTERS"><code>CALL_REALLY_USED_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CALL_005fUSED_005fREGISTERS"><code>CALL_USED_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005fused_005fregs"><code>call_used_regs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005fvalue-instruction-pattern"><code>call_value</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-call_005fvalue_005fpop-instruction-pattern"><code>call_value_pop</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-canadian">canadian</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Configure-Terms">Configure Terms</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CANNOT_005fCHANGE_005fMODE_005fCLASS"><code>CANNOT_CHANGE_MODE_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CANNOT_005fCHANGE_005fMODE_005fCLASS-and-subreg-semantics"><code>CANNOT_CHANGE_MODE_CLASS</code> and subreg semantics</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-canonicalization-of-instructions">canonicalization of instructions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-canonicalize_005ffuncptr_005ffor_005fcompare-instruction-pattern"><code>canonicalize_funcptr_for_compare</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-can_005fcreate_005fpseudo_005fp"><code>can_create_pseudo_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-can_005ffallthru"><code>can_fallthru</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-casesi-instruction-pattern"><code>casesi</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CASE_005fVECTOR_005fMODE"><code>CASE_VECTOR_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CASE_005fVECTOR_005fPC_005fRELATIVE"><code>CASE_VECTOR_PC_RELATIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CASE_005fVECTOR_005fSHORTEN_005fMODE"><code>CASE_VECTOR_SHORTEN_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cbranchmode4-instruction-pattern"><code>cbranch<var>mode</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cc0"><code>cc0</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cc0-1"><code>cc0</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#CC0-Condition-Codes">CC0 Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cc0_002c-RTL-sharing"><code>cc0</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cc0_005frtx"><code>cc0_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CC1PLUS_005fSPEC"><code>CC1PLUS_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CC1_005fSPEC"><code>CC1_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CCmode"><code>CCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CCmode-1"><code>CCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cc_005fstatus"><code>cc_status</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#CC0-Condition-Codes">CC0 Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CC_005fSTATUS_005fMDEP"><code>CC_STATUS_MDEP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#CC0-Condition-Codes">CC0 Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CC_005fSTATUS_005fMDEP_005fINIT"><code>CC_STATUS_MDEP_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#CC0-Condition-Codes">CC0 Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CDImode"><code>CDImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ceilm2-instruction-pattern"><code>ceil<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CEIL_005fDIV_005fEXPR"><code>CEIL_DIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CEIL_005fMOD_005fEXPR"><code>CEIL_MOD_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CFA_005fFRAME_005fBASE_005fOFFSET"><code>CFA_FRAME_BASE_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CFG-verification">CFG verification</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CFG_002c-Control-Flow-Graph">CFG, Control Flow Graph</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Control-Flow">Control Flow</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cfghooks_002eh"><code>cfghooks.h</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cgraph_005ffinalize_005ffunction"><code>cgraph_finalize_function</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-chain_005fcircular"><code>chain_circular</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-chain_005fnext"><code>chain_next</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-chain_005fprev"><code>chain_prev</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-change_005faddress"><code>change_address</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CHAR_005fTYPE_005fSIZE"><code>CHAR_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-check_005fstack-instruction-pattern"><code>check_stack</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CHImode"><code>CHImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-class-definitions_002c-register">class definitions, register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-class-preference-constraints">class preference constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Class-Preferences">Class Preferences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-class_002c-scope">class, scope</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-classes-of-RTX-codes">classes of RTX codes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLASSTYPE_005fDECLARED_005fCLASS"><code>CLASSTYPE_DECLARED_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLASSTYPE_005fHAS_005fMUTABLE"><code>CLASSTYPE_HAS_MUTABLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLASSTYPE_005fNON_005fPOD_005fP"><code>CLASSTYPE_NON_POD_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLASS_005fMAX_005fNREGS"><code>CLASS_MAX_NREGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLASS_005fTYPE_005fP"><code>CLASS_TYPE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Cleanups">Cleanups</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Cleanups">Cleanups</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEANUP_005fDECL"><code>CLEANUP_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEANUP_005fEXPR"><code>CLEANUP_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEANUP_005fPOINT_005fEXPR"><code>CLEANUP_POINT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEANUP_005fSTMT"><code>CLEANUP_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEAR_005fBY_005fPIECES_005fP"><code>CLEAR_BY_PIECES_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-clear_005fcache-instruction-pattern"><code>clear_cache</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEAR_005fINSN_005fCACHE"><code>CLEAR_INSN_CACHE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLEAR_005fRATIO"><code>CLEAR_RATIO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-clobber"><code>clobber</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-clrsb"><code>clrsb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-clz"><code>clz</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-clzm2-instruction-pattern"><code>clz<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CLZ_005fDEFINED_005fVALUE_005fAT_005fZERO"><code>CLZ_DEFINED_VALUE_AT_ZERO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cmpmemm-instruction-pattern"><code>cmpmem<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cmpstrm-instruction-pattern"><code>cmpstr<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cmpstrnm-instruction-pattern"><code>cmpstrn<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-code-generation-RTL-sequences">code generation RTL sequences</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-code-iterators-in-_002emd-files">code iterators in <samp>.md</samp> files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Code-Iterators">Code Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-codes_002c-RTL-expression">codes, RTL expression</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-code_005flabel"><code>code_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CODE_005fLABEL"><code>CODE_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-code_005flabel-and-_002fi"><code>code_label</code> and &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-code_005flabel-and-_002fv"><code>code_label</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CODE_005fLABEL_005fNUMBER"><code>CODE_LABEL_NUMBER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COImode"><code>COImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COLLECT2_005fHOST_005fINITIALIZATION"><code>COLLECT2_HOST_INITIALIZATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COLLECT_005fEXPORT_005fLIST"><code>COLLECT_EXPORT_LIST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COLLECT_005fSHARED_005fFINI_005fFUNC"><code>COLLECT_SHARED_FINI_FUNC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COLLECT_005fSHARED_005fINIT_005fFUNC"><code>COLLECT_SHARED_INIT_FUNC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-commit_005fedge_005finsertions"><code>commit_edge_insertions</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-compare"><code>compare</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-compare_002c-canonicalization-of"><code>compare</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-comparison_005foperator"><code>comparison_operator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-compiler-passes-and-files">compiler passes and files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Passes">Passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-complement_002c-bitwise">complement, bitwise</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPLEX_005fCST"><code>COMPLEX_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPLEX_005fEXPR"><code>COMPLEX_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPLEX_005fTYPE"><code>COMPLEX_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPONENT_005fREF"><code>COMPONENT_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Compound-Expressions">Compound Expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Compound-Expressions">Compound Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Compound-Lvalues">Compound Lvalues</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Compound-Lvalues">Compound Lvalues</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPOUND_005fEXPR"><code>COMPOUND_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPOUND_005fLITERAL_005fEXPR"><code>COMPOUND_LITERAL_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPOUND_005fLITERAL_005fEXPR_005fDECL"><code>COMPOUND_LITERAL_EXPR_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COMPOUND_005fLITERAL_005fEXPR_005fDECL_005fEXPR"><code>COMPOUND_LITERAL_EXPR_DECL_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-computed-jump">computed jump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-computing-the-length-of-an-insn">computing the length of an insn</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-concat"><code>concat</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-concatn"><code>concatn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cond"><code>cond</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cond-and-attributes"><code>cond</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-condition-code-register">condition code register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-condition-code-status">condition code status</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Condition-Code">Condition Code</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-condition-codes">condition codes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-conditional-execution">conditional execution</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conditional-Execution">Conditional Execution</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Conditional-Expressions">Conditional Expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conditional-Expressions">Conditional Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-conditions_002c-in-patterns">conditions, in patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cond_005fexec"><code>cond_exec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-COND_005fEXPR"><code>COND_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-configuration-file">configuration file</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-configuration-file-1">configuration file</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-configure-terms">configure terms</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Configure-Terms">Configure Terms</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONJ_005fEXPR"><code>CONJ_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const"><code>const</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const0_005frtx"><code>const0_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST0_005fRTX"><code>CONST0_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const1_005frtx"><code>const1_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST1_005fRTX"><code>CONST1_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const2_005frtx"><code>const2_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST2_005fRTX"><code>CONST2_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constant-attributes">constant attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Attributes">Constant Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constant-definitions">constant definitions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constants-in-constraints">constants in constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTANT_005fADDRESS_005fP"><code>CONSTANT_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTANT_005fALIGNMENT"><code>CONSTANT_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTANT_005fP"><code>CONSTANT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTANT_005fPOOL_005fADDRESS_005fP"><code>CONSTANT_POOL_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTANT_005fPOOL_005fBEFORE_005fFUNCTION"><code>CONSTANT_POOL_BEFORE_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constm1_005frtx"><code>constm1_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraint-modifier-characters">constraint modifier characters</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraint_002c-matching">constraint, matching</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraints">constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constraints">Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraints_002c-defining">constraints, defining</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraints_002c-defining_002c-obsolete-method">constraints, defining, obsolete method</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraints_002c-machine-specific">constraints, machine specific</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Constraints">Machine Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraints_002c-testing">constraints, testing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTRAINT_005fLEN"><code>CONSTRAINT_LEN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraint_005fnum"><code>constraint_num</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constraint_005fsatisfied_005fp"><code>constraint_satisfied_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONSTRUCTOR"><code>CONSTRUCTOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constructors_002c-automatic-calls">constructors, automatic calls</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Collect2">Collect2</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-constructors_002c-output-of">constructors, output of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fDECL"><code>CONST_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fdouble"><code>const_double</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fdouble_002c-RTL-sharing"><code>const_double</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fDOUBLE_005fLOW"><code>CONST_DOUBLE_LOW</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fDOUBLE_005fOK_005fFOR_005fCONSTRAINT_005fP"><code>CONST_DOUBLE_OK_FOR_CONSTRAINT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fDOUBLE_005fOK_005fFOR_005fLETTER_005fP"><code>CONST_DOUBLE_OK_FOR_LETTER_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fdouble_005foperand"><code>const_double_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005ffixed"><code>const_fixed</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fint"><code>const_int</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fint-and-attribute-tests"><code>const_int</code> and attribute tests</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fint-and-attributes"><code>const_int</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fint_002c-RTL-sharing"><code>const_int</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fint_005foperand"><code>const_int_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fOK_005fFOR_005fCONSTRAINT_005fP"><code>CONST_OK_FOR_CONSTRAINT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONST_005fOK_005fFOR_005fLETTER_005fP"><code>CONST_OK_FOR_LETTER_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fstring"><code>const_string</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fstring-and-attributes"><code>const_string</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005ftrue_005frtx"><code>const_true_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fvector"><code>const_vector</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-const_005fvector_002c-RTL-sharing"><code>const_vector</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-container">container</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONTINUE_005fSTMT"><code>CONTINUE_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-contributors">contributors</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Contributors">Contributors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-controlling-register-usage">controlling register usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-controlling-the-compilation-driver">controlling the compilation driver</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-conventions_002c-run_002dtime">conventions, run-time</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-conversions">conversions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CONVERT_005fEXPR"><code>CONVERT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-copysignm3-instruction-pattern"><code>copysign<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-copy_005frtx"><code>copy_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-copy_005frtx_005fif_005fshared"><code>copy_rtx_if_shared</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cosm2-instruction-pattern"><code>cos<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-costs-of-instructions">costs of instructions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CPLUSPLUS_005fCPP_005fSPEC"><code>CPLUSPLUS_CPP_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CPP_005fSPEC"><code>CPP_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CP_005fINTEGRAL_005fTYPE"><code>CP_INTEGRAL_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cp_005fnamespace_005fdecls"><code>cp_namespace_decls</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CP_005fTYPE_005fCONST_005fNON_005fVOLATILE_005fP"><code>CP_TYPE_CONST_NON_VOLATILE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CP_005fTYPE_005fCONST_005fP"><code>CP_TYPE_CONST_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cp_005ftype_005fquals"><code>cp_type_quals</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cp_005ftype_005fquals-1"><code>cp_type_quals</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CP_005fTYPE_005fRESTRICT_005fP"><code>CP_TYPE_RESTRICT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CP_005fTYPE_005fVOLATILE_005fP"><code>CP_TYPE_VOLATILE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CQImode"><code>CQImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cross-compilation-and-floating-point">cross compilation and floating point</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-crtl_002d_003eargs_002epops_005fargs"><code>crtl-&gt;args.pops_args</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-crtl_002d_003eargs_002epretend_005fargs_005fsize"><code>crtl-&gt;args.pretend_args_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-crtl_002d_003eoutgoing_005fargs_005fsize"><code>crtl-&gt;outgoing_args_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CRTSTUFF_005fT_005fCFLAGS"><code>CRTSTUFF_T_CFLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CRTSTUFF_005fT_005fCFLAGS_005fS"><code>CRTSTUFF_T_CFLAGS_S</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CRT_005fCALL_005fSTATIC_005fFUNCTION"><code>CRT_CALL_STATIC_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CSImode"><code>CSImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-cstoremode4-instruction-pattern"><code>cstore<var>mode</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CTImode"><code>CTImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ctrapMM4-instruction-pattern"><code>ctrap<var>MM</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ctz"><code>ctz</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ctzm2-instruction-pattern"><code>ctz<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CTZ_005fDEFINED_005fVALUE_005fAT_005fZERO"><code>CTZ_DEFINED_VALUE_AT_ZERO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-CUMULATIVE_005fARGS"><code>CUMULATIVE_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-current_005ffunction_005fis_005fleaf"><code>current_function_is_leaf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-current_005ffunction_005fuses_005fonly_005fleaf_005fregs"><code>current_function_uses_only_leaf_regs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-current_005finsn_005fpredicate"><code>current_insn_predicate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conditional-Execution">Conditional Execution</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-C_005fCOMMON_005fOVERRIDE_005fOPTIONS"><code>C_COMMON_OVERRIDE_OPTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-c_005fregister_005fpragma"><code>c_register_pragma</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-c_005fregister_005fpragma_005fwith_005fexpansion"><code>c_register_pragma_with_expansion</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-D">D</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-DAmode"><code>DAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-data-bypass">data bypass</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-data-bypass-1">data bypass</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-data-dependence-delays">data dependence delays</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Data-Dependency-Analysis">Data Dependency Analysis</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dependency-analysis">Dependency analysis</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-data-structures">data structures</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Per_002dFunction-Data">Per-Function Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DATA_005fALIGNMENT"><code>DATA_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DATA_005fSECTION_005fASM_005fOP"><code>DATA_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBR_005fOUTPUT_005fSEQEND"><code>DBR_OUTPUT_SEQEND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-dbr_005fsequence_005flength"><code>dbr_sequence_length</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fBLOCKS_005fFUNCTION_005fRELATIVE"><code>DBX_BLOCKS_FUNCTION_RELATIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fCONTIN_005fCHAR"><code>DBX_CONTIN_CHAR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fCONTIN_005fLENGTH"><code>DBX_CONTIN_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fDEBUGGING_005fINFO"><code>DBX_DEBUGGING_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fFUNCTION_005fFIRST"><code>DBX_FUNCTION_FIRST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fLINES_005fFUNCTION_005fRELATIVE"><code>DBX_LINES_FUNCTION_RELATIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fNO_005fXREFS"><code>DBX_NO_XREFS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fOUTPUT_005fMAIN_005fSOURCE_005fFILENAME"><code>DBX_OUTPUT_MAIN_SOURCE_FILENAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Names-and-DBX">File Names and DBX</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fOUTPUT_005fMAIN_005fSOURCE_005fFILE_005fEND"><code>DBX_OUTPUT_MAIN_SOURCE_FILE_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Names-and-DBX">File Names and DBX</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fOUTPUT_005fNULL_005fN_005fSO_005fAT_005fMAIN_005fSOURCE_005fFILE_005fEND"><code>DBX_OUTPUT_NULL_N_SO_AT_MAIN_SOURCE_FILE_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Names-and-DBX">File Names and DBX</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fOUTPUT_005fSOURCE_005fLINE"><code>DBX_OUTPUT_SOURCE_LINE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Hooks">DBX Hooks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fREGISTER_005fNUMBER"><code>DBX_REGISTER_NUMBER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#All-Debuggers">All Debuggers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fREGPARM_005fSTABS_005fCODE"><code>DBX_REGPARM_STABS_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fREGPARM_005fSTABS_005fLETTER"><code>DBX_REGPARM_STABS_LETTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fSTATIC_005fCONST_005fVAR_005fCODE"><code>DBX_STATIC_CONST_VAR_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fSTATIC_005fSTAB_005fDATA_005fSECTION"><code>DBX_STATIC_STAB_DATA_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fTYPE_005fDECL_005fSTABS_005fCODE"><code>DBX_TYPE_DECL_STABS_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DBX_005fUSE_005fBINCL"><code>DBX_USE_BINCL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DCmode"><code>DCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DDmode"><code>DDmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-De-Morgan_0027s-law">De Morgan&rsquo;s law</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-dead_005for_005fset_005fp"><code>dead_or_set_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole">define_peephole</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEBUGGER_005fARG_005fOFFSET"><code>DEBUGGER_ARG_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#All-Debuggers">All Debuggers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEBUGGER_005fAUTO_005fOFFSET"><code>DEBUGGER_AUTO_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#All-Debuggers">All Debuggers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-debug_005fexpr"><code>debug_expr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Debug-Information">Debug Information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEBUG_005fEXPR_005fDECL"><code>DEBUG_EXPR_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-debug_005finsn"><code>debug_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEBUG_005fSYMS_005fTEXT"><code>DEBUG_SYMS_TEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-decimal-float-library">decimal float library</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-declaration">declaration</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-declarations_002c-RTL">declarations, RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Declarations">RTL Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECLARE_005fLIBRARY_005fRENAMES"><code>DECLARE_LIBRARY_RENAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fALIGN"><code>DECL_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fANTICIPATED"><code>DECL_ANTICIPATED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fARGUMENTS"><code>DECL_ARGUMENTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fARRAY_005fDELETE_005fOPERATOR_005fP"><code>DECL_ARRAY_DELETE_OPERATOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fARTIFICIAL"><code>DECL_ARTIFICIAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Working-with-declarations">Working with declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fARTIFICIAL-1"><code>DECL_ARTIFICIAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fARTIFICIAL-2"><code>DECL_ARTIFICIAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fASSEMBLER_005fNAME"><code>DECL_ASSEMBLER_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fASSEMBLER_005fNAME-1"><code>DECL_ASSEMBLER_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fATTRIBUTES"><code>DECL_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Attributes">Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fBASE_005fCONSTRUCTOR_005fP"><code>DECL_BASE_CONSTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCOMPLETE_005fCONSTRUCTOR_005fP"><code>DECL_COMPLETE_CONSTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCOMPLETE_005fDESTRUCTOR_005fP"><code>DECL_COMPLETE_DESTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCONSTRUCTOR_005fP"><code>DECL_CONSTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCONST_005fMEMFUNC_005fP"><code>DECL_CONST_MEMFUNC_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCONTEXT"><code>DECL_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCONV_005fFN_005fP"><code>DECL_CONV_FN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fCOPY_005fCONSTRUCTOR_005fP"><code>DECL_COPY_CONSTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fDESTRUCTOR_005fP"><code>DECL_DESTRUCTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fEXTERNAL"><code>DECL_EXTERNAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fEXTERNAL-1"><code>DECL_EXTERNAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fEXTERN_005fC_005fFUNCTION_005fP"><code>DECL_EXTERN_C_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fFUNCTION_005fMEMBER_005fP"><code>DECL_FUNCTION_MEMBER_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fFUNCTION_005fSPECIFIC_005fOPTIMIZATION"><code>DECL_FUNCTION_SPECIFIC_OPTIMIZATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fFUNCTION_005fSPECIFIC_005fOPTIMIZATION-1"><code>DECL_FUNCTION_SPECIFIC_OPTIMIZATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fFUNCTION_005fSPECIFIC_005fTARGET"><code>DECL_FUNCTION_SPECIFIC_TARGET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fFUNCTION_005fSPECIFIC_005fTARGET-1"><code>DECL_FUNCTION_SPECIFIC_TARGET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fGLOBAL_005fCTOR_005fP"><code>DECL_GLOBAL_CTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fGLOBAL_005fDTOR_005fP"><code>DECL_GLOBAL_DTOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fINITIAL"><code>DECL_INITIAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fINITIAL-1"><code>DECL_INITIAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fLINKONCE_005fP"><code>DECL_LINKONCE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fLOCAL_005fFUNCTION_005fP"><code>DECL_LOCAL_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fMAIN_005fP"><code>DECL_MAIN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAME"><code>DECL_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Working-with-declarations">Working with declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAME-1"><code>DECL_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAME-2"><code>DECL_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAME-3"><code>DECL_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAMESPACE_005fALIAS"><code>DECL_NAMESPACE_ALIAS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNAMESPACE_005fSTD_005fP"><code>DECL_NAMESPACE_STD_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNONCONVERTING_005fP"><code>DECL_NONCONVERTING_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNONSTATIC_005fMEMBER_005fFUNCTION_005fP"><code>DECL_NONSTATIC_MEMBER_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fNON_005fTHUNK_005fFUNCTION_005fP"><code>DECL_NON_THUNK_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fOVERLOADED_005fOPERATOR_005fP"><code>DECL_OVERLOADED_OPERATOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fPURE_005fP"><code>DECL_PURE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fRESULT"><code>DECL_RESULT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fSAVED_005fTREE"><code>DECL_SAVED_TREE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fSIZE"><code>DECL_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fSTATIC_005fFUNCTION_005fP"><code>DECL_STATIC_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fSTMT"><code>DECL_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fSTMT_005fDECL"><code>DECL_STMT_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fTHUNK_005fP"><code>DECL_THUNK_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fVIRTUAL_005fP"><code>DECL_VIRTUAL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DECL_005fVOLATILE_005fMEMFUNC_005fP"><code>DECL_VOLATILE_MEMFUNC_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-decrement_005fand_005fbranch_005funtil_005fzero-instruction-pattern"><code>decrement_and_branch_until_zero</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-default"><code>default</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-default_005ffile_005fstart"><code>default_file_start</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEFAULT_005fGDB_005fEXTENSIONS"><code>DEFAULT_GDB_EXTENSIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEFAULT_005fPCC_005fSTRUCT_005fRETURN"><code>DEFAULT_PCC_STRUCT_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DEFAULT_005fSIGNED_005fCHAR"><code>DEFAULT_SIGNED_CHAR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005faddress_005fconstraint"><code>define_address_constraint</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fasm_005fattributes"><code>define_asm_attributes</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fattr"><code>define_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Attributes">Defining Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fautomaton"><code>define_automaton</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fbypass"><code>define_bypass</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fcode_005fattr"><code>define_code_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Code-Iterators">Code Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fcode_005fiterator"><code>define_code_iterator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Code-Iterators">Code Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fcond_005fexec"><code>define_cond_exec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conditional-Execution">Conditional Execution</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fconstants"><code>define_constants</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fconstraint"><code>define_constraint</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fcpu_005funit"><code>define_cpu_unit</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fc_005fenum"><code>define_c_enum</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fdelay"><code>define_delay</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Delay-Slots">Delay Slots</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fenum"><code>define_enum</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fenum_005fattr"><code>define_enum_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Attributes">Defining Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fenum_005fattr-1"><code>define_enum_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fexpand"><code>define_expand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005finsn"><code>define_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005finsn-example"><code>define_insn</code> example</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Example">Example</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005finsn_005fand_005fsplit"><code>define_insn_and_split</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Splitting">Insn Splitting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005finsn_005freservation"><code>define_insn_reservation</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fint_005fattr"><code>define_int_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Int-Iterators">Int Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fint_005fiterator"><code>define_int_iterator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Int-Iterators">Int Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fmemory_005fconstraint"><code>define_memory_constraint</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fmode_005fattr"><code>define_mode_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Substitutions">Substitutions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fmode_005fiterator"><code>define_mode_iterator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Mode-Iterators">Defining Mode Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fpeephole"><code>define_peephole</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole">define_peephole</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fpeephole2"><code>define_peephole2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole2">define_peephole2</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fpredicate"><code>define_predicate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Predicates">Defining Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fquery_005fcpu_005funit"><code>define_query_cpu_unit</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fregister_005fconstraint"><code>define_register_constraint</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005freservation"><code>define_reservation</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fspecial_005fpredicate"><code>define_special_predicate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Predicates">Defining Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsplit"><code>define_split</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Splitting">Insn Splitting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst">define_subst</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Subst">Define Subst</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst-1">define_subst</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Subst-Example">Define Subst Example</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst-2">define_subst</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Subst-Pattern-Matching">Define Subst Pattern Matching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst-3">define_subst</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Subst-Output-Template">Define Subst Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst-4"><code>define_subst</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Subst">Define Subst</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst-5"><code>define_subst</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Subst-Iterators">Subst Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst_005fattr"><code>define_subst_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Subst-Iterators">Subst Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-define_005fsubst_005fattr-1"><code>define_subst_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Subst-Iterators">Subst Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-attributes-and-their-values">defining attributes and their values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Attributes">Defining Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-constraints">defining constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Define-Constraints">Define Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-constraints_002c-obsolete-method">defining constraints, obsolete method</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-jump-instruction-patterns">defining jump instruction patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Jump-Patterns">Jump Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-looping-instruction-patterns">defining looping instruction patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Looping-Patterns">Looping Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-peephole-optimizers">defining peephole optimizers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Peephole-Definitions">Peephole Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-predicates">defining predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Predicates">Defining Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-defining-RTL-sequences-for-code-generation">defining RTL sequences for code generation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-delay-slots_002c-defining">delay slots, defining</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Delay-Slots">Delay Slots</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-deletable"><code>deletable</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DELETE_005fIF_005fORDINARY"><code>DELETE_IF_ORDINARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Dependent-Patterns">Dependent Patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dependent-Patterns">Dependent Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-desc"><code>desc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-destructors_002c-output-of">destructors, output of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-deterministic-finite-state-automaton">deterministic finite state automaton</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-deterministic-finite-state-automaton-1">deterministic finite state