/usr/share/doc/gcc-4.6-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="/gnu.css" />
    <link rel="stylesheet" type="text/css" href="http://gcc.gnu.org/gcc.css" />
  
 <title>
GCC 4.6 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 bgcolor="#FFFFFF" text="#000000" link="#1F00FF" alink="#FF0000" vlink="#9900DD">



<h1 align="center">
GCC 4.6 Release Series<br />Changes, New Features, and Fixes
</h1>

<h2>Caveats</h2>
  <ul>

    <li>The options <code>-b &lt;<em>machine</em>&gt;</code> and
    <code>-V &lt;<em>version</em>&gt;</code> have been removed because
    they were unreliable. Instead, users should directly run
    <code>&lt;<em>machine</em>&gt;-gcc</code> when cross-compiling, or
    <code>&lt;<em>machine</em>&gt;-gcc-&lt;<em>version</em>&gt;</code>
    to run a different version of <code>gcc</code>. </li>

    <li>GCC now has stricter checks for invalid command-line options.
    In particular, when <code>gcc</code> was called to link object
    files rather than compile source code, it would previously accept
    and ignore all options starting with <code>--</code>, including
    linker options such as <code>--as-needed</code>
    and <code>--export-dynamic</code>, although such options would
    result in errors if any source code was compiled.  Such options,
    if unknown to the compiler, are now rejected in all cases; if the
    intent was to pass them to the linker, options such
    as <code>-Wl,--as-needed</code> should be used.</li>

    <li>Versions of the GNU C library up to and including 2.11.1
    included an <a
    href="http://sourceware.org/bugzilla/show_bug.cgi?id=10401">incorrect
    implementation of the <code>cproj</code> function</a>.  GCC
    optimizes its builtin <code>cproj</code> according to the behavior
    specified and allowed by the ISO C99 standard.  If you want to
    avoid discrepancies between the C library and GCC's builtin
    transformations when using <code>cproj</code> in your code, use
    GLIBC 2.12 or later.  If you are using an older GLIBC and actually
    rely on the incorrect behavior of cproj, then you can disable
    GCC's transformations using <code>-fno-builtin-cproj</code>. </li>

    <li>The C-only intermodule optimization framework (IMA, enabled by
    <code>-combine</code>) has been removed in favor of the new
    generic link-time optimization framework (LTO) introduced
    in <a href="../gcc-4.5/changes.html">GCC 4.5.0</a>.</li>

    <li id="libquadmath">GCC now ships with the LGPL-licensed
    <code>libquadmath</code> library, which provides quad-precision
    mathematical functions for targets with a <code>__float128</code>
    datatype. <code>__float128</code> is available for targets on
    32-bit x86, x86-64 and Itanium architectures. The
    <code>libquadmath</code> library is automatically built on
    such targets when building the Fortran compiler.</li>

    <li>New <code>-Wunused-but-set-variable</code> and
    <code>-Wunused-but-set-parameter</code> warnings were added
    for C, C++, Objective-C and Objective-C++.
    These warnings diagnose variables respective parameters which
    are only set in the code and never otherwise used.
    Usually such variables are useless and often even the value
    assigned to them is computed needlessly, sometimes expensively.
    The <code>-Wunused-but-set-variable</code> warning is enabled by
    default by <code>-Wall</code> flag and <code>-Wunused-but-set-parameter</code>
    by <code>-Wall -Wextra</code> flags.</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 4.6.  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>Argonaut ARC (<code>arc-*</code>)</li>
      <li>National Semiconductor CRX (<code>crx-*</code>)</li>
      <li>Motorola 68HC11 and 68HC12
      (<code>m68hc11-*-*</code>, <code>m6811-*-*</code>,
      <code>m68hc12-*-*</code>, <code>m6812-*-*</code>)</li>
      <li>Sunplus S+core (<code>score-*</code>)</li>
    </ul>

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

    <ul>
      <li>Interix (<code>i[34567]86-*-interix3*</code>)</li>
      <li>Generic ARM PE (<code>arm-*-pe*</code> other
      than <code>arm*-wince-pe*</code>)</li>
      <li>MCore PE (<code>mcore-*-pe*</code>)</li>
      <li>SH SymbianOS (<code>sh*-*-symbianelf*</code>)</li>
      <li>GNU Hurd on Alpha and PowerPC
      (<code>alpha*-*-gnu*</code>, <code>powerpc*-*-gnu*</code>)</li>
      <li>M68K uClinux old ABI
      (<code>m68k-*-uclinuxoldabi*</code>)</li>
      <li>a.out NetBSD
      (<code>arm*-*-netbsd*</code>, <code>i[34567]86-*-netbsd*</code>,
      <code>vax-*-netbsd*</code>, but
      not <code>*-*-netbsdelf*</code>)</li>
    </ul>

    <p>The <code>i[34567]86-*-pe</code> alias for Cygwin targets has
    also been obsoleted; users should configure
    for <code>i[34567]86-*-cygwin*</code> instead.</p>

    <p>Certain configure options to control the set of libraries built
    with GCC on some targets have been obsoleted.  On ARM targets, the
    options <code>--disable-fpu</code>, <code>--disable-26bit</code>,
    <code>--disable-underscore</code>, <code>--disable-interwork</code>,
    <code>--disable-biendian</code> and <code>--disable-nofmult</code>
    have been obsoleted.  On MIPS targets, the options
    <code>--disable-single-float</code>, <code>--disable-biendian</code>
    and <code>--disable-softfloat</code> have been obsoleted.</p>

    </li>

    <li>Support has been removed for all the 
    <a href="../gcc-4.5/changes.html#obsoleted">configurations obsoleted
    in GCC 4.5</a>.</li>

