https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117192
--- Comment #8 from Andrew Pinski <pinskia at gcc dot gnu.org> --- (In reply to Uroš Bizjak from comment #7) > (In reply to Andrew Pinski from comment #6) > > before is before r15-4397-g70f59d2a1c51bd > > - movq 16(%rsp), %xmm2 > > + movq 16(%rsp), %xmm1 > > Isn't this reading from uninitialized stack slot? > > movq 16(%rsp), %xmm2 # MEM <vector(2) int> [(int *)&e + 16B], > vect__ifc__49.29 Yes but in this case the mask will 0 forcing it all to 0 . But the mask is swapping around to -1 for some reason.
