/usr/share/doc/gcc-4.7-base/NEWS.html

<?xml version="1.0" encoding="ISO-8859-1"?>
  <!DOCTYPE html
            PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
            "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
 


 










     <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
  
  
   <head>
 
    <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
    <link rev="made" href="mailto:gcc@gcc.gnu.org" />
    <link rel="shortcut icon" href="http://gcc.gnu.org/favicon.ico" />
    <link rel="stylesheet" type="text/css" href="http://gcc.gnu.org/gcc.css" />
  
 <title>
GCC 4.7 Release Series &mdash; Changes, New Features, and Fixes
- GNU Project - Free Software Foundation (FSF)</title>
   </head>
 

<!-- GCC maintainers, please do not hesitate to update/contribute entries
     concerning those part of GCC you maintain!  2002-03-23, Gerald.
-->

<body>



<h1>GCC 4.7 Release Series<br />Changes, New Features, and Fixes</h1>


<h2>Caveats</h2>

  <ul>
    <li><p>The <code>-fconserve-space</code> flag has been
    deprecated.  The flag had no effect for most targets: only
    targets without a global <code>.bss</code> section and without
    support for switchable sections.  Furthermore, the flag only
    had an effect for G++, where it could result in wrong semantics
    (please refer to the GCC manual for further details).
    The flag will be removed in GCC 4.8</p></li>

    <li><p>Support for a number of older systems and recently
    unmaintained or untested target ports of GCC has been declared
    obsolete in GCC 4.7.  Unless there is activity to revive them, the
    next release of GCC will have their sources permanently
    <strong>removed</strong>.</p>

    <p id="obsoleted">All GCC ports for the following processor
    architectures have been declared obsolete:</p>

    <ul>
	  <li>picoChip (<code>picochip-*</code>)</li>
    </ul>

    <p>The following ports for individual systems on
    particular architectures have been obsoleted:</p>

    <ul>
	  <li>IRIX 6.5 (mips-sgi-irix6.5)</li>
	  <li>MIPS OpenBSD (mips*-*-openbsd*)</li>
	  <li>Solaris 8 (*-*-solaris2.8).  Details can be found in the
	  <a href="http://gcc.gnu.org/ml/gcc-patches/2011-03/msg01263.html">
	      announcement</a>.</li>
	  <li>Tru64 UNIX V5.1 (alpha*-dec-osf5.1*)</li>
    </ul>

    </li>

    <li>On ARM, when compiling for ARMv6 (but not ARMv6-M), ARMv7-A,
    ARMv7-R, or ARMv7-M, the new option
    <code>-munaligned-access</code> is active by default, which for
    some source codes generates code that accesses memory on unaligned
    addresses.  This will require the kernel of those systems to enable
    such accesses (controlled by CP15 register <code>c1</code>, refer
    to ARM documentation).  Alternatively or for compatibility with
    kernels where unaligned accesses are not supported, all code has
    to be compiled with <code>-mno-unaligned-access</code>.
    Linux/ARM in official releases has automatically and
    unconditionally supported unaligned accesses as emitted by GCC due
    to this option being active, since Linux version 2.6.28.</li>

    <li>Support on ARM for the legacy floating-point accelerator (FPA) and
    the mixed-endian floating-point format that it used has been obsoleted.
    The ports that still use this format have been obsoleted as well.
    Many legacy ARM ports already provide an alternative that uses the
    VFP floating-point format.  The obsolete ports will be deleted in the
    next release.

    <p>The obsolete ports with alternatives are:</p>

    <ul>
	  <li>arm*-*-rtems (use arm*-*-rtemseabi)</li>
	  <li>arm*-*-linux-gnu (use arm*-*-linux-gnueabi)</li>
	  <li>arm*-*-elf (use arm*-*-eabi)</li>
	  <li>arm*-*-uclinux* (use arm*-*-uclinux*eabi)</li>
    </ul>
    <p>Note, however, that these alternatives are not binary compatible
    with their legacy counterparts (although some can support running legacy
    applications).</p>
    <p>The obsolete ports that currently lack a modern alternative are:</p>

    <ul>
	  <li>arm*-*-ecos-elf</li>
	  <li>arm*-*-freebsd</li>
	  <li>arm*-wince-pe*</li>
    </ul>

    New ports that support more recent versions of the architecture
    are welcome.</li>
    <li>Support for the Maverick co-processor on ARM has been obsoleted.
    Code to support it will be deleted in the next release.</li>
    <li>Support has been removed for Unix International threads on Solaris
    2, so the <code>--enable-threads=solaris</code> configure option and
    the <code>-threads</code> compiler option don't work any longer.</li>
    <li>Support has been removed for the Solaris BSD Compatibility Package,
    which lives in <code>/usr/ucbinclude</code> and
    <code>/usr/ucblib</code>.  It has been removed from Solaris 11, and was
    only intended as a migration aid from SunOS 4 to SunOS 5.  The
    <code>-compat-bsd</code> compiler option is not recognized any
    longer.</li>

    <li>The AVR port's libgcc has been improved and its multilib structure
      has been enhanced.  As a result, all objects contributing to an
      application must either be compiled with GCC versions up to 4.6.x or
      with GCC versions 4.7.0 or later.</li>
      
    <li>The ARM port's <code>-mwords-little-endian</code> option has
    been deprecated.  It will be removed in a future release.</li>

    <li>Support has been removed for the NetWare x86 configuration
    obsoleted in GCC 4.6.</li>

    <li>It is no longer possible to use the <code>"l"</code>
    constraint in MIPS16 <code>asm</code> statements.</li>

