https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114057
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |14.0
--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
If it is a FP rounding issue then my guess would be the extra reduc_scal_*
patterns in the x86 backend to do more BB reduction vectorization.
I guess it doesn't miscompare with -O3 or -O3 -fno-signed-zeros
-fno-trapping-math -ffinite-math-only -ffp-contract=fast?
