77 …CeedScalar rU[1][BASIS_NUM_COMP][BASIS_P] = {0.0};  // for a non-fused operator BASIS_DIM is alway…  in __launch_bounds__()
87   CeedScalar *sTmp = sT + BASIS_P * BASIS_Q;  in __launch_bounds__()
88   sTmp += ty * (BASIS_P * BASIS_MAX_P_Q);  in __launch_bounds__()
92     read_T_notrans_gm2sm<BASIS_P, BASIS_Q>(tx, dT, sT);  in __launch_bounds__()
96   read_U_2d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dU, cstrdU, rU, sTmp, tx);  in __launch_bounds__()
99 …magma_interp_2d_device<CeedScalar, 1, 1, BASIS_NUM_COMP, BASIS_P, BASIS_Q, BASIS_P, BASIS_Q>(sT, r…  in __launch_bounds__()
119 …CeedScalar rV[1][BASIS_NUM_COMP][BASIS_P] = {0.0};  // for a non-fused operator BASIS_DIM is alway…  in __launch_bounds__()
128   CeedScalar *sTmp = sT + BASIS_Q * BASIS_P;  in __launch_bounds__()
133     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dT, sT);  in __launch_bounds__()
140 …magma_interp_2d_device<CeedScalar, 1, 1, BASIS_NUM_COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P>(sT, r…  in __launch_bounds__()
144   write_V_2d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dV, cstrdV, rV, tx);  in __launch_bounds__()
160 …CeedScalar rV[1][BASIS_NUM_COMP][BASIS_P] = {0.0};  // for a non-fused operator BASIS_DIM is alway…  in __launch_bounds__()
169   CeedScalar *sTmp = sT + BASIS_Q * BASIS_P;  in __launch_bounds__()
174     read_T_trans_gm2sm<BASIS_Q, BASIS_P>(tx, dT, sT);  in __launch_bounds__()
181 …magma_interp_2d_device<CeedScalar, 1, 1, BASIS_NUM_COMP, BASIS_Q, BASIS_P, BASIS_Q, BASIS_P>(sT, r…  in __launch_bounds__()
185   sum_V_2d<CeedScalar, BASIS_P, 1, BASIS_NUM_COMP, BASIS_P, 0>(dV, cstrdV, rV, tx);  in __launch_bounds__()