    <li>GCC versions 4.7.0 and 4.7.1 had changes to the C++ standard library
    which affected the ABI in C++11 mode: a data member was added to
    <code>std::list</code> changing its size and altering the definitions of
    some member functions, and <code>std::pair</code>'s move constructor was
    non-trivial which altered the calling convention for functions with
    <code>std::pair</code> arguments or return types.  The ABI incompatibilities
    have been fixed for GCC version 4.7.2 but as a result C++11 code compiled
    with GCC 4.7.0 or 4.7.1 may be incompatible with C++11 code compiled with
    different GCC versions and with C++98/C++03 code compiled with any version.
    </li>

    <li>On ARM, a bug has been fixed in GCC's implementation of the AAPCS
    rules for the layout of vectors that could lead to wrong code being
    generated.  Vectors larger than 8 bytes in size are now by default
    aligned to an 8-byte boundary.  This is an ABI change: code that makes
    explicit use of vector types may be incompatible with binary objects
    built with older versions of GCC.  Auto-vectorized code is not affected
    by this change.  (This change affects GCC versions 4.7.2 and later.)</li>

    <li>More information on porting to GCC 4.7 from previous versions
    of GCC can be found in
    the <a href="http://gcc.gnu.org/gcc-4.7/porting_to.html">porting
    guide</a> for this release.</li>
  </ul>


<h2>General Optimizer Improvements</h2>

  <ul>
    <li>Support for a new parameter <code>--param case-values-threshold=n</code>
    was added to allow users to control the cutoff between doing switch statements
    as a series of if statements and using a jump table.
    </li>

    <li>Link-time optimization (LTO) improvements:
    <ul>
      <li>Improved scalability and reduced memory usage.  Link time
      optimization of Firefox now requires 3GB of RAM on a 64-bit system,
      while over 8GB was needed previously. Linking time has been improved,
      too. The serial stage of linking Firefox has been sped up by about a
      factor of 10.</li>
      <li>Reduced size of object files and temporary storage used during linking.</li>
      <li>Streaming performance (both outbound and inbound) has been improved.</li>
      <li><code>ld -r</code> is now supported with LTO.</li>
      <li>Several bug fixes, especially in symbol table handling and merging.</li>
    </ul></li>

    <li>Interprocedural optimization improvements:
    <ul>
      <li>Heuristics now take into account that after inlining code will
      be optimized out because of known values (or properties) of function
      parameters.
      For example:
      <pre>
void foo(int a)
{
  if (a &gt; 10)
    ... huge code ...
}
void bar (void)
{
  foo (0);
}
      </pre>
      The call of <code>foo</code> will be inlined into <code>bar</code> even when
      optimizing for code size. Constructs based on <code>__builtin_constant_p</code>
      are now understood by the inliner and code size estimates are evaluated a lot
      more realistically.</li>
      <li>The representation of C++ virtual thunks and aliases (both implicit and defined
      via the <code>alias</code> attribute) has been re-engineered. Aliases no
      longer pose optimization barriers and calls to an alias can be inlined
      and otherwise optimized.</li>
      <li>The inter-procedural constant propagation pass has been rewritten.
      It now performs generic function specialization.  For example when
      compiling the following:
      <pre>
void foo(bool flag)
{
  if (flag)
    ... do something ...
  else
    ... do something else ...
}
void bar (void)
{
  foo (false);
  foo (true);
  foo (false);
  foo (true);
  foo (false);
  foo (true);
}
      </pre>
      GCC will now produce two copies of <code>foo</code>. One with <code>flag</code> being
      <code>true</code>, while other with <code>flag</code> being
      <code>false</code>.  This leads to performance improvements previously
      possible only by inlining all calls.  Cloning causes a lot less code size
      growth.</li>
    </ul></li>

    <li>A string length optimization pass has been added.  It attempts
      to track string lengths and optimize various standard C string functions
      like <code>strlen</code>, <code>strchr</code>, <code>strcpy</code>,
      <code>strcat</code>, <code>stpcpy</code> and their
      <code>_FORTIFY_SOURCE</code> counterparts into faster alternatives.
      This pass is enabled by default at <code>-O2</code> or above, unless
      optimizing for size, and can be disabled by the
      <code>-fno-optimize-strlen</code> option.  The pass can e.g. optimize
      <pre>
char *bar (const char *a)
{
  size_t l = strlen (a) + 2;
  char *p = malloc (l); if (p == NULL) return p;
  strcpy (p, a); strcat (p, "/"); return p;
}
      </pre>
      into:
      <pre>
char *bar (const char *a)
{
  size_t tmp = strlen (a);
  char *p = malloc (tmp + 2); if (p == NULL) return p;
  memcpy (p, a, tmp); memcpy (p + tmp, "/", 2); return p;
}
      </pre>
      or for hosted compilations where <code>stpcpy</code> is available in the
      runtime and headers provide its prototype, e.g.
      <pre>
void foo (char *a, const char *b, const char *c, const char *d)
{
  strcpy (a, b); strcat (a, c); strcat (a, d);
}
      </pre>
      can be optimized into:
      <pre>
void foo (char *a, const char *b, const char *c, const char *d)
{
  strcpy (stpcpy (stpcpy (a, b), c), d);
}
      </pre>
    </li>
  </ul>


<h2>New Languages and Language specific improvements</h2>

  <ul>
    <li>Version 3.1 of the <a
    href="http://openmp.org/wp/openmp-specifications/">OpenMP specification</a>
    is now supported for the C, C++, and Fortran compilers.</li>
  </ul>

