[all-commits] [llvm/llvm-project] fa1012: [mlir][MLProgram] Add MLProgram to MemRef bufferiz...

[Ryan Holt via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: fa10121415adfea57954eddcec92a13c7ebab0ac
      https://github.com/llvm/llvm-project/commit/fa10121415adfea57954eddcec92a13c7ebab0ac
  Author: Ryan Holt <ryanholt at mathworks.com>
  Date:   2024-01-30 (Tue, 30 Jan 2024)

  Changed paths:
    A mlir/include/mlir/Dialect/MLProgram/Transforms/BufferizableOpInterfaceImpl.h
    M mlir/include/mlir/InitAllDialects.h
    M mlir/lib/Dialect/Bufferization/Transforms/Bufferize.cpp
    M mlir/lib/Dialect/Bufferization/Transforms/OneShotModuleBufferize.cpp
    A mlir/lib/Dialect/MLProgram/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/MLProgram/Transforms/CMakeLists.txt
    A mlir/test/Dialect/MLProgram/one-shot-bufferize.mlir

  Log Message:
  -----------
  [mlir][MLProgram] Add MLProgram to MemRef bufferization pass (#75103)

There is currently no lowering out of `ml_program` in the LLVM
repository. This change adds a lowering to `memref` so that it can be
lowered all the way to LLVM. This lowering was taken from the [reference
backend in
torch-mlir](https://github.com/llvm/torch-mlir/commit/f41695360019bde71d52ca7548944d5488779e12
).

I had tried implementing the `BufferizableOpInterface` for `ml_program`
instead of adding a new pass but that did not work because
`OneShotBufferize` does not visit module-level ops like
`ml_program.global`.