| /external/opencv3/modules/core/include/opencv2/core/cuda/ |
| D | emulation.hpp | 84 struct smem struct
86 enum { TAG_MASK = (1U << ( (sizeof(unsigned int) << 3) - 5U)) - 1U };
89 static __device__ __forceinline__ T atomicInc(T* address, T val) in atomicInc()
108 static __device__ __forceinline__ T atomicAdd(T* address, T val) in atomicAdd()
127 static __device__ __forceinline__ T atomicMin(T* address, T val) in atomicMin()
|
| D | vec_distance.hpp | 160 …Global(const T1* vec1, const T2* vec2, int len, Dist& dist, typename Dist::result_type* smem, int … in calcVecDiffGlobal()
178 …onst T1* vecCached, const T2* vecGlob, int len, Dist& dist, typename Dist::result_type* smem, int … in calcVecDiffCached()
|
| D | reduce.hpp | 59 __device__ __forceinline__ void reduce(volatile T* smem, T& val, unsigned int tid, const Op& op) in reduce()
67 …ce__ __forceinline__ void reduce(const thrust::tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in reduce()
|
| /external/opencv3/modules/cudev/include/opencv2/cudev/block/detail/ |
| D | reduce.hpp | 85 get<I>(smem)[tid] = get<I>(val); in loadToSmem() local
102 get<I>(smem)[tid] = get<I>(val) = get<I>(op)(get<I>(val), reg); in merge() local
146 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& val, uint tid) in loadToSmem()
152 __device__ __forceinline__ void loadFromSmem(volatile T* smem, T& val, uint tid) in loadFromSmem()
159 …device__ __forceinline__ void loadToSmem(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in loadToSmem()
168 …vice__ __forceinline__ void loadFromSmem(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in loadFromSmem()
178 …__device__ __forceinline__ void merge(volatile T* smem, T& val, uint tid, uint delta, const Op& op) in merge()
187 __device__ __forceinline__ void merge(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in merge()
|
| D | reduce_key_val.hpp | 84 get<I>(smem)[tid] = get<I>(data); in loadToSmem() local
152 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& data, uint tid) in loadToSmem()
158 __device__ __forceinline__ void loadFromSmem(volatile T* smem, T& data, uint tid) in loadFromSmem()
165 …_forceinline__ void loadToSmem(const tuple<VP0, VP1, VP2, VP3, VP4, VP5, VP6, VP7, VP8, VP9>& smem, in loadToSmem()
174 …orceinline__ void loadFromSmem(const tuple<VP0, VP1, VP2, VP3, VP4, VP5, VP6, VP7, VP8, VP9>& smem, in loadFromSmem()
|
| /external/opencv3/modules/cudev/include/opencv2/cudev/block/ |
| D | scan.hpp | 58 __device__ T blockScanInclusive(T data, volatile T* smem, uint tid) in blockScanInclusive()
97 __device__ __forceinline__ T blockScanExclusive(T data, volatile T* smem, uint tid) in blockScanExclusive()
|
| D | reduce.hpp | 63 __device__ __forceinline__ void blockReduce(volatile T* smem, T& val, uint tid, const Op& op) in blockReduce()
72 …evice__ __forceinline__ void blockReduce(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in blockReduce()
|
| /external/opencv3/modules/cudev/include/opencv2/cudev/warp/ |
| D | scan.hpp | 59 __device__ T warpScanInclusive(T data, volatile T* smem, uint tid) in warpScanInclusive()
95 __device__ __forceinline__ T warpScanExclusive(T data, volatile T* smem, uint tid) in warpScanExclusive()
|
| D | reduce.hpp | 62 __device__ __forceinline__ void warpReduce(volatile T* smem, T& val, uint tid, const Op& op) in warpReduce()
70 …device__ __forceinline__ void warpReduce(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in warpReduce()
|
| /external/opencv3/modules/core/include/opencv2/core/cuda/detail/ |
| D | reduce.hpp | 76 thrust::get<I>(smem)[tid] = thrust::get<I>(val); in loadToSmem() local
92 … thrust::get<I>(smem)[tid] = thrust::get<I>(val) = thrust::get<I>(op)(thrust::get<I>(val), reg); in merge() local
128 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& val, unsigned int tid) in loadToSmem()
133 __device__ __forceinline__ void loadFromSmem(volatile T* smem, T& val, unsigned int tid) in loadFromSmem()
139 … __forceinline__ void loadToSmem(const thrust::tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in loadToSmem()
147 …_forceinline__ void loadFromSmem(const thrust::tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in loadFromSmem()
