[all-commits] [llvm/llvm-project] d8bc4d: [X86] Fold cmpeq/ne(or(X, Y), X) --> cmpeq/ne(and(~X...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d8bc4de3cfe045cf393f6928e359a6f05deffd55
      https://github.com/llvm/llvm-project/commit/d8bc4de3cfe045cf393f6928e359a6f05deffd55
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-04-09 (Fri, 09 Apr 2021)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/avx512-cmp.ll
    M llvm/test/CodeGen/X86/pr27202.ll
    M llvm/test/CodeGen/X86/ragreedy-hoist-spill.ll
    M llvm/test/CodeGen/X86/setcc-logic.ll
    M llvm/test/CodeGen/X86/switch-or.ll

  Log Message:
  -----------
  [X86] Fold cmpeq/ne(or(X,Y),X) --> cmpeq/ne(and(~X,Y),0) on non-BMI targets (PR44136)

Followup to D100177, enable the fold for non-BMI targets as well.