[all-commits] [llvm/llvm-project] 96e94b: [AMDGPU] Add a new amdgcn.load.to.lds intrinsic

[Krzysztof Drewniak via All-commits]

  Branch: refs/heads/users/krzysz00/load-lds-everywhere
  Home:   https://github.com/llvm/llvm-project
  Commit: 96e94b5662c613fd80f712080751076254a73524
      https://github.com/llvm/llvm-project/commit/96e94b5662c613fd80f712080751076254a73524
  Author: Krzysztof Drewniak <Krzysztof.Drewniak at amd.com>
  Date:   2025-04-26 (Sat, 26 Apr 2025)

  Changed paths:
    M clang/include/clang/Basic/BuiltinsAMDGPU.def
    M clang/lib/CodeGen/TargetBuiltins/AMDGPU.cpp
    M clang/lib/Sema/SemaAMDGPU.cpp
    M clang/test/CodeGenOpenCL/builtins-amdgcn-gfx950.cl
    A clang/test/CodeGenOpenCL/builtins-amdgcn-load-to-lds.cl
    M llvm/docs/ReleaseNotes.md
    M llvm/include/llvm/IR/IntrinsicsAMDGPU.td
    M llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
    M llvm/lib/Target/AMDGPU/AMDGPULowerBufferFatPointers.cpp
    M llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
    M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
    A llvm/test/CodeGen/AMDGPU/llvm.amdgcn.load.to.lds.gfx950.ll
    A llvm/test/CodeGen/AMDGPU/llvm.amdgcn.load.to.lds.ll
    M llvm/test/CodeGen/AMDGPU/lower-buffer-fat-pointers-mem-transfer.ll
    M mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td
    M mlir/include/mlir/Dialect/LLVMIR/ROCDLOps.td
    M mlir/lib/Conversion/AMDGPUToROCDL/AMDGPUToROCDL.cpp
    M mlir/lib/Dialect/AMDGPU/IR/AMDGPUDialect.cpp
    M mlir/test/Conversion/AMDGPUToROCDL/load_lds.mlir
    M mlir/test/Dialect/LLVMIR/rocdl.mlir
    M mlir/test/Target/LLVMIR/rocdl.mlir

  Log Message:
  -----------
  [AMDGPU] Add a new amdgcn.load.to.lds intrinsic

This PR adds a amdgns_load_to_lds intrinsic that abstracts over loads
to LDS from global (address space 1) pointers and buffer fat
pointers (address space 7), since they use the saem API and "gather
from a pointer to LDS" is something of an abstract operation.

This commet adds the intrinsic and its lowerings for addrspaces 1 and
7, and updates the MLIR wrappers to use it (loosening up the
restrictions on loads to LDS along the way to match the ground truth
from target features).

It also plumbs the intrinsic through to clang.



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