[all-commits] [llvm/llvm-project] f1172c: AMDGPU: Simplify demanded bits on readlane/writeli...

[Matt Arsenault via All-commits]

  Branch: refs/heads/users/arsenm/amdgpu-simplify-demanded-readlane-index
  Home:   https://github.com/llvm/llvm-project
  Commit: f1172c1063af418ee93c0a62d9243afa74d1bd9d
      https://github.com/llvm/llvm-project/commit/f1172c1063af418ee93c0a62d9243afa74d1bd9d
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2024-12-02 (Mon, 02 Dec 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
    M llvm/test/Transforms/InstCombine/AMDGPU/lane-index-simplify-demanded-bits.ll

  Log Message:
  -----------
  AMDGPU: Simplify demanded bits on readlane/writeline index arguments

The main goal is to fold away wave64 code when compiled for wave32.
If we have out of bounds indexing, these will now clamp down to
a low bit which may CSE with the operations on the low half of the
wave.


  Commit: 1894d976e8f852575a2ef6b808d4ae13ee494c0c
      https://github.com/llvm/llvm-project/commit/1894d976e8f852575a2ef6b808d4ae13ee494c0c
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2024-12-02 (Mon, 02 Dec 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp

  Log Message:
  -----------
  Just use getWavefrontSizeLog2


  Commit: 45498a1816045d78a2eda9ce13a083fbedec47a4
      https://github.com/llvm/llvm-project/commit/45498a1816045d78a2eda9ce13a083fbedec47a4
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2024-12-02 (Mon, 02 Dec 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp

  Log Message:
  -----------
  Reword comment


Compare: https://github.com/llvm/llvm-project/compare/5a617f85379e...45498a181604

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