[all-commits] [llvm/llvm-project] 104310: [mlir][bufferize] Insert memref.cast ops during fi...

[Matthias Springer via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 1043107ce5e2dee38f6a9bf459549a75f78a83b2
      https://github.com/llvm/llvm-project/commit/1043107ce5e2dee38f6a9bf459549a75f78a83b2
  Author: Matthias Springer <springerm at google.com>
  Date:   2022-01-27 (Thu, 27 Jan 2022)

  Changed paths:
    M mlir/include/mlir/Dialect/Bufferization/IR/Bufferization.h
    M mlir/lib/Dialect/Bufferization/IR/BufferizationOps.cpp
    M mlir/lib/Dialect/Bufferization/Transforms/Bufferize.cpp
    M mlir/test/Dialect/Bufferization/Transforms/finalizing-bufferize.mlir

  Log Message:
  -----------
  [mlir][bufferize] Insert memref.cast ops during finalizing pass

The pass can currently not handle to_memref(to_tensor(x)) folding where a cast is necessary. This is required with the new unified bufferization. There is already a canonicalization pattern that handles such foldings and it should be used during this pass.

Differential Revision: https://reviews.llvm.org/D117988