diff options
| author | Lei Zhang <antiagainst@google.com> | 2023-04-18 13:57:35 -0700 |
|---|---|---|
| committer | Lei Zhang <antiagainst@google.com> | 2023-04-18 13:57:44 -0700 |
| commit | d343a395431f70f63d66ef31cb69c8c4babdb21f (patch) | |
| tree | 00d306b6c239d15144855ea61997dd6e51211a40 /llvm/lib/ExecutionEngine/Orc/PerfSupportPlugin.cpp | |
| parent | b51a03e1bbe7ea8868ffb24472fc532d0c00943c (diff) | |
| download | llvm-d343a395431f70f63d66ef31cb69c8c4babdb21f.zip llvm-d343a395431f70f63d66ef31cb69c8c4babdb21f.tar.gz llvm-d343a395431f70f63d66ef31cb69c8c4babdb21f.tar.bz2 | |
[mlir][spirv][vector] Enable vector3 when converting to dot product
It's common to see such cases for contraction from convolution with
input channel as 3. Although we aren't utilizing all 4 lanes for
dot product, it should still be better than performing the multiply
and reduction separately.
Reviewed By: kuhar
Differential Revision: https://reviews.llvm.org/D148642
Diffstat (limited to 'llvm/lib/ExecutionEngine/Orc/PerfSupportPlugin.cpp')
0 files changed, 0 insertions, 0 deletions