[all-commits] [llvm/llvm-project] 6af81e: [mlir][std] Fold load(tensor_to_memref) into extra...

[Stephan Herhut via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 6af81ea1d6d36c7151a61f65e21b5c4ad9cf859d
      https://github.com/llvm/llvm-project/commit/6af81ea1d6d36c7151a61f65e21b5c4ad9cf859d
  Author: Stephan Herhut <herhut at google.com>
  Date:   2020-11-20 (Fri, 20 Nov 2020)

  Changed paths:
    M mlir/include/mlir/Dialect/StandardOps/IR/Ops.td
    M mlir/lib/Dialect/StandardOps/IR/Ops.cpp
    M mlir/test/Dialect/Standard/canonicalize.mlir

  Log Message:
  -----------
  [mlir][std] Fold load(tensor_to_memref) into extract_element

This canonicalization is useful to resolve loads into scalar values when
doing partial bufferization.

Differential Revision: https://reviews.llvm.org/D91855