<h3>Ada</h3>

  <ul>
    <li>The command-line option <code>-feliminate-unused-debug-types</code>
      has been re-enabled by default, as it is for the other languages,
      leading to a reduction in debug info size of 12.5% and more for
      relevant cases, as well as to a small compilation speedup.</li>
  </ul>

<h3>C family</h3>

<ul>
  <li>A new built-in, <code>__builtin_assume_aligned</code>, has been added,
      through which the compiler can be hinted about pointer alignment
      and can use it to improve generated code.
  </li>

  <li>A new -Wunused-local-typedefs warning was added for C, C++,
      Objective-C and Objective-C++.  This warning diagnoses typedefs
      locally defined in a function, and otherwise not used.
  </li>

  <li>A new experimental -ftrack-macro-expansion option was added for
      C, C++, Objective-C, Objective-C++ and Fortran.  It allows the
      compiler to emit diagnostic about the current macro expansion
      stack when a compilation error occurs in a macro expansion.
  </li>

  <li>
    <p>
      Experimental support for transactional memory has been added.
      It includes support in the compiler, as well as a supporting
      runtime library called <code>libitm</code>.  To compile code
      with transactional memory constructs, use
      the <code>-fgnu-tm</code> option.
    </p>

    <p>
      Support is currently available for Alpha, ARM, PowerPC, SH, SPARC,
      and 32-bit/64-bit x86 platforms.
    </p>

    <p>
      For more details on transactional memory
      see <a href="http://gcc.gnu.org/wiki/TransactionalMemory">the GCC
      WiKi</a>.
    </p>
  </li>

  <li>
    <p>
      Support for atomic operations specifying the C++11/C11 memory model
      has been added.  These new <code>__atomic</code> routines replace the 
      existing <code>__sync</code> built-in routines.
    </p>
    <p>
      Atomic support is also available for memory blocks.  Lock-free
      instructions will be used if a memory block is the same size and 
      alignment as a supported integer type.  Atomic operations which do not
      have lock-free support are left as function calls.  A set of library 
      functions is available on the GCC atomic wiki in the "External 
      Atomics Library" section.
    </p>
    <p>
      For more details on the memory models and features, see the 
      <a href="http://gcc.gnu.org/wiki/Atomic/GCCMM">atomic wiki</a>.
    </p>
  </li>

  <li>When a binary operation is performed on vector types and one of the operands
      is a uniform vector, it is possible to replace the vector with the
      generating element. For example:
      <pre>
typedef int v4si __attribute__ ((vector_size (16)));
v4si res, a = {1,2,3,4};
int x;

res = 2 + a;  /* means {2,2,2,2} + a  */
res = a - x;  /* means a - {x,x,x,x}  */
      </pre>
  </li>
</ul>

<h3>C</h3>

  <ul>
    <li>There is support for some more features from the C11 revision
    of the ISO C standard.  GCC now accepts the
    options <code>-std=c11</code> and <code>-std=gnu11</code>, in
    addition to the previous <code>-std=c1x</code>
    and <code>-std=gnu1x</code>.
    <ul>
      <li>Unicode strings (previously supported only with options such
      as <code>-std=gnu11</code>, now supported
      with <code>-std=c11</code>), and the predefined
      macros <code>__STDC_UTF_16__</code>
      and <code>__STDC_UTF_32__</code>.</li>
      <li>Nonreturning functions (<code>_Noreturn</code>
      and <code>&lt;stdnoreturn.h&gt;</code>).</li>
      <li>Alignment support
      (<code>_Alignas</code>, <code>_Alignof</code>,
      <code>max_align_t</code>, <code>&lt;stdalign.h&gt;</code>).</li>
      <li>A built-in function <code>__builtin_complex</code> is
      provided to support C library implementation of
      the <code>CMPLX</code> family of macros.</li>
    </ul>
    </li>
  </ul>

<a name="cxx" />
<h3>C++</h3>

<ul>
  <li>G++ now accepts the <code>-std=c++11</code>,
    <code>-std=gnu++11</code>, and <code>-Wc++11-compat</code> options,
    which are equivalent to <code>-std=c++0x</code>,
    <code>-std=gnu++0x</code>, and <code>-Wc++0x-compat</code>,
    respectively.</li>
  
  <li>G++ now implements <a href="cxx0x_status.html">C++11</a> extended friend syntax:
    <blockquote><pre>
template&lt;class W&gt;
class Q
{
  static const int I = 2;
public:
  friend W;
};

struct B
{
  int ar[Q&lt;B&gt;::I];
};</pre></blockquote></li>

  <li>Thanks to Ville Voutilainen, G++ now implements <a href="cxx0x_status.html">C++11</a> explicit override control.
    <blockquote><pre>
struct B {
  virtual void f() const final;
  virtual void f(int);
};

struct D : B {
  void f() const;            // error: D::f attempts to override final B::f
  void f(long) override;     // error: doesn't override anything
  void f(int) override;      // ok
};

struct E final { };
struct F: E { }; // error: deriving from final class</pre></blockquote></li>

  <li>G++ now implements <a href="cxx0x_status.html">C++11</a> non-static data member initializers.
    <blockquote><pre>
struct A {
  int i = 42;
} a; // initializes a.i to 42</pre></blockquote></li>

  <li>Thanks to Ed Smith-Rowland, G++ now implements 
    <a href="cxx0x_status.html">C++11</a> user-defined literals.
    <blockquote><pre>
// Not actually a good approximation.  :)
constexpr long double operator"" _degrees (long double d) { return d * 0.0175; }
long double pi = 180.0_degrees;</pre></blockquote></li>

  <li>G++ now implements
    <a href="cxx0x_status.html">C++11</a> alias-declarations.
    <blockquote><pre>
template &lt;class T> using Ptr = T*;
Ptr&lt;int> ip;  // decltype(ip) is int*</pre></blockquote></li>

  <li>Thanks to Ville Voutilainen and Pedro Lamar&atilde;o, G++ now implements <a href="cxx0x_status.html">C++11</a> delegating constructors.
    <blockquote><pre>
struct A {
  A(int);
  A(): A(42) { } // delegate to the A(int) constructor
};</pre></blockquote></li>

  <li>G++ now fully implements C++11 atomic classes rather than just integer 
    derived classes.
    <blockquote><pre>
class POD {
  int a;
  int b;
};
std::atomic&lt;POD> my_atomic_POD;
</pre></blockquote></li>

  <li>G++ now sets the predefined macro <code>__cplusplus</code> to the
    correct value, <code>199711L</code> for C++98/03,
    and <code>201103L</code> for C++11.
  </li>

  <li>G++ now correctly implements the two-phase lookup rules such that an
  unqualified name used in a template must have an appropriate declaration
  found either in scope at the point of definition of the template or by
  argument-dependent lookup at the point of instantiation.  As a result,
  code that relies on a second unqualified lookup at the point of
  instantiation to find functions declared after the template or in
  dependent bases will be rejected.  The compiler will suggest ways to fix
  affected code, and using the <code>-fpermissive</code> compiler flag will
  allow the code to compile with a warning.

    <blockquote><pre>
template &lt;class T&gt;
void f() { g(T()); } // error, g(int) not found by argument-dependent lookup
void g(int) { } // fix by moving this declaration before the declaration of f

template &lt;class T&gt;
struct A: T {
  // error, B::g(B) not found by argument-dependent lookup
  void f() { g(T()); } // fix by using this-&gt;g or A::g
};

struct B { void g(B); };

int main()
{
  f&lt;int&gt;();
  A&lt;B&gt;().f();
}</pre></blockquote></li>

  <li>G++ now properly re-uses stack space allocated for temporary objects
when their lifetime ends, which can significantly lower stack consumption
for some C++ functions.  As a result of this, some code with undefined
behavior will now break:
<blockquote><pre>
const int &amp;f(const int &amp;i) { return i; }
....
const int &amp;x = f(1);
const int &amp;y = f(2);</pre></blockquote>
Here, <code>x</code> refers to the temporary allocated to hold the
<code>1</code> argument, which only lives until the end of the
initialization; it immediately becomes a dangling reference.  So the
next statement re-uses the stack slot to hold the <code>2</code>
argument, and users of <code>x</code> get that value instead.

<p>Note that this should not cause any change of behavior for temporaries
of types with non-trivial destructors, as they are already destroyed at end
of full-expression; the change is that now the storage is released as
well.</p></li>

  <li>A new command-line option <code>-Wdelete-non-virtual-dtor</code>
      has been added to warn when <code>delete</code> is used to destroy
      an instance of a class which has virtual functions and non-virtual
      destructor. It is unsafe to delete an instance of a derived class
      through a pointer to a base class if the base class does not have a
      virtual destructor.  This warning is enabled by <code>-Wall</code>.
  </li>

  <li>A new command-line option <code>-Wzero-as-null-pointer-constant</code>
      has been added to warn when a literal '0' is used as null pointer
      constant.  It can be useful to facilitate the conversion to 
      <code>nullptr</code> in C++11.
  </li>

  <li>As per C++98, access-declarations are now deprecated by
      G++. Using-declarations are to be used instead. Furthermore,
      some efforts have been made to improve the support of class
      scope using-declarations. In particular, using-declarations
      referring to a dependent type now work as expected
      (<a href="http://gcc.gnu.org/PR14258">bug c++/14258</a>).
  </li>

  <li>The ELF symbol visibility of a template instantiation is now properly
    constrained by the visibility of its template arguments 
    (<a href="http://gcc.gnu.org/PR35688">bug c++/35688</a>).</li>

</ul>

  <h4>Runtime Library (libstdc++)</h4>

  <ul>
    <li><a href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/libstdc++/manual/manual/status.html#status.iso.2011">
       Improved experimental support for the new ISO C++ standard, C++11</a>,
       including:
       <ul>
         <li> using <code>noexcept</code> in most of the library;</li>
         <li> implementations of <code>pointer_traits</code>, <code>allocator_traits</code>
              and <code>scoped_allocator_adaptor</code>; </li>
         <li> uses-allocator construction for <code>tuple</code>; </li>
         <li> <code>vector</code> meets the allocator-aware container requirements; </li>
         <li> replacing <code>monotonic_clock</code> with <code>steady_clock</code>; </li>
         <li> enabling the thread support library on most POSIX targets; </li>
         <li> many small improvements to conform to the FDIS. </li>
       </ul>
     </li>
     <li>Added <code>--enable-clocale=newlib</code> configure option. </li>
     <li>Debug Mode iterators for unordered associative containers. </li>
     <li>Avoid polluting the global namespace and do not include
	<code>&lt;unistd.h&gt;</code>.</li>


