[all-commits] [llvm/llvm-project] 56b2be: [X86] Fold scalar_to_vector(funnel(x, y, imm)) -> fu...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 56b2be4a7608770bae5db9d467f50c232c3cf19a
      https://github.com/llvm/llvm-project/commit/56b2be4a7608770bae5db9d467f50c232c3cf19a
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-09-05 (Thu, 05 Sep 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/vector-shuffle-combining.ll

  Log Message:
  -----------
  [X86] Fold scalar_to_vector(funnel(x,y,imm)) -> funnel(scalar_to_vector(x),scalar_to_vector(y),imm)

Limit this to cases where x, y are known to be extracted from a vector.

Addresses poor x86 codegen on #107289



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