projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 04fb9b7) | patch
[mlir][sparse] enable a few vectorized runs in integration tests
authorAart Bik <ajcbik@google.com>
Tue, 24 Aug 2021 22:18:29 +0000 (15:18 -0700)
committerAart Bik <ajcbik@google.com>
Tue, 24 Aug 2021 23:08:01 +0000 (16:08 -0700)
commitc5735fada4971c88647c67b608bf91619b48949c
tree0e0b782d6e0ed23c289cf2ed34822a428b06b21ctree | snapshot
parent04fb9b729a53a5f513328159590d86d28336a6dacommit | diff
[mlir][sparse] enable a few vectorized runs in integration tests

Recent changes outside sparse compiler exposed the requirement of running a
new pass (lower-affine) but this only became apparent with private testing.
By adding some vectorized runs to integration test, we will detect the need
for such changes earlier and also widen codegen coverage of course.

Reviewed By: gussmith23

Differential Revision: https://reviews.llvm.org/D108667
mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_matvec.mlir diff | blob | history
mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_sampled_matmul.mlir diff | blob | history
Domain: System / Toolchain;