projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a68ccda) | patch
[mlir][sparse] refine lexicographic insertion to any tensor
authorAart Bik <ajcbik@google.com>
Tue, 16 Nov 2021 20:46:55 +0000 (12:46 -0800)
committerAart Bik <ajcbik@google.com>
Thu, 18 Nov 2021 02:08:42 +0000 (18:08 -0800)
commit1ce77b562de4d9a6fc703d39c9242f9786082ef1
tree5c0219c19a54060ecd81e2aaee338ee16604f8fdtree | snapshot
parenta68ccda203aad2cc4fd386d5e893237f09fa1ffbcommit | diff
[mlir][sparse] refine lexicographic insertion to any tensor

First version was vectors only. With some clever "path" insertion,
we now support any d-dimensional tensor. Up next: reductions too

Reviewed By: bixia, wrengr

Differential Revision: https://reviews.llvm.org/D114024
mlir/lib/Dialect/SparseTensor/Transforms/Sparsification.cpp diff | blob | history
mlir/lib/ExecutionEngine/SparseTensorUtils.cpp diff | blob | history
mlir/test/Dialect/SparseTensor/sparse_out.mlir diff | blob | history
mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_matrix_ops.mlir [new file with mode: 0644] blob
mlir/test/Integration/Dialect/SparseTensor/CPU/sparse_tensor_ops.mlir [new file with mode: 0644] blob
Domain: System / Toolchain;