[all-commits] [llvm/llvm-project] 8d0c34: [AMDGPU] Omit unnecessary waitcnt before barriers

[Austin Kerbow via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 8d0c34fd4fb66ea0d19563154a59658e4b7f35d4
      https://github.com/llvm/llvm-project/commit/8d0c34fd4fb66ea0d19563154a59658e4b7f35d4
  Author: Austin Kerbow <Austin.Kerbow at amd.com>
  Date:   2022-03-07 (Mon, 07 Mar 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPU.td
    M llvm/lib/Target/AMDGPU/GCNSubtarget.h
    M llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
    A llvm/test/CodeGen/AMDGPU/back-off-barrier-subtarget-feature.ll
    M llvm/test/CodeGen/AMDGPU/waitcnt-preexisting-vscnt.mir
    M llvm/test/CodeGen/AMDGPU/waitcnt-vscnt.ll

  Log Message:
  -----------
  [AMDGPU] Omit unnecessary waitcnt before barriers

It is not necessary to wait for all outstanding memory operations before
barriers on hardware that can back off of the barrier in the event of an
exception when traps are enabled. Add a new subtarget feature which
tracks which HW has this ability.

Reviewed By: #amdgpu, rampitec

Differential Revision: https://reviews.llvm.org/D120544