automaton</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DFmode"><code>DFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DF_005fSIZE"><code>DF_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-digits-in-constraint">digits in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DImode"><code>DImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-directory-options-_002emd">directory options .md</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Including-Patterns">Including Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DIR_005fSEPARATOR"><code>DIR_SEPARATOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DIR_005fSEPARATOR_005f2"><code>DIR_SEPARATOR_2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-disabling-certain-registers">disabling certain registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-dispatch-table">dispatch table</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-div"><code>div</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-div-and-attributes"><code>div</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-division">division</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-division-1">division</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-division-2">division</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-divm3-instruction-pattern"><code>div<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-divmodm4-instruction-pattern"><code>divmod<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DOLLARS_005fIN_005fIDENTIFIERS"><code>DOLLARS_IN_IDENTIFIERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-doloop_005fbegin-instruction-pattern"><code>doloop_begin</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-doloop_005fend-instruction-pattern"><code>doloop_end</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DONE"><code>DONE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DONT_005fUSE_005fBUILTIN_005fSETJMP"><code>DONT_USE_BUILTIN_SETJMP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DOUBLE_005fTYPE_005fSIZE"><code>DOUBLE_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DO_005fBODY"><code>DO_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DO_005fCOND"><code>DO_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DO_005fSTMT"><code>DO_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DQmode"><code>DQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-driver">driver</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DRIVER_005fSELF_005fSPECS"><code>DRIVER_SELF_SPECS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DUMPFILE_005fFORMAT"><code>DUMPFILE_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF2_005fASM_005fLINE_005fDEBUG_005fINFO"><code>DWARF2_ASM_LINE_DEBUG_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF2_005fDEBUGGING_005fINFO"><code>DWARF2_DEBUGGING_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF2_005fFRAME_005fINFO"><code>DWARF2_FRAME_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF2_005fFRAME_005fREG_005fOUT"><code>DWARF2_FRAME_REG_OUT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF2_005fUNWIND_005fINFO"><code>DWARF2_UNWIND_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fALT_005fFRAME_005fRETURN_005fCOLUMN"><code>DWARF_ALT_FRAME_RETURN_COLUMN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fCIE_005fDATA_005fALIGNMENT"><code>DWARF_CIE_DATA_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fFRAME_005fREGISTERS"><code>DWARF_FRAME_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fFRAME_005fREGNUM"><code>DWARF_FRAME_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fREG_005fTO_005fUNWIND_005fCOLUMN"><code>DWARF_REG_TO_UNWIND_COLUMN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DWARF_005fZERO_005fREG"><code>DWARF_ZERO_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-DYNAMIC_005fCHAIN_005fADDRESS"><code>DYNAMIC_CHAIN_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-E">E</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-E-in-constraint">&lsquo;<samp>E</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-earlyclobber-operand">earlyclobber operand</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-edge"><code>edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-edge-in-the-flow-graph">edge in the flow graph</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-edge-iterators">edge iterators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-edge-splitting">edge splitting</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDGE_005fABNORMAL"><code>EDGE_ABNORMAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDGE_005fABNORMAL_002c-EDGE_005fABNORMAL_005fCALL"><code>EDGE_ABNORMAL, EDGE_ABNORMAL_CALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDGE_005fABNORMAL_002c-EDGE_005fEH"><code>EDGE_ABNORMAL, EDGE_EH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDGE_005fABNORMAL_002c-EDGE_005fSIBCALL"><code>EDGE_ABNORMAL, EDGE_SIBCALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDGE_005fFALLTHRU_002c-force_005fnonfallthru"><code>EDGE_FALLTHRU, force_nonfallthru</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EDOM_002c-implicit-usage"><code>EDOM</code>, implicit usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fFRAME_005fIN_005fDATA_005fSECTION"><code>EH_FRAME_IN_DATA_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fFRAME_005fSECTION_005fNAME"><code>EH_FRAME_SECTION_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-eh_005freturn-instruction-pattern"><code>eh_return</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fRETURN_005fDATA_005fREGNO"><code>EH_RETURN_DATA_REGNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fRETURN_005fHANDLER_005fRTX"><code>EH_RETURN_HANDLER_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fRETURN_005fSTACKADJ_005fRTX"><code>EH_RETURN_STACKADJ_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fTABLES_005fCAN_005fBE_005fREAD_005fONLY"><code>EH_TABLES_CAN_BE_READ_ONLY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EH_005fUSES"><code>EH_USES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fedge"><code>ei_edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fend_005fp"><code>ei_end_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005flast"><code>ei_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fnext"><code>ei_next</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fone_005fbefore_005fend_005fp"><code>ei_one_before_end_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fprev"><code>ei_prev</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fsafe_005fsafe"><code>ei_safe_safe</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ei_005fstart"><code>ei_start</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ELIMINABLE_005fREGS"><code>ELIMINABLE_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ELSE_005fCLAUSE"><code>ELSE_CLAUSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Embedded-C">Embedded C</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EMIT_005fMODE_005fSET"><code>EMIT_MODE_SET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Empty-Statements">Empty Statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Empty-Statements">Empty Statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EMPTY_005fCLASS_005fEXPR"><code>EMPTY_CLASS_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EMPTY_005fFIELD_005fBOUNDARY"><code>EMPTY_FIELD_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Emulated-TLS">Emulated TLS</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-enabled">enabled</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Disable-Insn-Alternatives">Disable Insn Alternatives</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ENDFILE_005fSPEC"><code>ENDFILE_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-endianness">endianness</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Portability">Portability</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ENTRY_005fBLOCK_005fPTR_002c-EXIT_005fBLOCK_005fPTR"><code>ENTRY_BLOCK_PTR, EXIT_BLOCK_PTR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-enum-machine_005fmode"><code>enum machine_mode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-enum-reg_005fclass"><code>enum reg_class</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ENUMERAL_005fTYPE"><code>ENUMERAL_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-enumerations">enumerations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-epilogue">epilogue</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-epilogue-instruction-pattern"><code>epilogue</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EPILOGUE_005fUSES"><code>EPILOGUE_USES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-eq"><code>eq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-eq-and-attributes"><code>eq</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-equal">equal</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-eq_005fattr"><code>eq_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EQ_005fEXPR"><code>EQ_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-errno_002c-implicit-usage"><code>errno</code>, implicit usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXACT_005fDIV_005fEXPR"><code>EXACT_DIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-examining-SSA_005fNAMEs">examining SSA_NAMEs</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exception-handling">exception handling</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exception-handling-1">exception handling</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exception_005freceiver-instruction-pattern"><code>exception_receiver</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exclamation-point">exclamation point</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Multi_002dAlternative">Multi-Alternative</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exclusion_005fset"><code>exclusion_set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-exclusive_002dor_002c-bitwise">exclusive-or, bitwise</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXIT_005fEXPR"><code>EXIT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXIT_005fIGNORE_005fSTACK"><code>EXIT_IGNORE_STACK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-expander-definitions">expander definitions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-expm2-instruction-pattern"><code>exp<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-expression">expression</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expression-trees">Expression trees</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-expression-codes">expression codes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXPR_005fFILENAME"><code>EXPR_FILENAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Working-with-declarations">Working with declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXPR_005fLINENO"><code>EXPR_LINENO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Working-with-declarations">Working with declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-expr_005flist"><code>expr_list</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXPR_005fSTMT"><code>EXPR_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXPR_005fSTMT_005fEXPR"><code>EXPR_STMT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extendmn2-instruction-pattern"><code>extend<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extensible-constraints">extensible constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXTRA_005fADDRESS_005fCONSTRAINT"><code>EXTRA_ADDRESS_CONSTRAINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXTRA_005fCONSTRAINT"><code>EXTRA_CONSTRAINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXTRA_005fCONSTRAINT_005fSTR"><code>EXTRA_CONSTRAINT_STR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXTRA_005fMEMORY_005fCONSTRAINT"><code>EXTRA_MEMORY_CONSTRAINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-EXTRA_005fSPECS"><code>EXTRA_SPECS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extv-instruction-pattern"><code>extv</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extvm-instruction-pattern"><code>extv<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extvmisalignm-instruction-pattern"><code>extvmisalign<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extzv-instruction-pattern"><code>extzv</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extzvm-instruction-pattern"><code>extzv<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-extzvmisalignm-instruction-pattern"><code>extzvmisalign<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-F">F</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-F-in-constraint">&lsquo;<samp>F</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FAIL"><code>FAIL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fall_002dthru"><code>fall-thru</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FATAL_005fEXIT_005fCODE"><code>FATAL_EXIT_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FDL_002c-GNU-Free-Documentation-License">FDL, GNU Free Documentation License</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GNU-Free-Documentation-License">GNU Free Documentation License</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-features_002c-optional_002c-in-system-conventions">features, optional, in system conventions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ffs"><code>ffs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ffsm2-instruction-pattern"><code>ffs<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIELD_005fDECL"><code>FIELD_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-files-and-passes-of-the-compiler">files and passes of the compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Passes">Passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-files_002c-generated">files, generated</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Files">Files</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-file_005fend_005findicate_005fexec_005fstack"><code>file_end_indicate_exec_stack</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-final_005fabsence_005fset"><code>final_absence_set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FINAL_005fPRESCAN_005fINSN"><code>FINAL_PRESCAN_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-final_005fpresence_005fset"><code>final_presence_set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-final_005fsequence"><code>final_sequence</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIND_005fBASE_005fTERM"><code>FIND_BASE_TERM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-finite-state-automaton-minimization">finite state automaton minimization</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FINI_005fARRAY_005fSECTION_005fASM_005fOP"><code>FINI_ARRAY_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FINI_005fSECTION_005fASM_005fOP"><code>FINI_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIRST_005fPARM_005fOFFSET"><code>FIRST_PARM_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIRST_005fPARM_005fOFFSET-and-virtual-registers"><code>FIRST_PARM_OFFSET</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIRST_005fPSEUDO_005fREGISTER"><code>FIRST_PSEUDO_REGISTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIRST_005fSTACK_005fREG"><code>FIRST_STACK_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Registers">Stack Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIRST_005fVIRTUAL_005fREGISTER"><code>FIRST_VIRTUAL_REGISTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fix"><code>fix</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixed-register">fixed register</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixed_002dpoint-fractional-library">fixed-point fractional library</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIXED_005fCONVERT_005fEXPR"><code>FIXED_CONVERT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIXED_005fCST"><code>FIXED_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIXED_005fPOINT_005fTYPE"><code>FIXED_POINT_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIXED_005fREGISTERS"><code>FIXED_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixed_005fregs"><code>fixed_regs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixmn2-instruction-pattern"><code>fix<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixunsmn2-instruction-pattern"><code>fixuns<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fixuns_005ftruncmn2-instruction-pattern"><code>fixuns_trunc<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fix_005ftruncmn2-instruction-pattern"><code>fix_trunc<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FIX_005fTRUNC_005fEXPR"><code>FIX_TRUNC_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-flags-in-RTL-expression">flags in RTL expression</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-float"><code>float</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-floating-point-and-cross-compilation">floating point and cross compilation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-floatmn2-instruction-pattern"><code>float<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-floatunsmn2-instruction-pattern"><code>floatuns<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fEXPR"><code>FLOAT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-float_005fextend"><code>float_extend</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fLIB_005fCOMPARE_005fRETURNS_005fBOOL"><code>FLOAT_LIB_COMPARE_RETURNS_BOOL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fSTORE_005fFLAG_005fVALUE"><code>FLOAT_STORE_FLAG_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-float_005ftruncate"><code>float_truncate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fTYPE_005fSIZE"><code>FLOAT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fWORDS_005fBIG_005fENDIAN"><code>FLOAT_WORDS_BIG_ENDIAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOAT_005fWORDS_005fBIG_005fENDIAN_002c-_0028lack-of_0029-effect-on-subreg"><code>FLOAT_WORDS_BIG_ENDIAN</code>, (lack of) effect on <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-floorm2-instruction-pattern"><code>floor<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOOR_005fDIV_005fEXPR"><code>FLOOR_DIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FLOOR_005fMOD_005fEXPR"><code>FLOOR_MOD_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-flow_002dinsensitive-alias-analysis">flow-insensitive alias analysis</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alias-analysis">Alias analysis</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-flow_002dsensitive-alias-analysis">flow-sensitive alias analysis</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alias-analysis">Alias analysis</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fma"><code>fma</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fmam4-instruction-pattern"><code>fma<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fmodm3-instruction-pattern"><code>fmod<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fmsm4-instruction-pattern"><code>fms<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fnmam4-instruction-pattern"><code>fnma<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fnmsm4-instruction-pattern"><code>fnms<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FORCE_005fCODE_005fSECTION_005fALIGN"><code>FORCE_CODE_SECTION_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-force_005freg"><code>force_reg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FOR_005fBODY"><code>FOR_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FOR_005fCOND"><code>FOR_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FOR_005fEXPR"><code>FOR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FOR_005fINIT_005fSTMT"><code>FOR_INIT_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FOR_005fSTMT"><code>FOR_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fractional-types">fractional types</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fixed_002dpoint-fractional-library-routines">Fixed-point fractional library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fractmn2-instruction-pattern"><code>fract<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fractunsmn2-instruction-pattern"><code>fractuns<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fract_005fconvert"><code>fract_convert</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRACT_005fTYPE_005fSIZE"><code>FRACT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame-layout">frame layout</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fADDR_005fRTX"><code>FRAME_ADDR_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fGROWS_005fDOWNWARD"><code>FRAME_GROWS_DOWNWARD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fGROWS_005fDOWNWARD-and-virtual-registers"><code>FRAME_GROWS_DOWNWARD</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fPOINTER_005fCFA_005fOFFSET"><code>FRAME_POINTER_CFA_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005fpointer_005fneeded"><code>frame_pointer_needed</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fPOINTER_005fREGNUM"><code>FRAME_POINTER_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FRAME_005fPOINTER_005fREGNUM-and-virtual-registers"><code>FRAME_POINTER_REGNUM</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005fpointer_005frtx"><code>frame_pointer_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005frelated"><code>frame_related</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005frelated_002c-in-insn_002c-call_005finsn_002c-jump_005finsn_002c-barrier_002c-and-set"><code>frame_related</code>, in <code>insn</code>, <code>call_insn</code>, <code>jump_insn</code>, <code>barrier</code>, and <code>set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005frelated_002c-in-mem"><code>frame_related</code>, in <code>mem</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005frelated_002c-in-reg"><code>frame_related</code>, in <code>reg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frame_005frelated_002c-in-symbol_005fref"><code>frame_related</code>, in <code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-frequency_002c-count_002c-BB_005fFREQ_005fBASE"><code>frequency, count, BB_FREQ_BASE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ftruncm2-instruction-pattern"><code>ftrunc<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function">function</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions">Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function-1">function</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function-call-conventions">function call conventions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function-entry-and-exit">function entry and exit</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function-entry-point_002c-alternate-function-entry-point">function entry point, alternate function entry point</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function-properties">function properties</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-function_002dcall-insns">function-call insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Calls">Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-functions_002c-leaf">functions, leaf</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fARG_005fOFFSET"><code>FUNCTION_ARG_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fARG_005fPADDING"><code>FUNCTION_ARG_PADDING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fARG_005fREGNO_005fP"><code>FUNCTION_ARG_REGNO_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fBOUNDARY"><code>FUNCTION_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fDECL"><code>FUNCTION_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions">Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fDECL-1"><code>FUNCTION_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fMODE"><code>FUNCTION_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fPROFILER"><code>FUNCTION_PROFILER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fTYPE"><code>FUNCTION_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fVALUE"><code>FUNCTION_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-FUNCTION_005fVALUE_005fREGNO_005fP"><code>FUNCTION_VALUE_REGNO_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-fundamental-type">fundamental type</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-G">G</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-G-in-constraint">&lsquo;<samp>G</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-g-in-constraint">&lsquo;<samp>g</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-garbage-collector_002c-invocation">garbage collector, invocation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Invoking-the-garbage-collector">Invoking the garbage collector</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-garbage-collector_002c-troubleshooting">garbage collector, troubleshooting</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Troubleshooting">Troubleshooting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GCC-and-portability">GCC and portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Portability">Portability</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GCC_005fDRIVER_005fHOST_005fINITIALIZATION"><code>GCC_DRIVER_HOST_INITIALIZATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gcov_005ftype"><code>gcov_type</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ge"><code>ge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ge-and-attributes"><code>ge</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gencodes"><code>gencodes</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-general_005foperand"><code>general_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GENERAL_005fREGS"><code>GENERAL_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-generated-files">generated files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Files">Files</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-generating-assembler-output">generating assembler output</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-generating-insns">generating insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GENERIC">GENERIC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GENERIC-1">GENERIC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GENERIC">GENERIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-generic-predicates">generic predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-genflags"><code>genflags</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GEN_005fERRNO_005fRTX"><code>GEN_ERRNO_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005fattr"><code>get_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005fattr_005flength"><code>get_attr_length</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fCLASS_005fNARROWEST_005fMODE"><code>GET_CLASS_NARROWEST_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fCODE"><code>GET_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005fframe_005fsize"><code>get_frame_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005finsns"><code>get_insns</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005flast_005finsn"><code>get_last_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE"><code>GET_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fALIGNMENT"><code>GET_MODE_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fBITSIZE"><code>GET_MODE_BITSIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fCLASS"><code>GET_MODE_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fFBIT"><code>GET_MODE_FBIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fIBIT"><code>GET_MODE_IBIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fMASK"><code>GET_MODE_MASK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fNAME"><code>GET_MODE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fNUNITS"><code>GET_MODE_NUNITS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fSIZE"><code>GET_MODE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fUNIT_005fSIZE"><code>GET_MODE_UNIT_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fMODE_005fWIDER_005fMODE"><code>GET_MODE_WIDER_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fRTX_005fCLASS"><code>GET_RTX_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fRTX_005fFORMAT"><code>GET_RTX_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GET_005fRTX_005fLENGTH"><code>GET_RTX_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-get_005fthread_005fpointermode-instruction-pattern"><code>get_thread_pointer<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-geu"><code>geu</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-geu-and-attributes"><code>geu</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GE_005fEXPR"><code>GE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GGC">GGC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Information">Type Information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ggc_005fcollect"><code>ggc_collect</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Invoking-the-garbage-collector">Invoking the garbage collector</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE">GIMPLE</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-1">GIMPLE</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Gimplification-pass">Gimplification pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-2">GIMPLE</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE">GIMPLE</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-Exception-Handling">GIMPLE Exception Handling</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-Exception-Handling">GIMPLE Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-instruction-set">GIMPLE instruction set</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-instruction-set">GIMPLE instruction set</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-sequences">GIMPLE sequences</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-statement-iterators">GIMPLE statement iterators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE-statement-iterators-1">GIMPLE statement iterators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005faddresses_005ftaken"><code>gimple_addresses_taken</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fASM"><code>GIMPLE_ASM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fclear_005fvolatile"><code>gimple_asm_clear_volatile</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fclobber_005fop"><code>gimple_asm_clobber_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005finput_005fop"><code>gimple_asm_input_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fnclobbers"><code>gimple_asm_nclobbers</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fninputs"><code>gimple_asm_ninputs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fnoutputs"><code>gimple_asm_noutputs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005foutput_005fop"><code>gimple_asm_output_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fset_005fclobber_005fop"><code>gimple_asm_set_clobber_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fset_005finput_005fop"><code>gimple_asm_set_input_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fset_005foutput_005fop"><code>gimple_asm_set_output_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fset_005fvolatile"><code>gimple_asm_set_volatile</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fstring"><code>gimple_asm_string</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fasm_005fvolatile_005fp"><code>gimple_asm_volatile_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fcast_005fp"><code>gimple_assign_cast_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fcast_005fp-1"><code>gimple_assign_cast_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005flhs"><code>gimple_assign_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005flhs_005fptr"><code>gimple_assign_lhs_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs1"><code>gimple_assign_rhs1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs1_005fptr"><code>gimple_assign_rhs1_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs2"><code>gimple_assign_rhs2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs2_005fptr"><code>gimple_assign_rhs2_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs3"><code>gimple_assign_rhs3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs3_005fptr"><code>gimple_assign_rhs3_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs_005fclass"><code>gimple_assign_rhs_class</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005frhs_005fcode"><code>gimple_assign_rhs_code</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fset_005flhs"><code>gimple_assign_set_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fset_005frhs1"><code>gimple_assign_set_rhs1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fset_005frhs2"><code>gimple_assign_set_rhs2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fassign_005fset_005frhs3"><code>gimple_assign_set_rhs3</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbb"><code>gimple_bb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fadd_005fseq"><code>gimple_bind_add_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fadd_005fstmt"><code>gimple_bind_add_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fappend_005fvars"><code>gimple_bind_append_vars</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fblock"><code>gimple_bind_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fbody"><code>gimple_bind_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fset_005fblock"><code>gimple_bind_set_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fset_005fbody"><code>gimple_bind_set_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fset_005fvars"><code>gimple_bind_set_vars</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbind_005fvars"><code>gimple_bind_vars</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fblock"><code>gimple_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fasm"><code>gimple_build_asm</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fasm_005fvec"><code>gimple_build_asm_vec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASM"><code>GIMPLE_ASM</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fassign"><code>gimple_build_assign</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fassign_005fwith_005fops"><code>gimple_build_assign_with_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fbind"><code>gimple_build_bind</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fBIND"><code>GIMPLE_BIND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcall"><code>gimple_build_call</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcall_005ffrom_005ftree"><code>gimple_build_call_from_tree</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcall_005fvec"><code>gimple_build_call_vec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcatch"><code>gimple_build_catch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcond"><code>gimple_build_cond</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fcond_005ffrom_005ftree"><code>gimple_build_cond_from_tree</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fdebug_005fbind"><code>gimple_build_debug_bind</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005feh_005ffilter"><code>gimple_build_eh_filter</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fgoto"><code>gimple_build_goto</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005flabel"><code>gimple_build_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fnop"><code>gimple_build_nop</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fNOP"><code>GIMPLE_NOP</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fatomic_005fload"><code>gimple_build_omp_atomic_load</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fatomic_005fstore"><code>gimple_build_omp_atomic_store</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fcontinue"><code>gimple_build_omp_continue</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fcritical"><code>gimple_build_omp_critical</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005ffor"><code>gimple_build_omp_for</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fmaster"><code>gimple_build_omp_master</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fMASTER"><code>GIMPLE_OMP_MASTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fordered"><code>gimple_build_omp_ordered</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fORDERED"><code>GIMPLE_OMP_ORDERED</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fparallel"><code>gimple_build_omp_parallel</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005freturn"><code>gimple_build_omp_return</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fsection"><code>gimple_build_omp_section</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fsections"><code>gimple_build_omp_sections</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fsections_005fswitch"><code>gimple_build_omp_sections_switch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fomp_005fsingle"><code>gimple_build_omp_single</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fresx"><code>gimple_build_resx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005freturn"><code>gimple_build_return</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fswitch"><code>gimple_build_switch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005ftry"><code>gimple_build_try</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fbuild_005fwce"><code>gimple_build_wce</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005farg"><code>gimple_call_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005farg_005fptr"><code>gimple_call_arg_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fcannot_005finline_005fp"><code>gimple_call_cannot_inline_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fchain"><code>gimple_call_chain</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fcopy_005fskip_005fargs"><code>gimple_call_copy_skip_args</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005ffn"><code>gimple_call_fn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005ffndecl"><code>gimple_call_fndecl</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005flhs"><code>gimple_call_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005flhs_005fptr"><code>gimple_call_lhs_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fmark_005funinlinable"><code>gimple_call_mark_uninlinable</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fnoreturn_005fp"><code>gimple_call_noreturn_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fnum_005fargs"><code>gimple_call_num_args</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005freturn_005ftype"><code>gimple_call_return_type</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005farg"><code>gimple_call_set_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005fchain"><code>gimple_call_set_chain</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005ffn"><code>gimple_call_set_fn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005ffndecl"><code>gimple_call_set_fndecl</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005flhs"><code>gimple_call_set_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005fset_005ftail"><code>gimple_call_set_tail</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcall_005ftail_005fp"><code>gimple_call_tail_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCALL"><code>GIMPLE_CALL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcatch_005fhandler"><code>gimple_catch_handler</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcatch_005fset_005fhandler"><code>gimple_catch_set_handler</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcatch_005fset_005ftypes"><code>gimple_catch_set_types</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcatch_005ftypes"><code>gimple_catch_types</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcatch_005ftypes_005fptr"><code>gimple_catch_types_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCATCH"><code>GIMPLE_CATCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcode"><code>gimple_code</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fCOND"><code>GIMPLE_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fcode"><code>gimple_cond_code</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005ffalse_005flabel"><code>gimple_cond_false_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005flhs"><code>gimple_cond_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fmake_005ffalse"><code>gimple_cond_make_false</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fmake_005ftrue"><code>gimple_cond_make_true</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005frhs"><code>gimple_cond_rhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fset_005fcode"><code>gimple_cond_set_code</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fset_005ffalse_005flabel"><code>gimple_cond_set_false_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fset_005flhs"><code>gimple_cond_set_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fset_005frhs"><code>gimple_cond_set_rhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005fset_005ftrue_005flabel"><code>gimple_cond_set_true_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcond_005ftrue_005flabel"><code>gimple_cond_true_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fCOND"><code>GIMPLE_COND</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fcopy"><code>gimple_copy</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fDEBUG_005fBIND"><code>GIMPLE_DEBUG_BIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fget_005fvalue"><code>gimple_debug_bind_get_value</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fget_005fvalue_005fptr"><code>gimple_debug_bind_get_value_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fget_005fvar"><code>gimple_debug_bind_get_var</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fhas_005fvalue_005fp"><code>gimple_debug_bind_has_value_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fp"><code>gimple_debug_bind_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005freset_005fvalue"><code>gimple_debug_bind_reset_value</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fset_005fvalue"><code>gimple_debug_bind_set_value</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdebug_005fbind_005fset_005fvar"><code>gimple_debug_bind_set_var</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fDEBUG"><code>GIMPLE_DEBUG</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fdef_005fops"><code>gimple_def_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005ffailure"><code>gimple_eh_filter_failure</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005fmust_005fnot_005fthrow"><code>gimple_eh_filter_must_not_throw</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005fset_005ffailure"><code>gimple_eh_filter_set_failure</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005fset_005fmust_005fnot_005fthrow"><code>gimple_eh_filter_set_must_not_throw</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005fset_005ftypes"><code>gimple_eh_filter_set_types</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005ftypes"><code>gimple_eh_filter_types</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005feh_005ffilter_005ftypes_005fptr"><code>gimple_eh_filter_types_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fEH_005fFILTER"><code>GIMPLE_EH_FILTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fexpr_005fcode"><code>gimple_expr_code</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fexpr_005ftype"><code>gimple_expr_type</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fgoto_005fdest"><code>gimple_goto_dest</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fgoto_005fset_005fdest"><code>gimple_goto_set_dest</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fhas_005fmem_005fops"><code>gimple_has_mem_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fhas_005fops"><code>gimple_has_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fhas_005fvolatile_005fops"><code>gimple_has_volatile_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005flabel_005flabel"><code>gimple_label_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005flabel_005fset_005flabel"><code>gimple_label_set_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fLABEL"><code>GIMPLE_LABEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005floaded_005fsyms"><code>gimple_loaded_syms</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005flocus"><code>gimple_locus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005flocus_005fempty_005fp"><code>gimple_locus_empty_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fmodified_005fp"><code>gimple_modified_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fNOP"><code>GIMPLE_NOP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fNOP"><code>GIMPLE_NOP</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fnop_005fp"><code>gimple_nop_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fNOP"><code>GIMPLE_NOP</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fno_005fwarning_005fp"><code>gimple_no_warning_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fnum_005fops"><code>gimple_num_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fnum_005fops-1"><code>gimple_num_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fload_005flhs"><code>gimple_omp_atomic_load_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fload_005frhs"><code>gimple_omp_atomic_load_rhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fload_005fset_005flhs"><code>gimple_omp_atomic_load_set_lhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fload_005fset_005frhs"><code>gimple_omp_atomic_load_set_rhs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fLOAD"><code>GIMPLE_OMP_ATOMIC_LOAD</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fstore_005fset_005fval"><code>gimple_omp_atomic_store_set_val</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fatomic_005fstore_005fval"><code>gimple_omp_atomic_store_val</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fATOMIC_005fSTORE"><code>GIMPLE_OMP_ATOMIC_STORE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fbody"><code>gimple_omp_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fcontrol_005fdef"><code>gimple_omp_continue_control_def</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fcontrol_005fdef_005fptr"><code>gimple_omp_continue_control_def_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fcontrol_005fuse"><code>gimple_omp_continue_control_use</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fcontrol_005fuse_005fptr"><code>gimple_omp_continue_control_use_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fset_005fcontrol_005fdef"><code>gimple_omp_continue_set_control_def</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcontinue_005fset_005fcontrol_005fuse"><code>gimple_omp_continue_set_control_use</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCONTINUE"><code>GIMPLE_OMP_CONTINUE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcritical_005fname"><code>gimple_omp_critical_name</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcritical_005fname_005fptr"><code>gimple_omp_critical_name_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fcritical_005fset_005fname"><code>gimple_omp_critical_set_name</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fCRITICAL"><code>GIMPLE_OMP_CRITICAL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fclauses"><code>gimple_omp_for_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fclauses_005fptr"><code>gimple_omp_for_clauses_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fcond"><code>gimple_omp_for_cond</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005ffinal"><code>gimple_omp_for_final</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005ffinal_005fptr"><code>gimple_omp_for_final_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fincr"><code>gimple_omp_for_incr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fincr_005fptr"><code>gimple_omp_for_incr_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005findex"><code>gimple_omp_for_index</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005findex_005fptr"><code>gimple_omp_for_index_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005finitial"><code>gimple_omp_for_initial</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005finitial_005fptr"><code>gimple_omp_for_initial_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fpre_005fbody"><code>gimple_omp_for_pre_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005fclauses"><code>gimple_omp_for_set_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005fcond"><code>gimple_omp_for_set_cond</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005ffinal"><code>gimple_omp_for_set_final</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005fincr"><code>gimple_omp_for_set_incr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005findex"><code>gimple_omp_for_set_index</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005finitial"><code>gimple_omp_for_set_initial</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005ffor_005fset_005fpre_005fbody"><code>gimple_omp_for_set_pre_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fFOR"><code>GIMPLE_OMP_FOR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fMASTER"><code>GIMPLE_OMP_MASTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fMASTER"><code>GIMPLE_OMP_MASTER</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fORDERED"><code>GIMPLE_OMP_ORDERED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fORDERED"><code>GIMPLE_OMP_ORDERED</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fchild_005ffn"><code>gimple_omp_parallel_child_fn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fchild_005ffn_005fptr"><code>gimple_omp_parallel_child_fn_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fclauses"><code>gimple_omp_parallel_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fclauses_005fptr"><code>gimple_omp_parallel_clauses_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fcombined_005fp"><code>gimple_omp_parallel_combined_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fdata_005farg"><code>gimple_omp_parallel_data_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fdata_005farg_005fptr"><code>gimple_omp_parallel_data_arg_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fset_005fchild_005ffn"><code>gimple_omp_parallel_set_child_fn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fset_005fclauses"><code>gimple_omp_parallel_set_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fset_005fcombined_005fp"><code>gimple_omp_parallel_set_combined_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fparallel_005fset_005fdata_005farg"><code>gimple_omp_parallel_set_data_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005freturn_005fnowait_005fp"><code>gimple_omp_return_nowait_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005freturn_005fset_005fnowait"><code>gimple_omp_return_set_nowait</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fRETURN"><code>GIMPLE_OMP_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fclauses"><code>gimple_omp_sections_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fclauses_005fptr"><code>gimple_omp_sections_clauses_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fcontrol"><code>gimple_omp_sections_control</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fcontrol_005fptr"><code>gimple_omp_sections_control_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fset_005fclauses"><code>gimple_omp_sections_set_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsections_005fset_005fcontrol"><code>gimple_omp_sections_set_control</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTIONS"><code>GIMPLE_OMP_SECTIONS</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsection_005flast_005fp"><code>gimple_omp_section_last_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsection_005fset_005flast"><code>gimple_omp_section_set_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSECTION"><code>GIMPLE_OMP_SECTION</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fset_005fbody"><code>gimple_omp_set_body</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsingle_005fclauses"><code>gimple_omp_single_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsingle_005fclauses_005fptr"><code>gimple_omp_single_clauses_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fomp_005fsingle_005fset_005fclauses"><code>gimple_omp_single_set_clauses</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fSINGLE"><code>GIMPLE_OMP_SINGLE</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fop"><code>gimple_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fop-1"><code>gimple_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fops"><code>gimple_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fops-1"><code>gimple_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fop_005fptr"><code>gimple_op_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005farg"><code>gimple_phi_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fcapacity"><code>gimple_phi_capacity</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fnum_005fargs"><code>gimple_phi_num_args</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fresult"><code>gimple_phi_result</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fresult_005fptr"><code>gimple_phi_result_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fset_005farg"><code>gimple_phi_set_arg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fphi_005fset_005fresult"><code>gimple_phi_set_result</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fPHI"><code>GIMPLE_PHI</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fplf"><code>gimple_plf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fresx_005fregion"><code>gimple_resx_region</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fresx_005fset_005fregion"><code>gimple_resx_set_region</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRESX"><code>GIMPLE_RESX</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005freturn_005fretval"><code>gimple_return_retval</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005freturn_005fset_005fretval"><code>gimple_return_set_retval</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fRETURN"><code>GIMPLE_RETURN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fadd_005fseq"><code>gimple_seq_add_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fadd_005fstmt"><code>gimple_seq_add_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005falloc"><code>gimple_seq_alloc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fcopy"><code>gimple_seq_copy</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fdeep_005fcopy"><code>gimple_seq_deep_copy</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fempty_005fp"><code>gimple_seq_empty_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005ffirst"><code>gimple_seq_first</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005finit"><code>gimple_seq_init</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005flast"><code>gimple_seq_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005freverse"><code>gimple_seq_reverse</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fset_005ffirst"><code>gimple_seq_set_first</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fset_005flast"><code>gimple_seq_set_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fseq_005fsingleton_005fp"><code>gimple_seq_singleton_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fblock"><code>gimple_set_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fdef_005fops"><code>gimple_set_def_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fhas_005fvolatile_005fops"><code>gimple_set_has_volatile_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005flocus"><code>gimple_set_locus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fop"><code>gimple_set_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fplf"><code>gimple_set_plf</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fuse_005fops"><code>gimple_set_use_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fvdef_005fops"><code>gimple_set_vdef_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fvisited"><code>gimple_set_visited</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fset_005fvuse_005fops"><code>gimple_set_vuse_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fstatement_005fbase">gimple_statement_base</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tuple-representation">Tuple representation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fstatement_005fwith_005fops">gimple_statement_with_ops</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tuple-representation">Tuple representation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fstored_005fsyms"><code>gimple_stored_syms</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fdefault_005flabel"><code>gimple_switch_default_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005findex"><code>gimple_switch_index</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005flabel"><code>gimple_switch_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fnum_005flabels"><code>gimple_switch_num_labels</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fset_005fdefault_005flabel"><code>gimple_switch_set_default_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fset_005findex"><code>gimple_switch_set_index</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fset_005flabel"><code>gimple_switch_set_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fswitch_005fset_005fnum_005flabels"><code>gimple_switch_set_num_labels</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fSWITCH"><code>GIMPLE_SWITCH</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fcatch_005fis_005fcleanup"><code>gimple_try_catch_is_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fcleanup"><code>gimple_try_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005feval"><code>gimple_try_eval</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fkind"><code>gimple_try_kind</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fset_005fcatch_005fis_005fcleanup"><code>gimple_try_set_catch_is_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fset_005fcleanup"><code>gimple_try_set_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005ftry_005fset_005feval"><code>gimple_try_set_eval</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fTRY"><code>GIMPLE_TRY</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fuse_005fops"><code>gimple_use_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fvdef_005fops"><code>gimple_vdef_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fvisited_005fp"><code>gimple_visited_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fvuse_005fops"><code>gimple_vuse_ops</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fwce_005fcleanup"><code>gimple_wce_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fwce_005fcleanup_005feh_005fonly"><code>gimple_wce_cleanup_eh_only</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fwce_005fset_005fcleanup"><code>gimple_wce_set_cleanup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimple_005fwce_005fset_005fcleanup_005feh_005fonly"><code>gimple_wce_set_cleanup_eh_only</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fWITH_005fCLEANUP_005fEXPR"><code>GIMPLE_WITH_CLEANUP_EXPR</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplification">gimplification</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplification-1">gimplification</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Gimplification-pass">Gimplification pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplifier">gimplifier</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplify_005fassign"><code>gimplify_assign</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fASSIGN"><code>GIMPLE_ASSIGN</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplify_005fexpr"><code>gimplify_expr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Gimplification-pass">Gimplification pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gimplify_005ffunction_005ftree"><code>gimplify_function_tree</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Gimplification-pass">Gimplification pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GLOBAL_005fINIT_005fPRIORITY"><code>GLOBAL_INIT_PRIORITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-global_005fregs"><code>global_regs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GO_005fIF_005fLEGITIMATE_005fADDRESS"><code>GO_IF_LEGITIMATE_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-greater-than">greater than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-greater-than-1">greater than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-greater-than-2">greater than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fafter_005flabels"><code>gsi_after_labels</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fbb"><code>gsi_bb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fcommit_005fedge_005finserts"><code>gsi_commit_edge_inserts</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fcommit_005fedge_005finserts-1"><code>gsi_commit_edge_inserts</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fcommit_005fone_005fedge_005finsert"><code>gsi_commit_one_edge_insert</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fend_005fp"><code>gsi_end_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fend_005fp-1"><code>gsi_end_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005ffor_005fstmt"><code>gsi_for_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fafter"><code>gsi_insert_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fafter-1"><code>gsi_insert_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fbefore"><code>gsi_insert_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fbefore-1"><code>gsi_insert_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fon_005fedge"><code>gsi_insert_on_edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fon_005fedge-1"><code>gsi_insert_on_edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fon_005fedge_005fimmediate"><code>gsi_insert_on_edge_immediate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fseq_005fafter"><code>gsi_insert_seq_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fseq_005fbefore"><code>gsi_insert_seq_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005finsert_005fseq_005fon_005fedge"><code>gsi_insert_seq_on_edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flast"><code>gsi_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flast-1"><code>gsi_last</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flast_005fbb"><code>gsi_last_bb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flink_005fafter"><code>gsi_link_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flink_005fbefore"><code>gsi_link_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flink_005fseq_005fafter"><code>gsi_link_seq_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005flink_005fseq_005fbefore"><code>gsi_link_seq_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fmove_005fafter"><code>gsi_move_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fmove_005fbefore"><code>gsi_move_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fmove_005fto_005fbb_005fend"><code>gsi_move_to_bb_end</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fnext"><code>gsi_next</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fnext-1"><code>gsi_next</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fone_005fbefore_005fend_005fp"><code>gsi_one_before_end_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fprev"><code>gsi_prev</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fprev-1"><code>gsi_prev</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fremove"><code>gsi_remove</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fremove-1"><code>gsi_remove</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005freplace"><code>gsi_replace</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fseq"><code>gsi_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fsplit_005fseq_005fafter"><code>gsi_split_seq_after</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fsplit_005fseq_005fbefore"><code>gsi_split_seq_before</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fstart"><code>gsi_start</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fstart-1"><code>gsi_start</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fstart_005fbb"><code>gsi_start_bb</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fstmt"><code>gsi_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gsi_005fstmt_005fptr"><code>gsi_stmt_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gt"><code>gt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gt-and-attributes"><code>gt</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gtu"><code>gtu</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-gtu-and-attributes"><code>gtu</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GTY"><code>GTY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Information">Type Information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-GT_005fEXPR"><code>GT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-H">H</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-H-in-constraint">&lsquo;<samp>H</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAmode"><code>HAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HANDLER"><code>HANDLER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HANDLER_005fBODY"><code>HANDLER_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HANDLER_005fPARMS"><code>HANDLER_PARMS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HANDLE_005fPRAGMA_005fPACK_005fWITH_005fEXPANSION"><code>HANDLE_PRAGMA_PACK_WITH_EXPANSION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-hard-registers">hard registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fFRAME_005fPOINTER_005fIS_005fARG_005fPOINTER"><code>HARD_FRAME_POINTER_IS_ARG_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fFRAME_005fPOINTER_005fIS_005fFRAME_005fPOINTER"><code>HARD_FRAME_POINTER_IS_FRAME_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fFRAME_005fPOINTER_005fREGNUM"><code>HARD_FRAME_POINTER_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fCALLER_005fSAVE_005fMODE"><code>HARD_REGNO_CALLER_SAVE_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Caller-Saves">Caller Saves</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fCALL_005fPART_005fCLOBBERED"><code>HARD_REGNO_CALL_PART_CLOBBERED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fMODE_005fOK"><code>HARD_REGNO_MODE_OK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fNREGS"><code>HARD_REGNO_NREGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fNREGS_005fHAS_005fPADDING"><code>HARD_REGNO_NREGS_HAS_PADDING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fNREGS_005fWITH_005fPADDING"><code>HARD_REGNO_NREGS_WITH_PADDING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HARD_005fREGNO_005fRENAME_005fOK"><code>HARD_REGNO_RENAME_OK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAS_005fINIT_005fSECTION"><code>HAS_INIT_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAS_005fLONG_005fCOND_005fBRANCH"><code>HAS_LONG_COND_BRANCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAS_005fLONG_005fUNCOND_005fBRANCH"><code>HAS_LONG_UNCOND_BRANCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fDOS_005fBASED_005fFILE_005fSYSTEM"><code>HAVE_DOS_BASED_FILE_SYSTEM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPOST_005fDECREMENT"><code>HAVE_POST_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPOST_005fINCREMENT"><code>HAVE_POST_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPOST_005fMODIFY_005fDISP"><code>HAVE_POST_MODIFY_DISP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPOST_005fMODIFY_005fREG"><code>HAVE_POST_MODIFY_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPRE_005fDECREMENT"><code>HAVE_PRE_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPRE_005fINCREMENT"><code>HAVE_PRE_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPRE_005fMODIFY_005fDISP"><code>HAVE_PRE_MODIFY_DISP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HAVE_005fPRE_005fMODIFY_005fREG"><code>HAVE_PRE_MODIFY_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HCmode"><code>HCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HFmode"><code>HFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-high"><code>high</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HImode"><code>HImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HImode_002c-in-insn"><code>HImode</code>, in <code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HONOR_005fREG_005fALLOC_005fORDER"><code>HONOR_REG_ALLOC_ORDER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-host-configuration">host configuration</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Config">Host Config</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-host-functions">host functions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-host-hooks">host hooks</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-host-makefile-fragment">host makefile fragment</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Fragment">Host Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fBIT_005fBUCKET"><code>HOST_BIT_BUCKET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fEXECUTABLE_005fSUFFIX"><code>HOST_EXECUTABLE_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fHOOKS_005fEXTRA_005fSIGNALS"><code>HOST_HOOKS_EXTRA_SIGNALS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fHOOKS_005fGT_005fPCH_005fALLOC_005fGRANULARITY"><code>HOST_HOOKS_GT_PCH_ALLOC_GRANULARITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fHOOKS_005fGT_005fPCH_005fGET_005fADDRESS"><code>HOST_HOOKS_GT_PCH_GET_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fHOOKS_005fGT_005fPCH_005fUSE_005fADDRESS"><code>HOST_HOOKS_GT_PCH_USE_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Common">Host Common</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fLACKS_005fINODE_005fNUMBERS"><code>HOST_LACKS_INODE_NUMBERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fLONG_005fFORMAT"><code>HOST_LONG_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fLONG_005fLONG_005fFORMAT"><code>HOST_LONG_LONG_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fOBJECT_005fSUFFIX"><code>HOST_OBJECT_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOST_005fPTR_005fPRINTF"><code>HOST_PTR_PRINTF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HOT_005fTEXT_005fSECTION_005fNAME"><code>HOT_TEXT_SECTION_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-HQmode"><code>HQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-I">I</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-i-in-constraint">&lsquo;<samp>i</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-I-in-constraint">&lsquo;<samp>I</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-identifier">identifier</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IDENTIFIER_005fLENGTH"><code>IDENTIFIER_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IDENTIFIER_005fNODE"><code>IDENTIFIER_NODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IDENTIFIER_005fOPNAME_005fP"><code>IDENTIFIER_OPNAME_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IDENTIFIER_005fPOINTER"><code>IDENTIFIER_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IDENTIFIER_005fTYPENAME_005fP"><code>IDENTIFIER_TYPENAME_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IEEE-754_002d2008">IEEE 754-2008</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Decimal-float-library-routines">Decimal float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMACHDEP_005fINIT"><code>IFCVT_MACHDEP_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMODIFY_005fCANCEL"><code>IFCVT_MODIFY_CANCEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMODIFY_005fFINAL"><code>IFCVT_MODIFY_FINAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMODIFY_005fINSN"><code>IFCVT_MODIFY_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMODIFY_005fMULTIPLE_005fTESTS"><code>IFCVT_MODIFY_MULTIPLE_TESTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IFCVT_005fMODIFY_005fTESTS"><code>IFCVT_MODIFY_TESTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IF_005fCOND"><code>IF_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-if_005fmarked"><code>if_marked</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IF_005fSTMT"><code>IF_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-if_005fthen_005felse"><code>if_then_else</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-if_005fthen_005felse-and-attributes"><code>if_then_else</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-if_005fthen_005felse-usage"><code>if_then_else</code> usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IMAGPART_005fEXPR"><code>IMAGPART_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Immediate-Uses">Immediate Uses</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-immediate_005foperand"><code>immediate_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IMMEDIATE_005fPREFIX"><code>IMMEDIATE_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-include"><code>include</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Including-Patterns">Including Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCLUDE_005fDEFAULTS"><code>INCLUDE_DEFAULTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-inclusive_002dor_002c-bitwise">inclusive-or, bitwise</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCOMING_005fFRAME_005fSP_005fOFFSET"><code>INCOMING_FRAME_SP_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCOMING_005fREGNO"><code>INCOMING_REGNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCOMING_005fREG_005fPARM_005fSTACK_005fSPACE"><code>INCOMING_REG_PARM_STACK_SPACE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCOMING_005fRETURN_005fADDR_005fRTX"><code>INCOMING_RETURN_ADDR_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INCOMING_005fSTACK_005fBOUNDARY"><code>INCOMING_STACK_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INDEX_005fREG_005fCLASS"><code>INDEX_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-indirect_005fjump-instruction-pattern"><code>indirect_jump</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-indirect_005foperand"><code>indirect_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INDIRECT_005fREF"><code>INDIRECT_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-initialization-routines">initialization routines</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INITIAL_005fELIMINATION_005fOFFSET"><code>INITIAL_ELIMINATION_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INITIAL_005fFRAME_005fADDRESS_005fRTX"><code>INITIAL_FRAME_ADDRESS_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INITIAL_005fFRAME_005fPOINTER_005fOFFSET"><code>INITIAL_FRAME_POINTER_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fARRAY_005fSECTION_005fASM_005fOP"><code>INIT_ARRAY_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fCUMULATIVE_005fARGS"><code>INIT_CUMULATIVE_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fCUMULATIVE_005fINCOMING_005fARGS"><code>INIT_CUMULATIVE_INCOMING_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fCUMULATIVE_005fLIBCALL_005fARGS"><code>INIT_CUMULATIVE_LIBCALL_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fENVIRONMENT"><code>INIT_ENVIRONMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fEXPANDERS"><code>INIT_EXPANDERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Per_002dFunction-Data">Per-Function Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fEXPR"><code>INIT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-init_005fmachine_005fstatus"><code>init_machine_status</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Per_002dFunction-Data">Per-Function Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-init_005fone_005flibfunc"><code>init_one_libfunc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fSECTION_005fASM_005fOP"><code>INIT_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INIT_005fSECTION_005fASM_005fOP-1"><code>INIT_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-inlining">inlining</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insert_005finsn_005fon_005fedge"><code>insert_insn_on_edge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn"><code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002ff"><code>insn</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002fj"><code>insn</code> and &lsquo;<samp>/j</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002fs"><code>insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002fs-1"><code>insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002fu"><code>insn</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-and-_002fv"><code>insn</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-attributes">insn attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Attributes">Insn Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-canonicalization">insn canonicalization</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-includes">insn includes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Including-Patterns">Including Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-lengths_002c-computing">insn lengths, computing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-notes_002c-notes">insn notes, notes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn-splitting">insn splitting</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Splitting">Insn Splitting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn_002dattr_002eh"><code>insn-attr.h</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Defining-Attributes">Defining Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insns">insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insns_002c-generating">insns, generating</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insns_002c-recognizing">insns, recognizing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fANNULLED_005fBRANCH_005fP"><code>INSN_ANNULLED_BRANCH_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fCODE"><code>INSN_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fDELETED_005fP"><code>INSN_DELETED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fFROM_005fTARGET_005fP"><code>INSN_FROM_TARGET_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insn_005flist"><code>insn_list</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fREFERENCES_005fARE_005fDELAYED"><code>INSN_REFERENCES_ARE_DELAYED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fSETS_005fARE_005fDELAYED"><code>INSN_SETS_ARE_DELAYED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fUID"><code>INSN_UID</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INSN_005fVAR_005fLOCATION"><code>INSN_VAR_LOCATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-attributes">instruction attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Attributes">Insn Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-latency-time">instruction latency time</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-latency-time-1">instruction latency time</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-latency-time-2">instruction latency time</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-patterns">instruction patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-instruction-splitting">instruction splitting</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Splitting">Insn Splitting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insv-instruction-pattern"><code>insv</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insvm-instruction-pattern"><code>insv<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-insvmisalignm-instruction-pattern"><code>insvmisalign<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-int-iterators-in-_002emd-files">int iterators in <samp>.md</samp> files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Int-Iterators">Int Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT16_005fTYPE"><code>INT16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT32_005fTYPE"><code>INT32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT64_005fTYPE"><code>INT64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT8_005fTYPE"><code>INT8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INTEGER_005fCST"><code>INTEGER_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INTEGER_005fTYPE"><code>INTEGER_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Interdependence-of-Patterns">Interdependence of Patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dependent-Patterns">Dependent Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-interfacing-to-GCC-output">interfacing to GCC output</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-interlock-delays">interlock delays</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-intermediate-representation-lowering">intermediate representation lowering</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INTMAX_005fTYPE"><code>INTMAX_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INTPTR_005fTYPE"><code>INTPTR_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-introduction">introduction</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Top">Top</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fFAST16_005fTYPE"><code>INT_FAST16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fFAST32_005fTYPE"><code>INT_FAST32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fFAST64_005fTYPE"><code>INT_FAST64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fFAST8_005fTYPE"><code>INT_FAST8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fLEAST16_005fTYPE"><code>INT_LEAST16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fLEAST32_005fTYPE"><code>INT_LEAST32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fLEAST64_005fTYPE"><code>INT_LEAST64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fLEAST8_005fTYPE"><code>INT_LEAST8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INT_005fTYPE_005fSIZE"><code>INT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-INVOKE_005f_005fmain"><code>INVOKE__main</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-in_005fstruct"><code>in_struct</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-in_005fstruct_002c-in-code_005flabel-and-note"><code>in_struct</code>, in <code>code_label</code> and <code>note</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-in_005fstruct_002c-in-insn-and-jump_005finsn-and-call_005finsn"><code>in_struct</code>, in <code>insn</code> and <code>jump_insn</code> and <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-in_005fstruct_002c-in-insn_002c-jump_005finsn-and-call_005finsn"><code>in_struct</code>, in <code>insn</code>, <code>jump_insn</code> and <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-in_005fstruct_002c-in-subreg"><code>in_struct</code>, in <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ior"><code>ior</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ior-and-attributes"><code>ior</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ior_002c-canonicalization-of"><code>ior</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-iorm3-instruction-pattern"><code>ior<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IRA_005fHARD_005fREGNO_005fADD_005fCOST_005fMULTIPLIER"><code>IRA_HARD_REGNO_ADD_COST_MULTIPLIER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IS_005fASM_005fLOGICAL_005fLINE_005fSEPARATOR"><code>IS_ASM_LOGICAL_LINE_SEPARATOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005faddressable"><code>is_gimple_addressable</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fasm_005fval"><code>is_gimple_asm_val</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fassign"><code>is_gimple_assign</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fcall"><code>is_gimple_call</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fcall_005faddr"><code>is_gimple_call_addr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fconstant"><code>is_gimple_constant</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fdebug"><code>is_gimple_debug</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fip_005finvariant"><code>is_gimple_ip_invariant</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fip_005finvariant_005faddress"><code>is_gimple_ip_invariant_address</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fmem_005fref_005faddr"><code>is_gimple_mem_ref_addr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fmin_005finvariant"><code>is_gimple_min_invariant</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fomp"><code>is_gimple_omp</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE_005fOMP_005fPARALLEL"><code>GIMPLE_OMP_PARALLEL</code></a></td></tr>
