diff options
| author | Quinn Dawkins <quinn@nod-labs.com> | 2023-03-18 15:02:07 -0400 |
|---|---|---|
| committer | Quinn Dawkins <quinn@nod-labs.com> | 2023-07-31 16:15:25 -0400 |
| commit | e444dcef7cadb29490deca726cb743e38d0412b3 (patch) | |
| tree | 66585c011fcefc1972936b11657d0ba2f46adcf3 /llvm/lib/CodeGen/TargetLoweringBase.cpp | |
| parent | aeef9c0274c6419d15df5cb2bb885aa3072423d6 (diff) | |
| download | llvm-e444dcef7cadb29490deca726cb743e38d0412b3.zip llvm-e444dcef7cadb29490deca726cb743e38d0412b3.tar.gz llvm-e444dcef7cadb29490deca726cb743e38d0412b3.tar.bz2 | |
[mlir][Transform] Add classify_convolution_dims op
Includes `inferConvolutionDims` based on the existing helper for
contractions, `inferContractionDims`. This allows matching and
identifying the relevant dims for a convolution sub-computation of
a linalg operation.
Additionally adds stride/dilations inference to the captures and
convolution interface matcher.
Differential Revision: https://reviews.llvm.org/D156080
Diffstat (limited to 'llvm/lib/CodeGen/TargetLoweringBase.cpp')
0 files changed, 0 insertions, 0 deletions