(wwwdocs) changes.html vs. index.html
Christian Cornelssen
ccorn@cs.tu-berlin.de
Fri Feb 7 13:53:00 GMT 2003
On Fri, 7 Feb 2003, Gerald Pfeifer wrote:
> The downgrading of <hx> to <h(x+1)> is fine, as is moving the first two
> paragraphs from changes.html to index.html; please go ahead and commit
> these.
I have checked in the move (two paragraphs only). For the downgrading in
`changes.html', I have to offer these refinements:
* Group the initial (general) sections under the headline
"Important Characteristics" and downgrade their own headers accordingly.
* Downgrade the headers of the release sections such that
there is only one <h1> in the page which applies to the whole body.
* Remove the outer <ul>s so that there are no longer <li><h4> constructs.
(These have always looked bad with my browsers because there was no
vertical space above the headers, but below them.)
* Separate the four main sections with <hr /> to help scrolling
readers avoid overshooting. (Since there is no actual "story flow"
between the sections, inserting such barriers is ok.)
The result looks nice. I have attached the corresponding patch.
Regards,
Christian Cornelssen
-------------- next part --------------
--- htdocs/gcc-3.2/changes.html_ Fri Feb 7 14:12:43 2003
+++ htdocs/gcc-3.2/changes.html Fri Feb 7 13:22:35 2003
@@ -12,7 +12,9 @@
<h1>GCC 3.2 Release Series — Changes, New Features, and Fixes</h1>
-<h2>Caveats</h2>
+<h2>Important Characteristics</h2>
+
+<h3>Caveats</h3>
<ul>
<li>The C++ compiler does not correctly zero-initialize
pointers-to-data members. You must explicitly initialize
@@ -28,9 +30,9 @@
and GCC 3.1.1.</li>
</ul>
-<h2>Frontend Enhancements</h2>
-<ul>
-<li><h3>C/C++/Objective-C</h3>
+<h3>Frontend Enhancements</h3>
+
+<h4>C/C++/Objective-C</h4>
<ul>
<li>The method of constructing the list of directories to be searched
for header files has been revised. If a directory named by a
@@ -47,8 +49,9 @@
caused the compiler to crash since version 3.0 and no one
noticed until very recently. Thus we conclude it is not in
widespread use.)</li>
- </ul></li>
-<li><h3>C++</h3>
+ </ul>
+
+<h4>C++</h4>
<ul>
<li>There are now no known differences between the C++ ABI
implemented in GCC and the multi-vendor standard. We believe that
@@ -62,11 +65,10 @@
GNU/Linux systems, check the <a href="c++-abi.html">common C++ ABI
page</a>.</li>
</ul>
-</li></ul>
-<h2>New Targets and Target Specific Improvements</h2>
-<ul>
-<li><h3>IA-32</h3>
+<h3>New Targets and Target Specific Improvements</h3>
+
+<h4>IA-32</h4>
<ul>
<li>Fixed a number of bugs in SSE and MMX intrinsics.</li>
<li>Fixed common compiler crashes with SSE instruction set enabled
@@ -74,8 +76,9 @@
<code>athlon-xp</code>)</li>
<li><code>__m128</code> and <code>__m128i</code> is not 128bit
aligned when used in structures.</li>
- </ul></li>
-<li><h3>x86-64</h3>
+ </ul>
+
+<h4>x86-64</h4>
<ul>
<li>A bug whereby the compiler could generate bad code for
<code>bzero</code> has been fixed.</li>
@@ -83,11 +86,11 @@
corner cases)</li>
<li>Fixed prefetch code generation</li>
</ul>
-</li></ul>
<!-- Changes for later point releases should be added at the top. -->
-<h1><a name="3.2.2">GCC 3.2.2</a></h1>
+<hr />
+<h2><a name="3.2.2">GCC 3.2.2</a></h2>
<p>Beginning with 3.2.2, GCC's Makefile suite supports redirection of
<code>make install</code> by means of the <code>DESTDIR</code>
@@ -97,7 +100,7 @@
<p>Other than that, GCC 3.2.2 is a bug fix release only; there are no
new features that were not present in GCC 3.2.1.</p>
-<h2>Bug Fixes</h2>
+<h3>Bug Fixes</h3>
<p>This section lists the problem reports (PRs) from GCC's bug tracking
system (GNATS) that are known to be fixed in the 3.2.2 release. This
@@ -107,8 +110,7 @@
<p>The sorting is slightly different from the GNATS categories, and some of
the titles have been changed to make them more clear.</p>
-<ul>
-<li><h3>Internal Compiler Errors (multi-platform)</h3>
+<h4>Internal Compiler Errors (multi-platform)</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5919">5919</a>: (c++) ICE when passing variable array to template function</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7129">7129</a>: (c++) ICE with min/max assignment operators (<code><?=</code> and <code>>?=</code>)</li>
@@ -127,8 +129,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8799">8799</a>: (c++) ICE: error reporting routines re-entered</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9328">9328</a>: (c++) ICE with <code>typeof(X)</code> for overloaded <code>X</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9465">9465</a>: (preprocessor) <code>cpp -traditional</code> ICE on null bytes</li>
-</ul></li>
-<li><h3>C++ (compiler and library) bugs</h3>
+</ul>
+
