https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115483
--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> --- it's again the interaction of store motion, re-assoc and then invariant motion.
rguenth at gcc dot gnu.org via Gcc-bugs Thu, 13 Jun 2024 22:48:32 -0700
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115483
--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> --- it's again the interaction of store motion, re-assoc and then invariant motion.