https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103066
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |redi at gcc dot gnu.org,
| |rodgertq at gcc dot gnu.org
--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
I don't think so, but we'd need somebody more familiar with the C/C++ memory
model to say for sure.
In any case, even if it is, I'm not convinced it is a good idea, users should
be able to choose whether they do the separate atomic load first themselves
rather than compiler forcing them to do that unconditionally.
