[all-commits] [llvm/llvm-project] c1fabd: [llvm][AMDGPU] Enable FWD_PROGRESS bit for GFX10+ ...

[Alex Voicu via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: c1fabd681fed3e024879f64f2e122e667778b9be
      https://github.com/llvm/llvm-project/commit/c1fabd681fed3e024879f64f2e122e667778b9be
  Author: Alex Voicu <alexandru.voicu at amd.com>
  Date:   2025-03-17 (Mon, 17 Mar 2025)

  Changed paths:
    M llvm/docs/AMDGPUUsage.rst
    M llvm/docs/ReleaseNotes.md
    M llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
    M llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUMCKernelDescriptor.cpp
    M llvm/lib/Target/AMDGPU/SIProgramInfo.cpp
    M llvm/lib/Target/AMDGPU/SIProgramInfo.h
    M llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
    M llvm/test/MC/AMDGPU/hsa-gfx12-v4.s
    M llvm/test/MC/AMDGPU/hsa-sym-exprs-gfx10.s
    M llvm/test/MC/AMDGPU/hsa-sym-exprs-gfx11.s
    M llvm/test/MC/AMDGPU/hsa-sym-exprs-gfx12.s
    M llvm/test/tools/llvm-objdump/ELF/AMDGPU/kd-gfx10.s
    M llvm/test/tools/llvm-objdump/ELF/AMDGPU/kd-gfx11.s
    M llvm/test/tools/llvm-objdump/ELF/AMDGPU/kd-gfx12.s

  Log Message:
  -----------
  [llvm][AMDGPU] Enable FWD_PROGRESS bit for GFX10+ (#128367)

>From GFX10 onwards it is possible to employ benevolent scheduling of
waves. This patch unconditionally enables, for the `amdhsa` OS, the bit
which controls that capability, as it is beneficial for algorithms that
rely on more complex concurrent coordination and it is generally
performance neutral otherwise.



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