+<h4>C++ (compiler and library) bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=47">47</a>: scoping in nested classes is broken</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6745">6745</a>: problems with iostream rdbuf() member function</li>
@@ -150,8 +153,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9269">9269</a>: libstdc++ headers: explicit specialization of function must precede its first use</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9322">9322</a>: return value of <code>basic_streambuf<>::getloc</code> affected by <code>locale::global</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9433">9433</a>: segfault in runtime support for <code>dynamic_cast</code></li>
-</ul></li>
-<li><h3>C and optimizer bugs</h3>
+</ul>
+
+<h4>C and optimizer bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8032">8032</a>: GCC incorrectly initializes static structs that have flexible arrays</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8639">8639</a>: simple arithmetic expression broken</li>
@@ -159,55 +163,64 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8832">8832</a>: traditional "<code>asm volatile</code>" code is illegally optimized</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8988">8988</a>: loop optimizer bug: with <code>-O2</code>, code is generated that segfaults (found on i386, bug present for all platforms)</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9492">9492</a>: structure copy clobbers subsequent stores to structure</li>
-</ul></li>
-<li><h3>Objective-C bugs</h3>
+</ul>
+
+<h4>Objective-C bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9267">9267</a>: Objective-C parser won't build with newer bison versions (e.g. 1.875)</li>
-</ul></li>
-<li><h3>Ada bugs</h3>
+</ul>
+
+<h4>Ada bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8344">8344</a>: Ada build problem due to conflict between <code>gcc/final.o</code>, <code>gcc/ada/final.o</code></li>
-</ul></li>
-<li><h3>Preprocessor bugs</h3>
+</ul>
+
+<h4>Preprocessor bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8524">8524</a>: <code>_Pragma</code> within macros is improperly expanded</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8880">8880</a>: <code>__WCHAR_TYPE__</code> macro incorrectly set to "<code>long int</code>" with <code>-fshort-wchar</code></li>
-</ul></li>
-<li><h3>ARM-specific</h3>
+</ul>
+
+<h4>ARM-specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9090">9090</a>: arm ICE with <code>>= -O2</code>; regression from gcc-2.95</li>
-</ul></li>
-<li><h3>x86-specific (Intel/AMD)</h3>
+</ul>
+
+<h4>x86-specific (Intel/AMD)</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8588">8588</a>: ICE in <code>extract_insn</code>, at <code>recog.c:NNNN</code> (shift instruction)</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8599">8599</a>: loop unroll bug with <code>-march=k6-3</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9506">9506</a>: ABI breakage in structure return (affects BSD and Cygwin, but not GNU/Linux)</li>
-</ul></li>
-<li><h3>FreeBSD 5.0 specific</h3>
+</ul>
+
+<h4>FreeBSD 5.0 specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9484">9484</a>: GCC 3.2.1 Bootstrap failure on FreeBSD 5.0</li>
-</ul></li>
-<li><h3>RTEMS-specific</h3>
+</ul>
+
+<h4>RTEMS-specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9292">9292</a>: hppa1.1-rtems configurery problems</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9293">9293</a>: [m68k-elf/rtems] <code>config/m68k/t-crtstuff</code> bug</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9295">9295</a>: [mips-rtems] <code>config/mips/rtems.h</code> <code>init</code>/<code>fini</code> issue</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9296">9296</a>: gthr-rtems regression</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9316">9316</a>: powerpc-rtems: extending multilibs</li>
-</ul></li>
-<li><h3>HP-PA specific</h3>
+</ul>
+
+<h4>HP-PA specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=9493">9493</a>: ICE with <code>-O2</code> when building a simple function</li>
-</ul></li>
-<li><h3>Documentation</h3>
+</ul>
+
+<h4>Documentation</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7341">7341</a>: hyperlink to <code>gcov</code> in GCC documentation doesn't work</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8947">8947</a>: Please add a warning about "<code>-malign-double</code>" in docs</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7448">7448</a>, <a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8882">8882</a>: typo cleanups</li>
-</ul></li>
</ul>
-<h1><a name="3.2.1">GCC 3.2.1</a></h1>
+<hr />
+<h2><a name="3.2.1">GCC 3.2.1</a></h2>
<p>3.2.1 adds a new warning, <code>-Wabi</code>. This option warns when GNU
C++ generates code that is known not to be binary-compatible with the
