https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88786
--- Comment #2 from rsandifo at gcc dot gnu.org <rsandifo at gcc dot gnu.org> --- (In reply to Richard Biener from comment #1) > I think it's a bit too low-level and having the higher-level op benefits > later optimizations (match.pd patterns, etc.) match.pd will already have seen the scalar form though, so I think we should assume that any optimisations like that have already been applied. AFAIK the only copysign optimisation that happens after vectorisation is the xorsign one. But in a way that's working around the fact that we *haven't* exposed the individual operations. The xorsign_optab only exists as a way of representing the result of a particular fold.
