[all-commits] [llvm/llvm-project] 05a893: [mlir][Linalg] Allow folding to rank-zero tensor w...

[MaheshRavishankar via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 05a89312d812bb5dcec6deca8f1e28a198ce1167
      https://github.com/llvm/llvm-project/commit/05a89312d812bb5dcec6deca8f1e28a198ce1167
  Author: MaheshRavishankar <ravishankarm at google.com>
  Date:   2021-05-06 (Thu, 06 May 2021)

  Changed paths:
    M mlir/lib/Dialect/Linalg/Transforms/DropUnitDims.cpp
    M mlir/test/Dialect/Linalg/drop-unit-extent-dims.mlir

  Log Message:
  -----------
  [mlir][Linalg] Allow folding to rank-zero tensor when using rank-reducing subtensors.

The pattern to convert subtensor ops to their rank-reduced versions
(by dropping unit-dims in the result) can also convert to a zero-rank
tensor. Handle that case.
This also fixes a OOB access bug in the existing pattern for such
cases.

Differential Revision: https://reviews.llvm.org/D101949