[PATCH] D14909: [X86][FMA] Optimize FNEG(FMUL) Patterns

[Simon Pilgrim via llvm-commits]

RKSimon updated this revision to Diff 41335.
RKSimon added a comment.

Updated patch to correctly support 512-bit float types on AVX512 and pre-AVX512 targets.

Note - it turns out we do need the hasAVX512() test to correctly lower on 128-256 bit vectors on AVX512 targets.


Repository:
  rL LLVM

http://reviews.llvm.org/D14909

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/CodeGen/X86/fma_patterns.ll
  test/CodeGen/X86/fma_patterns_wide.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D14909.41335.patch
Type: text/x-patch
Size: 7456 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20151128/e84edf0c/attachment.bin>