@@ -228,7 +241,7 @@
applications) was reverted ("unfixed"), because the "fix" was not
thread-safe.</p>
-<h2>Bug Fixes</h2>
+<h3>Bug Fixes</h3>
<p>This section lists the problem reports (PRs) from GCC's bug tracking
system (<a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl">GNATS</a>) that
@@ -238,8 +251,7 @@
strongly recommended that users of earlier gcc 3.x releases upgrade to
GCC 3.2.1.</p>
-<ul>
-<li><h3>Internal Compiler Errors (multi-platform)</h3>
+<h4>Internal Compiler Errors (multi-platform)</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=2521">2521</a>: (c++) ICE in <code>build_ptrmemfunc</code>, in <code>cp/typeck.c</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5661">5661</a>: (c++) ICE instantiating template on array of unknown size (bad code)</li>
@@ -263,9 +275,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8134">8134</a>: (c++) ICE in <code>force_store_init_value</code> on legal code</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8149">8149</a>: (c++) ICE on incomplete type</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8160">8160</a>: (c++) ICE in <code>build_modify_expr</code>, in <code>cp/typeck.c</code>: array initialization</li>
-</ul></li>
+</ul>
-<li><h3>C++ (compiler and library) bugs</h3>
+<h4>C++ (compiler and library) bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5607">5607</a>: No pointer adjustment in covariant return types</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6579">6579</a>: Infinite loop with statement expressions in member initialization</li>
@@ -290,8 +302,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8347">8347</a>: empty vector range used in string construction causes core dump</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8348">8348</a>: <code>fail()</code> flag is set in <code>istringstream</code> when <code>eof()</code> flag is set</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8391">8391</a>: regression: infinite loop in <code>cp/decl2.c(finish_file)</code></li>
-</ul></li>
-<li><h3>C and optimizer bugs</h3>
+</ul>
+
+<h4>C and optimizer bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6627">6627</a>: <code>-fno-align-functions</code> doesn't seem to disable function alignment</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6631">6631</a>: <code>life_analysis</code> misoptimizes code to initialize fields of a structure</li>
@@ -301,8 +314,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7515">7515</a>: invalid inlining of global function with <code>-O3</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7814">7814</a>: incorrect scheduling for glibc-2.2.92 strcpy test</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8467">8467</a>: bug in sibling call optimization</li>
-</ul></li>
-<li><h3>Preprocessor bugs</h3>
+</ul>
+
+<h4>Preprocessor bugs</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=4890">4890</a>: incorrect line markers from the traditional preprocessor</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7357">7357</a>: <code>-M</code> option omits system headers files (making it the same as <code>-MM</code>)</li>
@@ -311,8 +325,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7862">7862</a>: Interrupting <code>GCC -MD</code> removes <code>.d</code> file but not <code>.o</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8190">8190</a>: Failed compilation deletes <code>-MD</code> dependency file</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8524">8524</a>: <code>_Pragma</code> within macro is improperly expanded</li>
-</ul></li>
-<li><h3>x86 specific (Intel/AMD)</h3>
+</ul>
+
+<h4>x86 specific (Intel/AMD)</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5351">5351</a>: (i686-only) function pass-by-value structure copy corrupts stack
(<a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7591">7591</a> is a duplicate)</li>
@@ -332,8 +347,9 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7723">7723</a>: ICE - Pentium3 sse - GCC 3.2</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7951">7951</a>: ICE on <code>-march=pentium4 -O2 -mfpmath=sse</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8146">8146</a>: (i686 only) gcc 3.2 miscompiles gcc 2.95.3</li>
-</ul></li>
-<li><h3>PowerPC specific</h3>
+</ul>
+
+<h4>PowerPC specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5967">5967</a>: GCC bug when profiling nested functions on powerpc</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6984">6984</a>: wrong code generated with <code>-O2</code>, <code>-O3</code>, <code>-Os</code> for do-while loop on PowerPC</li>
