CUDPP
2.1
CUDA Data-Parallel Primitives Library
|
CUDPP kernel-level compact routines. More...
Functions | |
Compact Functions | |
template<class T , bool isBackward> | |
__global__ void | compactData (T *d_out, size_t *d_numValidElements, const unsigned int *d_indices, const unsigned int *d_isValid, const T *d_in, unsigned int numElements) |
Consolidate non-null elements - for each non-null element in d_in write it to d_out, in the position specified by d_isValid. Called by compactArray(). More... | |
CUDPP kernel-level compact routines.
compact_kernel.cu