[PATCH] D35978: [AVX-512] Don't use unmasked VMOVDQU8/16 for 8-bit or 16-bit element stores even when BWI instructions are supported. Always use VMOVDQA32/VMOVDQU32.

[Phabricator via Phabricator via llvm-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rL309693: [AVX-512] Don't use unmasked VMOVDQU8/16 for 8-bit or 16-bit element stores… (authored by ctopper).

Changed prior to commit:
  https://reviews.llvm.org/D35978?vs=109014&id=109129#toc

Repository:
  rL LLVM

https://reviews.llvm.org/D35978

Files:
  llvm/trunk/lib/Target/X86/X86InstrAVX512.td
  llvm/trunk/test/CodeGen/X86/avg.ll
  llvm/trunk/test/CodeGen/X86/avx512-insert-extract.ll
  llvm/trunk/test/CodeGen/X86/avx512-insert-extract_i1.ll
  llvm/trunk/test/CodeGen/X86/avx512bw-intrinsics-upgrade.ll
  llvm/trunk/test/CodeGen/X86/avx512bw-mov.ll
  llvm/trunk/test/CodeGen/X86/avx512bwvl-mov.ll
  llvm/trunk/test/CodeGen/X86/subvector-broadcast.ll
  llvm/trunk/test/CodeGen/X86/x86-interleaved-access.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D35978.109129.patch
Type: text/x-patch
Size: 26087 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170801/d6e5ad06/attachment.bin>