</ul>

<h2>General Optimizer Improvements</h2>

  <ul>
    <li>A new general optimization level, <code>-Ofast</code>, has been
      introduced.  It combines the existing optimization level <code>-O3</code>
      with options that can affect standards compliance but result in
      better optimized code.  For example, <code>-Ofast</code> enables
      <code>-ffast-math</code>.</li>
      <li>Link-time optimization improvements:
      <ul>
        <li>The <a href="../projects/lto/whopr.pdf">Scalable Whole
           Program Optimizer (WHOPR)</a> project has stabilized to the
           point of being usable.  It has become the default mode when
           using the LTO optimization model.  Link time optimization can
           now split itself into multiple parallel compilations. Parallelism
           is controlled with <code>-flto=<em>n</em></code> (where
           <em>n</em> specifies the number of compilations to execute in
           parallel).  GCC can also cooperate with a GNU make job server
           by specifying the <code>-flto=jobserver</code> option and
           adding <code>+</code> to the beginning of the
           Makefile rule executing the linker.<br />
	   Classical LTO mode can be enforced by
	   <code>-flto-partition=none</code>. This may result in small code
	   quality improvements.</li>
        <li>A large number of bugs were fixed.  GCC itself, Mozilla
           Firefox and other large applications can be built with
           LTO enabled.</li>
	<li>The linker plugin support improvements
	<ul>
	   <li>Linker plugin is now enabled by default when the linker is
	   detected to have plugin support.  This is the case for GNU
	   ld 2.21.51 or newer (on ELF and Cygwin targets) and the Gold
	   linker on ELF targets.  Plugin support of the Apple linker on
	   Darwin is not compatible with GCC.
	   The linker plugin can also be controlled by the
	   <code>-fuse-linker-plugin</code> command line option.</li>
	   <li>Resolution information from the linker plugin is used to drive
           whole program assumptions. Use of the linker plugin results in
           more aggressive optimization on binaries and on shared libraries
	   that use the <code>hidden</code> visibility
	   attribute. Consequently the use
	   of <code>-fwhole-program</code> is not neccesary in addition to
           LTO.</li>
	</ul>
	</li>
        <li>Hidden symbols used from non-LTO objects now have to be
           explicitly annotated with <code>externally_visible</code> when
           the linker plugin is not used.</li>
        <li>C++ inline functions and virtual tables are now privatized more
           aggressively, leading to better inter-procedural optimization
           and faster dynamic linking.</li>
        <li>Memory usage and intermediate language streaming performance
           have been improved.</li>
        <li>Static constructors and destructors from individual units are
           inlined into a single function.
           This can significantly improve startup times of large C++
           applications where static constructors are very common. For
           example, static constructors are used when including the
           <code>iostream</code> header.</li>
        <li>Support for the Ada language has been added.</li>
      </ul>
      </li>
      <li>Interprocedural optimization improvements
      <ul>
	  <li>The interprocedural framework was re-tuned for link time
	      optimization. Several scalability issues were resolved.</li>
	  <li>Improved auto-detection of <code>const</code> and <code>pure</code>
	      functions.  Newly, <code>noreturn</code> functions are auto-detected.
	      <p>The <a href="http://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html#Warning-Options"><code>-Wsuggest-attribute=[const|pure|noreturn]</code></a>
	      flag is available that informs users when adding
	      attributes to headers might improve code generation.</p></li>
	  <li>A number of inlining heuristic improvements. In particular:
	      <ul>
		 <li>Partial inlining is now supported and enabled by default
		     at <code>-O2</code> and greater.  The feature can be
		     controlled via <code>-fpartial-inlining</code>.
		     <p>
		     Partial inlining splits functions with short hot path
		     to return.  This allows more aggressive inlining of the
		     hot path leading to better performance and often to
		     code size reductions (because cold parts of functions
		     are not duplicated).
		     </p>
		 </li>
		 <li>Scalability for large compilation units was improved
		     significantly.</li>
		 <li>Inlining of callbacks is now more aggressive.</li>
		 <li>Virtual methods are considered for inlining when the
		     caller is
		     inlined and devirtualization is then possible.</li>
                 <li>Inlining when optimizing for size (either in cold
                    regions of a program or when compiling with
                    <code>-Os</code>) was improved to better handle C++
                    programs with larger abstraction penalty, leading
                    to smaller and faster code.</li>
	      </ul>
	  </li>
	  <li>The IPA reference optimization pass detecting global
            variables used or modified by functions was strengthened
            and sped up.</li>
          <li>Functions whose address was taken are now optimized out
            when all references to them are dead.</li>
          <li>A new inter-procedural static profile estimation pass detects
            functions that are executed once or unlikely to be executed.
            Unlikely executed functions are optimized for size.  Functions
            executed once are optimized for size except for the inner
            loops.</li>
	  <li>On most targets with named section support, functions used only
	    at startup (static constructors and <code>main</code>), functions
	    used only at exit and functions detected to be cold are placed into
	    separate text segment subsections.
	    This extends the <code>-freorder-functions</code> feature and is
	    controlled by the same switch. The goal is to improve the startup
	    time of large C++ programs.
	    <p>Proper function placement requires linker support.
	    GNU ld 2.21.51 on ELF targets was updated to place
	    those functions together within the text section leading to better code
	    locality and faster startup times of large C++ programs. The feature is
	    also supported in the Apple linker.
	    Support in the gold linker is planned.</p>
	    </li>
      </ul>
      </li>
      <li>A new switch <code>-fstack-usage</code> has been added.  It makes
      the compiler output stack usage information for the program, on a
      per-function basis, in an auxiliary file.</li>
      <li>A new switch <code>-fcombine-stack-adjustments</code> has been added.
      It can be used to enable or disable the compiler's stack-slot combining
      pass which before was enabled automatically at <code>-O1</code> and above,
      but could not be controlled on its own.</li>
      <li>A new switch <code>-fstrict-volatile-bitfields</code> has been
	added.  Using it indicates that accesses to volatile bitfields
	should use a single access of the width of the field's type.
	This option can be useful for precisely defining and accessing
	memory-mapped peripheral registers from C or C++.</li>
  </ul>

<h2>Compile time and memory usage improvements</h2>
  <ul>
      <li>Datastructures used by the dataflow framework in GCC were reorganized
	  for better memory usage and more cache locality.  Compile
	  time is improved especially on units with large functions (possibly
	  resulting from a lot of inlining) not fitting into the processor cache.
	  The compile time of the GCC C compiler binary with link-time
	  optimization went down by over 10% (benchmarked on x86-64 target).</li>
  </ul>

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

<h3>Ada</h3>

  <ul>
    <li>Stack checking has been improved on selected architectures (Alpha,
    IA-32/x86-64, RS/6000 and SPARC): it now will detect stack overflows
    in all cases on these architectures.</li>
    <li>Initial support for Ada 2012 has been added.</li>
  </ul>

