https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121908
--- Comment #4 from Richard Biener <rguenth at gcc dot gnu.org> --- Btw, loop_distribution::transform_reduction_loop which handles rawmemchr-like loops could be extended to cover such kind of loop. A optab for this is missing, of course. The vectorizer should now have support for mutually aligned peeling, so I expect this to be vectorized.
