SkePU
1.2
|
Contains the definitions of the Matrix transpose functions for CPU, CUDA, OpenMP and OpenCL. More...
#include <omp.h>
Namespaces | |
skepu | |
The main namespace for SkePU library. | |
Functions | |
template<typename T > | |
__global__ void | skepu::transposeNaive (T *odata, T *idata, size_t width, size_t height) |
A näive CUDA kernel to take Matrix transpose. | |
template<typename T > | |
__global__ void | skepu::transposeNoBankConflicts (T *odata, T *idata, size_t width, size_t height) |
An optimized CUDA kernel to take Matrix transpose. | |
Contains the definitions of the Matrix transpose functions for CPU, CUDA, OpenMP and OpenCL.