https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120957
--- Comment #5 from Hongtao Liu <liuhongt at gcc dot gnu.org> --- (In reply to Filip Kastl from comment #3) > I've bisected this on Zen2. It is possible that this is actually two > different slowdowns and only the Zen2 slowdown is caused by r16-1647. I'll > bisect on Zen3. I've tried with -march=zen2 -Ofast on AMD zen3 machine, also can't reproduce the issue. So it looks hit some micro-architecture specific issue.
