https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104493
--- Comment #2 from Tobias Burnus <burnus at gcc dot gnu.org> ---
(In reply to Ye Luo from comment #1)
> gcc 12 rejects the code but devel/omp/gcc-11 was happy.
Note: the devel/omp/gcc-* branch contains both patches backported ("git
cherry-picked") from the mainline (curent development) branch but also some
patches which are not yet on mainline for various reasons. – In particular,
this happens for features which are only ready shortly before the next release
("release freeze"). The closer to the release and the more invasive the
patches, the less likely to be included. — See https://gcc.gnu.org/develop.html
for "Schedule" with Stage 1/3/4 for an overview and towards the end of the
page, the timeline (i.e. when did Stage 1/(2/)/3/4 start and when were the GCC
releases).
If a patch is in OG11 then it is almost sure that OG12 will also contain it.
Additionally, there is a high probability that this patch will also be included
in mainline.
For GCC's git, its branches branches in general and the active branches,
including devel/omp/gcc-*, see also https://gcc.gnu.org/git.html (OG12, based
on the GCC 12 branch, will soon be available; at the same time, OG11 will then
be retired).
* * *
Regarding the issue at hand, I think the issue is fixed by patch:
https://gcc.gnu.org/pipermail/gcc-patches/2022-March/591449.html
is was reviewed at
https://gcc.gnu.org/pipermail/gcc-patches/2022-May/594082.html
with some minor comments. Thus, I assume that this patch will soon be committed
to
GCC 13 (current mainline).
* * *
I concur that it is a bit unfortunate that it missed the GCC 12 deadline, but
that's the general problem with deadlines and bug fixes and new features:
There will always be features/bug fixes, which are very important for some,
that missed the deadline by a very short time.
