https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98335
Jakub Jelinek <jakub at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |law at gcc dot gnu.org
--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
I agree it would be weird to try to undo the
- D.2365 = {};
+ MEM <char[7]> [(struct Data *)&D.2365 + 1B] = {};
transformation by DSE in store_merging instead of adjusting the DSE
optimization to take into account costs and likely ways how the clearing will
be expanded.
On the other side, the user could have written it that way.
Regressed with r9-1663-g99e87c0eef2f6020a3ded2c785389939c07ac04e aka PR86010
fix.
