42#include "../util_namespace.cuh"
133 void* d_temp_storage,
134 size_t& temp_storage_bytes,
137 int* d_column_indices,
143 cudaStream_t stream = 0,
144 bool debug_synchronous =
false)
Optional outer namespace(s)
OffsetT spmv_params
[in] SpMV input parameter bundle
ArrayWrapper< int, NUM_ACTIVE_CHANNELS > ArrayWrapper< int, NUM_ACTIVE_CHANNELS > ArrayWrapper< CounterT *, NUM_ACTIVE_CHANNELS > ArrayWrapper< CounterT *, NUM_ACTIVE_CHANNELS > ArrayWrapper< OutputDecodeOpT, NUM_ACTIVE_CHANNELS > ArrayWrapper< PrivatizedDecodeOpT, NUM_ACTIVE_CHANNELS > OffsetT OffsetT num_rows
The number of rows in the region of interest.
DeviceSpmv provides device-wide parallel operations for performing sparse-matrix * dense-vector multi...
static CUB_RUNTIME_FUNCTION cudaError_t CsrMV(void *d_temp_storage, size_t &temp_storage_bytes, ValueT *d_values, int *d_row_offsets, int *d_column_indices, ValueT *d_vector_x, ValueT *d_vector_y, int num_rows, int num_cols, int num_nonzeros, cudaStream_t stream=0, bool debug_synchronous=false)
This function performs the matrix-vector operation y = A*x.
CUB_RUNTIME_FUNCTION static __forceinline__ cudaError_t Dispatch(void *d_temp_storage, size_t &temp_storage_bytes, SpmvParamsT &spmv_params, cudaStream_t stream, bool debug_synchronous, Spmv1ColKernelT spmv_1col_kernel, SpmvSearchKernelT spmv_search_kernel, SpmvKernelT spmv_kernel, SegmentFixupKernelT segment_fixup_kernel, KernelConfig spmv_config, KernelConfig segment_fixup_config)
< Function type of cub::DeviceSegmentFixupKernelT
< Signed integer type for sequence offsets
ValueT * d_values
Pointer to the array of num_nonzeros values of the corresponding nonzero elements of matrix A.