https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93462
Tobias Burnus <burnus at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|REOPENED |RESOLVED
Resolution|--- |FIXED
--- Comment #9 from Tobias Burnus <burnus at gcc dot gnu.org> ---
FIXED – as the fix has now been backported to GCC 8 (= 8.4.1), the oldest open
branch.
(In reply to Thomas Schwinge from comment #6)
> Also needed on releases/gcc-8, to avoid introducing a regression, since […]
> "backport: re PR fortran/92977 […]"
I am not sure I understand this argument. The backported fix for OpenMP makes
it easier to fix OpenACC (as it is then only one line), but by itself, the
OpenMP commit does not affect the status of the OpenACC issue.
(Whether an ICE caused by a new feature is a regression only because it was
rejected before the new feature is also the question. But as the fix is
trivial, backporting was/is fine.)
