[all-commits] [llvm/llvm-project] 56796a: [mlir][linalg] Fix tensor tiling together with int...

[Lei Zhang via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 56796ae1a8db4c85dada28676f8303a5a3609c63
      https://github.com/llvm/llvm-project/commit/56796ae1a8db4c85dada28676f8303a5a3609c63
  Author: Lei Zhang <antiagainst at google.com>
  Date:   2022-07-15 (Fri, 15 Jul 2022)

  Changed paths:
    M mlir/lib/Dialect/Linalg/Transforms/FusionOnTensors.cpp
    M mlir/test/Dialect/Linalg/transform-op-fuse.mlir

  Log Message:
  -----------
  [mlir][linalg] Fix tensor tiling together with interchange

In `linalg::tileConsumerAndFuseProducers`, there are two levels of
tiling and fusion; we partition the tile sizes and only use one
half for each of them. The partition is using the first non-parallel
dimension *after* interchange as the boundary. However, concrete
tiling happens *together with* loop interchange, so we still need
to provide the partial tile sizes *before* the interchange.
Otherwise, there will be inconsistency, which is what this patch
is to fix.

Reviewed By: ThomasRaoux

Differential Revision: https://reviews.llvm.org/D129804