[PATCH] D50074: [X86][AVX2] Prefer VPBLENDW +VPBLENDD to VPBLENDVB for v16i16 blend shuffles

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon updated this revision to Diff 162716.
RKSimon retitled this revision from "[X86][AVX2] Prefer VPBLENDW+VPBLENDW+VPBLENDD to VPBLENDVB for v16i16 blend shuffles" to "[X86][AVX2] Prefer VPBLENDW +VPBLENDD to VPBLENDVB for v16i16 blend shuffles".
RKSimon edited the summary of this revision.
RKSimon added a comment.

Limit to only one PBLENDW until we have shuffle combining support for VSELECT.


Repository:
  rL LLVM

https://reviews.llvm.org/D50074

Files:
  lib/Target/X86/X86ISelLowering.cpp
  test/CodeGen/X86/insertelement-ones.ll
  test/CodeGen/X86/vector-shuffle-256-v16.ll
  test/CodeGen/X86/vector-shuffle-256-v32.ll
  test/CodeGen/X86/vector-shuffle-512-v32.ll
  test/CodeGen/X86/vector-shuffle-v48.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D50074.162716.patch
Type: text/x-patch
Size: 17992 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180827/29be3eb7/attachment.bin>