[PATCH] D81327: [X86] Teach combineBitcastvxi1 to prefer movmsk on avx512 in more cases

[Craig Topper via Phabricator via llvm-commits]

craig.topper updated this revision to Diff 269041.
craig.topper added a comment.

Add a reversing transform for bitcasts that appear during intrinsic lowering. This
required a new test file. The output for that file shown here is the current output
before this patch. No existing tests in tree demonstrated the need for the reversing
combine.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D81327/new/

https://reviews.llvm.org/D81327

Files:
  llvm/lib/Target/X86/X86ISelLowering.cpp
  llvm/test/CodeGen/X86/avx512-movmsk.ll
  llvm/test/CodeGen/X86/avx512bwvl-intrinsics-upgrade.ll
  llvm/test/CodeGen/X86/avx512dqvl-intrinsics-upgrade.ll
  llvm/test/CodeGen/X86/bitcast-setcc-256.ll
  llvm/test/CodeGen/X86/bitcast-vector-bool.ll
  llvm/test/CodeGen/X86/movmsk-cmp.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D81327.269041.patch
Type: text/x-patch
Size: 26198 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20200607/30f4c70f/attachment.bin>