[all-commits] [llvm/llvm-project] 4672ba: [AMDGPU] Introduce Strict WQM mode

[Piotr Sobczak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 4672bac1776e2245abfd671e208fd56d21cca1ad
      https://github.com/llvm/llvm-project/commit/4672bac1776e2245abfd671e208fd56d21cca1ad
  Author: Piotr Sobczak <Piotr.Sobczak at amd.com>
  Date:   2021-03-03 (Wed, 03 Mar 2021)

  Changed paths:
    M llvm/include/llvm/IR/IntrinsicsAMDGPU.td
    M llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
    M llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
    M llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
    M llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
    M llvm/lib/Target/AMDGPU/SIInstructions.td
    M llvm/lib/Target/AMDGPU/SIPreAllocateWWMRegs.cpp
    M llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
    M llvm/test/CodeGen/AMDGPU/wqm.ll

  Log Message:
  -----------
  [AMDGPU] Introduce Strict WQM mode

* Add amdgcn_strict_wqm intrinsic.
* Add a corresponding STRICT_WQM machine instruction.
* The semantic is similar to amdgcn_strict_wwm with a notable difference that not all threads will be forcibly enabled during the computations of the intrinsic's argument, but only all threads in quads that have at least one thread active.
* The difference between amdgc_wqm and amdgcn_strict_wqm, is that in the strict mode an inactive lane will always be enabled irrespective of control flow decisions.

Reviewed By: critson

Differential Revision: https://reviews.llvm.org/D96258