[all-commits] [llvm/llvm-project] 41849a: [mlir][Linalg] Avoid changing the rank of the resu...

[MaheshRavishankar via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 41849a91956755e15591240816d1d0c5ec402895
      https://github.com/llvm/llvm-project/commit/41849a91956755e15591240816d1d0c5ec402895
  Author: MaheshRavishankar <ravishankarm at google.com>
  Date:   2021-04-28 (Wed, 28 Apr 2021)

  Changed paths:
    M mlir/include/mlir/Dialect/StandardOps/Utils/Utils.h
    M mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp
    M mlir/lib/Dialect/MemRef/Transforms/FoldSubViewOps.cpp
    M mlir/lib/Dialect/StandardOps/IR/Ops.cpp
    M mlir/test/Dialect/MemRef/canonicalize.mlir
    M mlir/test/Dialect/Standard/canonicalize.mlir

  Log Message:
  -----------
  [mlir][Linalg] Avoid changing the rank of the result in canonicalizations of subtensor.

Canonicalizations for subtensor operations defaulted to use the
rank-reduced version of the operation, but the cast inserted to get
back the original type would be illegal if the rank was actually
reduced. Instead make the canonicalization not reduce the rank of the
operation.

Differential Revision: https://reviews.llvm.org/D101258