[all-commits] [llvm/llvm-project] 938599: [X86] combineMOVMSK - fold movmsk(logic(X, C)) -> l...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9385996ec6d18ec6c52991b40489bd3cd8766594
      https://github.com/llvm/llvm-project/commit/9385996ec6d18ec6c52991b40489bd3cd8766594
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-03-29 (Wed, 29 Mar 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/masked_store.ll
    M llvm/test/CodeGen/X86/vector-reduce-or-cmp.ll

  Log Message:
  -----------
  [X86] combineMOVMSK - fold movmsk(logic(X,C)) -> logic(movmsk(X),C)

Avoid a vector mask load for something that is likely to be able to fold into a scalar logic op + EFLAGS comparison result