https://gcc.gnu.org/bugzilla/show_bug.cgi?id=36010
--- Comment #8 from rguenther at suse dot de <rguenther at suse dot de> --- On Mon, 2 Sep 2024, tnfchris at gcc dot gnu.org wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=36010 > > Tamar Christina <tnfchris at gcc dot gnu.org> changed: > > What |Removed |Added > ---------------------------------------------------------------------------- > CC| |tnfchris at gcc dot gnu.org > > --- Comment #7 from Tamar Christina <tnfchris at gcc dot gnu.org> --- > It looks like today at -Ofast this is due to the full unrolling > > https://godbolt.org/z/3K4hPbWfG > > i.e. at -Ofast we fail due to the inner loop being fully unrolled. > > Would it make sense to perform loop distribution before cunrolli? > > in principle it should make any potential vect and SLP simpler no? cunrolli is to remove abstraction penalty and thus has to be done early. The option is to integrate that with value-numbering somehow to allow the "abstraction" to remain and still allow followup optimization.
