diff options
| author | Jakub Kuderski <kubak@google.com> | 2023-03-09 14:56:20 -0500 |
|---|---|---|
| committer | Jakub Kuderski <kubak@google.com> | 2023-03-09 14:56:21 -0500 |
| commit | fb7ef637a84652dbd3d973a1ba7db9470181b5aa (patch) | |
| tree | 7a594f47e8ff3a36ff6da3a6ede3b1380b4020f7 /flang/lib/Frontend/CompilerInvocation.cpp | |
| parent | e99b5ad38381ab263820b23a184d217a4112519c (diff) | |
| download | llvm-fb7ef637a84652dbd3d973a1ba7db9470181b5aa.zip llvm-fb7ef637a84652dbd3d973a1ba7db9470181b5aa.tar.gz llvm-fb7ef637a84652dbd3d973a1ba7db9470181b5aa.tar.bz2 | |
[mlir][vector][nvgpu] Move MMA contraction preparation to VectorUtils
This pattern is not specific to nvgpu; I intend to use in SPIR-V codegen. `VectorTransforms` seems like a more generally useful place.
In addition:
- Fix a bug in the second condition (the dimensions were swapped for RHS).
- Add tests.
- Add support for externally provided filter functions, similar to other vector transforms.
- Prefer to transpose before zero/sign-extending inputs.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D145638
Diffstat (limited to 'flang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions