Ups - now attached. Thanks Martin!
Martin Jambor wrote:
On Mon, Jan 08 2024, Tobias Burnus wrote:
The attached patch
there was no patch attached to your message.
Martin
does a tiny updated to the OpenMP features (AMD GCN
now also has an optimized memcpy_rect not only nvptx), but the main
change is some shifting around to make it more consistent and better
readable.
I intend to commit this relatively soon; like always, comments and
suggestions are welcome - be it before or after the commit.
Current version: http://gcc.gnu.org/gcc-14/changes.html
Thanks,
Tobiasgcc-14/changes.html: OpenMP - improve wording
This is mostly some shifting of items in bullet points to make it
more organized; it also improved the wording a bit. And there is one
new feature: the optimization for omp_target_memcpy_rect is now also
done for AMD GPUs and not only for nvptx.
diff --git a/htdocs/gcc-14/changes.html b/htdocs/gcc-14/changes.html
index e3a68998..2c64cf67 100644
--- a/htdocs/gcc-14/changes.html
+++ b/htdocs/gcc-14/changes.html
@@ -49,14 +49,24 @@ a work-in-progress.
https://gcc.gnu.org/projects/gomp/;>OpenMP
+
+ The https://gcc.gnu.org/onlinedocs/libgomp/;>GNU Offloading and
+ Multi Processing Runtime Library Manual has been updated and extended,
+ improving especially the description of ICVs, memory allocation, environment variables and OpenMP
+ routines.
+
The requires directive's unified_address
requirement is now fulfilled by both AMD GCN and nvptx devices.
AMD GCN and nvptx devices now support low-latency allocators as
https://gcc.gnu.org/onlinedocs/libgomp/Offload-Target-Specifics.html;
>detailed in the manual. Initial support for pinned-memory
- allocators has been added (https://gcc.gnu.org/onlinedocs/libgomp/Memory-allocation.html;
- >as detailed in the manual)
+ allocators has been added and, on Linux,
+ https://github.com/numactl/numactl;>libnuma is now used
+ for allocators requesting the nearest-partition trait (both is described
+ in the https://gcc.gnu.org/onlinedocs/libgomp/Memory-allocation.html;
+ >memory allocation section of the manual).
OpenMP 5.0: The allocate directive is now
supported for stack variables in C and Fortran, including the OpenMP 5.1
@@ -74,8 +84,8 @@ a work-in-progress.
using present as map-type modifier and in
defaultmap. The indirect clause is now supported
for C and C++. The performance of copying strided data from or to nvptx
- devices using the OpenMP 5.1 routine omp_target_memcpy_rect
- has been improved.
+ and AMD GPU devices using the OpenMP 5.1 routine
+ omp_target_memcpy_rect has been improved.
OpenMP 5.2: The OMP_TARGET_OFFLOAD=mandatory handling has
@@ -83,23 +93,14 @@ a work-in-progress.
For Fortran, the list of directives permitted in Fortran pure procedures
was extended. Additionally, the spec change has been implemented for
default implicit mapping of C/C++ pointers pointing to unmapped storage.
- The destroy now optionally accepts the depend object as
- argument.
+ The destroy clause now optionally accepts the depend object
+ as argument.
OpenMP 6.0 preview (TR11/TR12): The decl attribute is now
supported in C++ 11 and the directive, sequence
and decl attributes are now supported in C 23.
-
- The https://gcc.gnu.org/onlinedocs/libgomp/;>GNU Offloading and
- Multi Processing Runtime Library Manual has been updated and extended,
- improving especially the description of ICVs, memory allocation, environment variables and OpenMP
- routines. On Linux, https://github.com/numactl/numactl;>libnuma
- is now used for allocators requesting the nearest-partition trait as
- detailed in the manual.
-
https://gcc.gnu.org/wiki/OpenACC;>OpenACC
