[all-commits] [llvm/llvm-project] a834b7: [X86] foldMaskAndShiftToExtract - as long as we ca...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a834b79b7de3870045e039367775c988920d7d3a
      https://github.com/llvm/llvm-project/commit/a834b79b7de3870045e039367775c988920d7d3a
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-08-18 (Fri, 18 Aug 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelDAGToDAG.cpp

  Log Message:
  -----------
  [X86] foldMaskAndShiftToExtract - as long as we call insertDAGNode in the creation order then duplicates don't matter.


  Commit: bd9bf9cb6708ed193df23d09bafc16306f55d14e
      https://github.com/llvm/llvm-project/commit/bd9bf9cb6708ed193df23d09bafc16306f55d14e
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-08-18 (Fri, 18 Aug 2023)

  Changed paths:
    M llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp

  Log Message:
  -----------
  [X86] SimplifyDemandedBits - move MaskedValueIsZero as late as possible to avoid unnecessary (recursive) analysis costs. NFC.

Mentioned on D155472 for the SHL equivalent


Compare: https://github.com/llvm/llvm-project/compare/c931f2e6fd0c...bd9bf9cb6708