[Bug middle-end/84200] [8 Regression] r256888 causes 30% performance regression of 519.lbm_r at -Ofast generic tuning on Zen

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84200

--- Comment #6 from Martin Liška <marxin at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #5)
> This is the strange alignment case, right?  So we're not really
> understanding what goes wrong here and we're merely unlucky, correct? 
> Leaving at P3.

Yep, I'm fully happy with the classification. It's really weird regression.