[all-commits] [llvm/llvm-project] 2d2f7c: [X86] combineConcatVectorOps - pull out repeated O...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 2d2f7c1333ad1a9587f22c0505ce21d5106b9253
      https://github.com/llvm/llvm-project/commit/2d2f7c1333ad1a9587f22c0505ce21d5106b9253
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-08-11 (Fri, 11 Aug 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp

  Log Message:
  -----------
  [X86] combineConcatVectorOps - pull out repeated Ops.size() calls. NFCI.


  Commit: 6b36e0c544eaff501f70e991139af63fa5d57d21
      https://github.com/llvm/llvm-project/commit/6b36e0c544eaff501f70e991139af63fa5d57d21
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-08-11 (Fri, 11 Aug 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp

  Log Message:
  -----------
  [X86] combineConcatVectorOps - consistently use EltSizeInBits. NFCI.

Avoid repeated VT.getScalarSizeInBits() calls


  Commit: 2e79d0ad1c24d706fb60595bd624c1a13797a2d7
      https://github.com/llvm/llvm-project/commit/2e79d0ad1c24d706fb60595bd624c1a13797a2d7
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2023-08-11 (Fri, 11 Aug 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp

  Log Message:
  -----------
  [X86] combineConcatVectorOps - ensure X86ISD::SHUF128 nodes use vectors with 64-bit scalars

The actual shuffle always uses v8i64/v8f64 - the wrapping bitcast controls whether we use v16i32/v16f32 instead.


Compare: https://github.com/llvm/llvm-project/compare/14cfe929754a...2e79d0ad1c24