[all-commits] [llvm/llvm-project] 89d114: [mlir][gpu]Add GPUToXeVM lowering pipeline pass. (...

[Md Abdullah Shahneous Bari via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 89d1143a90b7bf9b9ebabab1d8b7ed519c6ec5eb
      https://github.com/llvm/llvm-project/commit/89d1143a90b7bf9b9ebabab1d8b7ed519c6ec5eb
  Author: Md Abdullah Shahneous Bari <98356296+mshahneo at users.noreply.github.com>
  Date:   2025-10-17 (Fri, 17 Oct 2025)

  Changed paths:
    M mlir/include/mlir/Dialect/GPU/Pipelines/Passes.h
    M mlir/lib/Dialect/GPU/Pipelines/CMakeLists.txt
    A mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp
    M mlir/lib/RegisterAllPasses.cpp
    A mlir/test/Integration/Dialect/XeGPU/LANE/lit.local.cfg
    A mlir/test/Integration/Dialect/XeGPU/LANE/simple_gemm.mlir
    A mlir/test/Integration/Dialect/XeGPU/SG/lit.local.cfg
    A mlir/test/Integration/Dialect/XeGPU/SG/simple_gemm.mlir
    A mlir/test/Integration/Dialect/XeGPU/WG/lit.local.cfg
    A mlir/test/Integration/Dialect/XeGPU/WG/simple_gemm.mlir

  Log Message:
  -----------
  [mlir][gpu]Add GPUToXeVM lowering pipeline pass. (#161216)

It's the default GPU to XeVM lowering pipeline. It starts by lowering
GPU code to the specified compilation target (default is fatbin), then
lowers the host code.
If XeGPU ops are used, it expects the MLIR code to have XeGPU ops
already embedded in gpu code.



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