<tr><td></td><td valign="top"><a href="#index-is_005fgimple_005fval"><code>is_gimple_val</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-iterators-in-_002emd-files">iterators in <samp>.md</samp> files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Iterators">Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IV-analysis-on-GIMPLE">IV analysis on GIMPLE</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-evolutions">Scalar evolutions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-IV-analysis-on-RTL">IV analysis on RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#loop_002div">loop-iv</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-J">J</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-JMP_005fBUF_005fSIZE"><code>JMP_BUF_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump"><code>jump</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump-instruction-pattern"><code>jump</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump-instruction-patterns">jump instruction patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Jump-Patterns">Jump Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump-instructions-and-set">jump instructions and <code>set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_002c-in-call_005finsn"><code>jump</code>, in <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_002c-in-insn"><code>jump</code>, in <code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_002c-in-mem"><code>jump</code>, in <code>mem</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Jumps">Jumps</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Jumps">Jumps</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-JUMP_005fALIGN"><code>JUMP_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn"><code>jump_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn-and-_002ff"><code>jump_insn</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn-and-_002fs"><code>jump_insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn-and-_002fs-1"><code>jump_insn</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn-and-_002fu"><code>jump_insn</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-jump_005finsn-and-_002fv"><code>jump_insn</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-JUMP_005fLABEL"><code>JUMP_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-JUMP_005fTABLES_005fIN_005fTEXT_005fSECTION"><code>JUMP_TABLES_IN_TEXT_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-L">L</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fALIGN"><code>LABEL_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fALIGN_005fAFTER_005fBARRIER"><code>LABEL_ALIGN_AFTER_BARRIER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fALTERNATE_005fNAME"><code>LABEL_ALTERNATE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fALT_005fENTRY_005fP"><code>LABEL_ALT_ENTRY_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fDECL"><code>LABEL_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fKIND"><code>LABEL_KIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fNUSES"><code>LABEL_NUSES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fPRESERVE_005fP"><code>LABEL_PRESERVE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-label_005fref"><code>label_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-label_005fref-and-_002fv"><code>label_ref</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-label_005fref_002c-RTL-sharing"><code>label_ref</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LABEL_005fREF_005fNONLOCAL_005fP"><code>LABEL_REF_NONLOCAL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-language_002ddependent-trees">language-dependent trees</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Language_002ddependent-trees">Language-dependent trees</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-language_002dindependent-intermediate-representation">language-independent intermediate representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lang_005fhooks_002egimplify_005fexpr"><code>lang_hooks.gimplify_expr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Gimplification-pass">Gimplification pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lang_005fhooks_002eparse_005ffile"><code>lang_hooks.parse_file</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-large-return-values">large return values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LARGEST_005fEXPONENT_005fIS_005fNORMAL"><code>LARGEST_EXPONENT_IS_NORMAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LAST_005fSTACK_005fREG"><code>LAST_STACK_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Registers">Stack Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LAST_005fVIRTUAL_005fREGISTER"><code>LAST_VIRTUAL_REGISTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lceilmn2"><code>lceil<var>m</var><var>n</var>2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LCSSA">LCSSA</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LCSSA">LCSSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LDD_005fSUFFIX"><code>LDD_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LD_005fFINI_005fSWITCH"><code>LD_FINI_SWITCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LD_005fINIT_005fSWITCH"><code>LD_INIT_SWITCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-le"><code>le</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-le-and-attributes"><code>le</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-leaf-functions">leaf functions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-leaf_005ffunction_005fp"><code>leaf_function_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LEAF_005fREGISTERS"><code>LEAF_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LEAF_005fREG_005fREMAP"><code>LEAF_REG_REMAP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Leaf-Functions">Leaf Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-left-rotate">left rotate</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-left-shift">left shift</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LEGITIMATE_005fPIC_005fOPERAND_005fP"><code>LEGITIMATE_PIC_OPERAND_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PIC">PIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LEGITIMIZE_005fRELOAD_005fADDRESS"><code>LEGITIMIZE_RELOAD_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-length"><code>length</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-less-than">less than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-less-than-or-equal">less than or equal</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-leu"><code>leu</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-leu-and-attributes"><code>leu</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LE_005fEXPR"><code>LE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lfloormn2"><code>lfloor<var>m</var><var>n</var>2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIB2FUNCS_005fEXTRA"><code>LIB2FUNCS_EXTRA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBCALL_005fVALUE"><code>LIBCALL_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-libgcc_002ea"><samp>libgcc.a</samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fCFLAGS"><code>LIBGCC2_CFLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fGNU_005fPREFIX"><code>LIBGCC2_GNU_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fHAS_005fDF_005fMODE"><code>LIBGCC2_HAS_DF_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fHAS_005fTF_005fMODE"><code>LIBGCC2_HAS_TF_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fHAS_005fXF_005fMODE"><code>LIBGCC2_HAS_XF_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fLONG_005fDOUBLE_005fTYPE_005fSIZE"><code>LIBGCC2_LONG_DOUBLE_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC2_005fUNWIND_005fATTRIBUTE"><code>LIBGCC2_UNWIND_ATTRIBUTE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBGCC_005fSPEC"><code>LIBGCC_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-library-subroutine-names">library subroutine names</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIBRARY_005fPATH_005fENV"><code>LIBRARY_PATH_ENV</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIB_005fSPEC"><code>LIB_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LIMIT_005fRELOAD_005fCLASS"><code>LIMIT_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Linear-loop-transformations-framework">Linear loop transformations framework</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Lambda">Lambda</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LINK_005fCOMMAND_005fSPEC"><code>LINK_COMMAND_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LINK_005fEH_005fSPEC"><code>LINK_EH_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LINK_005fGCC_005fC_005fSEQUENCE_005fSPEC"><code>LINK_GCC_C_SEQUENCE_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LINK_005fLIBGCC_005fSPECIAL_005f1"><code>LINK_LIBGCC_SPECIAL_1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LINK_005fSPEC"><code>LINK_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-list">list</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Liveness-representation">Liveness representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Liveness-information">Liveness information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-load-address-instruction">load address instruction</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOAD_005fEXTEND_005fOP"><code>LOAD_EXTEND_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-load_005fmultiple-instruction-pattern"><code>load_multiple</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Local-Register-Allocator-_0028LRA_0029">Local Register Allocator (LRA)</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fALIGNMENT"><code>LOCAL_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fCLASS_005fP"><code>LOCAL_CLASS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fDECL_005fALIGNMENT"><code>LOCAL_DECL_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fINCLUDE_005fDIR"><code>LOCAL_INCLUDE_DIR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fLABEL_005fPREFIX"><code>LOCAL_LABEL_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOCAL_005fREGNO"><code>LOCAL_REGNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Logical-Operators">Logical Operators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Logical-Operators">Logical Operators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-logical_002dand_002c-bitwise">logical-and, bitwise</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOGICAL_005fOP_005fNON_005fSHORT_005fCIRCUIT"><code>LOGICAL_OP_NON_SHORT_CIRCUIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-logm2-instruction-pattern"><code>log<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOG_005fLINKS"><code>LOG_LINKS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-longjmp-and-automatic-variables"><code>longjmp</code> and automatic variables</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fACCUM_005fTYPE_005fSIZE"><code>LONG_ACCUM_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fDOUBLE_005fTYPE_005fSIZE"><code>LONG_DOUBLE_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fFRACT_005fTYPE_005fSIZE"><code>LONG_FRACT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fLONG_005fACCUM_005fTYPE_005fSIZE"><code>LONG_LONG_ACCUM_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fLONG_005fFRACT_005fTYPE_005fSIZE"><code>LONG_LONG_FRACT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fLONG_005fTYPE_005fSIZE"><code>LONG_LONG_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LONG_005fTYPE_005fSIZE"><code>LONG_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Loop-analysis">Loop analysis</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Loop-representation">Loop representation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Loop-manipulation">Loop manipulation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Loop-manipulation">Loop manipulation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Loop-querying">Loop querying</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Loop-querying">Loop querying</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Loop-representation">Loop representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Loop-representation">Loop representation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Loop_002dclosed-SSA-form">Loop-closed SSA form</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LCSSA">LCSSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-looping-instruction-patterns">looping instruction patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Looping-Patterns">Looping Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOOP_005fALIGN"><code>LOOP_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LOOP_005fEXPR"><code>LOOP_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lowering_002c-language_002ddependent-intermediate-representation">lowering, language-dependent intermediate representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lo_005fsum"><code>lo_sum</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lrintmn2"><code>lrint<var>m</var><var>n</var>2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lroundmn2"><code>lround<var>m</var><var>n</var>2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lshiftrt"><code>lshiftrt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lshiftrt-and-attributes"><code>lshiftrt</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LSHIFT_005fEXPR"><code>LSHIFT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lshrm3-instruction-pattern"><code>lshr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lt"><code>lt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lt-and-attributes"><code>lt</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LTGT_005fEXPR"><code>LTGT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-lto">lto</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LTO">LTO</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ltrans">ltrans</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LTO">LTO</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ltu"><code>ltu</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-LT_005fEXPR"><code>LT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-M">M</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-m-in-constraint">&lsquo;<samp>m</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-attributes">machine attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-description-macros">machine description macros</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Macros">Target Macros</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-descriptions">machine descriptions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Desc">Machine Desc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-mode-conversions">machine mode conversions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-modes">machine modes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine-specific-constraints">machine specific constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Constraints">Machine Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-machine_002dindependent-predicates">machine-independent predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-macros_002c-target-description">macros, target description</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Macros">Target Macros</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-maddmn4-instruction-pattern"><code>madd<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-makefile-fragment">makefile fragment</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Fragments">Fragments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-makefile-targets">makefile targets</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Makefile">Makefile</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAKE_005fDECL_005fONE_005fONLY"><code>MAKE_DECL_ONE_ONLY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-make_005fsafe_005ffrom"><code>make_safe_from</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expander-Definitions">Expander Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MALLOC_005fABI_005fALIGNMENT"><code>MALLOC_ABI_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-marking-roots">marking roots</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GGC-Roots">GGC Roots</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mark_005fhook"><code>mark_hook</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MASK_005fRETURN_005fADDR"><code>MASK_RETURN_ADDR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-matching-constraint">matching constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-matching-operands">matching operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fdup"><code>match_dup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fdup-1"><code>match_dup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole2">define_peephole2</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fdup-and-attributes"><code>match_dup</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005foperand"><code>match_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005foperand-and-attributes"><code>match_operand</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005foperator"><code>match_operator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fop_005fdup"><code>match_op_dup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fparallel"><code>match_parallel</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fpar_005fdup"><code>match_par_dup</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fscratch"><code>match_scratch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005fscratch-1"><code>match_scratch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole2">define_peephole2</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-match_005ftest-and-attributes"><code>match_test</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-math-library">math library</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-math_002c-in-RTL">math, in RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-matherr"><code>matherr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MATH_005fLIBRARY"><code>MATH_LIBRARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-maxm3-instruction-pattern"><code>max<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fBITS_005fPER_005fWORD"><code>MAX_BITS_PER_WORD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fCONDITIONAL_005fEXECUTE"><code>MAX_CONDITIONAL_EXECUTE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fFIXED_005fMODE_005fSIZE"><code>MAX_FIXED_MODE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fMOVE_005fMAX"><code>MAX_MOVE_MAX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fOFILE_005fALIGNMENT"><code>MAX_OFILE_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fREGS_005fPER_005fADDRESS"><code>MAX_REGS_PER_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MAX_005fSTACK_005fALIGNMENT"><code>MAX_STACK_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-maybe_005fundef"><code>maybe_undef</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-may_005ftrap_005fp_002c-tree_005fcould_005ftrap_005fp"><code>may_trap_p, tree_could_trap_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mcount"><code>mcount</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fCAN_005fREDIRECT_005fBRANCH"><code>MD_CAN_REDIRECT_BRANCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fEXEC_005fPREFIX"><code>MD_EXEC_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fFALLBACK_005fFRAME_005fSTATE_005fFOR"><code>MD_FALLBACK_FRAME_STATE_FOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fHANDLE_005fUNWABI"><code>MD_HANDLE_UNWABI</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fSTARTFILE_005fPREFIX"><code>MD_STARTFILE_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MD_005fSTARTFILE_005fPREFIX_005f1"><code>MD_STARTFILE_PREFIX_1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem"><code>mem</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem-and-_002fc"><code>mem</code> and &lsquo;<samp>/c</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem-and-_002ff"><code>mem</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem-and-_002fj"><code>mem</code> and &lsquo;<samp>/j</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem-and-_002fu"><code>mem</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem-and-_002fv"><code>mem</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem_002c-RTL-sharing"><code>mem</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-memory-model">memory model</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Memory-model">Memory model</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-memory-reference_002c-nonoffsettable">memory reference, nonoffsettable</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-memory-references-in-constraints">memory references in constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-memory_005fbarrier-instruction-pattern"><code>memory_barrier</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEMORY_005fMOVE_005fCOST"><code>MEMORY_MOVE_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-memory_005foperand"><code>memory_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fADDR_005fSPACE"><code>MEM_ADDR_SPACE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fALIAS_005fSET"><code>MEM_ALIAS_SET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fALIGN"><code>MEM_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fEXPR"><code>MEM_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fKEEP_005fALIAS_005fSET_005fP"><code>MEM_KEEP_ALIAS_SET_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fNOTRAP_005fP"><code>MEM_NOTRAP_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fOFFSET"><code>MEM_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fOFFSET_005fKNOWN_005fP"><code>MEM_OFFSET_KNOWN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fPOINTER"><code>MEM_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fREADONLY_005fP"><code>MEM_READONLY_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fREF"><code>MEM_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem_005fsignal_005ffencemode-instruction-pattern"><code>mem_signal_fence<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fSIZE"><code>MEM_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fSIZE_005fKNOWN_005fP"><code>MEM_SIZE_KNOWN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mem_005fthread_005ffencemode-instruction-pattern"><code>mem_thread_fence<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MEM_005fVOLATILE_005fP"><code>MEM_VOLATILE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-METHOD_005fTYPE"><code>METHOD_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MINIMUM_005fALIGNMENT"><code>MINIMUM_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MINIMUM_005fATOMIC_005fALIGNMENT"><code>MINIMUM_ATOMIC_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-minm3-instruction-pattern"><code>min<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-minus"><code>minus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-minus-and-attributes"><code>minus</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-minus_002c-canonicalization-of"><code>minus</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MINUS_005fEXPR"><code>MINUS_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MIN_005fUNITS_005fPER_005fWORD"><code>MIN_UNITS_PER_WORD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MIPS-coprocessor_002ddefinition-macros">MIPS coprocessor-definition macros</a>:</td><td>&nbsp;</td><td valign="top"><a href="#MIPS-Coprocessors">MIPS Coprocessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mod"><code>mod</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mod-and-attributes"><code>mod</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mode-classes">mode classes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mode-iterators-in-_002emd-files">mode iterators in <samp>.md</samp> files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Iterators">Mode Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mode-switching">mode switching</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODES_005fTIEABLE_005fP"><code>MODES_TIEABLE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fACCUM"><code>MODE_ACCUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fAFTER"><code>MODE_AFTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fBASE_005fREG_005fCLASS"><code>MODE_BASE_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fBASE_005fREG_005fREG_005fCLASS"><code>MODE_BASE_REG_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fCC"><code>MODE_CC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fCC-1"><code>MODE_CC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fCODE_005fBASE_005fREG_005fCLASS"><code>MODE_CODE_BASE_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fCOMPLEX_005fFLOAT"><code>MODE_COMPLEX_FLOAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fCOMPLEX_005fINT"><code>MODE_COMPLEX_INT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fDECIMAL_005fFLOAT"><code>MODE_DECIMAL_FLOAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fENTRY"><code>MODE_ENTRY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fEXIT"><code>MODE_EXIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fFLOAT"><code>MODE_FLOAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fFRACT"><code>MODE_FRACT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fFUNCTION"><code>MODE_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fINT"><code>MODE_INT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fNEEDED"><code>MODE_NEEDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fPARTIAL_005fINT"><code>MODE_PARTIAL_INT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fPRIORITY_005fTO_005fMODE"><code>MODE_PRIORITY_TO_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fRANDOM"><code>MODE_RANDOM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fUACCUM"><code>MODE_UACCUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODE_005fUFRACT"><code>MODE_UFRACT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-modifiers-in-constraints">modifiers in constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Modifiers">Modifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODIFY_005fEXPR"><code>MODIFY_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MODIFY_005fJNI_005fMETHOD_005fCALL"><code>MODIFY_JNI_METHOD_CALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-modm3-instruction-pattern"><code>mod<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-modulo-scheduling">modulo scheduling</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MOVE_005fBY_005fPIECES_005fP"><code>MOVE_BY_PIECES_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MOVE_005fMAX"><code>MOVE_MAX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MOVE_005fMAX_005fPIECES"><code>MOVE_MAX_PIECES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MOVE_005fRATIO"><code>MOVE_RATIO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movm-instruction-pattern"><code>mov<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movmemm-instruction-pattern"><code>movmem<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movmisalignm-instruction-pattern"><code>movmisalign<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movmodecc-instruction-pattern"><code>mov<var>mode</var>cc</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movstr-instruction-pattern"><code>movstr</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-movstrictm-instruction-pattern"><code>movstrict<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-msubmn4-instruction-pattern"><code>msub<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mulhisi3-instruction-pattern"><code>mulhisi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mulm3-instruction-pattern"><code>mul<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mulqihi3-instruction-pattern"><code>mulqihi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mulsidi3-instruction-pattern"><code>mulsidi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mult"><code>mult</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mult-and-attributes"><code>mult</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mult_002c-canonicalization-of"><code>mult</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-mult_002c-canonicalization-of-1"><code>mult</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTIARCH_005fDIRNAME"><code>MULTIARCH_DIRNAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fDEFAULTS"><code>MULTILIB_DEFAULTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fDIRNAMES"><code>MULTILIB_DIRNAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fEXCEPTIONS"><code>MULTILIB_EXCEPTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fEXTRA_005fOPTS"><code>MULTILIB_EXTRA_OPTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fMATCHES"><code>MULTILIB_MATCHES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fOPTIONS"><code>MULTILIB_OPTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fOSDIRNAMES"><code>MULTILIB_OSDIRNAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fREQUIRED"><code>MULTILIB_REQUIRED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTILIB_005fREUSE"><code>MULTILIB_REUSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-multiple-alternative-constraints">multiple alternative constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Multi_002dAlternative">Multi-Alternative</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULTIPLE_005fSYMBOL_005fSPACES"><code>MULTIPLE_SYMBOL_SPACES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-multiplication">multiplication</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-multiplication-with-signed-saturation">multiplication with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-multiplication-with-unsigned-saturation">multiplication with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULT_005fEXPR"><code>MULT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-MULT_005fHIGHPART_005fEXPR"><code>MULT_HIGHPART_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-N">N</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-n-in-constraint">&lsquo;<samp>n</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-name">name</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Identifiers">Identifiers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-named-address-spaces">named address spaces</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-named-patterns-and-conditions">named patterns and conditions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-names_002c-pattern">names, pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-namespace_002c-scope">namespace, scope</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NAMESPACE_005fDECL"><code>NAMESPACE_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NAMESPACE_005fDECL-1"><code>NAMESPACE_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Namespaces">Namespaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NATIVE_005fSYSTEM_005fHEADER_005fCOMPONENT"><code>NATIVE_SYSTEM_HEADER_COMPONENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ne"><code>ne</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ne-and-attributes"><code>ne</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nearbyintm2-instruction-pattern"><code>nearbyint<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-neg"><code>neg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-neg-and-attributes"><code>neg</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-neg_002c-canonicalization-of"><code>neg</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NEGATE_005fEXPR"><code>NEGATE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-negation">negation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-negation-with-signed-saturation">negation with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-negation-with-unsigned-saturation">negation with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-negm2-instruction-pattern"><code>neg<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nested-functions_002c-trampolines-for">nested functions, trampolines for</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nested_005fptr"><code>nested_ptr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-next_005fbb_002c-prev_005fbb_002c-FOR_005fEACH_005fBB_002c-FOR_005fALL_005fBB"><code>next_bb, prev_bb, FOR_EACH_BB, FOR_ALL_BB</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NEXT_005fINSN"><code>NEXT_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NEXT_005fOBJC_005fRUNTIME"><code>NEXT_OBJC_RUNTIME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NE_005fEXPR"><code>NE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nil">nil</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NM_005fFLAGS"><code>NM_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nondeterministic-finite-state-automaton">nondeterministic finite state automaton</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonimmediate_005foperand"><code>nonimmediate_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonlocal-goto-handler">nonlocal goto handler</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonlocal_005fgoto-instruction-pattern"><code>nonlocal_goto</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonlocal_005fgoto_005freceiver-instruction-pattern"><code>nonlocal_goto_receiver</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonmemory_005foperand"><code>nonmemory_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nonoffsettable-memory-reference">nonoffsettable memory reference</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NON_005fLVALUE_005fEXPR"><code>NON_LVALUE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-nop-instruction-pattern"><code>nop</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOP_005fEXPR"><code>NOP_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-normal-predicates">normal predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-not"><code>not</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-not-and-attributes"><code>not</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-not-equal">not equal</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-not_002c-canonicalization-of"><code>not</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-note"><code>note</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-note-and-_002fi"><code>note</code> and &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-note-and-_002fv"><code>note</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fBASIC_005fBLOCK"><code>NOTE_INSN_BASIC_BLOCK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fBASIC_005fBLOCK-1"><code>NOTE_INSN_BASIC_BLOCK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fBLOCK_005fBEG"><code>NOTE_INSN_BLOCK_BEG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fBLOCK_005fEND"><code>NOTE_INSN_BLOCK_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fDELETED"><code>NOTE_INSN_DELETED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fDELETED_005fLABEL"><code>NOTE_INSN_DELETED_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fEH_005fREGION_005fBEG"><code>NOTE_INSN_EH_REGION_BEG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fEH_005fREGION_005fEND"><code>NOTE_INSN_EH_REGION_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fFUNCTION_005fBEG"><code>NOTE_INSN_FUNCTION_BEG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fLOOP_005fBEG"><code>NOTE_INSN_LOOP_BEG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fLOOP_005fCONT"><code>NOTE_INSN_LOOP_CONT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fLOOP_005fEND"><code>NOTE_INSN_LOOP_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fLOOP_005fVTOP"><code>NOTE_INSN_LOOP_VTOP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fINSN_005fVAR_005fLOCATION"><code>NOTE_INSN_VAR_LOCATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fLINE_005fNUMBER"><code>NOTE_LINE_NUMBER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fSOURCE_005fFILE"><code>NOTE_SOURCE_FILE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTE_005fVAR_005fLOCATION"><code>NOTE_VAR_LOCATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NOTICE_005fUPDATE_005fCC"><code>NOTICE_UPDATE_CC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#CC0-Condition-Codes">CC0 Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDBX_005fBNSYM_005fENSYM"><code>NO_DBX_BNSYM_ENSYM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Hooks">DBX Hooks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDBX_005fFUNCTION_005fEND"><code>NO_DBX_FUNCTION_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Hooks">DBX Hooks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDBX_005fGCC_005fMARKER"><code>NO_DBX_GCC_MARKER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Names-and-DBX">File Names and DBX</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDBX_005fMAIN_005fSOURCE_005fDIRECTORY"><code>NO_DBX_MAIN_SOURCE_DIRECTORY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Names-and-DBX">File Names and DBX</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDOLLAR_005fIN_005fLABEL"><code>NO_DOLLAR_IN_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fDOT_005fIN_005fLABEL"><code>NO_DOT_IN_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fFUNCTION_005fCSE"><code>NO_FUNCTION_CSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fIMPLICIT_005fEXTERN_005fC"><code>NO_IMPLICIT_EXTERN_C</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fPROFILE_005fCOUNTERS"><code>NO_PROFILE_COUNTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NO_005fREGS"><code>NO_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Number-of-iterations-analysis">Number of iterations analysis</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Number-of-iterations">Number of iterations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NUM_005fMACHINE_005fMODES"><code>NUM_MACHINE_MODES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-NUM_005fMODES_005fFOR_005fMODE_005fSWITCHING"><code>NUM_MODES_FOR_MODE_SWITCHING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-N_005fREG_005fCLASSES"><code>N_REG_CLASSES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-O">O</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-o-in-constraint">&lsquo;<samp>o</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OBJC_005fGEN_005fMETHOD_005fLABEL"><code>OBJC_GEN_METHOD_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OBJC_005fJBLEN"><code>OBJC_JBLEN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OBJECT_005fFORMAT_005fCOFF"><code>OBJECT_FORMAT_COFF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-offsettable-address">offsettable address</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OFFSET_005fTYPE"><code>OFFSET_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OImode"><code>OImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Omega-a-solver-for-linear-programming-problems">Omega a solver for linear programming