diff options
| author | Rob Suderman <rob.suderman@gmail.com> | 2021-09-15 12:44:17 -0700 |
|---|---|---|
| committer | Rob Suderman <rob.suderman@gmail.com> | 2021-09-15 14:09:15 -0700 |
| commit | 1ac2d195ecb5d4c549c11b9c1df00179f5fea7ed (patch) | |
| tree | 8fc4917bbded97e24eba4be1a2be44acdc549de5 /llvm/lib/CodeGen/SplitKit.cpp | |
| parent | 3273430406c186f1f2af597adeedf21d4fa52b18 (diff) | |
| download | llvm-1ac2d195ecb5d4c549c11b9c1df00179f5fea7ed.zip llvm-1ac2d195ecb5d4c549c11b9c1df00179f5fea7ed.tar.gz llvm-1ac2d195ecb5d4c549c11b9c1df00179f5fea7ed.tar.bz2 | |
[mlir][linalg] Add canonicalizers for depthwise conv
There are two main versions of depthwise conv depending whether the multiplier
is 1 or not. In cases where m == 1 we should use the version without the
multiplier channel as it can perform greater optimization.
Add lowering for the quantized/float versions to have a multiplier of one.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D108959
Diffstat (limited to 'llvm/lib/CodeGen/SplitKit.cpp')
0 files changed, 0 insertions, 0 deletions