https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120069
Filip Kastl <pheeck at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |pheeck at gcc dot gnu.org
--- Comment #7 from Filip Kastl <pheeck at gcc dot gnu.org> ---
The Intel slowdowns disappeared. The graphs are at their original values. The
Zen2 slowdowns also got better, but not entirely. I'm gonna bisect those.
There's also this Zen2 slowdown with -O2 -march=native -flto *without* PGO.
Maybe it is the same slowdown? But then this wouldn't be caused by PGO.
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=296.507.0
