https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79347
--- Comment #9 from amker at gcc dot gnu.org --- (In reply to Pat Haugen from comment #7) > See pr77536 for similar issue. Not sure if that should be marked as dup of > this one or left open as a more general bug on what should be done to > specifically fix the vectorized loop. Hi Pat, Thanks for pointing me to this. This report is mostly tracking mismatch issue introduced when I rewriting loop peeling in vectorizer. Issue that vectorized loop's frequency is wrong is scaled exists regardless of peeling change, so IMHO it would be better to close this one and tracking the issue in PR77536. Thanks.