[all-commits] [llvm/llvm-project] f42b76: [mlir][Vector] Add fold transpose(shape_cast) -> s...

[Benjamin Maxwell via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: f42b7615b862bb5f77981f619f92877eb20adf54
      https://github.com/llvm/llvm-project/commit/f42b7615b862bb5f77981f619f92877eb20adf54
  Author: Benjamin Maxwell <benjamin.maxwell at arm.com>
  Date:   2023-12-01 (Fri, 01 Dec 2023)

  Changed paths:
    M mlir/lib/Dialect/Vector/IR/VectorOps.cpp
    M mlir/test/Dialect/Vector/canonicalize.mlir

  Log Message:
  -----------
  [mlir][Vector] Add fold transpose(shape_cast) -> shape_cast (#73951)

This folds transpose(shape_cast) into a new shape_cast, when the
transpose just permutes a unit dim from the result of the shape_cast.

Example:

```
%0 = vector.shape_cast %vec : vector<[4]xf32> to vector<[4]x1xf32>
%1 = vector.transpose %0, [1, 0] : vector<[4]x1xf32> to vector<1x[4]xf32>
```

Folds to:
```
%0 = vector.shape_cast %vec : vector<[4]xf32> to vector<1x[4]xf32>
```

This is an (alternate) fix for lowering matmuls to ArmSME.