[all-commits] [llvm/llvm-project] 8a3d12: [AMDGPU] Add getLocalMemorySize to TargetParser

[Chinmay Deshpande via All-commits]

  Branch: refs/heads/users/chinmaydd/lds-local-memory-size
  Home:   https://github.com/llvm/llvm-project
  Commit: 8a3d12cec2539eb980ad3f28a7325e38e28c7dab
      https://github.com/llvm/llvm-project/commit/8a3d12cec2539eb980ad3f28a7325e38e28c7dab
  Author: Chinmay Deshpande <chdeshpa at amd.com>
  Date:   2026-09-12 (Sat, 12 Sep 2026)

  Changed paths:
    M llvm/include/llvm/TargetParser/AMDGPUTargetParser.h
    M llvm/lib/Target/AMDGPU/GCNSubtarget.cpp
    M llvm/lib/TargetParser/AMDGPUTargetParser.cpp
    M llvm/unittests/TargetParser/TargetParserTest.cpp

  Log Message:
  -----------
  [AMDGPU] Add getLocalMemorySize to TargetParser

Add getLocalMemorySize and getAddressableLocalMemorySize, both taking a
GPUKind or a Triple::SubArchType, so the LDS a work-group gets can be
queried from a GPU name alone without an MCSubtargetInfo. The first
returns the physical block available in the current mode, the second
caps it at what one work-group can address, mirroring the IsaInfo pair.

The number of SIMDs a work-group runs on is a per-kernel mode rather
than a property of the GPU, so it stays a parameter. GCNSubtarget
initializes its cached sizes from the new entry points. There is no
functional change.

Change-Id: Ib71428b66032a231ed491d6294122b359abfe7e2
Co-Authored-By: Claude Opus 5 (1M context) <noreply at anthropic.com>


  Commit: 97e29895c338ffb1546d2bc4f6e1748a3c77f440
      https://github.com/llvm/llvm-project/commit/97e29895c338ffb1546d2bc4f6e1748a3c77f440
  Author: Chinmay Deshpande <chdeshpa at amd.com>
  Date:   2026-09-12 (Sat, 12 Sep 2026)

  Changed paths:
    M llvm/include/llvm/TargetParser/AMDGPUTargetParser.h
    M llvm/lib/Target/AMDGPU/GCNSubtarget.cpp
    M llvm/lib/Target/AMDGPU/GCNSubtarget.h
    M llvm/lib/TargetParser/AMDGPUTargetParser.cpp

  Log Message:
  -----------
  [AMDGPU] Use SIMD-mode terminology for LDS queries

Change-Id: I6d6900868d58374d295126c52f67ffcdd8676902


Compare: https://github.com/llvm/llvm-project/compare/49ea2edc1c4a...97e29895c338

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