https://bugs.llvm.org/show_bug.cgi?id=35129
Simon Pilgrim <llvm-...@redking.me.uk> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution|--- |FIXED
--- Comment #5 from Simon Pilgrim <llvm-...@redking.me.uk> ---
Resolving the 'allof' MOVMSK(PCMPEQ(x,0)) == -1 pattern now uses PTEST if its
available.
The 'anyof' cases still must use MOVMSK as there isn't a PTEST equivalent that
wouldn't involve still doing the PCMPEQ, which is never quicker than the MOVMSK
path on any existing hardware.
--
You are receiving this mail because:
You are on the CC list for the bug._______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
