[all-commits] [llvm/llvm-project] a96ec0: [AMDGPU] Optimize out s_barrier_signal/_wait (#116...

[Piotr Sobczak via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a96ec01e1a269b663ccc1dadc2f4429fd0df887d
      https://github.com/llvm/llvm-project/commit/a96ec01e1a269b663ccc1dadc2f4429fd0df887d
  Author: Piotr Sobczak <piotr.sobczak at amd.com>
  Date:   2024-11-26 (Tue, 26 Nov 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    A llvm/test/CodeGen/AMDGPU/barrier-elimination-gfx12.ll

  Log Message:
  -----------
  [AMDGPU] Optimize out s_barrier_signal/_wait (#116993)

Extend the optimization that converts s_barrier to wave_barrier (nop)
when the number of work items is not larger than wave size.
    
This handles the "split barrier" form of s_barrier where the barrier
is represented by separate intrinsics (s_barrier_signal/s_barrier_wait).
Note: the version where s_barrier is used in gfx12 (and later split)
has the optimization already, but some front-ends may prefer to use
split intrinsics and this is being addressed by the patch.



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