[PATCH] D26019: [AVX-512] Use scalar vfmsub/vfnmsub mask3 intrinsics instead of inverting the mask argument of a vfmadd intrinsic.

[Craig Topper via cfe-commits]

craig.topper created this revision.
craig.topper added reviewers: delena, igorb.
craig.topper added a subscriber: cfe-commits.

Inverting the mask argument currently fails isel and even if that was fixed it does not reflect the intended semantics of the intrinsic.


https://reviews.llvm.org/D26019

Files:
  include/clang/Basic/BuiltinsX86.def
  lib/Headers/avx512fintrin.h
  test/CodeGen/avx512f-builtins.c

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D26019.75978.patch
Type: text/x-patch
Size: 7693 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20161027/6d9712f2/attachment.bin>