problems</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Omega">Omega</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fATOMIC"><code>OMP_ATOMIC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fCLAUSE"><code>OMP_CLAUSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fCONTINUE"><code>OMP_CONTINUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fCRITICAL"><code>OMP_CRITICAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fFOR"><code>OMP_FOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fMASTER"><code>OMP_MASTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fORDERED"><code>OMP_ORDERED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fPARALLEL"><code>OMP_PARALLEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fRETURN"><code>OMP_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fSECTION"><code>OMP_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fSECTIONS"><code>OMP_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OMP_005fSINGLE"><code>OMP_SINGLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#OpenMP">OpenMP</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-one_005fcmplm2-instruction-pattern"><code>one_cmpl<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operand-access">operand access</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Operand-Access-Routines">Operand Access Routines</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operand-constraints">operand constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constraints">Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Operand-Iterators">Operand Iterators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operand-predicates">operand predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operand-substitution">operand substitution</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Operands">Operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Operands">Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operands">operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operands-1"><code>operands</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-operator-predicates">operator predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-optc_002dgen_002eawk">&lsquo;<samp>optc-gen.awk</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Options">Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Optimization-infrastructure-for-GIMPLE">Optimization infrastructure for GIMPLE</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tree-SSA">Tree SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OPTIMIZE_005fMODE_005fSWITCHING"><code>OPTIMIZE_MODE_SWITCHING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Mode-Switching">Mode Switching</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-option-specification-files">option specification files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Options">Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-optional-hardware-or-system-features">optional hardware or system features</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-options_002c-directory-search">options, directory search</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Including-Patterns">Including Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OPTION_005fDEFAULT_005fSPECS"><code>OPTION_DEFAULT_SPECS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-order-of-register-allocation">order of register allocation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ordered_005fcomparison_005foperator"><code>ordered_comparison_operator</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ORDERED_005fEXPR"><code>ORDERED_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Ordering-of-Patterns">Ordering of Patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Pattern-Ordering">Pattern Ordering</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ORIGINAL_005fREGNO"><code>ORIGINAL_REGNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-other-register-constraints">other register constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-outgoing_005fargs_005fsize"><code>outgoing_args_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OUTGOING_005fREGNO"><code>OUTGOING_REGNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OUTGOING_005fREG_005fPARM_005fSTACK_005fSPACE"><code>OUTGOING_REG_PARM_STACK_SPACE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-output-of-assembler-code">output of assembler code</a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-output-statements">output statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-output-templates">output templates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-output_005fasm_005finsn"><code>output_asm_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OUTPUT_005fQUOTED_005fSTRING"><code>OUTPUT_QUOTED_STRING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OVERLAPPING_005fREGISTER_005fNAMES"><code>OVERLAPPING_REGISTER_NAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OVERLOAD"><code>OVERLOAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OVERRIDE_005fABI_005fFORMAT"><code>OVERRIDE_ABI_FORMAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OVL_005fCURRENT"><code>OVL_CURRENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-OVL_005fNEXT"><code>OVL_NEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-P">P</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-p-in-constraint">&lsquo;<samp>p</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PAD_005fVARARGS_005fDOWN"><code>PAD_VARARGS_DOWN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-parallel"><code>parallel</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-parameters_002c-c_002b_002b-abi">parameters, c++ abi</a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-parameters_002c-miscellaneous">parameters, miscellaneous</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-parameters_002c-precompiled-headers">parameters, precompiled headers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#PCH-Target">PCH Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-paramn_005fis"><code>param<var>n</var>_is</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-param_005fis"><code>param_is</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-parity"><code>parity</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-paritym2-instruction-pattern"><code>parity<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PARM_005fBOUNDARY"><code>PARM_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PARM_005fDECL"><code>PARM_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PARSE_005fLDD_005fOUTPUT"><code>PARSE_LDD_OUTPUT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-passes-and-files-of-the-compiler">passes and files of the compiler</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Passes">Passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-passing-arguments">passing arguments</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pass_005fduplicate_005fcomputed_005fgotos"><code>pass_duplicate_computed_gotos</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PATH_005fSEPARATOR"><code>PATH_SEPARATOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PATTERN"><code>PATTERN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pattern-conditions">pattern conditions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pattern-names">pattern names</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Pattern-Ordering">Pattern Ordering</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Pattern-Ordering">Pattern Ordering</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-patterns">patterns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Patterns">Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pc"><code>pc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pc-and-attributes"><code>pc</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Lengths">Insn Lengths</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pc_002c-RTL-sharing"><code>pc</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PCC_005fBITFIELD_005fTYPE_005fMATTERS"><code>PCC_BITFIELD_TYPE_MATTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PCC_005fSTATIC_005fSTRUCT_005fRETURN"><code>PCC_STATIC_STRUCT_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PC_005fREGNUM"><code>PC_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pc_005frtx"><code>pc_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PDImode"><code>PDImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-peephole-optimization_002c-RTL-representation">peephole optimization, RTL representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-peephole-optimizer-definitions">peephole optimizer definitions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Peephole-Definitions">Peephole Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-per_002dfunction-data">per-function data</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Per_002dFunction-Data">Per-Function Data</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-percent-sign">percent sign</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Template">Output Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI-nodes">PHI nodes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI_005fARG_005fDEF"><code>PHI_ARG_DEF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI_005fARG_005fEDGE"><code>PHI_ARG_EDGE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI_005fARG_005fELT"><code>PHI_ARG_ELT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI_005fNUM_005fARGS"><code>PHI_NUM_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PHI_005fRESULT"><code>PHI_RESULT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PIC">PIC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#PIC">PIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PIC_005fOFFSET_005fTABLE_005fREGNUM"><code>PIC_OFFSET_TABLE_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PIC">PIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PIC_005fOFFSET_005fTABLE_005fREG_005fCALL_005fCLOBBERED"><code>PIC_OFFSET_TABLE_REG_CALL_CLOBBERED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PIC">PIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pipeline-hazard-recognizer">pipeline hazard recognizer</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pipeline-hazard-recognizer-1">pipeline hazard recognizer</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Plugins">Plugins</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Plugins">Plugins</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-plus"><code>plus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-plus-and-attributes"><code>plus</code> and attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expressions">Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-plus_002c-canonicalization-of"><code>plus</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PLUS_005fEXPR"><code>PLUS_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Pmode"><code>Pmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pmode_005fregister_005foperand"><code>pmode_register_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pointer">pointer</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POINTERS_005fEXTEND_005fUNSIGNED"><code>POINTERS_EXTEND_UNSIGNED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POINTER_005fPLUS_005fEXPR"><code>POINTER_PLUS_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POINTER_005fSIZE"><code>POINTER_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POINTER_005fTYPE"><code>POINTER_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-popcount"><code>popcount</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-popcountm2-instruction-pattern"><code>popcount<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pops_005fargs"><code>pops_args</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pop_005foperand"><code>pop_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-portability">portability</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Portability">Portability</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-position-independent-code">position independent code</a>:</td><td>&nbsp;</td><td valign="top"><a href="#PIC">PIC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POSTDECREMENT_005fEXPR"><code>POSTDECREMENT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POSTINCREMENT_005fEXPR"><code>POSTINCREMENT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-post_005fdec"><code>post_dec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-post_005finc"><code>post_inc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-post_005fmodify"><code>post_modify</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-post_005forder_005fcompute_002c-inverted_005fpost_005forder_005fcompute_002c-walk_005fdominator_005ftree"><code>post_order_compute, inverted_post_order_compute, walk_dominator_tree</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-POWI_005fMAX_005fMULTS"><code>POWI_MAX_MULTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-powm3-instruction-pattern"><code>pow<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pragma"><code>pragma</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREDECREMENT_005fEXPR"><code>PREDECREMENT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-predefined-macros">predefined macros</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-predicates">predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-predicates-and-machine-modes">predicates and machine modes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-predication">predication</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conditional-Execution">Conditional Execution</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-predict_002edef"><code>predict.def</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREFERRED_005fDEBUGGING_005fTYPE"><code>PREFERRED_DEBUGGING_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#All-Debuggers">All Debuggers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREFERRED_005fRELOAD_005fCLASS"><code>PREFERRED_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREFERRED_005fSTACK_005fBOUNDARY"><code>PREFERRED_STACK_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prefetch"><code>prefetch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prefetch-and-_002fv"><code>prefetch</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prefetch-instruction-pattern"><code>prefetch</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREFETCH_005fSCHEDULE_005fBARRIER_005fP"><code>PREFETCH_SCHEDULE_BARRIER_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREINCREMENT_005fEXPR"><code>PREINCREMENT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-presence_005fset"><code>presence_set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-preserving-SSA-form">preserving SSA form</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-preserving-virtual-SSA-form">preserving virtual SSA form</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pretend_005fargs_005fsize"><code>pretend_args_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prev_005factive_005finsn"><code>prev_active_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#define_005fpeephole">define_peephole</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PREV_005fINSN"><code>PREV_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pre_005fdec"><code>pre_dec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PRE_005fGCC3_005fDWARF_005fFRAME_005fREGISTERS"><code>PRE_GCC3_DWARF_FRAME_REGISTERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pre_005finc"><code>pre_inc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pre_005fmodify"><code>pre_modify</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PRINT_005fOPERAND"><code>PRINT_OPERAND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PRINT_005fOPERAND_005fADDRESS"><code>PRINT_OPERAND_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PRINT_005fOPERAND_005fPUNCT_005fVALID_005fP"><code>PRINT_OPERAND_PUNCT_VALID_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-probe_005fstack-instruction-pattern"><code>probe_stack</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-probe_005fstack_005faddress-instruction-pattern"><code>probe_stack_address</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-processor-functional-units">processor functional units</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-processor-functional-units-1">processor functional units</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-processor-pipeline-description">processor pipeline description</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-product">product</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-profile-feedback">profile feedback</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-profile-representation">profile representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PROFILE_005fBEFORE_005fPROLOGUE"><code>PROFILE_BEFORE_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PROFILE_005fHOOK"><code>PROFILE_HOOK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-profiling_002c-code-generation">profiling, code generation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profiling">Profiling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-program-counter">program counter</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prologue">prologue</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-prologue-instruction-pattern"><code>prologue</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PROMOTE_005fMODE"><code>PROMOTE_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pseudo-registers">pseudo registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PSImode"><code>PSImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PTRDIFF_005fTYPE"><code>PTRDIFF_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-purge_005fdead_005fedges"><code>purge_dead_edges</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-purge_005fdead_005fedges-1"><code>purge_dead_edges</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-push-address-instruction">push address instruction</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-pushm1-instruction-pattern"><code>push<var>m</var>1</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUSH_005fARGS"><code>PUSH_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUSH_005fARGS_005fREVERSED"><code>PUSH_ARGS_REVERSED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-push_005foperand"><code>push_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-push_005freload"><code>push_reload</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUSH_005fROUNDING"><code>PUSH_ROUNDING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUT_005fCODE"><code>PUT_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUT_005fMODE"><code>PUT_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUT_005fREG_005fNOTE_005fKIND"><code>PUT_REG_NOTE_KIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-PUT_005fSDB_005f"><code>PUT_SDB_</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-Q">Q</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-QCmode"><code>QCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-QFmode"><code>QFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-QImode"><code>QImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-QImode_002c-in-insn"><code>QImode</code>, in <code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-QQmode"><code>QQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-qualified-type">qualified type</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-qualified-type-1">qualified type</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-querying-function-unit-reservations">querying function unit reservations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-question-mark">question mark</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Multi_002dAlternative">Multi-Alternative</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-quotient">quotient</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-R">R</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-r-in-constraint">&lsquo;<samp>r</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RDIV_005fEXPR"><code>RDIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-READONLY_005fDATA_005fSECTION_005fASM_005fOP"><code>READONLY_DATA_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-real-operands">real operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REALPART_005fEXPR"><code>REALPART_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fARITHMETIC"><code>REAL_ARITHMETIC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fCST"><code>REAL_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fLIBGCC_005fSPEC"><code>REAL_LIBGCC_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fNM_005fFILE_005fNAME"><code>REAL_NM_FILE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fTYPE"><code>REAL_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUES_005fEQUAL"><code>REAL_VALUES_EQUAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUES_005fLESS"><code>REAL_VALUES_LESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fABS"><code>REAL_VALUE_ABS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fATOF"><code>REAL_VALUE_ATOF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fFIX"><code>REAL_VALUE_FIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fFROM_005fINT"><code>REAL_VALUE_FROM_INT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fISINF"><code>REAL_VALUE_ISINF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fISNAN"><code>REAL_VALUE_ISNAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fNEGATE"><code>REAL_VALUE_NEGATE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fNEGATIVE"><code>REAL_VALUE_NEGATIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fINT"><code>REAL_VALUE_TO_INT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL128"><code>REAL_VALUE_TO_TARGET_DECIMAL128</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL32"><code>REAL_VALUE_TO_TARGET_DECIMAL32</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fDECIMAL64"><code>REAL_VALUE_TO_TARGET_DECIMAL64</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fDOUBLE"><code>REAL_VALUE_TO_TARGET_DOUBLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fLONG_005fDOUBLE"><code>REAL_VALUE_TO_TARGET_LONG_DOUBLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTO_005fTARGET_005fSINGLE"><code>REAL_VALUE_TO_TARGET_SINGLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fTYPE"><code>REAL_VALUE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REAL_005fVALUE_005fUNSIGNED_005fFIX"><code>REAL_VALUE_UNSIGNED_FIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Floating-Point">Floating Point</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-recognizing-insns">recognizing insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-recog_005fdata_002eoperand"><code>recog_data.operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RECORD_005fTYPE"><code>RECORD_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RECORD_005fTYPE-1"><code>RECORD_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-redirect_005fedge_005fand_005fbranch"><code>redirect_edge_and_branch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-redirect_005fedge_005fand_005fbranch_002c-redirect_005fjump"><code>redirect_edge_and_branch, redirect_jump</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fsmax_005fm-instruction-pattern"><code>reduc_smax_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fsmin_005fm-instruction-pattern"><code>reduc_smin_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fsplus_005fm-instruction-pattern"><code>reduc_splus_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fumax_005fm-instruction-pattern"><code>reduc_umax_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fumin_005fm-instruction-pattern"><code>reduc_umin_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reduc_005fuplus_005fm-instruction-pattern"><code>reduc_uplus_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reference">reference</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REFERENCE_005fTYPE"><code>REFERENCE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg"><code>reg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg-and-_002ff"><code>reg</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg-and-_002fi"><code>reg</code> and &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg-and-_002fv"><code>reg</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg_002c-RTL-sharing"><code>reg</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-regclass_005ffor_005fconstraint"><code>regclass_for_constraint</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register-allocation-order">register allocation order</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register-class-definitions">register class definitions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register-class-preference-constraints">register class preference constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Class-Preferences">Class Preferences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register-pairs">register pairs</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Register-Transfer-Language-_0028RTL_0029">Register Transfer Language (RTL)</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL">RTL</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register-usage">register usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Registers">Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-registers-arguments">registers arguments</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-registers-in-constraints">registers in constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGISTER_005fMOVE_005fCOST"><code>REGISTER_MOVE_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGISTER_005fNAMES"><code>REGISTER_NAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-register_005foperand"><code>register_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGISTER_005fPREFIX"><code>REGISTER_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGISTER_005fTARGET_005fPRAGMAS"><code>REGISTER_TARGET_PRAGMAS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGMODE_005fNATURAL_005fSIZE"><code>REGMODE_NATURAL_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fMODE_005fCODE_005fOK_005fFOR_005fBASE_005fP"><code>REGNO_MODE_CODE_OK_FOR_BASE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fMODE_005fOK_005fFOR_005fBASE_005fP"><code>REGNO_MODE_OK_FOR_BASE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fMODE_005fOK_005fFOR_005fREG_005fBASE_005fP"><code>REGNO_MODE_OK_FOR_REG_BASE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fOK_005fFOR_005fBASE_005fP"><code>REGNO_OK_FOR_BASE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fOK_005fFOR_005fINDEX_005fP"><code>REGNO_OK_FOR_INDEX_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REGNO_005fREG_005fCLASS"><code>REGNO_REG_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-regs_005fever_005flive"><code>regs_ever_live</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-regular-expressions">regular expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-regular-expressions-1">regular expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fALLOC_005fORDER"><code>REG_ALLOC_ORDER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Allocation-Order">Allocation Order</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fBR_005fPRED"><code>REG_BR_PRED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fBR_005fPROB"><code>REG_BR_PROB</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fBR_005fPROB_005fBASE_002c-BB_005fFREQ_005fBASE_002c-count"><code>REG_BR_PROB_BASE, BB_FREQ_BASE, count</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fBR_005fPROB_005fBASE_002c-EDGE_005fFREQUENCY"><code>REG_BR_PROB_BASE, EDGE_FREQUENCY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCC_005fSETTER"><code>REG_CC_SETTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCC_005fUSER"><code>REG_CC_USER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg_005fclass_005fcontents"><code>reg_class_contents</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCLASS_005fCONTENTS"><code>REG_CLASS_CONTENTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCLASS_005fFROM_005fCONSTRAINT"><code>REG_CLASS_FROM_CONSTRAINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCLASS_005fFROM_005fLETTER"><code>REG_CLASS_FROM_LETTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Old-Constraints">Old Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCLASS_005fNAMES"><code>REG_CLASS_NAMES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fCROSSING_005fJUMP"><code>REG_CROSSING_JUMP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fDEAD"><code>REG_DEAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fDEAD_002c-REG_005fUNUSED"><code>REG_DEAD, REG_UNUSED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Liveness-information">Liveness information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fDEP_005fANTI"><code>REG_DEP_ANTI</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fDEP_005fOUTPUT"><code>REG_DEP_OUTPUT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fDEP_005fTRUE"><code>REG_DEP_TRUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fEH_005fREGION_002c-EDGE_005fABNORMAL_005fCALL"><code>REG_EH_REGION, EDGE_ABNORMAL_CALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fEQUAL"><code>REG_EQUAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fEQUIV"><code>REG_EQUIV</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fEXPR"><code>REG_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fFRAME_005fRELATED_005fEXPR"><code>REG_FRAME_RELATED_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fFUNCTION_005fVALUE_005fP"><code>REG_FUNCTION_VALUE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fINC"><code>REG_INC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg_005flabel-and-_002fv"><code>reg_label</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fLABEL_005fOPERAND"><code>REG_LABEL_OPERAND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fLABEL_005fTARGET"><code>REG_LABEL_TARGET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg_005fnames"><code>reg_names</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reg_005fnames-1"><code>reg_names</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fNONNEG"><code>REG_NONNEG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fNOTES"><code>REG_NOTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fNOTE_005fKIND"><code>REG_NOTE_KIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fOFFSET"><code>REG_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fOK_005fSTRICT"><code>REG_OK_STRICT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fPARM_005fSTACK_005fSPACE"><code>REG_PARM_STACK_SPACE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fPARM_005fSTACK_005fSPACE_002c-and-TARGET_005fFUNCTION_005fARG"><code>REG_PARM_STACK_SPACE</code>, and <code>TARGET_FUNCTION_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fPOINTER"><code>REG_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fSETJMP"><code>REG_SETJMP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fUNUSED"><code>REG_UNUSED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fUSERVAR_005fP"><code>REG_USERVAR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fVALUE_005fIN_005fUNWIND_005fCONTEXT"><code>REG_VALUE_IN_UNWIND_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REG_005fWORDS_005fBIG_005fENDIAN"><code>REG_WORDS_BIG_ENDIAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-relative-costs">relative costs</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RELATIVE_005fPREFIX_005fNOT_005fLINKDIR"><code>RELATIVE_PREFIX_NOT_LINKDIR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reloading">reloading</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reload_005fcompleted"><code>reload_completed</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reload_005fin-instruction-pattern"><code>reload_in</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reload_005fin_005fprogress"><code>reload_in_progress</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reload_005fout-instruction-pattern"><code>reload_out</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-remainder">remainder</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-remainderm3-instruction-pattern"><code>remainder<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reorder"><code>reorder</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-representation-of-RTL">representation of RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL">RTL</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reservation-delays">reservation delays</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-restore_005fstack_005fblock-instruction-pattern"><code>restore_stack_block</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-restore_005fstack_005ffunction-instruction-pattern"><code>restore_stack_function</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-restore_005fstack_005fnonlocal-instruction-pattern"><code>restore_stack_nonlocal</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rest_005fof_005fdecl_005fcompilation"><code>rest_of_decl_compilation</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rest_005fof_005ftype_005fcompilation"><code>rest_of_type_compilation</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Parsing-pass">Parsing pass</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RESULT_005fDECL"><code>RESULT_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return"><code>return</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return-instruction-pattern"><code>return</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return-values-in-registers">return values in registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-returning-aggregate-values">returning aggregate values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-returning-structures-and-unions">returning structures and unions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fADDRESS_005fPOINTER_005fREGNUM"><code>RETURN_ADDRESS_POINTER_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fADDR_005fIN_005fPREVIOUS_005fFRAME"><code>RETURN_ADDR_IN_PREVIOUS_FRAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fADDR_005fOFFSET"><code>RETURN_ADDR_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fADDR_005fRTX"><code>RETURN_ADDR_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fEXPR"><code>RETURN_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RETURN_005fSTMT"><code>RETURN_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return_005fval"><code>return_val</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return_005fval_002c-in-call_005finsn"><code>return_val</code>, in <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return_005fval_002c-in-reg"><code>return_val</code>, in <code>reg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-return_005fval_002c-in-symbol_005fref"><code>return_val</code>, in <code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-reverse-probability">reverse probability</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REVERSE_005fCONDITION"><code>REVERSE_CONDITION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-REVERSIBLE_005fCC_005fMODE"><code>REVERSIBLE_CC_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-right-rotate">right rotate</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-right-shift">right shift</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rintm2-instruction-pattern"><code>rint<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RISC">RISC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RISC-1">RISC</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-roots_002c-marking">roots, marking</a>:</td><td>&nbsp;</td><td valign="top"><a href="#GGC-Roots">GGC Roots</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rotate">rotate</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rotate-1"><code>rotate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rotatert"><code>rotatert</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rotlm3-instruction-pattern"><code>rotl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-rotrm3-instruction-pattern"><code>rotr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-roundm2-instruction-pattern"><code>round<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ROUND_005fDIV_005fEXPR"><code>ROUND_DIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ROUND_005fMOD_005fEXPR"><code>ROUND_MOD_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ROUND_005fTOWARDS_005fZERO"><code>ROUND_TOWARDS_ZERO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ROUND_005fTYPE_005fALIGN"><code>ROUND_TYPE_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RSHIFT_005fEXPR"><code>RSHIFT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-addition">RTL addition</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-addition-with-signed-saturation">RTL addition with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-addition-with-unsigned-saturation">RTL addition with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-classes">RTL classes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-comparison">RTL comparison</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-comparison-operations">RTL comparison operations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-constant-expression-types">RTL constant expression types</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-constants">RTL constants</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-declarations">RTL declarations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Declarations">RTL Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-difference">RTL difference</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-expression">RTL expression</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-expressions-for-arithmetic">RTL expressions for arithmetic</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-format">RTL format</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-format-characters">RTL format characters</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-function_002dcall-insns">RTL function-call insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Calls">Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-insn-template">RTL insn template</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Template">RTL Template</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-integers">RTL integers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-memory-expressions">RTL memory expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-object-types">RTL object types</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-postdecrement">RTL postdecrement</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-postincrement">RTL postincrement</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-predecrement">RTL predecrement</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-preincrement">RTL preincrement</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Incdec">Incdec</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-register-expressions">RTL register expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-representation">RTL representation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL">RTL</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-side-effect-expressions">RTL side effect expressions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-strings">RTL strings</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-structure-sharing-assumptions">RTL structure sharing assumptions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-subtraction">RTL subtraction</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-subtraction-with-signed-saturation">RTL subtraction with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-subtraction-with-unsigned-saturation">RTL subtraction with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-sum">RTL sum</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL-vectors">RTL vectors</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL_005fCONST_005fCALL_005fP"><code>RTL_CONST_CALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL_005fCONST_005fOR_005fPURE_005fCALL_005fP"><code>RTL_CONST_OR_PURE_CALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL_005fLOOPING_005fCONST_005fOR_005fPURE_005fCALL_005fP"><code>RTL_LOOPING_CONST_OR_PURE_CALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTL_005fPURE_005fCALL_005fP"><code>RTL_PURE_CALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTX-_0028See-RTL_0029">RTX (See RTL)</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Objects">RTL Objects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTX-codes_002c-classes-of">RTX codes, classes of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Classes">RTL Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-RTX_005fFRAME_005fRELATED_005fP"><code>RTX_FRAME_RELATED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-run_002dtime-conventions">run-time conventions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-run_002dtime-target-specification">run-time target specification</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-S">S</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-s-in-constraint">&lsquo;<samp>s</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-same_005ftype_005fp"><code>same_type_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SAmode"><code>SAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-satfractmn2-instruction-pattern"><code>satfract<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-satfractunsmn2-instruction-pattern"><code>satfractuns<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-satisfies_005fconstraint_005f"><code>satisfies_constraint_</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sat_005ffract"><code>sat_fract</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SAVE_005fEXPR"><code>SAVE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-save_005fstack_005fblock-instruction-pattern"><code>save_stack_block</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-save_005fstack_005ffunction-instruction-pattern"><code>save_stack_function</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-save_005fstack_005fnonlocal-instruction-pattern"><code>save_stack_nonlocal</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SBSS_005fSECTION_005fASM_005fOP"><code>SBSS_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Scalar-evolutions">Scalar