  </ul>

<h3 id="fortran">Fortran</h3>
  <ul>
    <li>The compile flag <a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfstack-arrays_007d-254"
      ><code>-fstack-arrays</code></a> has been added, which causes
      all local arrays to be put on stack memory. For some
      programs this will improve the performance significantly. If your
      program uses very large local arrays, it is possible that you will
      have to extend your runtime limits for stack memory.</li>
    <li>The <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gcc/Optimize-Options.html#index-Ofast-689"
      >-Ofast</a></code> flag now also implies <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfno-protect-parens_007d-270"
      >-fno-protect-parens</a></code> and <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfstack-arrays_007d-254"
      >-fstack-arrays</a></code>.</li>
    <li>Front-end optimizations can now be selected by the
      <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfrontend-optimize_007d-275"
      >-ffrontend-optimize</a></code> option and deselected by
      the <code>-fno-frontend-optimize</code> option.</li>
    <li>When front-end optimization removes a function call,
      <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Error-and-Warning-Options.html#index-g_t_0040code_007bWfunction-elimination_007d-170"
      >-Wfunction-elimination</a></code> warns about that.</li>
    <li>When performing front-end-optimization, the
      <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfaggressive-function-elimination_007d-270">-faggressive-function-elimination</a></code> option
      allows the removal of duplicate function calls even for impure
      functions.</li>
    <li>The flag <code><a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Error-and-Warning-Options.html#index-g_t_0040code_007bWreal-q-constant_007d-149"
      >-Wreal-q-constant</a></code> has been added, which
      warns if floating-point literals have been specified using
      <code>q</code> (such as <code>1.0q0</code>); the <code>q</code>
      marker is now supported as a vendor extension to denote quad precision
      (<code>REAL(16)</code> or, if not available, <code>REAL(10)</code>).
      Consider using a kind parameter (such as in <code>1.0_qp</code>)
      instead, which can be obtained via <a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/SELECTED_005fREAL_005fKIND.html"
      ><code>SELECTED_REAL_KIND</code></a>.</li>
    <li>The <code>GFORTRAN_USE_STDERR</code> environment variable has
      been removed. GNU Fortran now always prints error messages to
      standard error. If you wish to redirect standard error, please
      consult the manual for your OS, shell, batch environment etc.
      as appropriate.</li> 
    <li>The <code>-fdump-core</code> option and
      <code>GFORTRAN_ERROR_DUMPCORE</code> environment variable have
      been removed. When encountering a serious error, gfortran will
      now always abort the program. Whether a core dump is generated
      depends on the user environment settings; see the <code>ulimit -c</code>
      setting for POSIX shells, <code>limit coredumpsize</code> for C shells,
      and the <a
      href="http://msdn.microsoft.com/en-us/library/bb787181%28v=vs.85%29.aspx"
      >WER user-mode dumps settings</a> on Windows.</li>
    <li>The <a
      href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Debugging-Options.html#index-g_t_0040code_007bfno-backtrace_007d-183"
      ><code>-fbacktrace</code></a> option is now enabled by default.
      When encountering a fatal error, gfortran will attempt to
      print a backtrace to standard error before aborting. It can be
      disabled with <code>-fno-backtrace</code>. Note: On POSIX targets
      with the <code>addr2line</code> utility from GNU binutils, GNU
      Fortran can print a backtrace with function name, file name,
      line number information in addition to the addresses; otherwise
      only the addresses are printed.</li>
    <li><a href="http://gcc.gnu.org/wiki/Fortran2003Status">Fortran 2003</a>:
      <ul>
       <li>Generic interface names which have the same name as derived types
         are now supported, which allows to write constructor functions. Note
         that Fortran does not support static constructor functions; only
         default initialization or an explicit structure-constructor
         initialization are available.</li>
       <li><a href="http://gcc.gnu.org/wiki/OOP">Polymorphic</a>
         (<code>class</code>) arrays are now supported.</li>
      </ul></li>
    <li><a href="http://gcc.gnu.org/wiki/Fortran2008Status">Fortran 2008</a>:
      <ul>
        <li>Support for the <code>DO CONCURRENT</code> construct has been
          added, which allows the user to specify that individual loop
          iterations have no interdependencies.</li>
        <li><a href="http://gcc.gnu.org/wiki/Coarray">Coarrays</a>:
          Full single-image support except for polymorphic coarrays.
          Additionally, preliminary support for multiple images via an
          MPI-based <a href="http://gcc.gnu.org/wiki/CoarrayLib">
          coarray communication library</a> has been added. Note:
          The library version is not yet usable as remote coarray
          access is not yet possible.</li>
      </ul></li>
    <li><a href="http://gcc.gnu.org/wiki/TS29113Status">TS 29113</a>:
      <ul>
        <li>New flag <code><a
          href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gfortran/Fortran-Dialect-Options.html#index-g_t_0040code_007bstd_003d_007d_0040var_007bstd_007d-option-53"
          >-std=f2008ts</a></code> permits programs that are expected
          to conform to the Fortran 2008 standard and the draft Technical
          Specification (TS) 29113 on Further Interoperability of Fortran
          with C.</li>
        <li>The <code>OPTIONAL</code> attribute is now allowed
          for dummy arguments of <code>BIND(C)</code> procedures.</li> 
        <li>The <code>RANK</code> intrinsic has been added.</li>
        <li>The implementation of the <code>ASYNCHRONOUS</code> attribute
          in GCC is compatible with the candidate draft of TS 29113
	  (since GCC 4.6).</li>
      </ul></li>
  </ul>