155 …__device__ __forceinline__ void merge(volatile T* smem, T& val, unsigned int tid, unsigned int del… in merge()
169 …ice__ __forceinline__ void merge(const thrust::tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in merge()
|
| D | reduce_key_val.hpp | 76 thrust::get<I>(smem)[tid] = thrust::get<I>(data); in loadToSmem() local
171 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& data, unsigned int tid) in loadToSmem()
176 __device__ __forceinline__ void loadFromSmem(volatile T* smem, T& data, unsigned int tid) in loadFromSmem()
182 …line__ void loadToSmem(const thrust::tuple<VP0, VP1, VP2, VP3, VP4, VP5, VP6, VP7, VP8, VP9>& smem, in loadToSmem()
190 …ne__ void loadFromSmem(const thrust::tuple<VP0, VP1, VP2, VP3, VP4, VP5, VP6, VP7, VP8, VP9>& smem, in loadFromSmem()
|
| /external/opencv3/modules/cudev/include/opencv2/cudev/warp/detail/ |
| D | reduce.hpp | 83 get<I>(smem)[tid] = get<I>(val); in loadToSmem() local
92 get<I>(smem)[tid] = get<I>(val) = get<I>(op)(get<I>(val), reg); in merge() local
132 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& val, uint tid) in loadToSmem()
139 …device__ __forceinline__ void loadToSmem(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in loadToSmem()
149 …__device__ __forceinline__ void merge(volatile T* smem, T& val, uint tid, uint delta, const Op& op) in merge()
158 __device__ __forceinline__ void merge(const tuple<P0, P1, P2, P3, P4, P5, P6, P7, P8, P9>& smem, in merge()
|
| D | reduce_key_val.hpp | 82 get<I>(smem)[tid] = get<I>(data); in loadToSmem() local
137 __device__ __forceinline__ void loadToSmem(volatile T* smem, T& data, uint tid) in loadToSmem()
144 …_forceinline__ void loadToSmem(const tuple<VP0, VP1, VP2, VP3, VP4, VP5, VP6, VP7, VP8, VP9>& smem, in loadToSmem()
|
| /external/opencv3/modules/cudev/include/opencv2/cudev/grid/detail/ |
| D | reduce.hpp | 71 __device__ __forceinline__ static volatile R* smem(R* ptr) in smem() function
92 __device__ __forceinline__ static tuple<volatile R*, volatile R*> smem(R* ptr) in smem() function
113 __device__ __forceinline__ static tuple<volatile R*, volatile R*, volatile R*> smem(R* ptr) in smem() function
136 …ice__ __forceinline__ static tuple<volatile R*, volatile R*, volatile R*, volatile R*> smem(R* ptr) in smem() function
279 __shared__ work_elem_type smem[BLOCK_SIZE * cn]; in reduceGrid() local
339 __shared__ work_type smem[BLOCK_SIZE]; in reduceGrid() local
|
| D | histogram.hpp | 60 __shared__ ResType smem[BIN_COUNT]; in histogram() local
|
| D | reduce_to_row.hpp | 62 __shared__ work_type smem[BLOCK_SIZE_X * BLOCK_SIZE_Y]; in reduceToRow() local
|
| D | integral.hpp | 63 __shared__ D smem[NUM_SCAN_THREADS * 2]; in horizontal_pass() local
105 __shared__ D smem[NUM_SCAN_THREADS * 2]; in horizontal_pass() local
|
| D | pyr_down.hpp | 68 __shared__ work_type smem[256 + 4]; in pyrDown() local
|
| D | reduce_to_column.hpp | 102 __shared__ work_elem_type smem[cn][BLOCK_SIZE]; in reduceToColumn() local
|
| /external/opencv3/modules/cudaoptflow/src/cuda/ |
| D | farneback.cu | 140 int smem = 3 * block.x * sizeof(float); in polynomialExpansionGpu() local
422 int smem = (block.x + 2*ksizeHalf) * 5 * block.y * sizeof(float); in boxFilter5Gpu() local
441 int smem = (block.x + 2*ksizeHalf) * 5 * block.y * sizeof(float); in boxFilter5Gpu_CC11() local
509 int smem = (block.x + 2*ksizeHalf) * block.y * sizeof(float); in gaussianBlurCaller() local
607 int smem = (block.x + 2*ksizeHalf) * 5 * block.y * sizeof(float); in gaussianBlur5Caller() local
|
| /external/opencv3/modules/cudawarping/src/cuda/ |
| D | pyr_down.cu | 59 __shared__ work_t smem[256 + 4]; in pyrDown() local
|
| /external/opencv3/modules/cudafilters/src/cuda/ |
| D | column_filter.hpp | 74 __shared__ sum_t smem[(PATCH_PER_BLOCK + 2 * HALO_SIZE) * BLOCK_DIM_Y][BLOCK_DIM_X]; in linearColumnFilter() local
|
| D | row_filter.hpp | 74 __shared__ sum_t smem[BLOCK_DIM_Y][(PATCH_PER_BLOCK + 2 * HALO_SIZE) * BLOCK_DIM_X]; in linearRowFilter() local
|
| /external/opencv3/modules/cudalegacy/src/cuda/ |
| D | needle_map.cu | 56 __shared__ float smem[2 * NEEDLE_MAP_SCALE]; in NeedleMapAverageKernel() local
|
| /external/opencv3/modules/cudaimgproc/src/cuda/ |
| D | clahe.cu | 61 __shared__ int smem[512]; in calcLutKernel() local
|