https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103372
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Ever confirmed|0 |1
Status|UNCONFIRMED |NEW
Last reconfirmed| |2021-11-30
--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
So clang in this case is wrong in that there is no reason to do figure out
failure since succ is memory_order_relaxed.
In this case, with a non-constant in failure, GCC should just choise the same
as success. Likewise the opposite way around. Except for
release/memory_order_acq_rel which should get special cases.
ยง7.17.7.4 "The failure argument shall not be memory_order_release nor
memory_order_acq_rel. The failure argument shall be no stronger than the
success argument."
