[all-commits] [llvm/llvm-project] ee01c7: [mlir] [VectorOps] Add choice between dot and axpy...

[Aart Bik via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: ee01c7a7406345a50176216216ca384fb60e0267
      https://github.com/llvm/llvm-project/commit/ee01c7a7406345a50176216216ca384fb60e0267
  Author: aartbik <ajcbik at google.com>
  Date:   2020-07-02 (Thu, 02 Jul 2020)

  Changed paths:
    M mlir/include/mlir/Dialect/Vector/VectorOps.h
    M mlir/include/mlir/Dialect/Vector/VectorTransforms.h
    M mlir/lib/Dialect/Vector/VectorTransforms.cpp
    A mlir/test/Dialect/Vector/vector-contract-matvec-transforms.mlir
    M mlir/test/lib/Transforms/TestVectorTransforms.cpp

  Log Message:
  -----------
  [mlir] [VectorOps] Add choice between dot and axpy lowering of vector.contract

Default vector.contract lowering essentially yields a series of sdot/ddot
operations. However, for some layouts a series of saxpy/daxpy operations,
chained through fma are more efficient. This CL introduces a choice between
the two lowering paths. A default heuristic is to follow.

Some preliminary avx2 performance numbers for matrix-times-vector.
Here, dot performs best for 64x64 A x b and saxpy for 64x64 A^T x b.

```
------------------------------------------------------------
            A x b                          A^T x b
------------------------------------------------------------
GFLOPS    sdot (reassoc)    saxpy    sdot (reassoc)    saxpy
------------------------------------------------------------
1x1        0.6               0.9       0.6             0.9
2x2        2.5               3.2       2.4             3.5
4x4        6.4               8.4       4.9             11.8
8x8       11.7               6.1       5.0             29.6
16x16     20.7              10.8       7.3             43.3
32x32     29.3               7.9       6.4             51.8
64x64     38.9                                         79.3
128x128   32.4                                         40.7
------------------------------------------------------------
```

Reviewed By: nicolasvasilache, ftynse

Differential Revision: https://reviews.llvm.org/D83012