CUDPP 1.1
|
CUDPP kernel-level scan routines. More...
Functions | |
Sparse Matrix-Vector multiply Functions | |
template<class T , bool isFullBlock> | |
__global__ void | sparseMatrixVectorFetchAndMultiply (unsigned int *d_flags, T *d_prod, const T *d_A, const T *d_x, const unsigned int *d_indx, unsigned int numNZElts) |
Fetch and multiply kernel. | |
__global__ void | sparseMatrixVectorSetFlags (unsigned int *d_flags, const unsigned int *d_rowindx, unsigned int numRows) |
Set Flags kernel. | |
template<class T > | |
__global__ void | yGather (T *d_y, const T *d_prod, const unsigned int *d_rowFindx, unsigned int numRows) |
Gather final y values kernel. |
CUDPP kernel-level scan routines.