<h3>C family</h3>

  <ul>
    <li>A new warning, enabled by <code>-Wdouble-promotion</code>,
      has been added that warns about cases where a value of type
      <code>float</code> is implicitly promoted to <code>double</code>.
      This is especially helpful for CPUs that handle the former in
      hardware, but emulate the latter in software.</li>
    <li>A new function attribute <code>leaf</code> was introduced.
      This attribute allows better inter-procedural optimization across
      calls to functions that return to the current unit only via returning
      or exception handling.  This is the case for most library functions
      that have no callbacks.</li>
    <li>Support for a new data type <code>__int128</code> for targets having
    wide enough machine-mode support.</li>
    <li>The new function attribute <code>callee_pop_aggregate</code> allows
    to specify if the caller or callee is responsible for popping the
    aggregate return pointer value from the stack.</li>
    <li>Support for selectively enabling and disabling warnings
      via <code>#pragma GCC diagnostic</code> has been added.  For instance:
<pre>#pragma GCC diagnostic error "-Wuninitialized"
  foo(a);			/* error is given for this one */
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wuninitialized"
  foo(b);			/* no diagnostic for this one */
#pragma GCC diagnostic pop
  foo(c);			/* error is given for this one */
#pragma GCC diagnostic pop
  foo(d);			/* depends on command line options */
</pre></li>
    <li>The <code>-fmax-errors=N</code> option is now supported.  Using
    this option causes the compiler to exit after <code>N</code> errors
    have been issued.</li>
  </ul>

<h3 id="c">C</h3>

  <ul>
    <li>There is now experimental support for some features from the
    upcoming C1X revision of the ISO C standard.  This support may be
    selected with <code>-std=c1x</code>, or <code>-std=gnu1x</code>
    for C1X with GNU extensions.  Note that this support is
    experimental and may change incompatibly in future releases for
    consistency with changes to the C1X standard draft.  The following
    features are newly supported as described in the N1539 draft of
    C1X (with changes agreed at the March 2011 WG14 meeting); some
    other features were already supported with no compiler
    changes being needed, or have some support but not in full accord
    with N1539 (as amended).
    <ul>
      <li>Static assertions (<code>_Static_assert</code> keyword)</li>
      <li>Typedef redefinition</li>
      <li>New macros in <code>&lt;float.h&gt;</code></li>
      <li>Anonymous structures and unions</li>
    </ul>
    </li>
    <li>The new <code>-fplan9-extensions</code> option directs the
      compiler to support some extensions for anonymous struct fields
      which are implemented by the Plan 9 compiler.  A pointer to a
      struct may be automatically converted to a pointer to an
      anonymous field when calling a function, in order to make the
      types match.  An anonymous struct field whose type is a typedef
      name may be referred to using the typedef name.</li>
  </ul>

<h3 id="cplusplus">C++</h3>

  <ul>
    <li>Improved <a href="cxx0x_status.html">experimental support for the
    upcoming C++0x</a> ISO C++ standard, including support for
    <code>constexpr</code> (thanks to Gabriel Dos Reis and Jason Merrill),
    <code>nullptr</code> (thanks to Magnus Fromreide), <code>noexcept</code>,
    unrestricted unions, range-based <code>for</code> loops (thanks to Rodrigo Rivas Costa),
    opaque enum declarations (thanks also to Rodrigo), implicitly deleted
    functions and implicit move constructors.</li>

    <li>When an extern declaration within a function does not match a
    declaration in the enclosing context, G++ now properly declares the
    name within the namespace of the function rather than the namespace
    which was open just before the function definition
    (<a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43145">c++/43145</a>).</li>

    <li>GCC now warns by default when casting integers to larger
    pointer types.  These warnings can be disabled with the option
    <code>-Wno-int-to-pointer-cast</code>, which is now also available
    in C++.</li>

    <li>G++ no longer optimizes using the assumption that a value of
    enumeration type will fall within the range specified by the standard,
    since that assumption is easily violated with a conversion from integer
    type (<a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43680">c++/43680</a>).
    The old behavior can be restored with <code>-fstrict-enums</code>.</li>

    <li>The new <code>-fnothrow-opt</code> flag changes the semantics of
    a <code>throw()</code> exception specification to match the proposed
    semantics of the <code>noexcept</code> specification: just call
    <code>terminate</code> if an exception tries to propagate out of a
    function with such an exception specification.  This dramatically
    reduces or eliminates the code size overhead from adding the exception
    specification.</li>

    <li>The new <code>-Wnoexcept</code> flag will suggest adding
    a <code>noexcept</code> qualifier to a function that the compiler can
    tell doesn't throw if it would change the value of
    a <code>noexcept</code> expression.</li>

    <li>The <code>-Wshadow</code> option now warns if a local variable or
    type declaration shadows another type in C++. Note that the compiler will
    not warn if a local variable shadows a struct/class/enum, but will warn
    if it shadows an explicit typedef. </li>

    <li>When an identifier is not found in the current scope, G++ now
    offers suggestions about which identifier might have been
    intended.</li>

    <li>G++ now issues clearer diagnostics for missing semicolons
    after <code>class</code>, <code>struct</code>,
    and <code>union</code> definitions.</li>

    <li>G++ now issues clearer diagnostics for missing semicolons after
    class member declarations.</li>

    <li>G++ now issues clearer diagnostics when a colon is used in a
    place where a double-colon was intended.</li>

    <li>G++ no longer accepts <code>mutable</code> on reference members
    (<a href="http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33558">c++/33558</a>).
    Use <code>-fpermissive</code> to allow the old, non-conforming behaviour.
    </li>

    <li>A few mangling fixes have been made, to attribute const/volatile on
      function pointer types, decltype of a plain decl, and use of a
      function parameter in the declaration of another parameter.  By
      default the compiler still uses the old mangling, but emits aliases
      with the new mangling on targets that support strong aliases.  Users
      can switch over entirely to the new mangling
      with <code>-fabi-version=5</code> or <code>-fabi-version=0</code>.
      <code>-Wabi</code> will now warn about code that uses the old
      mangling.</li>

    <li>G++ no longer allows objects of const-qualified type to be default
    initialized unless the type has a user-declared default constructor.
    Code that fails to compile can be fixed by providing an initializer e.g.
    <pre>
    struct A { A(); };
    struct B : A { };
    const B b = B();
    </pre>
    Use <code>-fpermissive</code> to allow the old, non-conforming behaviour.
    </li>

