[all-commits] [llvm/llvm-project] 7de7f5: [InstCombine][X86] Fold blendv(x, y, shuffle(bitcast...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 7de7f50fc9d76ced7a971a66abf59aab6f9e5be6
      https://github.com/llvm/llvm-project/commit/7de7f50fc9d76ced7a971a66abf59aab6f9e5be6
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2024-07-03 (Wed, 03 Jul 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86InstCombineIntrinsic.cpp
    M llvm/test/Transforms/PhaseOrdering/X86/blendv-select.ll
    M llvm/test/Transforms/PhaseOrdering/X86/pr67803.ll

  Log Message:
  -----------
  [InstCombine][X86] Fold blendv(x,y,shuffle(bitcast(sext(m)))) -> select(shuffle(m),x,y) (#96882)

We already handle blendv(x,y,bitcast(sext(m))) -> select(m,x,y) cases, but this adds support for peeking through one-use shuffles as well. VectorCombine should already have canonicalized the IR to shuffle(bitcast(...)) for us.

The particular use case is where we have split generic 256/512-bit code to use target-specific blendv intrinsics (e.g. AVX1 spoofing AVX2 256-bit ops).

Fixes #58895



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