https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118125
--- Comment #7 from Martin Jambor <jamborm at gcc dot gnu.org> --- (In reply to Martin Jambor from comment #4) > [...] Unfortunately when I > then looked at SLP vectorization when all IPA-VR propagations were > allowed again, this particular case was not there (but there were > plenty of others). I must have messed up the LTO intermediate files or something, the SLP vectorization does take place also with all IPA-VR propagations. But when I disabled it with a #pragma GCC novector, the code is still slower.