  </ul>

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

  <ul>
    <li><a href="http://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html#status.iso.200x">
       Improved experimental support for the upcoming ISO C++ standard, 
       C++0x</a>, including using <code>constexpr</code> and
       <code>nullptr</code>. </li>
    <li>Performance improvements to the
      <a href="http://gcc.gnu.org/onlinedocs/libstdc++/manual/debug_mode.html">Debug
      Mode</a>, thanks to Fran&ccedil;ois Dumont.
    </li>
    <li>Atomic operations used for reference-counting are annotated so that
      they can be understood by race detectors such as Helgrind, see
      <a href="http://gcc.gnu.org/onlinedocs/libstdc++/manual/debug.html#debug.races">Data
      Race Hunting</a>.</li>
    <li>Most libstdc++ standard headers have been changed to no longer include
      the <code>cstddef</code> header as an implementation detail.  Code that
      relied on that header being included as side-effect of including other
      standard headers will need to include <code>cstddef</code> explicitly.</li>
  </ul>

<h3 id="fortran">Fortran</h3>
  <ul>
    <li>On systems supporting the <code>libquadmath</code> library, GNU Fortran
    now also supports a quad-precision, <code>kind=16</code> floating-point
    data type (<code>REAL(16)</code>, <code>COMPLEX(16)</code>). As the data
    type is not fully supported in hardware, calculations might be one to
    two orders of magnitude slower than with the 4, 8 or 10 bytes
    floating-point data types. This change does not affect systems which
    support <code>REAL(16)</code> in hardware nor those which do not support
    <code>libquadmath</code>.</li>
    <li>Much improved compile time for large array constructors.</li>
    <li>In order to reduce execution time and memory consumption, use of
      temporary arrays in assignment expressions is avoided for
      many cases.  The compiler now reverses loops in order to avoid
      generating a temporary array where possible.</li>
    <li>Improved diagnostics, especially with
      <code>-fwhole-file</code>.</li>
    <li>The <code>-fwhole-file</code> flag is now enabled by default. This
    improves code generation and diagnostics. It can be
    disabled using the deprecated <code>-fno-whole-file</code> flag.</li>
    <li>Support the generation of Makefile dependencies via the <a
    href="http://gcc.gnu.org/onlinedocs/gcc/Preprocessor-Options.html">
    <code>-M</code>...</a> flags of GCC; you may need to specify the
    <code>-cpp</code> option in addition. The dependencies take
    modules, Fortran's <code>include</code>, and CPP's <code>#include</code>
    into account. Note: Using <code>-M</code> for the module path is no
    longer supported, use <code>-J</code> instead.</li>
    <li>The flag <code>-Wconversion</code> has been modified to only issue
    warnings where a conversion leads to information loss.  This drastically
    reduces the number of warnings; <code>-Wconversion</code> is thus now
    enabled with <code>-Wall</code>. The flag <code>-Wconversion-extra</code>
    has been added and also warns about other conversions;
    <code>-Wconversion-extra</code> typically issues a huge number of
    warnings, most of which can be ignored.</li>
    <li>A new command-line option <code>-Wunused-dummy-argument</code> warns
	about unused dummy arguments and is included in <code>-Wall</code>.
	Before, <code>-Wunused-variable</code> also warned about unused dummy
	arguments.</li>
    <li>Fortran 2003 support has been extended:
      <ul>
	<li>Improved support for polymorphism between libraries and
	programs and for complicated inheritance patterns (cf. <a
	href="http://gcc.gnu.org/wiki/OOP">object-oriented programming</a>).
	</li>
	<li>Experimental support of the <code>ASSOCIATE</code> construct.</li>
	<li>In pointer assignments it is now possible to specify the lower
	bounds of the pointer and, for a rank-1 or a simply contiguous
	data-target, to remap the bounds.</li>
	<li>Automatic (re)allocation: In intrinsic assignments to
	allocatable variables the left-hand side will be automatically
	allocated (if unallocated) or reallocated (if the shape or type
	parameter is different). To avoid the small performance penalty,
	you can use <code>a(:) = ...</code> instead of <code>a = ...</code>
        for arrays and character strings &ndash; or disable the feature using
        <code>-std=f95</code> or <code>-fno-realloc-lhs</code>.</li>
	<li>Deferred type parameter: For scalar allocatable and pointer
	variables the character length can be deferred.</li>
	<li>Namelist variables with allocatable and pointer attribute and
	nonconstant length type parameter are supported.</li>
      </ul>
    </li>
    <li>Fortran 2008 support has been extended:
      <ul>
	<li>Experimental <a href="http://gcc.gnu.org/wiki/Coarray">
	coarray support</a> (for one image only, i.e. <code>num_images() ==
	1</code>); use the <a
	href="http://gcc.gnu.org/onlinedocs/gfortran/Code-Gen-Options.html#index-g_t_0040code_007bfcoarray_007d-233">
	<code>-fcoarray=single</code></a> flag to enable it.</li>
	<li>The <code>STOP</code> and the new <code>ERROR STOP</code>
	statements now support all constant expressions.</li>
	<li>Support for the <code>CONTIGUOUS</code> attribute.</li>
	<li>Support for <code>ALLOCATE</code> with <code>MOLD</code>.</li>
	<li>Support for the <code>STORAGE_SIZE</code> intrinsic inquiry
	function.</li>
	<li>Support of the <code>NORM2</code> and <code>PARITY</code>
	intrinsic functions.</li>
	<li>The following bit intrinsics were added: <code>POPCNT</code>
	and <code>POPPAR</code> for counting the number of 1 bits and
	returning the parity; <code>BGE</code>, <code>BGT</code>,
	<code>BLE</code>, and <code>BLT</code> for bitwise comparisons;
	<code>DSHIFTL</code> and <code>DSHIFTR</code> for combined left
	and right shifts, <code>MASKL</code> and <code>MASKR</code> for
	simple left and right justified masks, <code>MERGE_BITS</code>
	for a bitwise merge using a mask, <code>SHIFTA</code>,
	<code>SHIFTL</code> and <code>SHIFTR</code> for shift operations,
	and the transformational bit intrinsics <code>IALL</code>,
	<code>IANY</code> and <code>IPARITY</code>.</li>
	<li>Support of the <code>EXECUTE_COMMAND_LINE</code> intrinsic
	subroutine.</li>
	<li>Support for the <code>IMPURE</code> attribute for procedures,
	which allows for <code>ELEMENTAL</code> procedures without the
	restrictions of <code>PURE</code>.</li>
	<li>Null pointers (including <code>NULL()</code>) and not
	allocated variables can be used as actual argument to optional
	non-pointer, non-allocatable dummy arguments, denoting an absent
	argument.</li>
	<li>Non-pointer variables with <code>TARGET</code> attribute can
	be used as actual argument to <code>POINTER</code> dummies with
	<code>INTENT(IN)</code></li>
	<li>Pointers including procedure pointers and those in a derived
	type (pointer components) can now be initialized by a target
	instead of only by <code>NULL</code>.</li>
	<li>The <code>EXIT</code> statement (with construct-name) can
	now be used to leave not only the <code>DO</code> but also the
	<code>ASSOCIATE</code>, <code>BLOCK</code>, <code>IF</code>,
	<code>SELECT CASE</code> and <code>SELECT TYPE</code> constructs.</li>
	<li>Internal procedures can now be used as actual argument.</li>
	<li>The named constants <code>INTEGER_KINDS</code>,
	<code>LOGICAL_KINDS</code>, <code>REAL_KINDS</code> and
	<code>CHARACTER_KINDS</code> of the intrinsic module
	<code>ISO_FORTRAN_ENV</code> have been added; these arrays contain
	the supported kind values for the respective types.</li>
	<li>The module procedures <code>C_SIZEOF</code> of the intrinsic
	module <code>ISO_C_BINDINGS</code> and <code>COMPILER_VERSION</code>
	and <code>COMPILER_OPTIONS</code> of <code>ISO_FORTRAN_ENV</code>
	have been implemented.</li>
	<li>Minor changes: obsolescence diagnostics for <code>ENTRY</code>
	was added for <code>-std=f2008</code>;
	a line may start with a semicolon;
	for internal and module procedures <code>END</code> can be used
	instead of <code>END SUBROUTINE</code> and <code>END
	FUNCTION</code>; <code>SELECTED_REAL_KIND</code> now also takes a
	<code>RADIX</code> argument; intrinsic types are supported for
	<code>TYPE(<i>intrinsic-type-spec</i>)</code>; multiple type-bound
	procedures can be declared in a single <code>PROCEDURE</code>
	statement; implied-shape arrays are supported for named constants
	(<code>PARAMETER</code>). The transformational, three argument
	versions of <code>BESSEL_JN</code> and <code>BESSEL_YN</code>
	were added &ndash; the elemental, two-argument version had been
	added in GCC 4.4; note that the transformational functions use
	a recurrence algorithm.</li>
      </ul>
    </li>
  </ul>

