[PATCH] D36854: [AVX512] Don't use 32-bit elements version of AND/OR/XOR/ANDN during isel unless we're matching a masked op or broadcast

[Craig Topper via Phabricator via llvm-commits]

craig.topper updated this revision to Diff 111591.
craig.topper added a comment.

Add the complete patch this time


https://reviews.llvm.org/D36854

Files:
  lib/Target/X86/X86InstrAVX512.td
  lib/Target/X86/X86InstrSSE.td
  test/CodeGen/X86/avx512-arith.ll
  test/CodeGen/X86/avx512-intrinsics-upgrade.ll
  test/CodeGen/X86/avx512-logic.ll
  test/CodeGen/X86/broadcast-elm-cross-splat-vec.ll
  test/CodeGen/X86/cast-vsel.ll
  test/CodeGen/X86/vector-bitreverse.ll
  test/CodeGen/X86/vector-lzcnt-512.ll
  test/CodeGen/X86/vector-rotate-512.ll
  test/CodeGen/X86/vector-trunc-math.ll
  test/CodeGen/X86/vector-tzcnt-512.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D36854.111591.patch
Type: text/x-patch
Size: 24651 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170818/fe2d65fe/attachment.bin>