[PATCH] D31398: [X86][X86 intrinsics]Folding cmp(sub(a, b), 0) into cmp(a, b) optimization

[michael zuckerman via Phabricator via llvm-commits]

m_zuckerman updated this revision to Diff 93545.
m_zuckerman marked 4 inline comments as done.

https://reviews.llvm.org/D31398

Files:
  lib/Transforms/InstCombine/InstCombineCalls.cpp
  test/Transforms/InstCombine/X86FsubCmppd256Combine.ll
  test/Transforms/InstCombine/X86FsubCmppd512Combine.ll
  test/Transforms/InstCombine/X86FsubCmppdCombine.ll
  test/Transforms/InstCombine/X86FsubCmpps256Combine.ll
  test/Transforms/InstCombine/X86FsubCmpps512Combine.ll
  test/Transforms/InstCombine/X86FsubCmppsCombine.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D31398.93545.patch
Type: text/x-patch
Size: 10736 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170330/288ca052/attachment.bin>