[all-commits] [llvm/llvm-project] bdcce4: [X86] combineMaskBitOp - fold vXi1 logicop(truncat...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: bdcce4a12c828d7cf5475e5f39dde32280c29a3c
      https://github.com/llvm/llvm-project/commit/bdcce4a12c828d7cf5475e5f39dde32280c29a3c
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2026-05-29 (Fri, 29 May 2026)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/avx512-mask-op.ll
    M llvm/test/CodeGen/X86/fold-select.ll

  Log Message:
  -----------
  [X86] combineMaskBitOp - fold vXi1 logicop(truncate(N0),truncate(N1)) -> truncate(logicop(X,Y)) (#200385)

Replace costly predicate logic when possible - generic 128/256/512 vector types are always cheaper

Part of #198162



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