https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126515
Haochen Jiang <haochen.jiang at intel dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |haochen.jiang at intel dot com
--- Comment #1 from Haochen Jiang <haochen.jiang at intel dot com> ---
I have met regression on 736.ocio_r on multiple x86 platforms, under -O2, but
with an extra -march=x86-64-v3, w/o -flto, Granite Rapids -4.17%, Clearwater
Forest -5.27%, Sierra Forest -2.92%, Sapphire Rapids -4.78%, Ice Lake -6.51%,
Zen5 Client -7.13%.
For -O2 -march=x86-64-v3, there also ~2% regression on Sierra Forest on both
525.x265_r and 521.wrf_r. For Zen5 Client, 508.namd_r also have ~2% regression.
There are also regressions with -O3 -ffast-math -march=native -flto, I will
reply to another Ofast bugzilla.
