|
template<typename T > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::loadToSmem (volatile T *smem, T &val, unsigned int tid) |
|
template<typename T > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::loadFromSmem (volatile T *smem, T &val, unsigned int tid) |
|
template<typename P0 , typename P1 , typename P2 , typename P3 , typename P4 , typename P5 , typename P6 , typename P7 , typename P8 , typename P9 , typename R0 , typename R1 , typename R2 , typename R3 , typename R4 , typename R5 , typename R6 , typename R7 , typename R8 , typename R9 > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::loadToSmem (const thrust::tuple< P0, P1, P2, P3, P4, P5, P6, P7, P8, P9 > &smem, const thrust::tuple< R0, R1, R2, R3, R4, R5, R6, R7, R8, R9 > &val, unsigned int tid) |
|
template<typename P0 , typename P1 , typename P2 , typename P3 , typename P4 , typename P5 , typename P6 , typename P7 , typename P8 , typename P9 , typename R0 , typename R1 , typename R2 , typename R3 , typename R4 , typename R5 , typename R6 , typename R7 , typename R8 , typename R9 > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::loadFromSmem (const thrust::tuple< P0, P1, P2, P3, P4, P5, P6, P7, P8, P9 > &smem, const thrust::tuple< R0, R1, R2, R3, R4, R5, R6, R7, R8, R9 > &val, unsigned int tid) |
|
template<typename T , class Op > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::merge (volatile T *smem, T &val, unsigned int tid, unsigned int delta, const Op &op) |
|
template<typename T , class Op > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::mergeShfl (T &val, unsigned int delta, unsigned int width, const Op &op) |
|
template<typename P0 , typename P1 , typename P2 , typename P3 , typename P4 , typename P5 , typename P6 , typename P7 , typename P8 , typename P9 , typename R0 , typename R1 , typename R2 , typename R3 , typename R4 , typename R5 , typename R6 , typename R7 , typename R8 , typename R9 , class Op0 , class Op1 , class Op2 , class Op3 , class Op4 , class Op5 , class Op6 , class Op7 , class Op8 , class Op9 > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::merge (const thrust::tuple< P0, P1, P2, P3, P4, P5, P6, P7, P8, P9 > &smem, const thrust::tuple< R0, R1, R2, R3, R4, R5, R6, R7, R8, R9 > &val, unsigned int tid, unsigned int delta, const thrust::tuple< Op0, Op1, Op2, Op3, Op4, Op5, Op6, Op7, Op8, Op9 > &op) |
|
template<typename R0 , typename R1 , typename R2 , typename R3 , typename R4 , typename R5 , typename R6 , typename R7 , typename R8 , typename R9 , class Op0 , class Op1 , class Op2 , class Op3 , class Op4 , class Op5 , class Op6 , class Op7 , class Op8 , class Op9 > |
__device__ __forceinline__ void | cv::gpu::device::reduce_detail::mergeShfl (const thrust::tuple< R0, R1, R2, R3, R4, R5, R6, R7, R8, R9 > &val, unsigned int delta, unsigned int width, const thrust::tuple< Op0, Op1, Op2, Op3, Op4, Op5, Op6, Op7, Op8, Op9 > &op) |
|