https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127746
--- Comment #3 from Drea Pinski <pinskia at gcc dot gnu.org> --- > this is in GIMPLE (for which operands of a COND_EXPR would by construction > already have been evaluated, with the same consequences as for a vector > condition) This is not 100% true due to phiopt usage. Where the rewrite does happen but only after the transformation does happen.
