http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56788
--- Comment #12 from Marc Glisse <glisse at gcc dot gnu.org> --- (In reply to Uroš Bizjak from comment #11) > I left the prototype the way it was. Makes sense to me. > Marc, since it looks you have access to XOP target (I don't have one), can > you please write an XOP testcase for frcz insns? Sorry, I don't have access to a XOP target, that's why I've been relying on the AMD documentation (I would have just tested it otherwise). I only noticed this issue because I was writing a patch (long discussion with rth) changing the pattern of a lot of the vec_merge scalar operations (that got postponed, if not abandoned). The authors of this code are in Cc:, Ganesh added himself some time ago, I am hoping someone from AMD will eventually react.
