This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "gcc-wwwdocs".
The branch, master has been updated
via 127a4caf32709a506fb1c947f6fac8d0cbf32e39 (commit)
from 0d4fd54a2901ff9dab42274aa065748af0367e02 (commit)
Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.
- Log -----------------------------------------------------------------
commit 127a4caf32709a506fb1c947f6fac8d0cbf32e39
Author: Sandra Loosemore <sloosem...@baylibre.com>
Date: Wed Feb 12 21:39:49 2025 +0000
gcc-15: Update OpenMP release notes
diff --git a/htdocs/gcc-15/changes.html b/htdocs/gcc-15/changes.html
index e273693a..d7919379 100644
--- a/htdocs/gcc-15/changes.html
+++ b/htdocs/gcc-15/changes.html
@@ -65,37 +65,57 @@ a work-in-progress.</p>
<!-- .................................................................. -->
<h2 id="languages">New Languages and Language specific improvements</h2>
+<h3 id="openmp"><a href="../projects/gomp/">OpenMP</a></h3>
<ul>
- <li id="openmp"><a href="../projects/gomp/">OpenMP</a>
- <ul>
- <li>
- Support for unified-shared memory has been added for some AMD and Nvidia
- GPU devices, enabled when using the <code>unified_shared_memory</code>
- clause to the <code>requires</code> directive. For details,
- see the offload-target specifics section in the
- <a
href="https://gcc.gnu.org/onlinedocs/libgomp/Offload-Target-Specifics.html";
- >GNU Offloading and Multi Processing Runtime Library Manual</a>.
- GCC added <code>ompx_gnu_pinned_mem_alloc</code> as <a
-
href="https://gcc.gnu.org/onlinedocs/libgomp/OMP_005fALLOCATOR.html";>predefined
- allocator</a>. On <a
href="https://gcc.gnu.org/onlinedocs/libgomp/nvptx.html";>Nvidia
- GPUs, writing to the terminal</a> from OpenMP target regions (but not
from
- OpenACC compute regions) is now also supported in Fortran; in C/C++ and
- on AMD GPUs this was already supported before with both OpenMP and
OpenACC.
- Constructors and destructors on the device side for <code>declare
target</code>
- static aggregates are now handled.
- </li>
- <li>
- OpenMP 5.1: The <code>unroll</code> and <code>tile</code>
- loop-transformation constructs are now supported.
- </li>
- <li>
- OpenMP 6.0: The <a
-
href="https://gcc.gnu.org/onlinedocs/libgomp/omp_005fget_005fdevice_005ffrom_005fuid.html";
- ><code>get_device_from_uid</code></a> and <a
-
href="https://gcc.gnu.org/onlinedocs/libgomp/omp_005fget_005fuid_005ffrom_005fdevice.html";>
- <code>omp_get_uid_from_device</code></a> API routines have been added.
- </li>
- </ul>
+ <li>
+ Support for unified-shared memory has been added for some AMD and Nvidia
+ GPU devices, enabled when using the <code>unified_shared_memory</code>
+ clause to the <code>requires</code> directive.
+ The OpenMP 6.0 <code>self_maps</code> clause is also now supported.
+ For details,
+ see the offload-target specifics section in the
+ <a
href="https://gcc.gnu.org/onlinedocs/libgomp/Offload-Target-Specifics.html";
+ >GNU Offloading and Multi Processing Runtime Library Manual</a>.
+ </li>
+ <li>
+ GCC added <code>ompx_gnu_pinned_mem_alloc</code> as a <a
+
href="https://gcc.gnu.org/onlinedocs/libgomp/OMP_005fALLOCATOR.html";>predefined
+ allocator</a>.
+ </li>
+ <li>
+ In C and Fortran, the <code>allocate</code> directive now supports
+ static variables; stack variables were previously supported in
+ those languages. C++ support is not available yet.
+ </li>
+ <li>
+ Offloading improvements:
+ On <a href="https://gcc.gnu.org/onlinedocs/libgomp/nvptx.html";>Nvidia
+ GPUs, writing to the terminal</a> from OpenMP target regions (but not from
+ OpenACC compute regions) is now also supported in Fortran; in C/C++ and
+ on AMD GPUs this was already supported before with both OpenMP and OpenACC.
+ Constructors and destructors on the device side for
+ <code>declare target</code> static aggregates are now handled.
+ </li>
+ <li>
+ The OpenMP 5.1 <code>unroll</code> and <code>tile</code>
+ loop-transforming constructs are now supported.
+ </li>
+ <li>OpenMP 5.0 metadirectives are now supported, as are OpenMP 5.1 dynamic
+ selectors in both <code>metadirective</code> and
+ <code>declare variant</code> (the latter with some restrictions).
+ </li>
+ <li>
+ The OpenMP 5.1 <code>dispatch</code> construct has been implemented
+ with support for the <code>adjust_args</code> clause to the
+ <code>declare variant</code> directive.
+ </li>
+ <li>
+ OpenMP 6.0: The <a
+
href="https://gcc.gnu.org/onlinedocs/libgomp/omp_005fget_005fdevice_005ffrom_005fuid.html";
+ ><code>get_device_from_uid</code></a> and <a
+
href="https://gcc.gnu.org/onlinedocs/libgomp/omp_005fget_005fuid_005ffrom_005fdevice.html";>
+ <code>omp_get_uid_from_device</code></a> API routines have been added.
+ </li>
</ul>
<!-- <h3 id="ada">Ada</h3> -->
-----------------------------------------------------------------------
Summary of changes:
htdocs/gcc-15/changes.html | 80 +++++++++++++++++++++++++++++-----------------
1 file changed, 50 insertions(+), 30 deletions(-)
hooks/post-receive
--
gcc-wwwdocs
