Exposing the C versions of the methods of the sparse runtime support lib
through header files will enable using the same methods in an MLIR program
as well as a C++ program, which will simplify future benchmarking comparisons
(e.g. comparing MLIR generated code with eigen for Matrix Market sparse matrices).
Details
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
mlir/include/mlir/ExecutionEngine/CRunnerUtils.h | ||
---|---|---|
213 | Yes, we are. Thanks! | |
mlir/lib/ExecutionEngine/SparseUtils.cpp | ||
151 | Yes, and for scalars only the data pointer will work. Note that if we want to support more complex memrefs (like assigning directly into compound buffers), we may want to refine this a bit, but that is for now not the intended behavior. |
integration_test/Sparse/CPU/matrix-market-example.mlir
(note that the lib was already added, I am just adding a header so that cpp programs can use the same code by linking to the lib)
Nit: Are we missing a for? s/support/support for/ ?