<h3 id="go">Go</h3>
  <ul>
    <li>GCC 4.7 implements
      the <a href="http://weekly.golang.org/doc/go1.html">Go 1
      language standard.</a>  The library support in 4.7.0 is not
      quite complete, due to release timing.  Release 4.7.1 includes
      complete support for Go 1.  The Go library is from the Go 1.0.1
      release.</li>
    <li>Go has been tested on GNU/Linux and Solaris platforms.  It may
      work on other platforms as well.</li>
  </ul>

<!--
<h3>Java (GCJ)</h3>
-->


<h2 id="targets">New Targets and Target Specific Improvements</h2>

<h3 id="arm">ARM</h3>
  <ul>
    <li>GCC now supports the Cortex-A7 processor implementing the
      v7-a version of the architecture using the option
      <code>-mcpu=cortex-a7</code>.</li>
    <li>The default vector size in auto-vectorization for NEON is now 128 bits.
      If vectorization fails thusly, the vectorizer tries again with
      64-bit vectors.</li>
    <li>A new option <code>-mvectorize-with-neon-double</code> was added to
      allow users to change the vector size to 64 bits.</li>

  </ul>

<h3 id="avr">AVR</h3>
  <ul>
    <li>GCC now supports the XMEGA architecture.
      This requires GNU binutils 2.22 or later.</li>
    <li>Support for the
      <a href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gcc/Named-Address-Spaces.html">named address spaces</a>
      <code>__flash</code>, <code>__flash1</code>, &hellip;,
      <code>__flash5</code> and <code>__memx</code> has been added.
      These address spaces locate read-only data in
      flash memory and allow reading from flash memory by means of ordinary
      C code, i.e. without the need of (inline) assembler code:
      <blockquote><pre>
const __flash int values[] = { 42, 31 };

int add_values (const __flash int *p, int i)
{
    return values[i] + *p;
}</pre></blockquote></li>
    <li>Support has been added for a new AVR-specific configure option
      <code>--with-avrlibc=yes</code> in order to arrange for better
      integration of <a href="http://nongnu.org/avr-libc/">AVR-Libc</a>.
      This configure option is supported in avr-gcc 4.7.2 and newer and will
      only take effect in non-RTEMS configurations.  If avr-gcc is configured
      for RTEMS, the option will be ignored which is the same as
      specifying <code>--with-avrlibc=no</code>.
      See <a href="http://gcc.gnu.org/PR54461">PR54461</a> for more technical 
      details.</li>
    <li>Support for AVR-specific <a href="http://gcc.gnu.org/onlinedocs/gcc-4.7.1/gcc/AVR-Built%5f002din-Functions.html">built-in functions</a>
      has been added.</li>
    <li>Support has been added for the signed and unsigned 24-bit scalar
      integer types <code>__int24</code> and <code>__uint24</code>.</li>
    <li>New command-line options <code>-maccumulate-args</code>,
      <code>-mbranch-cost=<i>cost</i></code> and <code>-mstrict-X</code>
      were added to allow better fine-tuning of code optimization.</li>
    <li>The command option <code>-fdata-sections</code> now also takes affect
      on the section names of variables with the <code>progmem</code>
      attribute.</li>
    <li>A new inline assembler print modifier <code>%i</code> to print a RAM address as I/O
    address has been added:
      <blockquote><pre>
#include &lt;avr/io.h&gt; /* Port Definitions from AVR-LibC */

void set_portb (uint8_t value)
{
    asm volatile ("out %0, %i1" :: "r" (value), "n" (&amp;PORTB) : "memory");
}</pre></blockquote>
      The offset between an I/O address and the RAM address for that I/O
      location is device-specific.  This offset is taken into account when
      printing a RAM address with the <code>%i</code> modifier so that the
      address is suitable to be used as operand in an I/O command.
      The address must be a constant integer known at compile time.</li>
    <li>The inline assembler constraint <code>"R"</code> to represent integers
      in the range &minus;6&nbsp;&hellip;&nbsp;5 has been removed
      without replacement.</li>
    <li>Many optimizations to:
      <ul>
	<li>64-bit integer arithmetic</li>
	<li>Widening multiplication</li>
	<li>Integer division by a constant</li>
	<li>Avoid constant reloading in multi-byte instructions.</li>
	<li>Micro-optimizations for special instruction sequences.</li>
	<li>Generic built-in functions like
	  <code>__builtin_ffs*</code>, <code>__builtin_clz*</code>, etc.</li>
	<li>If-else decision trees generated by <code>switch</code>
	  instructions</li>
	<li>Merging of data located in flash memory</li>
	<li>New libgcc variants for devices with 8-bit wide stack pointer</li>
	<li>&hellip;</li>
      </ul>
    </li>
    <li>Better documentation:
      <ul>
	<li>Handling of <code>EIND</code> and indirect jumps on devices with
	  more than 128 KiB of program memory.</li>
	<li>Handling of	the <code>RAMPD</code>, <code>RAMPX</code>,
	  <code>RAMPY</code> and <code>RAMPZ</code> special function registers.</li>
	<li>Function attributes <code>OS_main</code> and <code>OS_task</code>.</li>
	<li>AVR-specific built-in macros.</li>
      </ul>
    </li>
  </ul>

<h3>C6X</h3>
  <ul>
    <li>Support has been added for the Texas Instruments C6X family of
      processors.</li>
  </ul>

<h3>CR16</h3>
  <ul>
    <li>Support has been added for National Semiconductor's CR16
      architecture.</li>
  </ul>

<h3>Epiphany</h3>
  <ul>
    <li>Support has been added for Adapteva's Epiphany architecture.</li>
  </ul>

