[all-commits] [llvm/llvm-project] 7a7eac: [MLIR][GPU] Implement a simple greedy loop mapper.

[Stephan Herhut via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 7a7eacc797f7cc603d50987883ea95aee99d6b22
      https://github.com/llvm/llvm-project/commit/7a7eacc797f7cc603d50987883ea95aee99d6b22
  Author: Stephan Herhut <herhut at google.com>
  Date:   2020-02-25 (Tue, 25 Feb 2020)

  Changed paths:
    A mlir/include/mlir/Dialect/GPU/ParallelLoopMapper.h
    M mlir/include/mlir/Dialect/LoopOps/LoopOps.td
    M mlir/lib/Conversion/LoopsToGPU/LoopsToGPU.cpp
    M mlir/lib/Dialect/GPU/CMakeLists.txt
    A mlir/lib/Dialect/GPU/Transforms/ParallelLoopMapper.cpp
    A mlir/test/Dialect/GPU/mapping.mlir
    M mlir/test/lib/Transforms/CMakeLists.txt
    A mlir/test/lib/Transforms/TestGpuParallelLoopMapping.cpp
    M mlir/tools/mlir-opt/mlir-opt.cpp

  Log Message:
  -----------
  [MLIR][GPU] Implement a simple greedy loop mapper.

Summary:
The mapper assigns annotations to loop.parallel operations that
are compatible with the loop to gpu mapping pass. The outermost
loop uses the grid dimensions, followed by block dimensions. All
remaining loops are mapped to sequential loops.

Differential Revision: https://reviews.llvm.org/D74963