https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65456
--- Comment #11 from Bill Schmidt <wschmidt at gcc dot gnu.org> --- (In reply to David Edelsohn from comment #10) > I believe that the choice to scalarize is based on the vector cost model. Hm, that would be interesting. The applied patch changes the cost model to favor the unaligned store on VSX. Might be a place where it should be using the cost model but perhaps isn't? I'll look into it later this week.