Lines Matching refs:tx

42 …                                  T rV[DIM_V][NUM_COMP][rV_SIZE], const int tx, T rTmp, T *swork) {
59     if (tx < (P * P)) {
60       const int batchid = tx;
75     if (tx < (P * Q)) {
76       const int batchid = tx / Q;
77       const int tx_     = tx % Q;
93     if (tx < (Q * Q)) {
102           rTmp += sTmp(tx, i, sld) * sT(i, j);
117   const int tx      = threadIdx.x;
139     read_T_notrans_gm2sm<BASIS_P, BASIS_Q>(tx, dinterp1d, sTinterp);
140     read_T_notrans_gm2sm<BASIS_P, BASIS_Q>(tx, dgrad1d, sTgrad);
146 …_U_3d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dU + (0 * dstrdU), cstrdU, rU, sTmp, tx);
150 …S_NUM_COMP, BASIS_P, BASIS_Q, BASIS_P, BASIS_Q, 0, 0, 0, false>(sTinterp, sTgrad, rU, rV, tx, rTmp,
153   write_V_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dV + (0 * dstrdV), cstrdV, rV, tx);
157 …S_NUM_COMP, BASIS_P, BASIS_Q, BASIS_P, BASIS_Q, 1, 0, 0, false>(sTinterp, sTgrad, rU, rV, tx, rTmp,
160   write_V_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dV + (1 * dstrdV), cstrdV, rV, tx);
164 …S_NUM_COMP, BASIS_P, BASIS_Q, BASIS_P, BASIS_Q, 2, 0, 0, false>(sTinterp, sTgrad, rU, rV, tx, rTmp,
167   write_V_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dV + (2 * dstrdV), cstrdV, rV, tx);
176   const int tx      = threadIdx.x;
198     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dinterp1d, sTinterp);
199     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dgrad1d, sTgrad);
205 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (0 * dstrdU), cstrdU, rU, sTmp, tx);
207 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 0, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
212 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (1 * dstrdU), cstrdU, rU, sTmp, tx);
214 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 1, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
219 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (2 * dstrdU), cstrdU, rU, sTmp, tx);
221 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 2, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
225   write_V_3d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dV + (0 * dstrdV), cstrdV, rV, tx);
234   const int tx      = threadIdx.x;
256     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dinterp1d, sTinterp);
257     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dgrad1d, sTgrad);
263 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (0 * dstrdU), cstrdU, rU, sTmp, tx);
265 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 0, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
270 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (1 * dstrdU), cstrdU, rU, sTmp, tx);
272 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 1, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
277 …_U_3d<CeedScalar, BASIS_Q, 1, BASIS_NUM_COMP, BASIS_Q, 0>(dU + (2 * dstrdU), cstrdU, rU, sTmp, tx);
279 …COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P, 2, 0, 0, true>(sTinterp, sTgrad, rU, rV, tx, rTmp, sTmp);
283   sum_V_3d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dV + (0 * dstrdV), cstrdV, rV, tx);