[all-commits] [llvm/llvm-project] 3d67d9: [MLIR] Remove unnecessary generated-header include...

[Mehdi Amini via All-commits]

  Branch: refs/heads/users/mamini/mlir-prune-generated-header-includes
  Home:   https://github.com/llvm/llvm-project
  Commit: 3d67d91eb71a60a9ef10a2ad143769ff920188b5
      https://github.com/llvm/llvm-project/commit/3d67d91eb71a60a9ef10a2ad143769ff920188b5
  Author: Mehdi Amini <joker.eph at gmail.com>
  Date:   2026-09-09 (Wed, 09 Sep 2026)

  Changed paths:
    M mlir/lib/Conversion/ArithAndMathToAPFloat/MathToAPFloat.cpp
    M mlir/lib/Conversion/ArithToAMDGPU/ArithToAMDGPU.cpp
    M mlir/lib/Dialect/AMDGPU/Transforms/MaskedloadToLoad.cpp
    M mlir/lib/Dialect/ArmNeon/IR/ArmNeonDialect.cpp
    M mlir/lib/Dialect/GPU/Pipelines/GPUToXeVMPipeline.cpp
    M mlir/lib/Dialect/LLVMIR/IR/VCIXDialect.cpp
    M mlir/lib/Dialect/SparseTensor/Pipelines/SparseTensorPipelines.cpp
    M mlir/test/lib/Dialect/NVGPU/TestNVGPUTransforms.cpp
    M mlir/test/lib/Dialect/Shard/TestReshardingPartition.cpp
    M mlir/test/lib/Dialect/Tosa/TosaTestPasses.cpp

  Log Message:
  -----------
  [MLIR] Remove unnecessary generated-header include dependencies

Remove unused dialect and transform headers, and replace broad pass aggregates
with the specific conversion headers used by GPU and SparseTensor pipelines.
Include the core pass and IR builder definitions directly where unrelated
dialect headers previously supplied them transitively.

These include-only changes avoid unnecessary TableGen header prerequisites
without depending on changes to the CMake generated-header dependency model.

Compile all ten affected translation units with PCH disabled in a fresh build
configured from this checkout. Check compiler dependency output to confirm
the eliminated provider header families are no longer included.

Assisted-by: Codex



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