Files
clang-p2996/mlir/lib/Dialect/Linalg/Transforms/BufferizableOpInterfaceImpl.cpp
Matthias Springer a27d886ce4 [mlir][linalg][bufferize] Fix element-wise access optimization for sparse tensors (#87305)
`linalg.generic` ops with sparse tensors do not necessarily bufferize to
element-wise access, because insertions into a sparse tensor may change
the layout of (or reallocate) the underlying sparse data structures.
2024-04-03 09:57:25 +09:00

6.7 KiB