Hello,
This patch adds news about AVX-512 to GCC main page
and entry to 4.9's changes.html.
Is it ok?
PS: I am not native speaker, any corrections are welcome!
--
Thanks, K
Index: htdocs/index.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/index.html,v
retrieving revision 1.903
diff -p -r1.903 index.html
*** htdocs/index.html 3 Feb 2014 10:02:51 -0000 1.903
--- htdocs/index.html 7 Feb 2014 07:46:01 -0000
*************** mission statement</a>.</p>
*** 53,58 ****
--- 53,63 ----
<dl class="news">
+ <dt><span>Intel AVX-512 support</span>
+ <span class="date">[2014-02-07]</span></dt>
+ <dd>Intel AVX-512 support was added to GCC. That includes inline assembly
+ support, extended existing and new registers, intrinsics set (covered by
+ corresponding testsuite), and basic autovectorization.</dd>
<dt><span>Altera Nios II support</span>
<span class="date">[2013-12-31]</span></dt>
<dd>A port for Altera Nios II has been contributed by Mentor
Graphics.</dd>
Index: htdocs/gcc-4.9/changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-4.9/changes.html,v
retrieving revision 1.54
diff -p -r1.54 changes.html
*** htdocs/gcc-4.9/changes.html 28 Jan 2014 23:57:49 -0000 1.54
--- htdocs/gcc-4.9/changes.html 7 Feb 2014 07:46:02 -0000
*************** auto incr = [](auto x) { return x++; };
*** 387,392 ****
--- 387,401 ----
<h3 id="x86">IA-32/x86-64</h3>
<ul>
+ <li>GCC now supports Intel Advanced Vector Extensions 512 instructions
+ (AVX-512), including inline assembly support, extended existing and
+ new registers, intrinsics set (covered by corresponding testsuite) and
+ basic autovectorization. AVX-512 instructions are available via
+ following GCC switches: AVX-512 foundamental instructions
+ <code>-mavx512f</code>, AVX-512 prefetch instructions
<code>-mavx512pf</code>,
+ AVX-512 exponential and reciprocal instructions <code>-mavx512er</code>,
+ AVX-512 conflict detection instructions <code>-mavx512cd</code>.
+ </li>
<li> It is now possible to call x86 intrinsics from select functions in
a file that are tagged with the corresponding target attribute without
having to compile the entire file with the <code>-mxxx</code> option.
