19 static __device__ __inline__ void read_1d(const T *devptr, const int compstride, T *sBuffer[NUM_COM…  in read_1d()  argument
22       sBuffer[comp][tx] = devptr[comp * compstride + tx];  in read_1d()
31 static __device__ __inline__ void write_1d(T *sBuffer[NUM_COMP], T *devptr, const int compstride, c…  in write_1d()  argument
34       devptr[comp * compstride + tx] = sBuffer[comp][tx];  in write_1d()
43 static __device__ __inline__ void sum_1d(T *sBuffer[NUM_COMP], T *devptr, const int compstride, con…  in sum_1d()  argument
46       devptr[comp * compstride + tx] += sBuffer[comp][tx];  in sum_1d()
59 static __device__ __inline__ void read_U_2d(const T *dU, const int compstride, T rU[DIM_U][NUM_COMP…  in read_U_2d()  argument
74         sTmp[i * P + tx] = dU[comp * compstride + i * P + tx];  in read_U_2d()
95 static __device__ __inline__ void read_V_2d(const T *dV, const int compstride, T rV[DIM_V][NUM_COMP…  in read_V_2d()  argument
99         rV[i_DIM][comp][j] = dV[comp * compstride + j * Q + tx];  in read_V_2d()
112 static __device__ __inline__ void write_V_2d(T *dV, const int compstride, T rV[DIM_V][NUM_COMP][rV_…  in write_V_2d()  argument
116         dV[comp * compstride + j * Q + tx] = rV[i_DIM][comp][j];  in write_V_2d()
129 static __device__ __inline__ void sum_V_2d(T *dV, const int compstride, T rV[DIM_V][NUM_COMP][rV_SI…  in sum_V_2d()  argument
133         dV[comp * compstride + j * Q + tx] += rV[i_DIM][comp][j];  in sum_V_2d()
147 static __device__ __inline__ void read_U_3d(const T *dU, const int compstride, T rU[DIM_U][NUM_COMP…  in read_U_3d()  argument
162         sTmp[i * P * P + tx] = dU[comp * compstride + i * P * P + tx];  in read_U_3d()
183 static __device__ __inline__ void read_V_3d(const T *dV, const int compstride, T rV[DIM_V][NUM_COMP…  in read_V_3d()  argument
187         rV[i_DIM][comp][j] = dV[comp * compstride + j * (Q * Q) + tx];  in read_V_3d()
200 static __device__ __inline__ void write_V_3d(T *dV, const int compstride, T rV[DIM_V][NUM_COMP][rV_…  in write_V_3d()  argument
204         dV[comp * compstride + j * (Q * Q) + tx] = rV[i_DIM][comp][j];  in write_V_3d()
217 static __device__ __inline__ void sum_V_3d(T *dV, const int compstride, T rV[DIM_V][NUM_COMP][rV_SI…  in sum_V_3d()  argument
221         dV[comp * compstride + j * (Q * Q) + tx] += rV[i_DIM][comp][j];  in sum_V_3d()