27 template <int NUM_COMP, int COMP_STRIDE, int NUM_PTS>
32   for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in ReadPoint()
40 template <int NUM_COMP, int COMP_STRIDE, int NUM_PTS>
46     for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in WritePoint()
59 template <int NUM_COMP, int P_1D>
72 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
79     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + COMP_STRIDE * comp];  in ReadLVecStandard1d()
86 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
93     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + comp * STRIDES_COMP];  in ReadLVecStrided1d()
100 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
107 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp…  in WriteLVecStandard1d()
111 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
128 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
133   const CeedInt e_vec_size = P_1D * NUM_COMP;  in WriteLVecStandard1d_Assembly()
138     for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in WriteLVecStandard1d_Assembly()
139 …d_v[elem * e_vec_size * e_vec_size + (in_comp * NUM_COMP + comp) * P_1D * P_1D + out_node * P_1D +…  in WriteLVecStandard1d_Assembly()
162 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
169     for (CeedInt comp = 0; comp < NUM_COMP; comp++) d_v[ind + comp * STRIDES_COMP] += r_v[comp];  in WriteLVecStrided1d()
180 template <int NUM_COMP, int P_1D>
194 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
201     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + COMP_STRIDE * comp];  in ReadLVecStandard2d()
208 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
215     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + comp * STRIDES_COMP];  in ReadLVecStrided2d()
222 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
229 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[comp…  in WriteLVecStandard2d()
233 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
252 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
259   const CeedInt e_vec_size = elem_size * NUM_COMP;  in WriteLVecStandard2d_Assembly()
265     for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in WriteLVecStandard2d_Assembly()
266 …const CeedInt index = (in_comp * NUM_COMP + comp) * elem_size * elem_size + out_node * elem_size +…  in WriteLVecStandard2d_Assembly()
291 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
298     for (CeedInt comp = 0; comp < NUM_COMP; comp++) d_v[ind + comp * STRIDES_COMP] += r_v[comp];  in WriteLVecStrided2d()
309 template <int NUM_COMP, int P_1D>
324 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
332 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[z + comp * P_1D] = d_u[ind + COMP_STRIDE * com…  in ReadLVecStandard3d()
340 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
348 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[z + comp * P_1D] = d_u[ind + comp * STRIDES_CO…  in ReadLVecStrided3d()
356 template <int NUM_COMP, int COMP_STRIDE, int Q_1D>
364     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + COMP_STRIDE * comp];  in ReadEVecSliceStandard3d()
371 template <int NUM_COMP, int Q_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
378     for (CeedInt comp = 0; comp < NUM_COMP; comp++) r_u[comp] = d_u[ind + comp * STRIDES_COMP];  in ReadEVecSliceStrided3d()
385 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
393 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) atomicAdd(&d_v[ind + COMP_STRIDE * comp], r_v[z + …  in WriteLVecStandard3d()
398 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
418 template <int NUM_COMP, int COMP_STRIDE, int P_1D>
426   const CeedInt e_vec_size = elem_size * NUM_COMP;  in WriteLVecStandard3d_Assembly()
433       for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in WriteLVecStandard3d_Assembly()
434 …const CeedInt index = (in_comp * NUM_COMP + comp) * elem_size * elem_size + out_node * elem_size +…  in WriteLVecStandard3d_Assembly()
462 template <int NUM_COMP, int P_1D, int STRIDES_NODE, int STRIDES_COMP, int STRIDES_ELEM>
470 …for (CeedInt comp = 0; comp < NUM_COMP; comp++) d_v[ind + comp * STRIDES_COMP] += r_v[z + comp * P…  in WriteLVecStrided3d()
478 template <int NUM_COMP, int Q_1D, int T_1D>
482     for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in GradColloSlice3d()
487       r_V[comp + 0 * NUM_COMP] = 0.0;  in GradColloSlice3d()
489 …      r_V[comp + 0 * NUM_COMP] += c_G[i + data.t_id_x * Q_1D] * data.slice[i + data.t_id_y * T_1D];  in GradColloSlice3d()
492       r_V[comp + 1 * NUM_COMP] = 0.0;  in GradColloSlice3d()
494 …      r_V[comp + 1 * NUM_COMP] += c_G[i + data.t_id_y * Q_1D] * data.slice[data.t_id_x + i * T_1D];  in GradColloSlice3d()
497       r_V[comp + 2 * NUM_COMP] = 0.0;  in GradColloSlice3d()
499         r_V[comp + 2 * NUM_COMP] += c_G[i + q * Q_1D] * r_U[i + comp * Q_1D];  in GradColloSlice3d()
508 template <int NUM_COMP, int Q_1D, int T_1D>
512     for (CeedInt comp = 0; comp < NUM_COMP; comp++) {  in GradColloSliceTranspose3d()
514       data.slice[data.t_id_x + data.t_id_y * T_1D] = r_U[comp + 0 * NUM_COMP];  in GradColloSliceTranspose3d()
522       data.slice[data.t_id_x + data.t_id_y * T_1D] = r_U[comp + 1 * NUM_COMP];  in GradColloSliceTranspose3d()
529         r_V[i + comp * Q_1D] += c_G[i + q * Q_1D] * r_U[comp + 2 * NUM_COMP];  in GradColloSliceTranspose3d()