<h3>IA-32/x86-64</h3>
  <ul>
    <li>Support for Intel AVX2 intrinsics, built-in functions and code generation is
	available via <code>-mavx2</code>.</li>
    <li>Support for Intel BMI2 intrinsics, built-in functions and code generation is
	available via <code>-mbmi2</code>.</li>
    <li>Implementation and automatic generation of <code>__builtin_clz*</code> 
      using the <code>lzcnt</code> instruction is available via <code>-mlzcnt</code>.</li>
    <li>Support for Intel FMA3 intrinsics and code generation is available via
      <code>-mfma</code>.</li>
    <li>A new <code>-mfsgsbase</code> command-line option is available that makes GCC
    generate new segment register read/write instructions through dedicated built-ins.</li>
    <li>Support for the new Intel <code>rdrnd</code> instruction is available via <code>-mrdrnd</code>.</li>
    <li>Two additional AVX vector conversion instructions are available via <code>-mf16c</code>.</li>
    <li>Support for new Intel processor codename IvyBridge with RDRND, FSGSBASE and F16C
      is available through <code>-march=core-avx-i</code>.</li>
    <li>Support for the new Intel processor codename Haswell with AVX2, FMA, BMI,
      BMI2, LZCNT is available through <code>-march=core-avx2</code>.</li>
    <li>Support for new AMD family 15h processors (Piledriver core) is now available
      through <code>-march=bdver2</code> and <code>-mtune=bdver2</code> options.</li>
    <li>Support for <a href="http://sites.google.com/site/x32abi/">the x32 psABI</a>
      is now available through the <code>-mx32</code> option.</li>
    <li>Windows mingw targets are using the <code>-mms-bitfields</code> option
      by default.</li>
    <li>Windows x86 targets are using the <code>__thiscall</code> calling
      convention for C++ class-member functions.</li>
    <li>Support for the configure option <code>--with-threads=posix</code>
      for Windows mingw targets.</li>
  </ul>

<h3 id="mips">MIPS</h3>
  <ul>
    <li>GCC now supports thread-local storage (TLS) for MIPS16.
        This requires GNU binutils 2.22 or later.</li>

    <li>GCC can now generate code specifically for the Cavium Octeon+
        and Octeon2 processors.  The associated command-line options are
        <code>-march=octeon+</code> and <code>-march=octeon2</code>
        respectively.  Both options require GNU binutils 2.22 or later.</li>

    <li>GCC can now work around certain 24k errata, under the control
        of the command-line option <code>-mfix-24k</code>.
        These workarounds require GNU binutils 2.20 or later.</li>

    <li>32-bit MIPS GNU/Linux targets such as <code>mips-linux-gnu</code>
        can now build n32 and n64 multilibs.  The result is effectively
        a 64-bit GNU/Linux toolchain that generates 32-bit code by default.
        Use the configure-time option <code>--enable-targets=all</code>
        to select these extra multilibs.</li>

    <li>Passing <code>-fno-delayed-branch</code> now also stops the
        assembler from automatically filling delay slots.</li>
  </ul>

<!--
<h3 id="picochip">picochip</h3>
-->

<h3>PowerPC/PowerPC64</h3>
  <ul>
    <li>Vectors of type <i>vector long long</i> or <i>vector long</i> are
	passed and returned using the same method as other vectors with the VSX
	instruction set.  Previously GCC did not adhere to the ABI
	for 128-bit vectors with 64-bit integer base types (PR 48857).
	This will also be fixed in the GCC 4.6.1 and 4.5.4 releases.</li>

     <li>A new option <code>-mno-pointers-to-nested-functions</code> was
	added to allow AIX 32-bit/64-bit and GNU/Linux 64-bit PowerPC users to
	specify that the compiler should not load up the chain register
	(<code>r11</code>) before calling a function through a pointer.
	If you use this option, you cannot call nested functions through a
	pointer, or call other languages that might use the static chain.</li>

     <li>A new option <code>msave-toc-indirect</code> was added to allow AIX
	32-bit/64-bit and GNU/Linux 64-bit PowerPC users control whether we
	save the TOC in the prologue for indirect calls or generate the save
	inline.  This can speed up some programs that call through a function
	pointer a lot, but it can slow down other functions that only call
	through a function pointer in exceptional cases.</li>

	<li>The PowerPC port will now enable machine-specific built-in
	functions when the user switches the target machine using the
	<code>#pragma GCC target</code> or
	<code>__attribute__ ((__target__ ("<em>target</em>")))</code>
	code sequences.  In addition, the target macros are updated.
	However, due to the way the <code>-save-temps</code> switch is
	implemented, you won't see the effect of these additional macros
	being defined in preprocessor output.</li>
  </ul>

<h3>SH</h3>
  <ul>
    <li>A new option <code>-msoft-atomic</code> has been added.  When it is
        specified, GCC will generate GNU/Linux-compatible gUSA atomic sequences
	for the new <code>__atomic</code> routines.</li>
    <li>Since it is neither supported by GAS nor officially documented, code
        generation for little endian SH2A has been disabled.  Specifying
        <code>-ml</code> with <code>-m2a*</code> will now result in a compiler
	error.</li>
    <li>The defunct <code>-mbranch-cost</code> option has been fixed.</li>
    <li>Some improvements to the generated code of:
      <ul>
        <li>Utilization of the <code>tst #imm,R0</code> instruction.</li>
	<li>Dynamic shift instructions on SH2A.</li>
        <li>Integer absolute value calculations.</li>
      </ul></li>
  </ul>

