projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 86a63b2) | patch
[mlir] Add vectorize_nd_extract attribute to the masked_vectorize Op
authorAndrzej Warzynski <andrzej.warzynski@gmail.com>
Thu, 26 Jan 2023 10:18:44 +0000 (10:18 +0000)
committerAndrzej Warzynski <andrzej.warzynski@gmail.com>
Wed, 15 Feb 2023 08:47:03 +0000 (08:47 +0000)
commit71712440bbb26b8e1789372024c3a14428b68eae
treeda6926a482439349553cc8e54bc778bb142a12d9tree | snapshot
parent86a63b2ae147e5a3edc39643783acfd39b059c92commit | diff
[mlir] Add vectorize_nd_extract attribute to the masked_vectorize Op

This patch simply adds `vectorize_nd_extract` (that's currently only
used for the `vectorize` Op) to `masked_vectorize`. A test is added to
verify that it works as expected - it prevents the masked vectorisation
of `tensor.extract`, which is currently not supported.

Differential Revision: https://reviews.llvm.org/D142634
mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td diff | blob | history
mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp diff | blob | history
mlir/test/Dialect/Linalg/vectorization-unsupported.mlir [new file with mode: 0644] blob
Domain: System / Toolchain;