Magic-Mirror/llvm-capstone
1
0
Fork 0
mirror of https://github.com/capstone-engine/llvm-capstone.git synced 2024-12-14 03:29:57 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

[mlir][sparse][gpu] refine type of workspace size variables (#66438)

Browse Source 
Rationale:
Some compiler settings don't like the size_t vs uint64_t setup.
This commit is contained in:
Aart Bik 2023-09-14 15:49:52 -07:00 committed by GitHub
parent dc4bf78e4e
commit 8998bcfbce
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
mlir/lib/ExecutionEngine/CudaRuntimeWrappers.cpp
View File

@ -813,9 +813,9 @@ mgpuCuSparseLtSpMMBufferSize(void *bs, int32_t ma, int32_t mb, void *a, void *b,
auto matA = reinterpret_cast<cusparseLtSpMatHandleAndData *>(a);
auto matB = reinterpret_cast<cusparseLtDnMatHandleAndData *>(b);
auto matC = reinterpret_cast<cusparseLtDnMatHandleAndData *>(c);
auto workspace_size = reinterpret_cast<int64_t *>(bs);
auto compressed_size = &(reinterpret_cast<int64_t *>(bs)[1]);
auto compressed_buffer_size = &(reinterpret_cast<int64_t *>(bs)[2]);
auto workspace_size = reinterpret_cast<size_t *>(bs);
auto compressed_size = &(reinterpret_cast<size_t *>(bs)[1]);
auto compressed_buffer_size = &(reinterpret_cast<size_t *>(bs)[2]);
auto cTp = static_cast<cusparseComputeType>(ctp);
cusparseOperation_t modeA = static_cast<cusparseOperation_t>(ma);