[all-commits] [llvm/llvm-project] d0ee09: [mlir][Bufferize] Fix incorrect bufferization of r...

[Nicolas Vasilache via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d0ee094d6acf72608e927bf2e9ba69c57da59a96
      https://github.com/llvm/llvm-project/commit/d0ee094d6acf72608e927bf2e9ba69c57da59a96
  Author: Nicolas Vasilache <ntv at google.com>
  Date:   2022-01-10 (Mon, 10 Jan 2022)

  Changed paths:
    M mlir/include/mlir/Interfaces/ViewLikeInterface.td
    M mlir/lib/Dialect/Linalg/ComprehensiveBufferize/BufferizableOpInterface.cpp
    M mlir/lib/Dialect/Linalg/ComprehensiveBufferize/LinalgInterfaceImpl.cpp
    M mlir/lib/Dialect/Linalg/ComprehensiveBufferize/SCFInterfaceImpl.cpp
    M mlir/lib/Dialect/Linalg/ComprehensiveBufferize/TensorInterfaceImpl.cpp
    M mlir/lib/Dialect/Linalg/Transforms/ComprehensiveBufferizePass.cpp
    M mlir/lib/Interfaces/ViewLikeInterface.cpp
    M mlir/test/Dialect/Linalg/comprehensive-function-bufferize.mlir
    M mlir/test/Dialect/Linalg/comprehensive-module-bufferize-analysis.mlir
    M mlir/test/Dialect/Linalg/comprehensive-module-bufferize.mlir
    M mlir/test/lib/Dialect/Linalg/TestComprehensiveBufferize.cpp

  Log Message:
  -----------
  [mlir][Bufferize] Fix incorrect bufferization of rank-reducing tensor ops.

This revision fixes SubviewOp, InsertSliceOp, ExtractSliceOp construction during bufferization
where not all offset/size/stride operands were properly specified.

A test that exhibited problematic behaviors related to incorrect memref casts is introduced.
Init tensor optimization is disabled in teh testing func bufferize pass.

Differential Revision: https://reviews.llvm.org/D116899