evolutions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-evolutions">Scalar evolutions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-scalars_002c-returned-as-values">scalars, returned as values</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SCHED_005fGROUP_005fP"><code>SCHED_GROUP_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SCmode"><code>SCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-scratch"><code>scratch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-scratch-operands">scratch operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-scratch_002c-RTL-sharing"><code>scratch</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-scratch_005foperand"><code>scratch_operand</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine_002dIndependent-Predicates">Machine-Independent Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDATA_005fSECTION_005fASM_005fOP"><code>SDATA_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDB_005fALLOW_005fFORWARD_005fREFERENCES"><code>SDB_ALLOW_FORWARD_REFERENCES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDB_005fALLOW_005fUNKNOWN_005fREFERENCES"><code>SDB_ALLOW_UNKNOWN_REFERENCES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDB_005fDEBUGGING_005fINFO"><code>SDB_DEBUGGING_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDB_005fDELIM"><code>SDB_DELIM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDB_005fOUTPUT_005fSOURCE_005fLINE"><code>SDB_OUTPUT_SOURCE_LINE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SDmode"><code>SDmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sdot_005fprodm-instruction-pattern"><code>sdot_prod<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-search-options">search options</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Including-Patterns">Including Patterns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fINPUT_005fRELOAD_005fCLASS"><code>SECONDARY_INPUT_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fMEMORY_005fNEEDED"><code>SECONDARY_MEMORY_NEEDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fMEMORY_005fNEEDED_005fMODE"><code>SECONDARY_MEMORY_NEEDED_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fMEMORY_005fNEEDED_005fRTX"><code>SECONDARY_MEMORY_NEEDED_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fOUTPUT_005fRELOAD_005fCLASS"><code>SECONDARY_OUTPUT_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SECONDARY_005fRELOAD_005fCLASS"><code>SECONDARY_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SELECT_005fCC_005fMODE"><code>SELECT_CC_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sequence"><code>sequence</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Sequence-iterators">Sequence iterators</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sequence-iterators">Sequence iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set"><code>set</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set-and-_002ff"><code>set</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-setmemm-instruction-pattern"><code>setmem<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SETUP_005fFRAME_005fADDRESSES"><code>SETUP_FRAME_ADDRESSES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fASM_005fOP"><code>SET_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fASM_005fOP-1"><code>SET_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set_005fattr"><code>set_attr</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set_005fattr_005falternative"><code>set_attr_alternative</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set_005fbb_005fseq"><code>set_bb_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GIMPLE-sequences">GIMPLE sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fBY_005fPIECES_005fP"><code>SET_BY_PIECES_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fDEST"><code>SET_DEST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fIS_005fRETURN_005fP"><code>SET_IS_RETURN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fLABEL_005fKIND"><code>SET_LABEL_KIND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set_005foptab_005flibfunc"><code>set_optab_libfunc</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fRATIO"><code>SET_RATIO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fSRC"><code>SET_SRC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-set_005fthread_005fpointermode-instruction-pattern"><code>set_thread_pointer<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fTYPE_005fSTRUCTURAL_005fEQUALITY"><code>SET_TYPE_STRUCTURAL_EQUALITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SET_005fTYPE_005fSTRUCTURAL_005fEQUALITY-1"><code>SET_TYPE_STRUCTURAL_EQUALITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SFmode"><code>SFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SF_005fSIZE"><code>SF_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sharing-of-RTL-components">sharing of RTL components</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-shift">shift</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHIFT_005fCOUNT_005fTRUNCATED"><code>SHIFT_COUNT_TRUNCATED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHLIB_005fSUFFIX"><code>SHLIB_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHORT_005fACCUM_005fTYPE_005fSIZE"><code>SHORT_ACCUM_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHORT_005fFRACT_005fTYPE_005fSIZE"><code>SHORT_FRACT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHORT_005fIMMEDIATES_005fSIGN_005fEXTEND"><code>SHORT_IMMEDIATES_SIGN_EXTEND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SHORT_005fTYPE_005fSIZE"><code>SHORT_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sibcall_005fepilogue-instruction-pattern"><code>sibcall_epilogue</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sibling-call">sibling call</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Edges">Edges</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SIBLING_005fCALL_005fP"><code>SIBLING_CALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-signed-division">signed division</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-signed-division-with-signed-saturation">signed division with signed saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-signed-maximum">signed maximum</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-signed-minimum">signed minimum</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sign_005fextend"><code>sign_extend</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sign_005fextract"><code>sign_extract</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Bit_002dFields">Bit-Fields</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sign_005fextract_002c-canonicalization-of"><code>sign_extract</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SIG_005fATOMIC_005fTYPE"><code>SIG_ATOMIC_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SImode"><code>SImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-simple-constraints">simple constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-simple_005freturn"><code>simple_return</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-simple_005freturn-instruction-pattern"><code>simple_return</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sincos-math-function_002c-implicit-usage">sincos math function, implicit usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sincosm3-instruction-pattern"><code>sincos<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sinm2-instruction-pattern"><code>sin<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SIZETYPE"><code>SIZETYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SIZE_005fASM_005fOP"><code>SIZE_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SIZE_005fTYPE"><code>SIZE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-skip"><code>skip</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SLOW_005fBYTE_005fACCESS"><code>SLOW_BYTE_ACCESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SLOW_005fUNALIGNED_005fACCESS"><code>SLOW_UNALIGNED_ACCESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-smax"><code>smax</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-smin"><code>smin</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sms_002c-swing_002c-software-pipelining">sms, swing, software pipelining</a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-passes">RTL passes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-smulm3_005fhighpart-instruction-pattern"><code>smul<var>m</var>3_highpart</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-soft-float-library">soft float library</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Soft-float-library-routines">Soft float library routines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-special"><code>special</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-special-predicates">special predicates</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Predicates">Predicates</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SPECS"><code>SPECS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-speed-of-instructions">speed of instructions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-splitting-instructions">splitting instructions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Splitting">Insn Splitting</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-split_005fblock"><code>split_block</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SQmode"><code>SQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sqrt"><code>sqrt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sqrtm2-instruction-pattern"><code>sqrt<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-square-root">square root</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SSA">SSA</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssaddm3-instruction-pattern"><code>ssadd<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssashlm3-instruction-pattern"><code>ssashl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SSA_005fNAME_005fDEF_005fSTMT"><code>SSA_NAME_DEF_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SSA_005fNAME_005fVERSION"><code>SSA_NAME_VERSION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssdivm3-instruction-pattern"><code>ssdiv<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssmaddmn4-instruction-pattern"><code>ssmadd<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssmsubmn4-instruction-pattern"><code>ssmsub<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssmulm3-instruction-pattern"><code>ssmul<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssnegm2-instruction-pattern"><code>ssneg<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sssubm3-instruction-pattern"><code>sssub<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ssum_005fwidenm3-instruction-pattern"><code>ssum_widen<var>m3</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fabs"><code>ss_abs</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fashift"><code>ss_ashift</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fdiv"><code>ss_div</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fminus"><code>ss_minus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fmult"><code>ss_mult</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fneg"><code>ss_neg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005fplus"><code>ss_plus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ss_005ftruncate"><code>ss_truncate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack-arguments">stack arguments</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack-frame-layout">stack frame layout</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack-smashing-protection">stack smashing protection</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Smashing-Protection">Stack Smashing Protection</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fALIGNMENT_005fNEEDED"><code>STACK_ALIGNMENT_NEEDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fBOUNDARY"><code>STACK_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fBUILTIN"><code>STACK_CHECK_BUILTIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fFIXED_005fFRAME_005fSIZE"><code>STACK_CHECK_FIXED_FRAME_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fMAX_005fFRAME_005fSIZE"><code>STACK_CHECK_MAX_FRAME_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fMAX_005fVAR_005fSIZE"><code>STACK_CHECK_MAX_VAR_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fMOVING_005fSP"><code>STACK_CHECK_MOVING_SP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fPROBE_005fINTERVAL_005fEXP"><code>STACK_CHECK_PROBE_INTERVAL_EXP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fPROTECT"><code>STACK_CHECK_PROTECT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fCHECK_005fSTATIC_005fBUILTIN"><code>STACK_CHECK_STATIC_BUILTIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Checking">Stack Checking</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fDYNAMIC_005fOFFSET"><code>STACK_DYNAMIC_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fDYNAMIC_005fOFFSET-and-virtual-registers"><code>STACK_DYNAMIC_OFFSET</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fGROWS_005fDOWNWARD"><code>STACK_GROWS_DOWNWARD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPARMS_005fIN_005fREG_005fPARM_005fAREA"><code>STACK_PARMS_IN_REG_PARM_AREA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPOINTER_005fOFFSET"><code>STACK_POINTER_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPOINTER_005fOFFSET-and-virtual-registers"><code>STACK_POINTER_OFFSET</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPOINTER_005fREGNUM"><code>STACK_POINTER_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPOINTER_005fREGNUM-and-virtual-registers"><code>STACK_POINTER_REGNUM</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack_005fpointer_005frtx"><code>stack_pointer_rtx</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack_005fprotect_005fset-instruction-pattern"><code>stack_protect_set</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stack_005fprotect_005ftest-instruction-pattern"><code>stack_protect_test</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fPUSH_005fCODE"><code>STACK_PUSH_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fREGS"><code>STACK_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Registers">Stack Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fREG_005fCOVER_005fCLASS"><code>STACK_REG_COVER_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Registers">Stack Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fSAVEAREA_005fMODE"><code>STACK_SAVEAREA_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fSIZE_005fMODE"><code>STACK_SIZE_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STACK_005fSLOT_005fALIGNMENT"><code>STACK_SLOT_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-standard-pattern-names">standard pattern names</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STANDARD_005fSTARTFILE_005fPREFIX"><code>STANDARD_STARTFILE_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STANDARD_005fSTARTFILE_005fPREFIX_005f1"><code>STANDARD_STARTFILE_PREFIX_1</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STANDARD_005fSTARTFILE_005fPREFIX_005f2"><code>STANDARD_STARTFILE_PREFIX_2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STARTFILE_005fSPEC"><code>STARTFILE_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STARTING_005fFRAME_005fOFFSET"><code>STARTING_FRAME_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STARTING_005fFRAME_005fOFFSET-and-virtual-registers"><code>STARTING_FRAME_OFFSET</code> and virtual registers</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Statement-and-operand-traversals">Statement and operand traversals</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statement-and-operand-traversals">Statement and operand traversals</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Statement-Sequences">Statement Sequences</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statement-Sequences">Statement Sequences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Statements">Statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements">Statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-statements">statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-statements-1">statements</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Static-profile-estimation">Static profile estimation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Profile-information">Profile information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-static-single-assignment">static single assignment</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STATIC_005fCHAIN_005fINCOMING_005fREGNUM"><code>STATIC_CHAIN_INCOMING_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STATIC_005fCHAIN_005fREGNUM"><code>STATIC_CHAIN_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-stdarg_002eh-and-register-arguments"><samp>stdarg.h</samp> and register arguments</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STDC_005f0_005fIN_005fSYSTEM_005fHEADERS"><code>STDC_0_IN_SYSTEM_HEADERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STMT_005fEXPR"><code>STMT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STMT_005fIS_005fFULL_005fEXPR_005fP"><code>STMT_IS_FULL_EXPR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-storage-layout">storage layout</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STORE_005fBY_005fPIECES_005fP"><code>STORE_BY_PIECES_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STORE_005fFLAG_005fVALUE"><code>STORE_FLAG_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-store_005fmultiple-instruction-pattern">&lsquo;<samp>store_multiple</samp>&rsquo; instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-strcpy"><code>strcpy</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STRICT_005fALIGNMENT"><code>STRICT_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-strict_005flow_005fpart"><code>strict_low_part</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Declarations">RTL Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-strict_005fmemory_005faddress_005fp"><code>strict_memory_address_p</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STRING_005fCST"><code>STRING_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STRING_005fPOOL_005fADDRESS_005fP"><code>STRING_POOL_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-strlenm-instruction-pattern"><code>strlen<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-structure-value-address">structure value address</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-structures_002c-returning">structures, returning</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-STRUCTURE_005fSIZE_005fBOUNDARY"><code>STRUCTURE_SIZE_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subm3-instruction-pattern"><code>sub<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBOBJECT"><code>SUBOBJECT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBOBJECT_005fCLEANUP"><code>SUBOBJECT_CLEANUP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subreg"><code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subreg-and-_002fs"><code>subreg</code> and &lsquo;<samp>/s</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subreg-and-_002fu"><code>subreg</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subreg-and-_002fu-and-_002fv"><code>subreg</code> and &lsquo;<samp>/u</samp>&rsquo; and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subreg_002c-in-strict_005flow_005fpart"><code>subreg</code>, in <code>strict_low_part</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#RTL-Declarations">RTL Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBREG_005fBYTE"><code>SUBREG_BYTE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBREG_005fPROMOTED_005fUNSIGNED_005fP"><code>SUBREG_PROMOTED_UNSIGNED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBREG_005fPROMOTED_005fUNSIGNED_005fSET"><code>SUBREG_PROMOTED_UNSIGNED_SET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBREG_005fPROMOTED_005fVAR_005fP"><code>SUBREG_PROMOTED_VAR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUBREG_005fREG"><code>SUBREG_REG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-subst-iterators-in-_002emd-files">subst iterators in <samp>.md</samp> files</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Subst-Iterators">Subst Iterators</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUCCESS_005fEXIT_005fCODE"><code>SUCCESS_EXIT_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUPPORTS_005fINIT_005fPRIORITY"><code>SUPPORTS_INIT_PRIORITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUPPORTS_005fONE_005fONLY"><code>SUPPORTS_ONE_ONLY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SUPPORTS_005fWEAK"><code>SUPPORTS_WEAK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SWITCHABLE_005fTARGET"><code>SWITCHABLE_TARGET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SWITCH_005fBODY"><code>SWITCH_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SWITCH_005fCOND"><code>SWITCH_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SWITCH_005fSTMT"><code>SWITCH_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbolic-label">symbolic label</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fANCHOR"><code>SYMBOL_FLAG_ANCHOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fEXTERNAL"><code>SYMBOL_FLAG_EXTERNAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fFUNCTION"><code>SYMBOL_FLAG_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fHAS_005fBLOCK_005fINFO"><code>SYMBOL_FLAG_HAS_BLOCK_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fLOCAL"><code>SYMBOL_FLAG_LOCAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fSMALL"><code>SYMBOL_FLAG_SMALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fFLAG_005fTLS_005fSHIFT"><code>SYMBOL_FLAG_TLS_SHIFT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref"><code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constants">Constants</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref-and-_002ff"><code>symbol_ref</code> and &lsquo;<samp>/f</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref-and-_002fi"><code>symbol_ref</code> and &lsquo;<samp>/i</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref-and-_002fu"><code>symbol_ref</code> and &lsquo;<samp>/u</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref-and-_002fv"><code>symbol_ref</code> and &lsquo;<samp>/v</samp>&rsquo;</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-symbol_005fref_002c-RTL-sharing"><code>symbol_ref</code>, RTL sharing</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fANCHOR_005fP"><code>SYMBOL_REF_ANCHOR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fBLOCK"><code>SYMBOL_REF_BLOCK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fBLOCK_005fOFFSET"><code>SYMBOL_REF_BLOCK_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fCONSTANT"><code>SYMBOL_REF_CONSTANT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fDATA"><code>SYMBOL_REF_DATA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fDECL"><code>SYMBOL_REF_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fEXTERNAL_005fP"><code>SYMBOL_REF_EXTERNAL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fFLAG"><code>SYMBOL_REF_FLAG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fFLAG_002c-in-TARGET_005fENCODE_005fSECTION_005fINFO"><code>SYMBOL_REF_FLAG</code>, in <code>TARGET_ENCODE_SECTION_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fFLAGS"><code>SYMBOL_REF_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fFUNCTION_005fP"><code>SYMBOL_REF_FUNCTION_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fHAS_005fBLOCK_005fINFO_005fP"><code>SYMBOL_REF_HAS_BLOCK_INFO_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fLOCAL_005fP"><code>SYMBOL_REF_LOCAL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fSMALL_005fP"><code>SYMBOL_REF_SMALL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fTLS_005fMODEL"><code>SYMBOL_REF_TLS_MODEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Special-Accessors">Special Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fUSED"><code>SYMBOL_REF_USED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYMBOL_005fREF_005fWEAK"><code>SYMBOL_REF_WEAK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005faddmode-instruction-pattern"><code>sync_add<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fandmode-instruction-pattern"><code>sync_and<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fcompare_005fand_005fswapmode-instruction-pattern"><code>sync_compare_and_swap<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fiormode-instruction-pattern"><code>sync_ior<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005flock_005freleasemode-instruction-pattern"><code>sync_lock_release<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005flock_005ftest_005fand_005fsetmode-instruction-pattern"><code>sync_lock_test_and_set<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnandmode-instruction-pattern"><code>sync_nand<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005faddmode-instruction-pattern"><code>sync_new_add<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005fandmode-instruction-pattern"><code>sync_new_and<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005fiormode-instruction-pattern"><code>sync_new_ior<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005fnandmode-instruction-pattern"><code>sync_new_nand<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005fsubmode-instruction-pattern"><code>sync_new_sub<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fnew_005fxormode-instruction-pattern"><code>sync_new_xor<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005faddmode-instruction-pattern"><code>sync_old_add<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005fandmode-instruction-pattern"><code>sync_old_and<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005fiormode-instruction-pattern"><code>sync_old_ior<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005fnandmode-instruction-pattern"><code>sync_old_nand<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005fsubmode-instruction-pattern"><code>sync_old_sub<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fold_005fxormode-instruction-pattern"><code>sync_old_xor<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fsubmode-instruction-pattern"><code>sync_sub<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-sync_005fxormode-instruction-pattern"><code>sync_xor<var>mode</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYSROOT_005fHEADERS_005fSUFFIX_005fSPEC"><code>SYSROOT_HEADERS_SUFFIX_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-SYSROOT_005fSUFFIX_005fSPEC"><code>SYSROOT_SUFFIX_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-T">T</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-t_002dtarget"><samp>t-<var>target</var></samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-table-jump">table jump</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Basic-Blocks">Basic Blocks</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tablejump-instruction-pattern"><code>tablejump</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tag"><code>tag</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tagging-insns">tagging insns</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tagging-Insns">Tagging Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tail-calls">tail calls</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tail-Calls">Tail Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TAmode"><code>TAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-attributes">target attributes</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-description-macros">target description macros</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Macros">Target Macros</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-functions">target functions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Structure">Target Structure</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-hooks">target hooks</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Structure">Target Structure</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-makefile-fragment">target makefile fragment</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Fragment">Target Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target-specifications">target specifications</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-targetm"><code>targetm</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Structure">Target Structure</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-targets_002c-makefile">targets, makefile</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Makefile">Makefile</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDRESS_005fCOST"><code>TARGET_ADDRESS_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fADDRESS_005fMODE"><code>TARGET_ADDR_SPACE_ADDRESS_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fCONVERT"><code>TARGET_ADDR_SPACE_CONVERT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fLEGITIMATE_005fADDRESS_005fP"><code>TARGET_ADDR_SPACE_LEGITIMATE_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fLEGITIMIZE_005fADDRESS"><code>TARGET_ADDR_SPACE_LEGITIMIZE_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fPOINTER_005fMODE"><code>TARGET_ADDR_SPACE_POINTER_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fSUBSET_005fP"><code>TARGET_ADDR_SPACE_SUBSET_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fADDR_005fSPACE_005fVALID_005fPOINTER_005fMODE"><code>TARGET_ADDR_SPACE_VALID_POINTER_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Named-Address-Spaces">Named Address Spaces</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fALIGN_005fANON_005fBITFIELD"><code>TARGET_ALIGN_ANON_BITFIELD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fALLOCATE_005fINITIAL_005fVALUE"><code>TARGET_ALLOCATE_INITIAL_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fALLOCATE_005fSTACK_005fSLOTS_005fFOR_005fARGS"><code>TARGET_ALLOCATE_STACK_SLOTS_FOR_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fALWAYS_005fSTRIP_005fDOTDOT"><code>TARGET_ALWAYS_STRIP_DOTDOT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fARG_005fPARTIAL_005fBYTES"><code>TARGET_ARG_PARTIAL_BYTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fARM_005fEABI_005fUNWINDER"><code>TARGET_ARM_EABI_UNWINDER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fARRAY_005fMODE_005fSUPPORTED_005fP"><code>TARGET_ARRAY_MODE_SUPPORTED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASAN_005fSHADOW_005fOFFSET"><code>TARGET_ASAN_SHADOW_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fALIGNED_005fDI_005fOP"><code>TARGET_ASM_ALIGNED_DI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fALIGNED_005fHI_005fOP"><code>TARGET_ASM_ALIGNED_HI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fALIGNED_005fSI_005fOP"><code>TARGET_ASM_ALIGNED_SI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fALIGNED_005fTI_005fOP"><code>TARGET_ASM_ALIGNED_TI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fASSEMBLE_005fVISIBILITY"><code>TARGET_ASM_ASSEMBLE_VISIBILITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fBYTE_005fOP"><code>TARGET_ASM_BYTE_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fCAN_005fOUTPUT_005fMI_005fTHUNK"><code>TARGET_ASM_CAN_OUTPUT_MI_THUNK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fCLOSE_005fPAREN"><code>TARGET_ASM_CLOSE_PAREN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fCODE_005fEND"><code>TARGET_ASM_CODE_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fCONSTRUCTOR"><code>TARGET_ASM_CONSTRUCTOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fDECLARE_005fCONSTANT_005fNAME"><code>TARGET_ASM_DECLARE_CONSTANT_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fDESTRUCTOR"><code>TARGET_ASM_DESTRUCTOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fEMIT_005fEXCEPT_005fPERSONALITY"><code>TARGET_ASM_EMIT_EXCEPT_PERSONALITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fEMIT_005fEXCEPT_005fTABLE_005fLABEL"><code>TARGET_ASM_EMIT_EXCEPT_TABLE_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fEMIT_005fUNWIND_005fLABEL"><code>TARGET_ASM_EMIT_UNWIND_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fEXTERNAL_005fLIBCALL"><code>TARGET_ASM_EXTERNAL_LIBCALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFILE_005fEND"><code>TARGET_ASM_FILE_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFILE_005fSTART"><code>TARGET_ASM_FILE_START</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFILE_005fSTART_005fAPP_005fOFF"><code>TARGET_ASM_FILE_START_APP_OFF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFILE_005fSTART_005fFILE_005fDIRECTIVE"><code>TARGET_ASM_FILE_START_FILE_DIRECTIVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFINAL_005fPOSTSCAN_005fINSN"><code>TARGET_ASM_FINAL_POSTSCAN_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fBEGIN_005fEPILOGUE"><code>TARGET_ASM_FUNCTION_BEGIN_EPILOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fEND_005fPROLOGUE"><code>TARGET_ASM_FUNCTION_END_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fEPILOGUE"><code>TARGET_ASM_FUNCTION_EPILOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fPROLOGUE"><code>TARGET_ASM_FUNCTION_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fRODATA_005fSECTION"><code>TARGET_ASM_FUNCTION_RODATA_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fSECTION"><code>TARGET_ASM_FUNCTION_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fFUNCTION_005fSWITCHED_005fTEXT_005fSECTIONS"><code>TARGET_ASM_FUNCTION_SWITCHED_TEXT_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fGLOBALIZE_005fDECL_005fNAME"><code>TARGET_ASM_GLOBALIZE_DECL_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fGLOBALIZE_005fLABEL"><code>TARGET_ASM_GLOBALIZE_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fINIT_005fSECTIONS"><code>TARGET_ASM_INIT_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fINTEGER"><code>TARGET_ASM_INTEGER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fINTERNAL_005fLABEL"><code>TARGET_ASM_INTERNAL_LABEL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fJUMP_005fALIGN_005fMAX_005fSKIP"><code>TARGET_ASM_JUMP_ALIGN_MAX_SKIP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fLABEL_005fALIGN_005fAFTER_005fBARRIER_005fMAX_005fSKIP"><code>TARGET_ASM_LABEL_ALIGN_AFTER_BARRIER_MAX_SKIP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fLABEL_005fALIGN_005fMAX_005fSKIP"><code>TARGET_ASM_LABEL_ALIGN_MAX_SKIP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fLOOP_005fALIGN_005fMAX_005fSKIP"><code>TARGET_ASM_LOOP_ALIGN_MAX_SKIP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Alignment-Output">Alignment Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fLTO_005fEND"><code>TARGET_ASM_LTO_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fLTO_005fSTART"><code>TARGET_ASM_LTO_START</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fMARK_005fDECL_005fPRESERVED"><code>TARGET_ASM_MARK_DECL_PRESERVED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fMERGEABLE_005fRODATA_005fPREFIX"><code>TARGET_ASM_MERGEABLE_RODATA_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fNAMED_005fSECTION"><code>TARGET_ASM_NAMED_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOPEN_005fPAREN"><code>TARGET_ASM_OPEN_PAREN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fADDR_005fCONST_005fEXTRA"><code>TARGET_ASM_OUTPUT_ADDR_CONST_EXTRA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fANCHOR"><code>TARGET_ASM_OUTPUT_ANCHOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fDWARF_005fDTPREL"><code>TARGET_ASM_OUTPUT_DWARF_DTPREL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fIDENT"><code>TARGET_ASM_OUTPUT_IDENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fMI_005fTHUNK"><code>TARGET_ASM_OUTPUT_MI_THUNK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Entry">Function Entry</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fOUTPUT_005fSOURCE_005fFILENAME"><code>TARGET_ASM_OUTPUT_SOURCE_FILENAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fRECORD_005fGCC_005fSWITCHES"><code>TARGET_ASM_RECORD_GCC_SWITCHES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fRECORD_005fGCC_005fSWITCHES_005fSECTION"><code>TARGET_ASM_RECORD_GCC_SWITCHES_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fRELOC_005fRW_005fMASK"><code>TARGET_ASM_RELOC_RW_MASK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fSELECT_005fRTX_005fSECTION"><code>TARGET_ASM_SELECT_RTX_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fSELECT_005fSECTION"><code>TARGET_ASM_SELECT_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fTM_005fCLONE_005fTABLE_005fSECTION"><code>TARGET_ASM_TM_CLONE_TABLE_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fTRAMPOLINE_005fTEMPLATE"><code>TARGET_ASM_TRAMPOLINE_TEMPLATE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fTTYPE"><code>TARGET_ASM_TTYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNALIGNED_005fDI_005fOP"><code>TARGET_ASM_UNALIGNED_DI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNALIGNED_005fHI_005fOP"><code>TARGET_ASM_UNALIGNED_HI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNALIGNED_005fSI_005fOP"><code>TARGET_ASM_UNALIGNED_SI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNALIGNED_005fTI_005fOP"><code>TARGET_ASM_UNALIGNED_TI_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Data-Output">Data Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNIQUE_005fSECTION"><code>TARGET_ASM_UNIQUE_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNWIND_005fEMIT"><code>TARGET_ASM_UNWIND_EMIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fASM_005fUNWIND_005fEMIT_005fBEFORE_005fINSN"><code>TARGET_ASM_UNWIND_EMIT_BEFORE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Dispatch-Tables">Dispatch Tables</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fATOMIC_005fTEST_005fAND_005fSET_005fTRUEVAL"><code>TARGET_ATOMIC_TEST_AND_SET_TRUEVAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fATTRIBUTE_005fTABLE"><code>TARGET_ATTRIBUTE_TABLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fATTRIBUTE_005fTAKES_005fIDENTIFIER_005fP"><code>TARGET_ATTRIBUTE_TAKES_IDENTIFIER_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBINDS_005fLOCAL_005fP"><code>TARGET_BINDS_LOCAL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBRANCH_005fTARGET_005fREGISTER_005fCALLEE_005fSAVED"><code>TARGET_BRANCH_TARGET_REGISTER_CALLEE_SAVED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBRANCH_005fTARGET_005fREGISTER_005fCLASS"><code>TARGET_BRANCH_TARGET_REGISTER_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBUILD_005fBUILTIN_005fVA_005fLIST"><code>TARGET_BUILD_BUILTIN_VA_LIST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBUILTIN_005fDECL"><code>TARGET_BUILTIN_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBUILTIN_005fRECIPROCAL"><code>TARGET_BUILTIN_RECIPROCAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fBUILTIN_005fSETJMP_005fFRAME_005fVALUE"><code>TARGET_BUILTIN_SETJMP_FRAME_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fC99_005fFUNCTIONS"><code>TARGET_C99_FUNCTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCALLEE_005fCOPIES"><code>TARGET_CALLEE_COPIES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCANNOT_005fFORCE_005fCONST_005fMEM"><code>TARGET_CANNOT_FORCE_CONST_MEM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCANNOT_005fMODIFY_005fJUMPS_005fP"><code>TARGET_CANNOT_MODIFY_JUMPS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCANONICALIZE_005fCOMPARISON"><code>TARGET_CANONICALIZE_COMPARISON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCANONICAL_005fVA_005fLIST_005fTYPE"><code>TARGET_CANONICAL_VA_LIST_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCAN_005fELIMINATE"><code>TARGET_CAN_ELIMINATE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCAN_005fFOLLOW_005fJUMP"><code>TARGET_CAN_FOLLOW_JUMP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCAN_005fINLINE_005fP"><code>TARGET_CAN_INLINE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCASE_005fVALUES_005fTHRESHOLD"><code>TARGET_CASE_VALUES_THRESHOLD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCC_005fMODES_005fCOMPATIBLE"><code>TARGET_CC_MODES_COMPATIBLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCHECK_005fPCH_005fTARGET_005fFLAGS"><code>TARGET_CHECK_PCH_TARGET_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PCH-Target">PCH Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCHECK_005fSTRING_005fOBJECT_005fFORMAT_005fARG"><code>TARGET_CHECK_STRING_OBJECT_FORMAT_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCLASS_005fLIKELY_005fSPILLED_005fP"><code>TARGET_CLASS_LIKELY_SPILLED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCLASS_005fMAX_005fNREGS"><code>TARGET_CLASS_MAX_NREGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCOMMUTATIVE_005fP"><code>TARGET_COMMUTATIVE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCOMPARE_005fVERSION_005fPRIORITY"><code>TARGET_COMPARE_VERSION_PRIORITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCOMP_005fTYPE_005fATTRIBUTES"><code>TARGET_COMP_TYPE_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCONDITIONAL_005fREGISTER_005fUSAGE"><code>TARGET_CONDITIONAL_REGISTER_USAGE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Basics">Register Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCONST_005fANCHOR"><code>TARGET_CONST_ANCHOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCONST_005fNOT_005fOK_005fFOR_005fDEBUG_005fP"><code>TARGET_CONST_NOT_OK_FOR_DEBUG_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCONVERT_005fTO_005fTYPE"><code>TARGET_CONVERT_TO_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCPU_005fCPP_005fBUILTINS"><code>TARGET_CPU_CPP_BUILTINS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fADJUST_005fCLASS_005fAT_005fDEFINITION"><code>TARGET_CXX_ADJUST_CLASS_AT_DEFINITION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fCDTOR_005fRETURNS_005fTHIS"><code>TARGET_CXX_CDTOR_RETURNS_THIS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fCLASS_005fDATA_005fALWAYS_005fCOMDAT"><code>TARGET_CXX_CLASS_DATA_ALWAYS_COMDAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fCOOKIE_005fHAS_005fSIZE"><code>TARGET_CXX_COOKIE_HAS_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fDECL_005fMANGLING_005fCONTEXT"><code>TARGET_CXX_DECL_MANGLING_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fDETERMINE_005fCLASS_005fDATA_005fVISIBILITY"><code>TARGET_CXX_DETERMINE_CLASS_DATA_VISIBILITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fGET_005fCOOKIE_005fSIZE"><code>TARGET_CXX_GET_COOKIE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fGUARD_005fMASK_005fBIT"><code>TARGET_CXX_GUARD_MASK_BIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fGUARD_005fTYPE"><code>TARGET_CXX_GUARD_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fIMPORT_005fEXPORT_005fCLASS"><code>TARGET_CXX_IMPORT_EXPORT_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fKEY_005fMETHOD_005fMAY_005fBE_005fINLINE"><code>TARGET_CXX_KEY_METHOD_MAY_BE_INLINE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fLIBRARY_005fRTTI_005fCOMDAT"><code>TARGET_CXX_LIBRARY_RTTI_COMDAT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fUSE_005fAEABI_005fATEXIT"><code>TARGET_CXX_USE_AEABI_ATEXIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fCXX_005fUSE_005fATEXIT_005fFOR_005fCXA_005fATEXIT"><code>TARGET_CXX_USE_ATEXIT_FOR_CXA_ATEXIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#C_002b_002b-ABI">C++ ABI</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fC_005fPREINCLUDE"><code>TARGET_C_PREINCLUDE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDEBUG_005fUNWIND_005fINFO"><code>TARGET_DEBUG_UNWIND_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDECIMAL_005fFLOAT_005fSUPPORTED_005fP"><code>TARGET_DECIMAL_FLOAT_SUPPORTED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDECLSPEC"><code>TARGET_DECLSPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDEFAULT_005fPACK_005fSTRUCT"><code>TARGET_DEFAULT_PACK_STRUCT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDEFAULT_005fSHORT_005fENUMS"><code>TARGET_DEFAULT_SHORT_ENUMS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDEFAULT_005fTARGET_005fFLAGS"><code>TARGET_DEFAULT_TARGET_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDEFERRED_005fOUTPUT_005fDEFS"><code>TARGET_DEFERRED_OUTPUT_DEFS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDELAY_005fSCHED2"><code>TARGET_DELAY_SCHED2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDELAY_005fVARTRACK"><code>TARGET_DELAY_VARTRACK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDELEGITIMIZE_005fADDRESS"><code>TARGET_DELEGITIMIZE_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDIFFERENT_005fADDR_005fDISPLACEMENT_005fP"><code>TARGET_DIFFERENT_ADDR_DISPLACEMENT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDLLIMPORT_005fDECL_005fATTRIBUTES"><code>TARGET_DLLIMPORT_DECL_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDWARF_005fCALLING_005fCONVENTION"><code>TARGET_DWARF_CALLING_CONVENTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDWARF_005fHANDLE_005fFRAME_005fUNSPEC"><code>TARGET_DWARF_HANDLE_FRAME_UNSPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Layout">Frame Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fDWARF_005fREGISTER_005fSPAN"><code>TARGET_DWARF_REGISTER_SPAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEDOM"><code>TARGET_EDOM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fDEBUG_005fFORM_005fTLS_005fADDRESS"><code>TARGET_EMUTLS_DEBUG_FORM_TLS_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fGET_005fADDRESS"><code>TARGET_EMUTLS_GET_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fREGISTER_005fCOMMON"><code>TARGET_EMUTLS_REGISTER_COMMON</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fTMPL_005fPREFIX"><code>TARGET_EMUTLS_TMPL_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fTMPL_005fSECTION"><code>TARGET_EMUTLS_TMPL_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fVAR_005fALIGN_005fFIXED"><code>TARGET_EMUTLS_VAR_ALIGN_FIXED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fVAR_005fFIELDS"><code>TARGET_EMUTLS_VAR_FIELDS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fVAR_005fINIT"><code>TARGET_EMUTLS_VAR_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fVAR_005fPREFIX"><code>TARGET_EMUTLS_VAR_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEMUTLS_005fVAR_005fSECTION"><code>TARGET_EMUTLS_VAR_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Emulated-TLS">Emulated TLS</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fENCODE_005fSECTION_005fINFO"><code>TARGET_ENCODE_SECTION_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fENCODE_005fSECTION_005fINFO-and-address-validation"><code>TARGET_ENCODE_SECTION_INFO</code> and address validation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fENCODE_005fSECTION_005fINFO-usage"><code>TARGET_ENCODE_SECTION_INFO</code> usage</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fENUM_005fVA_005fLIST_005fP"><code>TARGET_ENUM_VA_LIST_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXCEPT_005fUNWIND_005fINFO"><code>TARGET_EXCEPT_UNWIND_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXECUTABLE_005fSUFFIX"><code>TARGET_EXECUTABLE_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXPAND_005fBUILTIN"><code>TARGET_EXPAND_BUILTIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXPAND_005fBUILTIN_005fSAVEREGS"><code>TARGET_EXPAND_BUILTIN_SAVEREGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXPAND_005fTO_005fRTL_005fHOOK"><code>TARGET_EXPAND_TO_RTL_HOOK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXPR"><code>TARGET_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXTRA_005fINCLUDES"><code>TARGET_EXTRA_INCLUDES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXTRA_005fLIVE_005fON_005fENTRY"><code>TARGET_EXTRA_LIVE_ON_ENTRY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tail-Calls">Tail Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fEXTRA_005fPRE_005fINCLUDES"><code>TARGET_EXTRA_PRE_INCLUDES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFIXED_005fCONDITION_005fCODE_005fREGS"><code>TARGET_FIXED_CONDITION_CODE_REGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFIXED_005fPOINT_005fSUPPORTED_005fP"><code>TARGET_FIXED_POINT_SUPPORTED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-target_005fflags"><code>target_flags</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFLAGS_005fREGNUM"><code>TARGET_FLAGS_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#MODE_005fCC-Condition-Codes">MODE_CC Condition Codes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFLT_005fEVAL_005fMETHOD"><code>TARGET_FLT_EVAL_METHOD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFN_005fABI_005fVA_005fLIST"><code>TARGET_FN_ABI_VA_LIST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFOLD_005fBUILTIN"><code>TARGET_FOLD_BUILTIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFORCE_005fAT_005fCOMP_005fDIR"><code>TARGET_FORCE_AT_COMP_DIR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFORMAT_005fTYPES"><code>TARGET_FORMAT_TYPES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFRAME_005fPOINTER_005fREQUIRED"><code>TARGET_FRAME_POINTER_REQUIRED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Elimination">Elimination</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fARG"><code>TARGET_FUNCTION_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fARG_005fADVANCE"><code>TARGET_FUNCTION_ARG_ADVANCE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fARG_005fBOUNDARY"><code>TARGET_FUNCTION_ARG_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fARG_005fROUND_005fBOUNDARY"><code>TARGET_FUNCTION_ARG_ROUND_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fATTRIBUTE_005fINLINABLE_005fP"><code>TARGET_FUNCTION_ATTRIBUTE_INLINABLE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fINCOMING_005fARG"><code>TARGET_FUNCTION_INCOMING_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fOK_005fFOR_005fSIBCALL"><code>TARGET_FUNCTION_OK_FOR_SIBCALL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tail-Calls">Tail Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fVALUE"><code>TARGET_FUNCTION_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fFUNCTION_005fVALUE_005fREGNO_005fP"><code>TARGET_FUNCTION_VALUE_REGNO_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGENERATE_005fVERSION_005fDISPATCHER_005fBODY"><code>TARGET_GENERATE_VERSION_DISPATCHER_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGET_005fDRAP_005fRTX"><code>TARGET_GET_DRAP_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGET_005fFUNCTION_005fVERSIONS_005fDISPATCHER"><code>TARGET_GET_FUNCTION_VERSIONS_DISPATCHER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGET_005fPCH_005fVALIDITY"><code>TARGET_GET_PCH_VALIDITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PCH-Target">PCH Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGET_005fRAW_005fARG_005fMODE"><code>TARGET_GET_RAW_ARG_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGET_005fRAW_005fRESULT_005fMODE"><code>TARGET_GET_RAW_RESULT_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fGIMPLIFY_005fVA_005fARG_005fEXPR"><code>TARGET_GIMPLIFY_VA_ARG_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHANDLE_005fC_005fOPTION"><code>TARGET_HANDLE_C_OPTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHANDLE_005fOPTION"><code>TARGET_HANDLE_OPTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHARD_005fREGNO_005fSCRATCH_005fOK"><code>TARGET_HARD_REGNO_SCRATCH_OK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Values-in-Registers">Values in Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAS_005fSINCOS"><code>TARGET_HAS_SINCOS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fCONDITIONAL_005fEXECUTION"><code>TARGET_HAVE_CONDITIONAL_EXECUTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fCTORS_005fDTORS"><code>TARGET_HAVE_CTORS_DTORS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-for-Initialization">Macros for Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fNAMED_005fSECTIONS"><code>TARGET_HAVE_NAMED_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fSRODATA_005fSECTION"><code>TARGET_HAVE_SRODATA_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fSWITCHABLE_005fBSS_005fSECTIONS"><code>TARGET_HAVE_SWITCHABLE_BSS_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fHAVE_005fTLS"><code>TARGET_HAVE_TLS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINIT_005fBUILTINS"><code>TARGET_INIT_BUILTINS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINIT_005fDWARF_005fREG_005fSIZES_005fEXTRA"><code>TARGET_INIT_DWARF_REG_SIZES_EXTRA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINIT_005fLIBFUNCS"><code>TARGET_INIT_LIBFUNCS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINSERT_005fATTRIBUTES"><code>TARGET_INSERT_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINSTANTIATE_005fDECLS"><code>TARGET_INSTANTIATE_DECLS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fARG_005fFOR_005fUNPROTOTYPED_005fFN"><code>TARGET_INVALID_ARG_FOR_UNPROTOTYPED_FN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fBINARY_005fOP"><code>TARGET_INVALID_BINARY_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fCONVERSION"><code>TARGET_INVALID_CONVERSION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fPARAMETER_005fTYPE"><code>TARGET_INVALID_PARAMETER_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fRETURN_005fTYPE"><code>TARGET_INVALID_RETURN_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fUNARY_005fOP"><code>TARGET_INVALID_UNARY_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fINVALID_005fWITHIN_005fDOLOOP"><code>TARGET_INVALID_WITHIN_DOLOOP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fIN_005fSMALL_005fDATA_005fP"><code>TARGET_IN_SMALL_DATA_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLEGITIMATE_005fADDRESS_005fP"><code>TARGET_LEGITIMATE_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLEGITIMATE_005fCOMBINED_005fINSN"><code>TARGET_LEGITIMATE_COMBINED_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLEGITIMATE_005fCONSTANT_005fP"><code>TARGET_LEGITIMATE_CONSTANT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLEGITIMIZE_005fADDRESS"><code>TARGET_LEGITIMIZE_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIBCALL_005fVALUE"><code>TARGET_LIBCALL_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIBFUNC_005fGNU_005fPREFIX"><code>TARGET_LIBFUNC_GNU_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIBGCC_005fCMP_005fRETURN_005fMODE"><code>TARGET_LIBGCC_CMP_RETURN_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIBGCC_005fSDATA_005fSECTION"><code>TARGET_LIBGCC_SDATA_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIBGCC_005fSHIFT_005fCOUNT_005fMODE"><code>TARGET_LIBGCC_SHIFT_COUNT_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLIB_005fINT_005fCMP_005fBIASED"><code>TARGET_LIB_INT_CMP_BIASED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Library-Calls">Library Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLOOP_005fUNROLL_005fADJUST"><code>TARGET_LOOP_UNROLL_ADJUST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fLRA_005fP"><code>TARGET_LRA_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMACHINE_005fDEPENDENT_005fREORG"><code>TARGET_MACHINE_DEPENDENT_REORG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMANGLE_005fASSEMBLER_005fNAME"><code>TARGET_MANGLE_ASSEMBLER_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMANGLE_005fDECL_005fASSEMBLER_005fNAME"><code>TARGET_MANGLE_DECL_ASSEMBLER_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMANGLE_005fTYPE"><code>TARGET_MANGLE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMAX_005fANCHOR_005fOFFSET"><code>TARGET_MAX_ANCHOR_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMD_005fASM_005fCLOBBERS"><code>TARGET_MD_ASM_CLOBBERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMEMBER_005fTYPE_005fFORCES_005fBLK"><code>TARGET_MEMBER_TYPE_FORCES_BLK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMEMMODEL_005fCHECK"><code>TARGET_MEMMODEL_CHECK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMEMORY_005fMOVE_005fCOST"><code>TARGET_MEMORY_MOVE_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMEM_005fCONSTRAINT"><code>TARGET_MEM_CONSTRAINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMEM_005fREF"><code>TARGET_MEM_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-References">Storage References</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMERGE_005fDECL_005fATTRIBUTES"><code>TARGET_MERGE_DECL_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMERGE_005fTYPE_005fATTRIBUTES"><code>TARGET_MERGE_TYPE_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMIN_005fANCHOR_005fOFFSET"><code>TARGET_MIN_ANCHOR_OFFSET</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMIN_005fDIVISIONS_005fFOR_005fRECIP_005fMUL"><code>TARGET_MIN_DIVISIONS_FOR_RECIP_MUL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMODE_005fDEPENDENT_005fADDRESS_005fP"><code>TARGET_MODE_DEPENDENT_ADDRESS_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMODE_005fREP_005fEXTENDED"><code>TARGET_MODE_REP_EXTENDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMS_005fBITFIELD_005fLAYOUT_005fP"><code>TARGET_MS_BITFIELD_LAYOUT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMUST_005fPASS_005fIN_005fSTACK"><code>TARGET_MUST_PASS_IN_STACK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fMUST_005fPASS_005fIN_005fSTACK_002c-and-TARGET_005fFUNCTION_005fARG"><code>TARGET_MUST_PASS_IN_STACK</code>, and <code>TARGET_FUNCTION_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fNARROW_005fVOLATILE_005fBITFIELD"><code>TARGET_NARROW_VOLATILE_BITFIELD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fN_005fFORMAT_005fTYPES"><code>TARGET_N_FORMAT_TYPES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOBJC_005fCONSTRUCT_005fSTRING_005fOBJECT"><code>TARGET_OBJC_CONSTRUCT_STRING_OBJECT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOBJC_005fDECLARE_005fCLASS_005fDEFINITION"><code>TARGET_OBJC_DECLARE_CLASS_DEFINITION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOBJC_005fDECLARE_005fUNRESOLVED_005fCLASS_005fREFERENCE"><code>TARGET_OBJC_DECLARE_UNRESOLVED_CLASS_REFERENCE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOBJECT_005fSUFFIX"><code>TARGET_OBJECT_SUFFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOBJFMT_005fCPP_005fBUILTINS"><code>TARGET_OBJFMT_CPP_BUILTINS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTF"><code>TARGET_OPTF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fDEFAULT_005fPARAMS"><code>TARGET_OPTION_DEFAULT_PARAMS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fFUNCTION_005fVERSIONS"><code>TARGET_OPTION_FUNCTION_VERSIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fINIT_005fSTRUCT"><code>TARGET_OPTION_INIT_STRUCT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fOPTIMIZATION_005fTABLE"><code>TARGET_OPTION_OPTIMIZATION_TABLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fOVERRIDE"><code>TARGET_OPTION_OVERRIDE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fPRAGMA_005fPARSE"><code>TARGET_OPTION_PRAGMA_PARSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fPRINT"><code>TARGET_OPTION_PRINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fRESTORE"><code>TARGET_OPTION_RESTORE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fSAVE"><code>TARGET_OPTION_SAVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOPTION_005fVALID_005fATTRIBUTE_005fP"><code>TARGET_OPTION_VALID_ATTRIBUTE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOS_005fCPP_005fBUILTINS"><code>TARGET_OS_CPP_BUILTINS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOVERRIDES_005fFORMAT_005fATTRIBUTES"><code>TARGET_OVERRIDES_FORMAT_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOVERRIDES_005fFORMAT_005fATTRIBUTES_005fCOUNT"><code>TARGET_OVERRIDES_FORMAT_ATTRIBUTES_COUNT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOVERRIDES_005fFORMAT_005fINIT"><code>TARGET_OVERRIDES_FORMAT_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fOVERRIDE_005fOPTIONS_005fAFTER_005fCHANGE"><code>TARGET_OVERRIDE_OPTIONS_AFTER_CHANGE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPASS_005fBY_005fREFERENCE"><code>TARGET_PASS_BY_REFERENCE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPCH_005fVALID_005fP"><code>TARGET_PCH_VALID_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PCH-Target">PCH Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPOSIX_005fIO"><code>TARGET_POSIX_IO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPREFERRED_005fOUTPUT_005fRELOAD_005fCLASS"><code>TARGET_PREFERRED_OUTPUT_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPREFERRED_005fRELOAD_005fCLASS"><code>TARGET_PREFERRED_RELOAD_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPREFERRED_005fRENAME_005fCLASS"><code>TARGET_PREFERRED_RENAME_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPREPARE_005fPCH_005fSAVE"><code>TARGET_PREPARE_PCH_SAVE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#PCH-Target">PCH Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPRETEND_005fOUTGOING_005fVARARGS_005fNAMED"><code>TARGET_PRETEND_OUTGOING_VARARGS_NAMED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPROFILE_005fBEFORE_005fPROLOGUE"><code>TARGET_PROFILE_BEFORE_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPROMOTED_005fTYPE"><code>TARGET_PROMOTED_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPROMOTE_005fFUNCTION_005fMODE"><code>TARGET_PROMOTE_FUNCTION_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPROMOTE_005fPROTOTYPES"><code>TARGET_PROMOTE_PROTOTYPES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fPTRMEMFUNC_005fVBIT_005fLOCATION"><code>TARGET_PTRMEMFUNC_VBIT_LOCATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fREF_005fMAY_005fALIAS_005fERRNO"><code>TARGET_REF_MAY_ALIAS_ERRNO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fREGISTER_005fMOVE_005fCOST"><code>TARGET_REGISTER_MOVE_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fREGISTER_005fPRIORITY"><code>TARGET_REGISTER_PRIORITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRELAXED_005fORDERING"><code>TARGET_RELAXED_ORDERING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRESOLVE_005fOVERLOADED_005fBUILTIN"><code>TARGET_RESOLVE_OVERLOADED_BUILTIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRETURN_005fIN_005fMEMORY"><code>TARGET_RETURN_IN_MEMORY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRETURN_005fIN_005fMSB"><code>TARGET_RETURN_IN_MSB</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRETURN_005fPOPS_005fARGS"><code>TARGET_RETURN_POPS_ARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Arguments">Stack Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fRTX_005fCOSTS"><code>TARGET_RTX_COSTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCALAR_005fMODE_005fSUPPORTED_005fP"><code>TARGET_SCALAR_MODE_SUPPORTED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fADJUST_005fCOST"><code>TARGET_SCHED_ADJUST_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fADJUST_005fPRIORITY"><code>TARGET_SCHED_ADJUST_PRIORITY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fALLOC_005fSCHED_005fCONTEXT"><code>TARGET_SCHED_ALLOC_SCHED_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fCLEAR_005fSCHED_005fCONTEXT"><code>TARGET_SCHED_CLEAR_SCHED_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDEPENDENCIES_005fEVALUATION_005fHOOK"><code>TARGET_SCHED_DEPENDENCIES_EVALUATION_HOOK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDFA_005fNEW_005fCYCLE"><code>TARGET_SCHED_DFA_NEW_CYCLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDFA_005fPOST_005fADVANCE_005fCYCLE"><code>TARGET_SCHED_DFA_POST_ADVANCE_CYCLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDFA_005fPOST_005fCYCLE_005fINSN"><code>TARGET_SCHED_DFA_POST_CYCLE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDFA_005fPRE_005fADVANCE_005fCYCLE"><code>TARGET_SCHED_DFA_PRE_ADVANCE_CYCLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDFA_005fPRE_005fCYCLE_005fINSN"><code>TARGET_SCHED_DFA_PRE_CYCLE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDISPATCH"><code>TARGET_SCHED_DISPATCH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fDISPATCH_005fDO"><code>TARGET_SCHED_DISPATCH_DO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fEXPOSED_005fPIPELINE"><code>TARGET_SCHED_EXPOSED_PIPELINE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFINISH"><code>TARGET_SCHED_FINISH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFINISH_005fGLOBAL"><code>TARGET_SCHED_FINISH_GLOBAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fBACKTRACK"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_BACKTRACK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fBEGIN"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_BEGIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD_005fGUARD"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD_GUARD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fDFA_005fLOOKAHEAD_005fGUARD_005fSPEC"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD_GUARD_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fEND"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_END</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fFINI"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_FINI</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fINIT"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFIRST_005fCYCLE_005fMULTIPASS_005fISSUE"><code>TARGET_SCHED_FIRST_CYCLE_MULTIPASS_ISSUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fFREE_005fSCHED_005fCONTEXT"><code>TARGET_SCHED_FREE_SCHED_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fGEN_005fSPEC_005fCHECK"><code>TARGET_SCHED_GEN_SPEC_CHECK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fH_005fI_005fD_005fEXTENDED"><code>TARGET_SCHED_H_I_D_EXTENDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fINIT"><code>TARGET_SCHED_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fINIT_005fDFA_005fPOST_005fCYCLE_005fINSN"><code>TARGET_SCHED_INIT_DFA_POST_CYCLE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fINIT_005fDFA_005fPRE_005fCYCLE_005fINSN"><code>TARGET_SCHED_INIT_DFA_PRE_CYCLE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fINIT_005fGLOBAL"><code>TARGET_SCHED_INIT_GLOBAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fINIT_005fSCHED_005fCONTEXT"><code>TARGET_SCHED_INIT_SCHED_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fISSUE_005fRATE"><code>TARGET_SCHED_ISSUE_RATE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fIS_005fCOSTLY_005fDEPENDENCE"><code>TARGET_SCHED_IS_COSTLY_DEPENDENCE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fNEEDS_005fBLOCK_005fP"><code>TARGET_SCHED_NEEDS_BLOCK_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fREASSOCIATION_005fWIDTH"><code>TARGET_SCHED_REASSOCIATION_WIDTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fREORDER"><code>TARGET_SCHED_REORDER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fREORDER2"><code>TARGET_SCHED_REORDER2</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fSET_005fSCHED_005fCONTEXT"><code>TARGET_SCHED_SET_SCHED_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fSET_005fSCHED_005fFLAGS"><code>TARGET_SCHED_SET_SCHED_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fSMS_005fRES_005fMII"><code>TARGET_SCHED_SMS_RES_MII</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fSPECULATE_005fINSN"><code>TARGET_SCHED_SPECULATE_INSN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSCHED_005fVARIABLE_005fISSUE"><code>TARGET_SCHED_VARIABLE_ISSUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scheduling">Scheduling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSECONDARY_005fRELOAD"><code>TARGET_SECONDARY_RELOAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSECTION_005fTYPE_005fFLAGS"><code>TARGET_SECTION_TYPE_FLAGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#File-Framework">File Framework</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSETUP_005fINCOMING_005fVARARGS"><code>TARGET_SETUP_INCOMING_VARARGS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSET_005fCURRENT_005fFUNCTION"><code>TARGET_SET_CURRENT_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSET_005fDEFAULT_005fTYPE_005fATTRIBUTES"><code>TARGET_SET_DEFAULT_TYPE_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSET_005fUP_005fBY_005fPROLOGUE"><code>TARGET_SET_UP_BY_PROLOGUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tail-Calls">Tail Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSHIFT_005fTRUNCATION_005fMASK"><code>TARGET_SHIFT_TRUNCATION_MASK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSMALL_005fREGISTER_005fCLASSES_005fFOR_005fMODE_005fP"><code>TARGET_SMALL_REGISTER_CLASSES_FOR_MODE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSPILL_005fCLASS"><code>TARGET_SPILL_CLASS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Classes">Register Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSPLIT_005fCOMPLEX_005fARG"><code>TARGET_SPLIT_COMPLEX_ARG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTACK_005fPROTECT_005fFAIL"><code>TARGET_STACK_PROTECT_FAIL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Smashing-Protection">Stack Smashing Protection</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTACK_005fPROTECT_005fGUARD"><code>TARGET_STACK_PROTECT_GUARD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Smashing-Protection">Stack Smashing Protection</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTATIC_005fCHAIN"><code>TARGET_STATIC_CHAIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Frame-Registers">Frame Registers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTRICT_005fARGUMENT_005fNAMING"><code>TARGET_STRICT_ARGUMENT_NAMING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTRING_005fOBJECT_005fREF_005fTYPE_005fP"><code>TARGET_STRING_OBJECT_REF_TYPE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Run_002dtime-Target">Run-time Target</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTRIP_005fNAME_005fENCODING"><code>TARGET_STRIP_NAME_ENCODING</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSTRUCT_005fVALUE_005fRTX"><code>TARGET_STRUCT_VALUE_RTX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Aggregate-Return">Aggregate Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSUPPORTS_005fSPLIT_005fSTACK"><code>TARGET_SUPPORTS_SPLIT_STACK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Stack-Smashing-Protection">Stack Smashing Protection</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fSUPPORTS_005fWEAK"><code>TARGET_SUPPORTS_WEAK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fTERMINATE_005fDW2_005fEH_005fFRAME_005fINFO"><code>TARGET_TERMINATE_DW2_EH_FRAME_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fTRAMPOLINE_005fADJUST_005fADDRESS"><code>TARGET_TRAMPOLINE_ADJUST_ADDRESS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fTRAMPOLINE_005fINIT"><code>TARGET_TRAMPOLINE_INIT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUNSPEC_005fMAY_005fTRAP_005fP"><code>TARGET_UNSPEC_MAY_TRAP_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUNWIND_005fTABLES_005fDEFAULT"><code>TARGET_UNWIND_TABLES_DEFAULT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Region-Output">Exception Region Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUNWIND_005fWORD_005fMODE"><code>TARGET_UNWIND_WORD_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUPDATE_005fSTACK_005fBOUNDARY"><code>TARGET_UPDATE_STACK_BOUNDARY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUSES_005fWEAK_005fUNWIND_005fINFO"><code>TARGET_USES_WEAK_UNWIND_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Exception-Handling">Exception Handling</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUSE_005fANCHORS_005fFOR_005fSYMBOL_005fP"><code>TARGET_USE_ANCHORS_FOR_SYMBOL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Anchored-Addresses">Anchored Addresses</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUSE_005fBLOCKS_005fFOR_005fCONSTANT_005fP"><code>TARGET_USE_BLOCKS_FOR_CONSTANT_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUSE_005fBLOCKS_005fFOR_005fDECL_005fP"><code>TARGET_USE_BLOCKS_FOR_DECL_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fUSE_005fJCR_005fSECTION"><code>TARGET_USE_JCR_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVALID_005fDLLIMPORT_005fATTRIBUTE_005fP"><code>TARGET_VALID_DLLIMPORT_ATTRIBUTE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Attributes">Target Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVALID_005fPOINTER_005fMODE"><code>TARGET_VALID_POINTER_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fADD_005fSTMT_005fCOST"><code>TARGET_VECTORIZE_ADD_STMT_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fAUTOVECTORIZE_005fVECTOR_005fSIZES"><code>TARGET_VECTORIZE_AUTOVECTORIZE_VECTOR_SIZES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fCONVERSION"><code>TARGET_VECTORIZE_BUILTIN_CONVERSION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fGATHER"><code>TARGET_VECTORIZE_BUILTIN_GATHER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fMASK_005fFOR_005fLOAD"><code>TARGET_VECTORIZE_BUILTIN_MASK_FOR_LOAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fTM_005fLOAD"><code>TARGET_VECTORIZE_BUILTIN_TM_LOAD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fTM_005fSTORE"><code>TARGET_VECTORIZE_BUILTIN_TM_STORE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fVECTORIZATION_005fCOST"><code>TARGET_VECTORIZE_BUILTIN_VECTORIZATION_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fBUILTIN_005fVECTORIZED_005fFUNCTION"><code>TARGET_VECTORIZE_BUILTIN_VECTORIZED_FUNCTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fDESTROY_005fCOST_005fDATA"><code>TARGET_VECTORIZE_DESTROY_COST_DATA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fFINISH_005fCOST"><code>TARGET_VECTORIZE_FINISH_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fINIT_005fCOST"><code>TARGET_VECTORIZE_INIT_COST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fPREFERRED_005fSIMD_005fMODE"><code>TARGET_VECTORIZE_PREFERRED_SIMD_MODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fSUPPORT_005fVECTOR_005fMISALIGNMENT"><code>TARGET_VECTORIZE_SUPPORT_VECTOR_MISALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fVECTOR_005fALIGNMENT_005fREACHABLE"><code>TARGET_VECTORIZE_VECTOR_ALIGNMENT_REACHABLE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTORIZE_005fVEC_005fPERM_005fCONST_005fOK"><code>TARGET_VECTORIZE_VEC_PERM_CONST_OK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Addressing-Modes">Addressing Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTOR_005fALIGNMENT"><code>TARGET_VECTOR_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVECTOR_005fMODE_005fSUPPORTED_005fP"><code>TARGET_VECTOR_MODE_SUPPORTED_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Register-Arguments">Register Arguments</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVTABLE_005fDATA_005fENTRY_005fDISTANCE"><code>TARGET_VTABLE_DATA_ENTRY_DISTANCE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVTABLE_005fENTRY_005fALIGN"><code>TARGET_VTABLE_ENTRY_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fVTABLE_005fUSES_005fDESCRIPTORS"><code>TARGET_VTABLE_USES_DESCRIPTORS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fWANT_005fDEBUG_005fPUB_005fSECTIONS"><code>TARGET_WANT_DEBUG_PUB_SECTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SDB-and-DWARF">SDB and DWARF</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fWARN_005fFUNC_005fRETURN"><code>TARGET_WARN_FUNC_RETURN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tail-Calls">Tail Calls</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TARGET_005fWEAK_005fNOT_005fIN_005fARCHIVE_005fTOC"><code>TARGET_WEAK_NOT_IN_ARCHIVE_TOC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TCmode"><code>TCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TDmode"><code>TDmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TEMPLATE_005fDECL"><code>TEMPLATE_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Temporaries">Temporaries</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Temporaries">Temporaries</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-termination-routines">termination routines</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Initialization">Initialization</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-testing-constraints">testing constraints</a>:</td><td>&nbsp;</td><td valign="top"><a href="#C-Constraint-Interface">C Constraint Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TEXT_005fSECTION_005fASM_005fOP"><code>TEXT_SECTION_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TFmode"><code>TFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TF_005fSIZE"><code>TF_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-THEN_005fCLAUSE"><code>THEN_CLAUSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-THREAD_005fMODEL_005fSPEC"><code>THREAD_MODEL_SPEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-THROW_005fEXPR"><code>THROW_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-THUNK_005fDECL"><code>THUNK_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-THUNK_005fDELTA"><code>THUNK_DELTA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TImode"><code>TImode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TImode_002c-in-insn"><code>TImode</code>, in <code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insns">Insns</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TLS_005fCOMMON_005fASM_005fOP"><code>TLS_COMMON_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TLS_005fSECTION_005fASM_005fFLAG"><code>TLS_SECTION_ASM_FLAG</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tm_002eh-macros"><samp>tm.h</samp> macros</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Target-Macros">Target Macros</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TQFmode"><code>TQFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TQmode"><code>TQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-trampolines-for-nested-functions">trampolines for nested functions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRAMPOLINE_005fALIGNMENT"><code>TRAMPOLINE_ALIGNMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRAMPOLINE_005fSECTION"><code>TRAMPOLINE_SECTION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRAMPOLINE_005fSIZE"><code>TRAMPOLINE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRANSFER_005fFROM_005fTRAMPOLINE"><code>TRANSFER_FROM_TRAMPOLINE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Trampolines">Trampolines</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-trap-instruction-pattern"><code>trap</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tree">tree</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tree-overview">Tree overview</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tree-1">tree</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Tree-SSA">Tree SSA</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tree-SSA">Tree SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fCHAIN"><code>TREE_CHAIN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fCODE"><code>TREE_CODE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tree-overview">Tree overview</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tree_005fint_005fcst_005fequal"><code>tree_int_cst_equal</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fINT_005fCST_005fHIGH"><code>TREE_INT_CST_HIGH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fINT_005fCST_005fLOW"><code>TREE_INT_CST_LOW</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tree_005fint_005fcst_005flt"><code>tree_int_cst_lt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fLIST"><code>TREE_LIST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fOPERAND"><code>TREE_OPERAND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expression-trees">Expression trees</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fPUBLIC"><code>TREE_PUBLIC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fPUBLIC-1"><code>TREE_PUBLIC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fPURPOSE"><code>TREE_PURPOSE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fREADONLY"><code>TREE_READONLY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tree_005fsize"><code>tree_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fSTATIC"><code>TREE_STATIC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fSTRING_005fLENGTH"><code>TREE_STRING_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fSTRING_005fPOINTER"><code>TREE_STRING_POINTER</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTHIS_005fVOLATILE"><code>TREE_THIS_VOLATILE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Properties">Function Properties</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Macros-and-Functions">Macros and Functions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-1"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-2"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Working-with-declarations">Working with declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-3"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expression-trees">Expression trees</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-4"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Expression-trees">Expression trees</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-5"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Function-Basics">Function Basics</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fTYPE-6"><code>TREE_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fVALUE"><code>TREE_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fVEC"><code>TREE_VEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fVEC_005fELT"><code>TREE_VEC_ELT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TREE_005fVEC_005fLENGTH"><code>TREE_VEC_LENGTH</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRULY_005fNOOP_005fTRUNCATION"><code>TRULY_NOOP_TRUNCATION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-truncate"><code>truncate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-truncmn2-instruction-pattern"><code>trunc<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUNC_005fDIV_005fEXPR"><code>TRUNC_DIV_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUNC_005fMOD_005fEXPR"><code>TRUNC_MOD_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fANDIF_005fEXPR"><code>TRUTH_ANDIF_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fAND_005fEXPR"><code>TRUTH_AND_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fNOT_005fEXPR"><code>TRUTH_NOT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fORIF_005fEXPR"><code>TRUTH_ORIF_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fOR_005fEXPR"><code>TRUTH_OR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRUTH_005fXOR_005fEXPR"><code>TRUTH_XOR_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRY_005fBLOCK"><code>TRY_BLOCK</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRY_005fHANDLERS"><code>TRY_HANDLERS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TRY_005fSTMTS"><code>TRY_STMTS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Tuple-specific-accessors">Tuple