16 extern "C" __global__ void AtPointsTranspose(const CeedInt *__restrict__ indices, const CeedInt *__…  in AtPointsTranspose()
18 …for (CeedInt node = blockIdx.x * blockDim.x + threadIdx.x; node < RSTR_NUM_ELEM * RSTR_ELEM_SIZE; …  in AtPointsTranspose()
19     const CeedInt ind      = indices[node];  in AtPointsTranspose()
20     const CeedInt loc_node = node % RSTR_ELEM_SIZE;  in AtPointsTranspose()
21     const CeedInt elem     = node / RSTR_ELEM_SIZE;  in AtPointsTranspose()
24     for (CeedInt comp = 0; comp < RSTR_NUM_COMP; comp++) {  in AtPointsTranspose()
30 extern "C" __global__ void AtPointsTranspose(const CeedInt *__restrict__ l_vec_indices, const CeedI…  in AtPointsTranspose()
31 …                const CeedInt *__restrict__ points_per_elem, const CeedInt *__restrict__ t_offsets,  in AtPointsTranspose()
35 …for (CeedInt i = blockIdx.x * blockDim.x + threadIdx.x; i < RSTR_NUM_NODES; i += blockDim.x * grid…  in AtPointsTranspose()
36     const CeedInt ind     = l_vec_indices[i];  in AtPointsTranspose()
37     const CeedInt range_1 = t_offsets[i];  in AtPointsTranspose()
38     const CeedInt range_N = t_offsets[i + 1];  in AtPointsTranspose()
40     for (CeedInt comp = 0; comp < RSTR_NUM_COMP; comp++) value[comp] = 0.0;  in AtPointsTranspose()
42     for (CeedInt j = range_1; j < range_N; j++) {  in AtPointsTranspose()
43       const CeedInt t_ind    = t_indices[j];  in AtPointsTranspose()
44       const CeedInt loc_node = t_ind % RSTR_ELEM_SIZE;  in AtPointsTranspose()
45       const CeedInt elem     = t_ind / RSTR_ELEM_SIZE;  in AtPointsTranspose()
48       for (CeedInt comp = 0; comp < RSTR_NUM_COMP; comp++) {  in AtPointsTranspose()
53 …for (CeedInt comp = 0; comp < RSTR_NUM_COMP; comp++) v[ind + comp * RSTR_COMP_STRIDE] += value[com…  in AtPointsTranspose()