https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114635
--- Comment #7 from rguenther at suse dot de <rguenther at suse dot de> --- > Am 08.04.2024 um 16:55 schrieb tnfchris at gcc dot gnu.org > <gcc-bugzi...@gcc.gnu.org>: > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114635 > > --- Comment #6 from Tamar Christina <tnfchris at gcc dot gnu.org> --- > (In reply to Jakub Jelinek from comment #4) >> Now, with SVE/RISCV vectors the actual vectorization factor is a poly_int >> rather than constant. One possibility would be to use VLA arrays in those >> cases, but then it will be hard to undo that later, or allow both shrinking >> and growing the arrays and even turning them into VLA-like ones. > > I think they are already VLAs of some kind going into vect, unless I've > misunderstood the declaration: > > float D.28295[0:POLY_INT_CST [15, 16]]; > float D.28293[0:POLY_INT_CST [15, 16]]; > float D.28290[0:POLY_INT_CST [15, 16]]; > > it looks like during vectorization they are lowered though. Maybe it’s about upper vs lower bound when setting loop->safelen from ‚unlimited‘ > -- > You are receiving this mail because: > You are on the CC list for the bug.
