Hi all,
I have committed a quip for OpenACC and AMD GCN. Comments and
suggestions are welcome. Thanks for Andrew and Sandra for comments on an
early draft.
Tobias
PS: The linked OpenACC wiki page is unfortunately not yet fully up to dateā¦
commit 8c9fba2f83ef9bdace2fb6288f555c78949adfb6
Author: Tobias Burnus <tob...@codesourcery.com>
Date: Mon Feb 24 10:29:11 2020 +0100
OpenACC + AMD GCN entry for gcc-10/changes.html
diff --git a/htdocs/gcc-10/changes.html b/htdocs/gcc-10/changes.html
index 2920714a..83b1f071 100644
--- a/htdocs/gcc-10/changes.html
+++ b/htdocs/gcc-10/changes.html
@@ -142,7 +142,25 @@ a work-in-progress.</p>
</li>
</ul>
<!-- .................................................................. -->
-<h2 id="languages">New Languages and Language specific improvements</h2>
+<h2 id="languages">New Languages and Language-Specific Improvements</h2>
+
+<ul>
+ <li>
+ Version 2.6 of the <a href="https://www.openacc.org/">OpenACC</a>
+ specification is now supported in the C, C++ and Fortran compilers.
+ See the <a href="https://gcc.gnu.org/wiki/OpenACC#status-10">implementation
+ status</a> section on the OpenACC wiki page and the
+ <a href="https://gcc.gnu.org/onlinedocs/libgomp/#toc-Enabling-OpenACC-1">
+ run-time library documentation</a> for further information.
+ </li>
+ <li>
+ OpenMP and OpenACC now support
+ <a href="https://gcc.gnu.org/wiki/Offloading">offloading</a> to AMD Radeon
+ (GCN) GPUs; supported are the third-generation Fiji (<code>fiji</code>)
+ and the fifth-generation VEGA 10/VEGA 20 (<code>gfx900</code> or
+ <code>gfx906</code>).</li>
+ </li>
+</ul>
<!-- <h3 id="ada">Ada</h3> -->
@@ -533,6 +551,12 @@ a work-in-progress.</p>
</li>
</ul>
+<h3 id="amdgcn">AMD Radeon (GCN)</h3>
+<ul>
+ <li>The code generation and in particular the vectorization support has been
+ much improved.</li>
+</ul>
+
<!-- <h3 id="arc">ARC</h3> -->
<h3 id="arm">arm</h3>