https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123126

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=289.290.0

Shows a fixed result, but with PGO it isn't:

https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=785.290.0

possibly the vectorizer costing is imprecise or the PGO data is not a good
indicator for the actual expected niters (like if you iterate 80% 1 times
and 20% 20 times it will tell you we iterate an average of 5 times).

Reply via email to