<h3 id="go">Go</h3>

<p>Support for the <a href="http://golang.org/">Go programming
    language</a> has been added to GCC.  It is not enabled by default
    when you build GCC; use the <code>--enable-languages</code>
    configure option to build it.  The driver program for compiling Go
    code is <code>gccgo</code>.</p>

<p>Go is currently known to work on GNU/Linux and RTEMS.  Solaris
  support is in progress.  It may or may not work on other
  platforms.</p>

<h3>Java (GCJ)</h3>

<h3 id="objective-c">Objective-C and Objective-C++</h3>

  <ul>
    <li>The <code>-fobjc-exceptions</code> flag is now required to
    enable Objective-C exception and synchronization syntax
    (introduced by the keywords <code>@try</code>,
    <code>@catch</code>, <code>@finally</code> and
    <code>@synchronized</code>).</li>

    <li>A number of Objective-C 2.0 features and extensions are now
    supported by GCC.  These features are enabled by default; you can
    disable them by using the new <code>-fobjc-std=objc1</code>
    command-line option.</li>

    <li>The Objective-C 2.0 dot-syntax is now supported.  It is an
    alternative syntax for using getters and setters;
    <code>object.count</code> is automatically converted into
    <code>[object count]</code> or <code>[object setCount: ...]</code>
    depending on context; for example <code>if (object.count >
    0)</code> is automatically compiled into the equivalent of
    <code>if ([object count] > 0)</code> while <code>object.count =
    0;</code> is automatically compiled into the equivalent ot
    <code>[object setCount: 0];</code>.  The dot-syntax can be used
    with instance and class objects and with any setters or getters,
    no matter if they are part of a declared property or not.</li>

    <li>Objective-C 2.0 declared properties are now supported.  They
    are declared using the new <code>@property</code> keyword, and are
    most commonly used in conjunction with the new Objective-C 2.0
    dot-syntax.  The <code>nonatomic</code>, <code>readonly</code>,
    <code>readwrite</code>, <code>assign</code>, <code>retain</code>,
    <code>copy</code>, <code>setter</code> and <code>getter</code>
    attributes are all supported.  Marking declared properties with
    <code>__attribute__ ((deprecated))</code> is supported too.</li>

    <li>The Objective-C 2.0 <code>@synthesize</code> and
    <code>@dynamic</code> keywords are supported.
    <code>@synthesize</code> causes the compiler to automatically
    synthesize a declared property, while <code>@dynamic</code> is
    used to disable all warnings for a declared property for which no
    implementation is provided at compile time.  Synthesizing declared
    properties requires runtime support in most useful cases; to be
    able to use it with the GNU runtime, appropriate helper functions
    have been added to the GNU Objective-C runtime ABI, and are
    implemented by the GNU Objective-C runtime library shipped with
    GCC.</li>

    <li>The Objective-C 2.0 fast enumeration syntax is supported in
    Objective-C.  This is currently not yet available in
    Objective-C++.  Fast enumeration requires support in the runtime,
    and such support has been added to the GNU Objective-C runtime
    library (shipped with GCC).</li>

    <li>The Objective-C 2.0 <code>@optional</code> keyword is
    supported.  It allows you to mark methods or properties in a
    protocol as optional as opposed to required.</li>

    <li>The Objective-C 2.0 <code>@package</code> keyword is
    supported.  It has currently the same effect as the
    <code>@public</code> keyword.</li>

    <li>Objective-C 2.0 method attributes are supported.  Currently
    the supported attributes are <code>deprecated</code>,
    <code>sentinel</code>, <code>noreturn</code> and
    <code>format</code>.</li>

    <li>Objective-C 2.0 method argument attributes are supported.  The
    most widely used attribute is <code>unused</code>, to mark an
    argument as unused in the implementation.</li>

    <li>Objective-C 2.0 class and protocol attributes are supported.
    Currently the only supported attribute is
    <code>deprecated</code>.</li>

    <li>Objective-C 2.0 class extensions are supported.  A class
    extension has the same syntax as a category declaration with no
    category name, and the methods and properties declared in it are
    added directly to the main class.  It is mostly used as an
    alternative to a category to add methods to a class without
    advertising them in the public headers, with the advantage that
    for class extensions the compiler checks that all the privately
    declared methods are actually implemented.</li>

