squall/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Merge pull request #30 from NVIDIA/fix_utilities_example

Browse Source 
Fixed cutlass_utilities example.
This commit is contained in:
Andrew Kerr 2018-09-29 15:09:18 -07:00 committed by GitHub
commit 2332df492e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
examples/02_cutlass_utilities/cutlass_utilities.cu
View File

@ -144,18 +144,18 @@ cudaError_t Cutlass_FP16_SgemmNN(
typename Gemm::Params params;
int result = params.initialize(
M, // GEMM M dimension
N, // GEMM N dimension
K, // GEMM K dimension
half(float(alpha)), // scalar alpha - This is a legal conversion from cutlass::half_t to CUDA's half.
A, // matrix A operand
M, // GEMM M dimension
N, // GEMM N dimension
K, // GEMM K dimension
reinterpret_cast<half const &>(alpha), // scalar alpha - This is a legal conversion from cutlass::half_t to CUDA's half.
A, // matrix A operand
lda,
B, // matrix B operand
B, // matrix B operand
ldb,
half(float(beta)), // scalar beta - This is a legal conversion from cutlass::half_t to CUDA's half.
C, // source matrix C
reinterpret_cast<half const &>(beta), // scalar beta - This is a legal conversion from cutlass::half_t to CUDA's half.
C, // source matrix C
ldc,
C, // destination matrix C (may be different memory than source C matrix)
C, // destination matrix C (may be different memory than source C matrix)
ldc
);