specific accessors</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tuple-specific-accessors">Tuple specific accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-tuples">tuples</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Tuple-representation">Tuple representation</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-type">type</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-type-declaration">type declaration</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPENAME_005fTYPE"><code>TYPENAME_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPENAME_005fTYPE_005fFULLNAME"><code>TYPENAME_TYPE_FULLNAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPENAME_005fTYPE_005fFULLNAME-1"><code>TYPENAME_TYPE_FULLNAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPEOF_005fTYPE"><code>TYPEOF_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fALIGN"><code>TYPE_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fALIGN-1"><code>TYPE_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fALIGN-2"><code>TYPE_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fALIGN-3"><code>TYPE_ALIGN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fARG_005fTYPES"><code>TYPE_ARG_TYPES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fARG_005fTYPES-1"><code>TYPE_ARG_TYPES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fASM_005fOP"><code>TYPE_ASM_OP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fATTRIBUTES"><code>TYPE_ATTRIBUTES</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Attributes">Attributes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fBINFO"><code>TYPE_BINFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fBUILT_005fIN"><code>TYPE_BUILT_IN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fCANONICAL"><code>TYPE_CANONICAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fCANONICAL-1"><code>TYPE_CANONICAL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fCONTEXT"><code>TYPE_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fCONTEXT-1"><code>TYPE_CONTEXT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fDECL"><code>TYPE_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fFIELDS"><code>TYPE_FIELDS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fFIELDS-1"><code>TYPE_FIELDS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fFIELDS-2"><code>TYPE_FIELDS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fHAS_005fARRAY_005fNEW_005fOPERATOR"><code>TYPE_HAS_ARRAY_NEW_OPERATOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fHAS_005fDEFAULT_005fCONSTRUCTOR"><code>TYPE_HAS_DEFAULT_CONSTRUCTOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fHAS_005fMUTABLE_005fP"><code>TYPE_HAS_MUTABLE_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fHAS_005fNEW_005fOPERATOR"><code>TYPE_HAS_NEW_OPERATOR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMAIN_005fVARIANT"><code>TYPE_MAIN_VARIANT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMAIN_005fVARIANT-1"><code>TYPE_MAIN_VARIANT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMAIN_005fVARIANT-2"><code>TYPE_MAIN_VARIANT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMAX_005fVALUE"><code>TYPE_MAX_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMETHODS"><code>TYPE_METHODS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMETHOD_005fBASETYPE"><code>TYPE_METHOD_BASETYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMETHOD_005fBASETYPE-1"><code>TYPE_METHOD_BASETYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fMIN_005fVALUE"><code>TYPE_MIN_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fNAME"><code>TYPE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fNAME-1"><code>TYPE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fNAME-2"><code>TYPE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fNAME-3"><code>TYPE_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fNOTHROW_005fP"><code>TYPE_NOTHROW_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOFFSET_005fBASETYPE"><code>TYPE_OFFSET_BASETYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOFFSET_005fBASETYPE-1"><code>TYPE_OFFSET_BASETYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOPERAND_005fFMT"><code>TYPE_OPERAND_FMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Label-Output">Label Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOVERLOADS_005fARRAY_005fREF"><code>TYPE_OVERLOADS_ARRAY_REF</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOVERLOADS_005fARROW"><code>TYPE_OVERLOADS_ARROW</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fOVERLOADS_005fCALL_005fEXPR"><code>TYPE_OVERLOADS_CALL_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPOLYMORPHIC_005fP"><code>TYPE_POLYMORPHIC_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPRECISION"><code>TYPE_PRECISION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPRECISION-1"><code>TYPE_PRECISION</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTRDATAMEM_005fP"><code>TYPE_PTRDATAMEM_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTRDATAMEM_005fP-1"><code>TYPE_PTRDATAMEM_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTRFN_005fP"><code>TYPE_PTRFN_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTROBV_005fP"><code>TYPE_PTROBV_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTROB_005fP"><code>TYPE_PTROB_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fPTR_005fP"><code>TYPE_PTR_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fCONST"><code>TYPE_QUAL_CONST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fCONST-1"><code>TYPE_QUAL_CONST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fRESTRICT"><code>TYPE_QUAL_RESTRICT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fRESTRICT-1"><code>TYPE_QUAL_RESTRICT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fVOLATILE"><code>TYPE_QUAL_VOLATILE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fQUAL_005fVOLATILE-1"><code>TYPE_QUAL_VOLATILE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fRAISES_005fEXCEPTIONS"><code>TYPE_RAISES_EXCEPTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Functions-for-C_002b_002b">Functions for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSIZE"><code>TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSIZE-1"><code>TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSIZE-2"><code>TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSIZE-3"><code>TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSTRUCTURAL_005fEQUALITY_005fP"><code>TYPE_STRUCTURAL_EQUALITY_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fSTRUCTURAL_005fEQUALITY_005fP-1"><code>TYPE_STRUCTURAL_EQUALITY_P</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fUNQUALIFIED"><code>TYPE_UNQUALIFIED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fUNQUALIFIED-1"><code>TYPE_UNQUALIFIED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-TYPE_005fVFIELD"><code>TYPE_VFIELD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-U">U</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-UDAmode"><code>UDAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-udiv"><code>udiv</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-udivm3-instruction-pattern"><code>udiv<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-udivmodm4-instruction-pattern"><code>udivmod<var>m</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-udot_005fprodm-instruction-pattern"><code>udot_prod<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UDQmode"><code>UDQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UHAmode"><code>UHAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UHQmode"><code>UHQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT16_005fTYPE"><code>UINT16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT32_005fTYPE"><code>UINT32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT64_005fTYPE"><code>UINT64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT8_005fTYPE"><code>UINT8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINTMAX_005fTYPE"><code>UINTMAX_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINTPTR_005fTYPE"><code>UINTPTR_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fFAST16_005fTYPE"><code>UINT_FAST16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fFAST32_005fTYPE"><code>UINT_FAST32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fFAST64_005fTYPE"><code>UINT_FAST64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fFAST8_005fTYPE"><code>UINT_FAST8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fLEAST16_005fTYPE"><code>UINT_LEAST16_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fLEAST32_005fTYPE"><code>UINT_LEAST32_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fLEAST64_005fTYPE"><code>UINT_LEAST64_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UINT_005fLEAST8_005fTYPE"><code>UINT_LEAST8_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umaddmn4-instruction-pattern"><code>umadd<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umax"><code>umax</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umaxm3-instruction-pattern"><code>umax<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umin"><code>umin</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-uminm3-instruction-pattern"><code>umin<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umod"><code>umod</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umodm3-instruction-pattern"><code>umod<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umsubmn4-instruction-pattern"><code>umsub<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umulhisi3-instruction-pattern"><code>umulhisi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umulm3_005fhighpart-instruction-pattern"><code>umul<var>m</var>3_highpart</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umulqihi3-instruction-pattern"><code>umulqihi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-umulsidi3-instruction-pattern"><code>umulsidi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging"><code>unchanging</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-call_005finsn"><code>unchanging</code>, in <code>call_insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-jump_005finsn_002c-call_005finsn-and-insn"><code>unchanging</code>, in <code>jump_insn</code>, <code>call_insn</code> and <code>insn</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-mem"><code>unchanging</code>, in <code>mem</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-subreg"><code>unchanging</code>, in <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-subreg-1"><code>unchanging</code>, in <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unchanging_002c-in-symbol_005fref"><code>unchanging</code>, in <code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNEQ_005fEXPR"><code>UNEQ_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNGE_005fEXPR"><code>UNGE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNGT_005fEXPR"><code>UNGT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unions_002c-returning">unions, returning</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Interface">Interface</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNION_005fTYPE"><code>UNION_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNION_005fTYPE-1"><code>UNION_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Classes">Classes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNITS_005fPER_005fWORD"><code>UNITS_PER_WORD</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNKNOWN_005fTYPE"><code>UNKNOWN_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNKNOWN_005fTYPE-1"><code>UNKNOWN_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types-for-C_002b_002b">Types for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNLE_005fEXPR"><code>UNLE_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNLIKELY_005fEXECUTED_005fTEXT_005fSECTION_005fNAME"><code>UNLIKELY_EXECUTED_TEXT_SECTION_NAME</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNLT_005fEXPR"><code>UNLT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UNORDERED_005fEXPR"><code>UNORDERED_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unshare_005fall_005frtl"><code>unshare_all_rtl</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sharing">Sharing</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-division">unsigned division</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-division-with-unsigned-saturation">unsigned division with unsigned saturation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-greater-than">unsigned greater than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-greater-than-1">unsigned greater than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-less-than">unsigned less than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-less-than-1">unsigned less than</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Comparisons">Comparisons</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned-minimum-and-maximum">unsigned minimum and maximum</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned_005ffix"><code>unsigned_fix</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned_005ffloat"><code>unsigned_float</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned_005ffract_005fconvert"><code>unsigned_fract_convert</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unsigned_005fsat_005ffract"><code>unsigned_sat_fract</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unspec"><code>unspec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unspec-1"><code>unspec</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unspec_005fvolatile"><code>unspec_volatile</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-unspec_005fvolatile-1"><code>unspec_volatile</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-Definitions">Constant Definitions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-untyped_005fcall-instruction-pattern"><code>untyped_call</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-untyped_005freturn-instruction-pattern"><code>untyped_return</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UPDATE_005fPATH_005fHOST_005fCANONICALIZE-_0028path_0029"><code>UPDATE_PATH_HOST_CANONICALIZE (<var>path</var>)</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-update_005fssa"><code>update_ssa</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-update_005fstmt"><code>update_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-update_005fstmt-1"><code>update_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-update_005fstmt_005fif_005fmodified"><code>update_stmt_if_modified</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Manipulating-GIMPLE-statements">Manipulating GIMPLE statements</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UQQmode"><code>UQQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usaddm3-instruction-pattern"><code>usadd<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USAmode"><code>USAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usashlm3-instruction-pattern"><code>usashl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usdivm3-instruction-pattern"><code>usdiv<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-use"><code>use</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Side-Effects">Side Effects</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-used"><code>used</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-used_002c-in-symbol_005fref"><code>used</code>, in <code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-user"><code>user</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-user-gc">user gc</a>:</td><td>&nbsp;</td><td valign="top"><a href="#User-GC">User GC</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USER_005fLABEL_005fPREFIX"><code>USER_LABEL_PREFIX</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Instruction-Output">Instruction Output</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fC_005fALLOCA"><code>USE_C_ALLOCA</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fLD_005fAS_005fNEEDED"><code>USE_LD_AS_NEEDED</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Driver">Driver</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fLOAD_005fPOST_005fDECREMENT"><code>USE_LOAD_POST_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fLOAD_005fPOST_005fINCREMENT"><code>USE_LOAD_POST_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fLOAD_005fPRE_005fDECREMENT"><code>USE_LOAD_PRE_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fLOAD_005fPRE_005fINCREMENT"><code>USE_LOAD_PRE_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-use_005fparam"><code>use_param</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-use_005fparamn"><code>use_param<var>n</var></code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-use_005fparams"><code>use_params</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fSELECT_005fSECTION_005fFOR_005fFUNCTIONS"><code>USE_SELECT_SECTION_FOR_FUNCTIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Sections">Sections</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fSTORE_005fPOST_005fDECREMENT"><code>USE_STORE_POST_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fSTORE_005fPOST_005fINCREMENT"><code>USE_STORE_POST_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fSTORE_005fPRE_005fDECREMENT"><code>USE_STORE_PRE_DECREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USE_005fSTORE_005fPRE_005fINCREMENT"><code>USE_STORE_PRE_INCREMENT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Costs">Costs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USING_005fSTMT"><code>USING_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmaddmn4-instruction-pattern"><code>usmadd<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmsubmn4-instruction-pattern"><code>usmsub<var>m</var><var>n</var>4</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmulhisi3-instruction-pattern"><code>usmulhisi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmulm3-instruction-pattern"><code>usmul<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmulqihi3-instruction-pattern"><code>usmulqihi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usmulsidi3-instruction-pattern"><code>usmulsidi3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usnegm2-instruction-pattern"><code>usneg<var>m</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-USQmode"><code>USQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-ussubm3-instruction-pattern"><code>ussub<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-usum_005fwidenm3-instruction-pattern"><code>usum_widen<var>m3</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005fashift"><code>us_ashift</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005fminus"><code>us_minus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005fmult"><code>us_mult</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005fneg"><code>us_neg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005fplus"><code>us_plus</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-us_005ftruncate"><code>us_truncate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UTAmode"><code>UTAmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-UTQmode"><code>UTQmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-V">V</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-V-in-constraint">&lsquo;<samp>V</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-values_002c-returned-by-functions">values, returned by functions</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Scalar-Return">Scalar Return</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-varargs-implementation">varargs implementation</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Varargs">Varargs</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-variable">variable</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-Variable-Location-Debug-Information-in-RTL">Variable Location Debug Information in RTL</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Debug-Information">Debug Information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-variable_005fsize"><code>variable_size</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#GTY-Options">GTY Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VAR_005fDECL"><code>VAR_DECL</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Declarations">Declarations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-var_005flocation"><code>var_location</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Debug-Information">Debug Information</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vashlm3-instruction-pattern"><code>vashl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vashrm3-instruction-pattern"><code>vashr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VA_005fARG_005fEXPR"><code>VA_ARG_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Unary-and-Binary-Expressions">Unary and Binary Expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vcondmn-instruction-pattern"><code>vcond<var>m</var><var>n</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vector">vector</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Containers">Containers</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vector-operations">vector operations</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vector-Operations">Vector Operations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VECTOR_005fCST"><code>VECTOR_CST</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Constant-expressions">Constant expressions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VECTOR_005fSTORE_005fFLAG_005fVALUE"><code>VECTOR_STORE_FLAG_VALUE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fconcat"><code>vec_concat</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vector-Operations">Vector Operations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fduplicate"><code>vec_duplicate</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vector-Operations">Vector Operations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fextractm-instruction-pattern"><code>vec_extract<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005finitm-instruction-pattern"><code>vec_init<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fload_005flanesmn-instruction-pattern"><code>vec_load_lanes<var>m</var><var>n</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fLSHIFT_005fEXPR"><code>VEC_LSHIFT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fmerge"><code>vec_merge</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vector-Operations">Vector Operations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fPACK_005fFIX_005fTRUNC_005fEXPR"><code>VEC_PACK_FIX_TRUNC_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fPACK_005fSAT_005fEXPR"><code>VEC_PACK_SAT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpack_005fsfix_005ftrunc_005fm-instruction-pattern"><code>vec_pack_sfix_trunc_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpack_005fssat_005fm-instruction-pattern"><code>vec_pack_ssat_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fPACK_005fTRUNC_005fEXPR"><code>VEC_PACK_TRUNC_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpack_005ftrunc_005fm-instruction-pattern"><code>vec_pack_trunc_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpack_005fufix_005ftrunc_005fm-instruction-pattern"><code>vec_pack_ufix_trunc_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpack_005fusat_005fm-instruction-pattern"><code>vec_pack_usat_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fpermm-instruction-pattern"><code>vec_perm<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fperm_005fconstm-instruction-pattern"><code>vec_perm_const<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fRSHIFT_005fEXPR"><code>VEC_RSHIFT_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fselect"><code>vec_select</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vector-Operations">Vector Operations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fsetm-instruction-pattern"><code>vec_set<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fshl_005fm-instruction-pattern"><code>vec_shl_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fshr_005fm-instruction-pattern"><code>vec_shr_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fstore_005flanesmn-instruction-pattern"><code>vec_store_lanes<var>m</var><var>n</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacks_005ffloat_005fhi_005fm-instruction-pattern"><code>vec_unpacks_float_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacks_005ffloat_005flo_005fm-instruction-pattern"><code>vec_unpacks_float_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacks_005fhi_005fm-instruction-pattern"><code>vec_unpacks_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacks_005flo_005fm-instruction-pattern"><code>vec_unpacks_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacku_005ffloat_005fhi_005fm-instruction-pattern"><code>vec_unpacku_float_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacku_005ffloat_005flo_005fm-instruction-pattern"><code>vec_unpacku_float_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacku_005fhi_005fm-instruction-pattern"><code>vec_unpacku_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005funpacku_005flo_005fm-instruction-pattern"><code>vec_unpacku_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fUNPACK_005fFLOAT_005fHI_005fEXPR"><code>VEC_UNPACK_FLOAT_HI_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fUNPACK_005fFLOAT_005fLO_005fEXPR"><code>VEC_UNPACK_FLOAT_LO_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fUNPACK_005fHI_005fEXPR"><code>VEC_UNPACK_HI_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fUNPACK_005fLO_005fEXPR"><code>VEC_UNPACK_LO_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fWIDEN_005fMULT_005fHI_005fEXPR"><code>VEC_WIDEN_MULT_HI_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VEC_005fWIDEN_005fMULT_005fLO_005fEXPR"><code>VEC_WIDEN_MULT_LO_EXPR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Vectors">Vectors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsmult_005feven_005fm-instruction-pattern"><code>vec_widen_smult_even_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsmult_005fhi_005fm-instruction-pattern"><code>vec_widen_smult_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsmult_005flo_005fm-instruction-pattern"><code>vec_widen_smult_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsmult_005fodd_005fm-instruction-pattern"><code>vec_widen_smult_odd_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsshiftl_005fhi_005fm-instruction-pattern"><code>vec_widen_sshiftl_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fsshiftl_005flo_005fm-instruction-pattern"><code>vec_widen_sshiftl_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fumult_005feven_005fm-instruction-pattern"><code>vec_widen_umult_even_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fumult_005fhi_005fm-instruction-pattern"><code>vec_widen_umult_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fumult_005flo_005fm-instruction-pattern"><code>vec_widen_umult_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fumult_005fodd_005fm-instruction-pattern"><code>vec_widen_umult_odd_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fushiftl_005fhi_005fm-instruction-pattern"><code>vec_widen_ushiftl_hi_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vec_005fwiden_005fushiftl_005flo_005fm-instruction-pattern"><code>vec_widen_ushiftl_lo_<var>m</var></code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-verify_005fflow_005finfo"><code>verify_flow_info</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Maintaining-the-CFG">Maintaining the CFG</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-virtual-operands">virtual operands</a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA-Operands">SSA Operands</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VIRTUAL_005fINCOMING_005fARGS_005fREGNUM"><code>VIRTUAL_INCOMING_ARGS_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VIRTUAL_005fOUTGOING_005fARGS_005fREGNUM"><code>VIRTUAL_OUTGOING_ARGS_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VIRTUAL_005fSTACK_005fDYNAMIC_005fREGNUM"><code>VIRTUAL_STACK_DYNAMIC_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VIRTUAL_005fSTACK_005fVARS_005fREGNUM"><code>VIRTUAL_STACK_VARS_REGNUM</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VLIW">VLIW</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VLIW-1">VLIW</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Processor-pipeline-description">Processor pipeline description</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vlshrm3-instruction-pattern"><code>vlshr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VMS"><code>VMS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VMS_005fDEBUGGING_005fINFO"><code>VMS_DEBUGGING_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#VMS-Debug">VMS Debug</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VOIDmode"><code>VOIDmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-VOID_005fTYPE"><code>VOID_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Types">Types</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil"><code>volatil</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-insn_002c-call_005finsn_002c-jump_005finsn_002c-code_005flabel_002c-barrier_002c-and-note"><code>volatil</code>, in <code>insn</code>, <code>call_insn</code>, <code>jump_insn</code>, <code>code_label</code>, <code>barrier</code>, and <code>note</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-label_005fref-and-reg_005flabel"><code>volatil</code>, in <code>label_ref</code> and <code>reg_label</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-mem_002c-asm_005foperands_002c-and-asm_005finput"><code>volatil</code>, in <code>mem</code>, <code>asm_operands</code>, and <code>asm_input</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-reg"><code>volatil</code>, in <code>reg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-subreg"><code>volatil</code>, in <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-subreg-1"><code>volatil</code>, in <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatil_002c-in-symbol_005fref"><code>volatil</code>, in <code>symbol_ref</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatile-memory-references">volatile memory references</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-volatile_002c-in-prefetch"><code>volatile</code>, in <code>prefetch</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Flags">Flags</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-voting-between-constraint-alternatives">voting between constraint alternatives</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Class-Preferences">Class Preferences</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vrotlm3-instruction-pattern"><code>vrotl<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-vrotrm3-instruction-pattern"><code>vrotr<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-W">W</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-walk_005fdominator_005ftree"><code>walk_dominator_tree</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-walk_005fgimple_005fop"><code>walk_gimple_op</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statement-and-operand-traversals">Statement and operand traversals</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-walk_005fgimple_005fseq"><code>walk_gimple_seq</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statement-and-operand-traversals">Statement and operand traversals</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-walk_005fgimple_005fstmt"><code>walk_gimple_stmt</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statement-and-operand-traversals">Statement and operand traversals</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-walk_005fuse_005fdef_005fchains"><code>walk_use_def_chains</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#SSA">SSA</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WCHAR_005fTYPE"><code>WCHAR_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WCHAR_005fTYPE_005fSIZE"><code>WCHAR_TYPE_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-which_005falternative"><code>which_alternative</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Output-Statement">Output Statement</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WHILE_005fBODY"><code>WHILE_BODY</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WHILE_005fCOND"><code>WHILE_COND</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WHILE_005fSTMT"><code>WHILE_STMT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Statements-for-C_002b_002b">Statements for C++</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-whopr">whopr</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LTO">LTO</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WIDEST_005fHARDWARE_005fFP_005fSIZE"><code>WIDEST_HARDWARE_FP_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-window_005fsave-instruction-pattern"><code>window_save</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WINT_005fTYPE"><code>WINT_TYPE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WORDS_005fBIG_005fENDIAN"><code>WORDS_BIG_ENDIAN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Storage-Layout">Storage Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WORDS_005fBIG_005fENDIAN_002c-effect-on-subreg"><code>WORDS_BIG_ENDIAN</code>, effect on <code>subreg</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Regs-and-Memory">Regs and Memory</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-word_005fmode"><code>word_mode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-WORD_005fREGISTER_005fOPERATIONS"><code>WORD_REGISTER_OPERATIONS</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Misc">Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-wpa">wpa</a>:</td><td>&nbsp;</td><td valign="top"><a href="#LTO">LTO</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-X">X</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-X-in-constraint">&lsquo;<samp>X</samp>&rsquo; in constraint</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Simple-Constraints">Simple Constraints</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-x_002dhost"><samp>x-<var>host</var></samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Fragment">Host Fragment</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XCmode"><code>XCmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XCOFF_005fDEBUGGING_005fINFO"><code>XCOFF_DEBUGGING_INFO</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#DBX-Options">DBX Options</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XEXP"><code>XEXP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XFmode"><code>XFmode</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Machine-Modes">Machine Modes</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XF_005fSIZE"><code>XF_SIZE</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Type-Layout">Type Layout</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XINT"><code>XINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-xm_002dmachine_002eh"><samp>xm-<var>machine</var>.h</samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Filesystem">Filesystem</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-xm_002dmachine_002eh-1"><samp>xm-<var>machine</var>.h</samp></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Host-Misc">Host Misc</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-xor"><code>xor</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Arithmetic">Arithmetic</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-xor_002c-canonicalization-of"><code>xor</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-xorm3-instruction-pattern"><code>xor<var>m</var>3</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XSTR"><code>XSTR</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XVEC"><code>XVEC</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XVECEXP"><code>XVECEXP</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XVECLEN"><code>XVECLEN</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-XWINT"><code>XWINT</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Accessors">Accessors</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
<tr><th><a name="Concept-Index_cp_letter-Z">Z</a></th><td></td><td></td></tr>
<tr><td></td><td valign="top"><a href="#index-zero_005fextend"><code>zero_extend</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Conversions">Conversions</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-zero_005fextendmn2-instruction-pattern"><code>zero_extend<var>m</var><var>n</var>2</code> instruction pattern</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Standard-Names">Standard Names</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-zero_005fextract"><code>zero_extract</code></a>:</td><td>&nbsp;</td><td valign="top"><a href="#Bit_002dFields">Bit-Fields</a></td></tr>
<tr><td></td><td valign="top"><a href="#index-zero_005fextract_002c-canonicalization-of"><code>zero_extract</code>, canonicalization of</a>:</td><td>&nbsp;</td><td valign="top"><a href="#Insn-Canonicalizations">Insn Canonicalizations</a></td></tr>
<tr><td colspan="4"> <hr></td></tr>
</table>
<table><tr><th valign="top">Jump to: &nbsp; </th><td><a class="summary-letter" href="#Concept-Index_cp_symbol-1"><b>!</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-2"><b>#</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-3"><b>%</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-4"><b>&amp;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-5"><b>(</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-6"><b>*</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-7"><b>+</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-8"><b>-</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-9"><b>/</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-10"><b>0</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-11"><b>&lt;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-12"><b>=</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-13"><b>&gt;</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-14"><b>?</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-15"><b>\</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_symbol-16"><b>_</b></a>
 &nbsp; 
<br>
<a class="summary-letter" href="#Concept-Index_cp_letter-A"><b>A</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-B"><b>B</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-C"><b>C</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-D"><b>D</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-E"><b>E</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-F"><b>F</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-G"><b>G</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-H"><b>H</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-I"><b>I</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-J"><b>J</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-L"><b>L</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-M"><b>M</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-N"><b>N</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-O"><b>O</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-P"><b>P</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-Q"><b>Q</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-R"><b>R</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-S"><b>S</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-T"><b>T</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-U"><b>U</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-V"><b>V</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-W"><b>W</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-X"><b>X</b></a>
 &nbsp; 
<a class="summary-letter" href="#Concept-Index_cp_letter-Z"><b>Z</b></a>
 &nbsp; 
</td></tr></table>


<div class="footnote">
<hr>
<h4 class="footnotes-heading">Footnotes</h4>

<h3><a name="FOOT1" href="#DOCF1">(1)</a></h3>
<p>Except if the compiler was buggy and miscompiled
some of the files that were not modified.  In this case, it&rsquo;s best
to use <code>make restrap</code>.</p>
<h3><a name="FOOT2" href="#DOCF2">(2)</a></h3>
<p>Customarily, the system compiler
is also termed the <samp>stage0</samp> GCC.</p>
<h3><a name="FOOT3" href="#DOCF3">(3)</a></h3>
<p>These restrictions are derived
from those in Morgan 4.8.</p>
<h3><a name="FOOT4" href="#DOCF4">(4)</a></h3>
<p><code>note</code> insns can separate them, though.</p>
<h3><a name="FOOT5" href="#DOCF5">(5)</a></h3>
<p>However, the size of the automaton depends on
processor complexity.  To limit this effect, machine descriptions
can split orthogonal parts of the machine description among several
automata: but then, since each of these must be stepped independently,
this does cause a small decrease in the algorithm&rsquo;s performance.</p>
</div>
<hr>



</body>
</html>
gcc-4.8-doc 4.8.4-2ubuntu1~14.04.4 / usr / share / doc / gcc-4.8-base / gccint.html