    <li>As a result of these enhancements, GCC can now be used to
    build Objective-C and Objective-C++ software that uses Foundation
    and other important system frameworks with the NeXT runtime on
    Darwin 9 and Darwin 10 (Mac OS X 10.5 and 10.6).  Currently this is for
    m32 code only.</li>

    <li>Many bugs in the compiler have been fixed in this release; in
    particular, LTO can now be used when compiling Objective-C and
    Objective-C++ and the parser is much more robust in dealing with
    invalid code.</li>
  </ul>

  <h4>Runtime Library (libobjc)</h4>

  <ul>
    <li>The GNU Objective-C runtime library now defines the macro
    <code>__GNU_LIBOBJC__</code> (with a value that is increased at
    every release where there is any change to the API) in
    <code>objc/objc.h</code>, making it easy to determine if the GNU
    Objective-C runtime library is being used, and if so, which
    version.  Previous versions of the GNU Objective-C runtime library
    (and other Objective-C runtime libraries such as the Apple one) do
    not define this macro.</li>

    <li>A new Objective-C 2.0 API, almost identical to the one
    implemented by the Apple Objective-C runtime, has been implemented
    in the GNU Objective-C runtime library.  The new API hides the
    internals of most runtime structures but provides a more extensive
    set of functions to operate on them.  It is much easier, for
    example, to create or modify classes at runtime.  The new API also
    makes it easier to port software from Apple to GNU as almost no
    changes should be required.  The old API is still supported for
    backwards compatibility; including the old
    <code>objc/objc-api.h</code> header file automatically selects the
    old API, while including the new <code>objc/runtime.h</code>
    header file automatically selects the new API.  Support for the
    old API is being phased out and upgrading the software to use the
    new API is strongly recommended.  To check for the availability of
    the new API, the <code>__GNU_LIBOBJC__</code> macro can be used as
    older versions of the GNU Objective-C runtime library, which do
    not support the new API, do not define such a macro.</li>

    <li>Runtime support for <code>@synchronized</code> has been added.</li>

    <li>Runtime support for Objective-C 2.0 synthesized property
    accessors has been added.</li>

    <li>Runtime support for Objective-C 2.0 fast enumeration has been
    added.</li>
  </ul>

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

<h3 id="arm">ARM</h3>
    <ul>
    <li>GCC now supports the Cortex-M4 processor implementing
      the v7-em version of the architecture using the option
      <code>-mcpu=cortex-m4</code>.</li>

    <li>Scheduling descriptions for the Cortex-M4, the Neon and
      the floating point units of the Cortex-A9 and a pipeline 
      description for the Cortex-A5 have been added.</li>

    <li>Synchronization primitives such as <code>__sync_fetch_and_add
      </code> and friends are now inlined for supported architectures
      rather than calling into a kernel helper function.</li>

    <li>SSA loop prefetching is enabled by default for the
      Cortex-A9 at <code>-O3</code>.</li>

    <li>Several improvements were committed to improve code
      generation for the ARM architecture including a rewritten
      implementation for load and store multiples.</li>

    <li>Several enhancements were committed to improve SIMD code
      generation for NEON by adding support for widening instructions,
      misaligned loads and stores, vector conditionals and
      support for 64 bit arithmetic.</li>

    <li>Support was added for the Faraday cores fa526, fa606te,
      fa626te, fmp626te, fmp626 and fa726te and can be used with the
      respective names as parameters to the <code>-mcpu=</code>
      option.</li>

    <li>Basic support was added for Cortex-A15 and is available through
      <code>-mcpu=cortex-a15</code>.</li>

    <li>GCC for AAPCS configurations now more closely adheres to the AAPCS
      specification by enabling <code>-fstrict-volatile-bitfields</code> by
      default.</li>
  </ul>

<h3>IA-32/x86-64</h3>
  <ul>
    <li>
      The new <code>-fsplit-stack</code> option permits programs to
      use a discontiguous stack.  This is useful for threaded
      programs, in that it is no longer necessary to specify the
      maximum stack size when creating a thread.  This feature is
      currently only implemented for 32-bit and 64-bit x86 GNU/Linux
      targets.
    </li>
	<li>Support for emitting profiler counter calls before function
	prologues.  This is enabled via a new command-line option
	<code>-mfentry</code>.</li>
    <li>Optimization for the Intel Core 2 processors is now available through
	the <code>-march=core2</code> and <code>-mtune=core2</code>
	options.</li>
    <li>Support for Intel Core i3/i5/i7 processors is now available through
	the <code>-march=corei7</code> and <code>-mtune=corei7</code>
	options.</li>
    <li>Support for Intel Core i3/i5/i7 processors with AVX is now
	available through the <code>-march=corei7-avx</code> and
	<code>-mtune=corei7-avx</code> options.</li>
    <li>Support for AMD Bobcat (family 14) processors is now available through
	the <code>-march=btver1</code> and <code>-mtune=btver1</code>
	options.</li>
    <li>The default setting (when not optimizing for size) for 32-bit
      GNU/Linux and Darwin x86 targets has been changed to
      <code>-fomit-frame-pointer</code>.  The default can be reverted
      to <code>-fno-omit-frame-pointer</code> by configuring GCC with
      the <code>--enable-frame-pointer</code> configure option.</li>
    <li>Darwin, FreeBSD, Solaris 2, MinGW and Cygwin now all support
      <code>__float128</code> on 32-bit and 64-bit x86 targets.</li>
    <li>AVX floating-point arithmetic can now be enabled by default at
      configure time with the new <code>--with-fpmath=avx</code> option.</li>
    <li>The SSA loop prefetching pass is enabled when
      using <code>-O3</code> when optimizing for CPUs where prefetching
      is beneficial (AMD CPUs newer than K6).</li>
    <li>Support for TBM (Trailing Bit Manipulation) built-in functions
      and code generation is available via <code>-mtbm</code>.</li>
    <li>Support for AMD's BMI (Bit Manipulation) built-in functions and
      code generation is available via <code>-mbmi</code>.</li>
  </ul>

<h3 id="microblaze">MicroBlaze</h3>

  <ul>
    <li>Support has been added for the Xilinx MicroBlaze softcore processor
     (microblaze-elf) embedded target.  This configurable processor is
     supported on several Xilinx Spartan and Virtex FPGAs.
    </li>
  </ul>

