projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3e8eff3) | patch
[mlir][sparse] refine insertion code
authorAart Bik <ajcbik@google.com>
Tue, 18 Oct 2022 17:35:00 +0000 (10:35 -0700)
committerAart Bik <ajcbik@google.com>
Tue, 18 Oct 2022 21:16:38 +0000 (14:16 -0700)
commitd22df0ebba3f28700248c6de8a9d302628d0d2fb
tree7770a8463c55a0c9b796944c1cc1831bbca30ddbtree | snapshot
parent3e8eff3747e3478953dd5798f841a74eafb99e73commit | diff
[mlir][sparse] refine insertion code

builds SSA cycle for compress insertion loop
adds casting on index mismatch during push_back

Reviewed By: Peiming

Differential Revision: https://reviews.llvm.org/D136186
mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorCodegen.cpp diff | blob | history
mlir/test/Dialect/SparseTensor/codegen.mlir diff | blob | history
Domain: System / Toolchain;