[PATCH] D91515: [X86] Use GF2P8AFFINEQB to implement vector bitreverse.

[Craig Topper via Phabricator via llvm-commits]

craig.topper updated this revision to Diff 305668.
craig.topper added a comment.

Update cost model.

Not doing scalar yet. Making scalar Custom requires us to make vXi16/32/64 Custom as well as the LegalizeVectorOps Expand will scalarize if scalar is legal or custom.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D91515/new/

https://reviews.llvm.org/D91515

Files:
  llvm/lib/Target/X86/X86ISelLowering.cpp
  llvm/lib/Target/X86/X86TargetTransformInfo.cpp
  llvm/test/Analysis/CostModel/X86/bitreverse.ll
  llvm/test/CodeGen/X86/vector-bitreverse.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D91515.305668.patch
Type: text/x-patch
Size: 106871 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201117/e0c3b541/attachment.bin>