[all-commits] [llvm/llvm-project] 916bae: [VectorCombine] foldShuffleOfBinops - refactor to ...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 916bae2d921705c8ce78a4ddec4503c61bc8220c
      https://github.com/llvm/llvm-project/commit/916bae2d921705c8ce78a4ddec4503c61bc8220c
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-12-15 (Sun, 15 Dec 2024)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/VectorCombine.cpp

  Log Message:
  -----------
  [VectorCombine] foldShuffleOfBinops - refactor to make it easier to match icmp/fcmp patterns

NFC refactor to make it easier to also use the fold for icmp/fcmp patterns in a future patch - match the Shuffle with general Instruction operands and avoid explicit use of the BinaryOperator matches as much as possible for the general costing / fold.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications