diff options
| author | aartbik <ajcbik@google.com> | 2020-06-19 10:40:03 -0700 |
|---|---|---|
| committer | aartbik <ajcbik@google.com> | 2020-06-19 10:40:08 -0700 |
| commit | 0d82ab788534bfc05a4455292e27ed035caac639 (patch) | |
| tree | 9ea3f7939964bfa305f8b9ae2fdd08f735eb2bbd /llvm/unittests/Support/FileCheckTest.cpp | |
| parent | f3e8f9617365f129a0e9876fc3d38f15d6f2d4fd (diff) | |
| download | llvm-0d82ab788534bfc05a4455292e27ed035caac639.zip llvm-0d82ab788534bfc05a4455292e27ed035caac639.tar.gz llvm-0d82ab788534bfc05a4455292e27ed035caac639.tar.bz2 | |
[mlir] [VectorOps] Improve vector.constant_mask lowering
Use direct vector constants for the 1-D case. This approach
scales much better than generating elaborate insertion operations
that are eventually folded into a constant. We could of course
generalize the 1-D case to higher ranks, but this simplification
already helps in scaling some microbenchmarks that would formerly
crash on the intermediate IR length.
Reviewed By: reidtatge
Differential Revision: https://reviews.llvm.org/D82144
Diffstat (limited to 'llvm/unittests/Support/FileCheckTest.cpp')
0 files changed, 0 insertions, 0 deletions