@@ -343,52 +359,60 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7380">7380</a>: ICE in <code>extract_insn</code>, at <code>recog.c:2148</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8252">8252</a>: ICE on Altivec code with optimization turned on</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8451">8451</a>: Altivec ICE in GCC 3.2</li>
-</ul></li>
-<li><h3>HP/PA specific</h3>
+</ul>
+
+<h4>HP/PA specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7250">7250</a>: <code>__ashrdi3</code> returns wrong value on 32 bit hppa</li>
-</ul></li>
-<li><h3>SPARC specific</h3>
+</ul>
+
+<h4>SPARC specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6668">6668</a>: when using <code>--disable-multilib</code>, <code>libgcc_s.so</code> is installed in the wrong place on sparc-solaris</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7151">7151</a>: ICE when compiling for UltraSPARC</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7335">7335</a>: SPARC: ICE in <code>verify_wide_reg (flow.c:557)</code> with long double and <code>-O1</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7842">7842</a>: [REGRESSION] SPARC code gen bug</li>
-</ul></li>
-<li><h3>ARM specific</h3>
+</ul>
+
+<h4>ARM specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7856">7856</a>: [arm] invalid offset in constant pool reference</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7967">7967</a>: optimization produces wrong code (ARM)</li>
-</ul></li>
-<li><h3>Alpha specific</h3>
+</ul>
+
+<h4>Alpha specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7374">7374</a>: <code>__builtin_fabsl</code> broken on alpha</li>
-</ul></li>
-<li><h3>IBM s390 specific</h3>
+</ul>
+
+<h4>IBM s390 specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7370">7370</a>: ICE in <code>fixup_var_refs_1</code> on s390x</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7409">7409</a>: loop optimization bug on s390x-linux-gnu</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8232">8232</a>: s390x: ICE when using bcmp with <code>int</code> length argument</li>
-</ul></li>
-<li><h3>SCO specific</h3>
+</ul>
+
+<h4>SCO specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7623">7623</a>: SCO OpenServer build fails with <code>machmode.def: undefined symbol: BITS_PER_UNIT</code></li>
-</ul></li>
-<li><h3>m68k/Coldfire specific</h3>
+</ul>
+
+<h4>m68k/Coldfire specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8314">8314</a>: <code>crtbegin</code>, <code>crtend</code> need to be multilib'ed for this platform</li>
-</ul></li>
-<li><h3>Documentation</h3>
+</ul>
+
+<h4>Documentation</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=761">761</a>: Document some undocumented options</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=5610">5610</a>: Fix documentation about invoking SSE instructions (<code>-mfpmath=sse</code>)</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7484">7484</a>: List <code>-Wmissing-declarations</code> as C-only option</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7531">7531</a>: <code>-mcmodel</code> not documented for x86-64</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=8120">8120</a>: Update documentation of bad use of <code>##</code></li>
-</ul></li>
</ul>
-<h1><a name="3.2">GCC 3.2</a></h1>
+<hr />
+<h2><a name="3.2">GCC 3.2</a></h2>
<p>3.2 is a small bug fix release, but there is a change to the application
binary interface (ABI), hence the change to the second part of the version
@@ -398,16 +422,15 @@
the C++ ABI, with the intention of providing a stable interface going
forward. Accordingly, 3.2 is only a small change to 3.1.1.</p>
-<h2>Bug Fixes</h2>
+<h3>Bug Fixes</h3>
-<ul>
-<li><h3>C++</h3>
+<h4>C++</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7320">7320</a>: g++ 3.2 relocation problem</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7470">7470</a>: vtable: virtual function pointers not in declaration order</li>
-</ul></li>
+</ul>
-<li><h3>libstdc++</h3>
+<h4>libstdc++</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6410">6410</a>: Trouble with non-ASCII monetary symbols and <code>wchar_t</code></li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=6503">6503</a>,
@@ -419,12 +442,11 @@
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7286">7286</a>: placement operator delete issue</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7442">7442</a>: <code>cxxabi.h</code> does not match the C++ ABI</li>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7445">7445</a>: poor performance of <code>std::locale::classic()</code> in multi-threaded applications</li>
-</ul></li>
+</ul>
-<li><h3>x86-64 specific</h3>
+<h4>x86-64 specific</h4>
<ul>
<li><a href="http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=7291">7291</a>: off-by-one in generated inline <code>bzero</code> code for x86-64</li>
-</ul></li>
</ul>
</body>
More information about the Gcc-patches
mailing list