<h3>MIPS</h3>
  <ul>
    <li>GCC now supports the Loongson 3A processor.  Its canonical
	<code>-march=</code> and <code>-mtune=</code> name is
	<code>loongson3a</code>.
    </li>
  </ul>

<h3 id="mn10300">MN10300 / AM33</h3>
  <ul>
    <li>The inline assembly register constraint <code>"A"</code> has
        been renamed <code>"c"</code>.  This constraint is used to
        select a floating-point register that can be used as the
        destination of a multiply-accumulate instruction.
    </li>
    <li>New inline assembly register constraints <code>"A"</code> and
        <code>"D"</code> have been added.  These constraint letters
        resolve to all general registers when compiling for AM33, and
        resolve to address registers only or data registers only when
        compiling for MN10300.
    </li>
    <li>The <code>MDR</code> register is represented in the compiler.
        One can access the register via the <code>"z"</code> constraint
        in inline assembly.  It can be marked as clobbered or used as
        a local register variable via the <code>"mdr"</code> name.
        The compiler uses the <code>RETF</code> instruction if the
        function does not modify the <code>MDR</code> register, so it
        is important that inline assembly properly annotate any usage
        of the register.
    </li>
  </ul>

<h3>PowerPC/PowerPC64</h3>
  <ul>
    <li>GCC now supports the Applied Micro Titan processor
    with <code>-mcpu=titan</code>.</li>
    <li>The <code>-mrecip</code> option has been added, which indicates
    whether the reciprocal and reciprocal square root instructions
    should be used.</li>
    <li>The <code>-mveclibabi=mass</code> option can be used to enable
    the compiler to autovectorize mathematical functions using the
    Mathematical Acceleration Subsystem library.</li>
    <li>The <code>-msingle-pic-base</code> option has been added, which
    instructs the compiler to avoid loading the PIC base register in
    function prologues.  The PIC base register must be initialized by
    the runtime system.</li>
    <li>The <code>-mblock-move-inline-limit</code> option has been
    added, which enables the user to control the maximum size of
    inlined <code>memcpy</code> calls and similar.</li>
    <li>PowerPC64 GNU/Linux support for applications requiring a large
      TOC section has been improved.  A new command-line option,
      <code>-mcmodel=MODEL</code>, controls this feature; valid values
      for <code>MODEL</code>
      are <code>small</code>, <code>medium</code>,
      or <code>large</code>.</li>
    <li>The altivec builtin functions <code>vec_ld</code> and <code>vec_st</code>
      have been modified to generate the Altivec memory instructions
      <code>LVX</code> and <code>STVX</code>, even if the <code>-mvsx</code>
      option is used.  In the initial GCC 4.5 release, these builtin functions
      were changed to generate VSX memory reference instructions instead of
      Altivec memory instructions, but there are differences between the two
      instructions.  If the VSX instruction set is available, you can now use
      the new builtin functions <code>vec_vsx_ld</code> and <code>vec_vsx_st</code>
      which always generates the VSX memory instructions.</li>
    <li>The GCC compiler on AIX now defaults to a process layout with a
      larger data space allowing larger programs to be compiled.</li>
    <li>The GCC long double type on AIX 6.1 and above has reverted to 64 bit
      double precision, matching the AIX XL compiler default, because of
      missing C99 symbols required by the GCC runtime.</li>
    <li>The default processor scheduling model and tuning for PowerPC64
      GNU/Linux and for AIX 6.1 and above now is POWER7.</li>
  </ul>

<h3>S/390, zSeries and System z9/z10, IBM zEnterprise z196</h3>
  <ul>
    <li>Support for the zEnterprise z196 processor has been added.
      When using the <code>-march=z196</code> option, the compiler
      will generate code making use of the following instruction
      facilities:
      <ul>
	<li>Conditional load/store</li>
	<li>Distinct-operands</li>
	<li>Floating-point-extension</li>
	<li>Interlocked-access</li>
	<li>Population-count</li>
      </ul>
      The <code>-mtune=z196</code> option avoids the compare and
      branch instructions as well as the load address instruction
      with an index register as much as possible and performs
      instruction scheduling appropriate for the new out-of-order
      pipeline architecture.</li>
    <li>When using the <code>-m31 -mzarch</code> options the generated
      code still conforms to the 32-bit ABI but uses the general
      purpose registers as 64-bit registers internally.  This
      requires a Linux kernel saving the whole 64-bit registers when
      doing a context switch.  Kernels providing that feature
      indicate that by the 'highgprs' string
      in <code>/proc/cpuinfo</code>.</li>
    <li>The SSA loop prefetching pass is enabled when
      using <code>-O3</code>.</li>
  </ul>

<h3>SPARC</h3>
  <ul>
    <li>GCC now supports the LEON series of SPARC V8 processors.  The
     code generated by the compiler can either be tuned to it by means
     of the <code>--with-tune=leon</code> configure option and
     <code>-mtune=leon</code> compilation option, or the compiler can
     be built for the <code>sparc-leon-{elf,linux}</code> and
     <code>sparc-leon3-{elf,linux}</code> targets directly.</li>
    <li>GCC has stopped sign/zero-extending parameter registers in the
     callee for functions taking parameters with sub-word size in 32-bit
     mode, since this is redundant with the specification of the ABI.
     GCC has never done so in 64-bit mode since this is also redundant.</li>
  </ul>

<h2 id="os">Operating Systems</h2>

<h3 id="android">Android</h3>
  <ul>
    <li>GCC now supports the Bionic C library and provides a convenient
      way of building native libraries and applications for the Android
      platform.
      Refer to the documentation of the <code>-mandroid</code> and
      <code>-mbionic</code> options for details on building native code.
      At the moment, Android support is enabled only for ARM.</li>
  </ul>

<h3 id="darwin">Darwin/Mac OS X</h3>
  <ul>
      <li>General 
       <ul>
	<li>Initial support for <code>CFString</code> types has been
	added.<br /> This allows GCC to build projects including the system
	<em>Core Foundation</em> frameworks. The GCC Objective-C family
	supports <code>CFString</code> "toll-free bridged" as per the Mac 
	OS X system tools. <code>CFString</code> is also recognized in the
	context of <code>format</code> attributes and arguments (see the
	documentation for <code>format</code> attributes for limitations).
	At present, 8-bit character types are supported.</li>
	<li>LTO-support.<br />Darwin has benefited from ongoing work on
	LTO; support for this is now stable and enabled by default.</li>
	<li>Object file size reduction.<br /> The Darwin zeroed memory
	allocators have been re-written to make more use of
	<code>.zerofill</code> sections.  For non-debug code, this can
	reduce object file size significantly.</li>
       </ul></li>
      <li>x86 Architecture 
       <ul>
	<li>The <code>-mdynamic-no-pic</code> option has been 
	enabled.<br />Code supporting <code>-mdynamic-no-pic</code>
	optimization has been added and is applicable to <code>-m32</code>
	builds. The compiler bootstrap uses the option where 
	appropriate.</li>
	<li>The default value for <code>-mtune=</code> has been 
	changed.<br />Since Darwin systems are primarily Xeon, Core-2 or
	similar the default tuning has been changed to 
	<code>-mtune=core2</code>.</li>
      </ul></li>
    <li>PPC Architecture
      <ul>
       <li>Darwin64 ABI.<br />Several significant bugs have been fixed,
	such that GCC now produces code compatible with the Darwin64
	PowerPC ABI.</li>
	<li>libffi and boehm-gc.<br />The Darwin ports of the libffi and
	boehm-gc libraries have been upgraded to include a Darwin64
	implementation. This means that powerpc*-*-darwin9 platforms may
	now, for example, build Java applications with <code>-m64</code>
	enabled.</li>
	<li>Plug-in support has been enabled.</li>
	<li>The <code>-fsection-anchors</code> option is now available
	although, presently, not heavily tested.</li>
      </ul></li>
  </ul>

<h3 id="solaris">Solaris 2</h3>
    <h4>New Features</h4>
    <ul>
      <li>Support symbol versioning with the Sun linker.</li>
      <li>Allow libstdc++ to leverage full ISO C99 support on
        Solaris 10+.</li>
      <li>Support thread-local storage (TLS) with the Sun assembler on
	Solaris 2/x86.</li>
      <li>Support TLS on Solaris 8/9 if prerequisites are met.</li>
      <li>Support COMDAT group with the GNU assembler and recent Sun
	linker.</li>
      <li>Support the Sun assembler visibility syntax.</li>
      <li>Default Solaris 2/x86 to <code>-march=pentium4</code> (Solaris
	10+) resp. <code>-march=pentiumpro</code> (Solaris 8/9).</li>
      <li>Don't use SSE on Solaris 8/9 x86 by default.</li>
      <li>Enable 128-bit long double (<code>__float128</code>) support on
	Solaris 2/x86.</li>
      </ul>
    <h4>ABI Change</h4>
    <ul>
      <li>Change the ABI for returning 8-byte vectors like
	<code>__m64</code> in MMX registers on Solaris 10+/x86 to match the
	Sun Studio 12.1+ compilers.  <em>This is an incompatible change.
	If you use such types, you must either recompile all your code with
	the new compiler or use the new <code>-mvect8-ret-in-mem</code>
	option to remain compatible with previous versions of GCC and
	Sun Studio.</em></li>
    </ul>

<h3 id="windows">Windows x86/x86_64</h3>
  <ul>
    <li>Initial support for decimal floating point.</li>
    <li>Support for the <code>__thiscall</code> calling-convention.</li>
    <li>Support for hot-patchable function prologues via the
    <code>ms_hook_prologue</code> attribute for x86_64 in addition to 32-bit x86.</li>
    <li>Improvements of stack-probing and stack-allocation mechanisms.</li>
    <li>Support of push/pop-macro pragma as preprocessor command.<br />
    With <code>#pragma push_macro(&quot;macro-name&quot;)</code> the
    current definition of <code>macro-name</code> is saved and can be
    restored with <code>#pragma pop_macro(&quot;macro-name&quot;)</code>
    to its saved definition.</li>
    <li>Enable 128-bit long double (<code>__float128</code>) support on
      MinGW and Cygwin.</li>
  </ul>

<h2>Documentation improvements</h2>

<h2>Other significant improvements</h2>

<h3>Installation changes</h3>

  <ul>
    <li>
    An <code>install-strip</code> <code>make</code> target is provided
    that installs stripped executables, and may install libraries with
    unneeded or debugging sections stripped.
    </li>

    <li>
    On Power7 systems, there is a potential problem if you build the GCC
    compiler with a host compiler using options that enables the VSX
    instruction set generation.  If the host compiler has been patched so that
    the <code>vec_ld</code> and <code>vec_st</code> builtin functions
    generate Altivec memory instructions instead of VSX memory instructions,
    then you should be able to build the compiler with VSX instruction
    generation.
    </li>
  </ul>

<h2>Changes for GCC Developers</h2>

<p>Note: these changes concern developers that develop GCC itself or
  software that integrates with GCC, such as plugins, and not the
  general GCC users.</p>

<ul>
  <li>
    The <code>gengtype</code> utility, which previously was internal to
    the GCC build process, has been enchanced to provide GC root
    information for plugins as necessary.
  </li>
  <li>
    The old GC allocation interface of <code>ggc_alloc</code> and
    friends was replaced with a type-safe alternative.
  </li>
</ul>




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

<div class="copyright">

<p>Please send FSF &amp; GNU inquiries &amp; questions to
<a href="mailto:gnu@gnu.org">gnu@gnu.org</a>.
There are also <a href="http://www.gnu.org/home.html#ContactInfo">other ways
to contact</a> the FSF.</p>

<p>These pages are <a href="http://gcc.gnu.org/about.html">maintained by
the GCC team</a>.</p>

<address>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.<br />
Please send comments on these web pages and the development of GCC to our
developer list at <a href="mailto:gcc@gcc.gnu.org">gcc@gcc.gnu.org</a>.
All of our lists have
<a href="http://gcc.gnu.org/lists.html">public archives</a>.
</address>

<p>Copyright (C) Free Software Foundation, Inc.<br />
Verbatim copying and distribution of this entire article is
permitted in any medium, provided this notice is preserved.</p>

<table width="100%" border="0"><tr><td>
  <!-- IGNORE DIFF -->Last modified 2011-03-22
</td><td align="right" valign="bottom">
  <a href="http://validator.w3.org/check/referer">
      <img src="http://www.w3.org/Icons/valid-xhtml10"
     alt="Valid XHTML 1.0" border="0" width="88" height="31" />
   
  </a>
</td></tr></table>

</div>

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

</body>
     </html>
gcc-4.6 4.6.3-1ubuntu5 / usr / share / doc / gcc-4.6-base / NEWS.html
