[all-commits] [llvm/llvm-project] da3f35: AMDGPU: Don't look at dbg users for foldable operands

[Matt Arsenault via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: da3f357de67833ff2ded813a02158f3ec6dc25a7
      https://github.com/llvm/llvm-project/commit/da3f357de67833ff2ded813a02158f3ec6dc25a7
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2020-08-17 (Mon, 17 Aug 2020)

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

  Log Message:
  -----------
  AMDGPU: Don't look at dbg users for foldable operands

These would have always failed to fold, so checking them or adding
them to the fold candidates is useless.


  Commit: af162ac785dd6aa371afc0f111544c82d72c5242
      https://github.com/llvm/llvm-project/commit/af162ac785dd6aa371afc0f111544c82d72c5242
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2020-08-17 (Mon, 17 Aug 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
    A llvm/test/CodeGen/AMDGPU/GlobalISel/regbankselect-amdgcn.ballot.i64.mir

  Log Message:
  -----------
  AMDGPU/GlobalISel: Fix using readfirstlane with ballot intrinsics

This should use the default mapping and insert a copy to the vcc bank,
and not try to insert a readfirstlane.


  Commit: c7b9cd31bfb4b5bbce317b28e83d4e7ac2c746d1
      https://github.com/llvm/llvm-project/commit/c7b9cd31bfb4b5bbce317b28e83d4e7ac2c746d1
  Author: Matt Arsenault <Matthew.Arsenault at amd.com>
  Date:   2020-08-17 (Mon, 17 Aug 2020)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUGenRegisterBankInfo.def
    M llvm/test/CodeGen/AMDGPU/GlobalISel/regbankselect-build-vector.mir

  Log Message:
  -----------
  AMDGPU/GlobalISel: Fix missing 256-bit AGPR mapping


Compare: https://github.com/llvm/llvm-project/compare/9c4825ce282d...c7b9cd31bfb4