<h3>SPARC</h3>
  <ul>
    <li>The option <code>-mflat</code> has been reinstated.  When it is
        specified, the compiler will generate code for a single register
        window model.  This is essentially a new implementation and the
        corresponding debugger support has been added to GDB 7.4.</li>
    <li>Support for the options <code>-mtune=native</code> and
        <code>-mcpu=native</code> has been added on selected native platforms
        (GNU/Linux and Solaris).</li>
    <li>Support for the SPARC T3 (Niagara 3) processor has been added.</li>
    <li>VIS:
      <ul>
        <li>An intrinsics header <code>visintrin.h</code> has been added.</li>
        <li>Builtin intrinsics for the VIS 1.0 edge handling and pixel compare
            instructions have been added.</li>
        <li>The little-endian version of <code>alignaddr</code> is now
            supported.</li>
        <li>When possible, VIS builtins are marked <code>const</code>, which
            should increase the compiler's ability to optimize VIS
            operations.</li>
        <li>The compiler now properly tracks the <code>%gsr</code> register
            and how it behaves as an input for various VIS instructions.</li>
        <li>Akin to <code>fzero</code>, the compiler can now generate
            <code>fone</code> instructions in order to set all of the bits
            of a floating-point register to <code>1</code>.</li>
        <li>The documentation for the VIS intrinsics in the GCC manual has
            been brought up to date and many inaccuracies were fixed.</li>
        <li>Intrinsics for the VIS 2.0 <code>bmask</code>,
	    <code>bshuffle</code>, and non-condition-code 
	    setting edge instructions have been added.  Their availability
	    is controlled by the new <code>-mvis2</code> and
	    <code>-mno-vis2</code> options.  They are enabled by default
	    on UltraSPARC-III and later CPUs.</li>
      </ul>
    </li>
    <li>Support for UltraSPARC Fused Multiply-Add floating-point
        extensions has been added.  These instructions are enabled by
        default on SPARC T3 (Niagara 3) and later CPUs.</li>
  </ul>

<h3>TILE-Gx/TILEPro</h3>
  <ul>
    <li>Support has been added for the Tilera TILE-Gx and TILEPro families of
      processors.</li>
  </ul>

<!--
<h2>Documentation improvements</h2>
-->


<h2>Other significant improvements</h2>

<ul>
  <li>
    A new option (<code>-grecord-gcc-switches</code>) was added that
    appends compiler command-line options that might affect code
    generation to the <code>DW_AT_producer</code> attribute string in the
    DWARF debugging information.
  </li>

  <li>
    GCC now supports various new GNU extensions to the DWARF debugging
    information format, like
    <a
    href="http://www.dwarfstd.org/ShowIssue.php?issue=100909.1">entry
    value</a> and <a
    href="http://www.dwarfstd.org/ShowIssue.php?issue=100909.2">call
    site</a> information, <a
    href="http://www.dwarfstd.org/doc/040408.1.html">typed DWARF stack</a>
    or <a
    href="http://www.dwarfstd.org/ShowIssue.php?issue=110722.1">a
    more compact macro representation</a>.  Support for these extensions
    has been added to GDB 7.4. They can be disabled through the
    <code>-gstrict-dwarf</code> command-line option.
  </li>
</ul>

<h2><a name="4.7.1">GCC 4.7.1</a></h2>

<p>This is the <a
href="http://gcc.gnu.org/bugzilla/buglist.cgi?bug_status=RESOLVED&amp;resolution=FIXED&amp;target_milestone=4.7.1">list
of problem reports (PRs)</a> from GCC's bug tracking system that are
known to be fixed in the 4.7.1 release. This list might not be
complete (that is, it is possible that some PRs that have been fixed
are not listed here).</p>

<p>The Go frontend in the 4.7.1 release fully supports
the <a href="http://weekly.golang.org/doc/go1.html">Go 1 language
standard.</a></p>

<h2><a name="4.7.2">GCC 4.7.2</a></h2>

<p>This is the <a
href="http://gcc.gnu.org/bugzilla/buglist.cgi?bug_status=RESOLVED&amp;resolution=FIXED&amp;target_milestone=4.7.2">list
of problem reports (PRs)</a> from GCC's bug tracking system that are
known to be fixed in the 4.7.2 release. This list might not be
complete (that is, it is possible that some PRs that have been fixed
are not listed here).</p>




<!-- ==================================================================== -->

<div class="copyright">

<address style="margin-top:0;">For questions related to the use of GCC,
please consult these web pages and the
<a href="http://gcc.gnu.org/onlinedocs/">GCC manuals</a>. If that fails,
the <a href="mailto:gcc-help@gcc.gnu.org">gcc-help@gcc.gnu.org</a>
mailing list might help.
Comments on these web pages and the development of GCC are welcome on our
developer list at <a href="mailto:gcc@gcc.gnu.org">gcc@gcc.gnu.org</a>.
All of <a href="http://gcc.gnu.org/lists.html">our lists</a>
have public archives.
</address>

<p>Copyright (C)
<a href="http://www.fsf.org">Free Software Foundation, Inc.</a>
Verbatim copying and distribution of this entire article is
permitted in any medium, provided this notice is preserved.</p>

<p style="margin-bottom:0;">These pages are
<a href="http://gcc.gnu.org/about.html">maintained by the GCC team</a>.
Last modified 2012-09-20<!-- IGNORE DIFF
--><a href="http://validator.w3.org/check/referer">.</a></p>

</div>

<!-- ==================================================================== -->

</body>
     </html>
gcc-4.7 4.7.2-5 / usr / share / doc / gcc-4.7-base / NEWS.html
