diff mbox series

libgomp.texi: Update OpenMP implementation status

Message ID 80f1ab7b-ee69-cd1b-9d94-07da90e8aad1@codesourcery.com
State New
Headers show
Series libgomp.texi: Update OpenMP implementation status | expand

Commit Message

Tobias Burnus Jan. 26, 2022, 10:14 p.m. UTC 
]I saw that the following now is implemented:

* requires: dynamic_allocators is now also works
   (by assuming that it is supported, no longer giving 'sorry')
   https://gcc.gnu.org/r12-6641-g450c85b81f4dd67bf6211d307afdc0f3c07ef44f

* in_reduction on target: Now also supported for Fortran
   https://gcc.gnu.org/r12-4574-gd98626bf451dea6a28a42d953f7d0bd7659ad4d5
(I kept the "P" because of "nowait")


I also note – but did not change:

* omp_display_env runtime routine  P  Not inside target regions

I observe that OpenMP 5.2 added:
   "When called from within a target region the effect is unspecified."
Thus, regarding this as bug fix, it could also be "Y".
Thoughts?

[Some output modifications are also part of the pending device-specific
ICV patch, https://gcc.gnu.org/pipermail/gcc-patches/2022-January/588728.html
thus, as that patch are not yet it, one can also argue for "P" in terms of
the output.]


Otherwise: OK?

Tobias
-----------------
Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht München, HRB 106955

Comments

Jakub Jelinek Jan. 26, 2022, 10:19 p.m. UTC | #1 
On Wed, Jan 26, 2022 at 11:14:54PM +0100, Tobias Burnus wrote:
> ]I saw that the following now is implemented:
> 
> * requires: dynamic_allocators is now also works
>   (by assuming that it is supported, no longer giving 'sorry')
>   https://gcc.gnu.org/r12-6641-g450c85b81f4dd67bf6211d307afdc0f3c07ef44f
> 
> * in_reduction on target: Now also supported for Fortran
>   https://gcc.gnu.org/r12-4574-gd98626bf451dea6a28a42d953f7d0bd7659ad4d5
> (I kept the "P" because of "nowait")
> 
> 
> I also note – but did not change:
> 
> * omp_display_env runtime routine  P  Not inside target regions
> 
> I observe that OpenMP 5.2 added:
>   "When called from within a target region the effect is unspecified."
> Thus, regarding this as bug fix, it could also be "Y".
> Thoughts?

Yes, we should say Y instead of P for it.

> Otherwise: OK?

Yes.

	Jakub
diff mbox series

Patch

libgomp.texi: Update OpenMP implementation status

libgomp/
	* libgomp.texi (OpenMP 5.0): Update implementation status.

diff --git a/libgomp/libgomp.texi b/libgomp/libgomp.texi
index 3be9de51f11..73d9f4f76cb 100644
--- a/libgomp/libgomp.texi
+++ b/libgomp/libgomp.texi
@@ -187,7 +187,8 @@  The OpenMP 4.5 specification is fully supported.
       env variable @tab Y @tab
 @item Nested-parallel changes to @emph{max-active-levels-var} ICV @tab Y @tab
 @item @code{requires} directive @tab P
-      @tab Only fulfillable requirement is @code{atomic_default_mem_order}
+      @tab Only fulfillable requirement are @code{atomic_default_mem_order}
+      and @code{dynamic_allocators}
 @item @code{teams} construct outside an enclosing target region @tab Y @tab
 @item Non-rectangular loop nests @tab P @tab Only C/C++
 @item @code{!=} as relational-op in canonical loop form for C/C++ @tab Y @tab
@@ -203,7 +204,7 @@  The OpenMP 4.5 specification is fully supported.
       @code{reduction} clause @tab Y @tab
 @item @code{in_reduction} clause on @code{task} constructs @tab Y @tab
 @item @code{in_reduction} clause on @code{target} constructs @tab P
-      @tab Only C/C++, @code{nowait} only stub
+      @tab @code{nowait} only stub
 @item @code{task_reduction} clause with @code{taskgroup} @tab Y @tab
 @item @code{task} modifier to @code{reduction} clause @tab Y @tab
 @item @code{affinity} clause to @code{task} construct @tab Y @tab Stub only