44 …        const CeedScalar *restrict u, CeedScalar *restrict v, const CeedInt JJ, const CeedInt CC) {  in CeedTensorContract_Avx_Blocked()  argument
64               fmadd(vv[jj][cc], tqv, loadu(&u[(a * B + b) * C + c + cc * 4]));  in CeedTensorContract_Avx_Blocked()
88               fmadd(vv[jj][cc], tqv, loadu(&u[(a * B + b) * C + c + cc * 4]));  in CeedTensorContract_Avx_Blocked()
106 …        const CeedScalar *restrict u, CeedScalar *restrict v, const CeedInt JJ, const CeedInt CC) {  in CeedTensorContract_Avx_Remainder()  argument
127           if (C - c == 1) tqu = set(0.0, 0.0, 0.0, u[(a * B + b) * C + c + 0]);  in CeedTensorContract_Avx_Remainder()
128 …  else if (C - c == 2) tqu = set(0.0, 0.0, u[(a * B + b) * C + c + 1], u[(a * B + b) * C + c + 0]);  in CeedTensorContract_Avx_Remainder()
129 …else if (C - c == 3) tqu = set(0.0, u[(a * B + b) * C + c + 2], u[(a * B + b) * C + c + 1], u[(a *…  in CeedTensorContract_Avx_Remainder()
130           else tqu = loadu(&u[(a * B + b) * C + c]);  in CeedTensorContract_Avx_Remainder()
143 … for (CeedInt c = (C / CC) * CC; c < C; c++) v[(a * J + j) * C + c] += tq * u[(a * B + b) * C + c];  in CeedTensorContract_Avx_Remainder()
154 … CeedTransposeMode t_mode, const CeedInt add, const CeedScalar *restrict u, CeedScalar *restrict v,  in CeedTensorContract_Avx_Single()  argument
177             fmadd(vv[aa][jj], tqv, set1(u[(a + aa) * B + b]));  in CeedTensorContract_Avx_Single()
201           fmadd(vv[aa][jj], tqv, set1(u[(a + aa) * B + b]));  in CeedTensorContract_Avx_Single()
234           fmadd(vv[aa], tqv, set1(u[(a + aa) * B + b]));  in CeedTensorContract_Avx_Single()
245       for (CeedInt a = A_break; a < A; a++) v[a * J + j] += tq * u[a * B + b];  in CeedTensorContract_Avx_Single()
255 …CeedTransposeMode t_mode, const CeedInt add, const CeedScalar *restrict u, CeedScalar *restrict v)…  in CeedTensorContract_Avx_Blocked_4_8()  argument
256   return CeedTensorContract_Avx_Blocked(contract, A, B, C, J, t, t_mode, add, u, v, 4, 8);  in CeedTensorContract_Avx_Blocked_4_8()
259 …CeedTransposeMode t_mode, const CeedInt add, const CeedScalar *restrict u, CeedScalar *restrict v)…  in CeedTensorContract_Avx_Remainder_8_8()  argument
260   return CeedTensorContract_Avx_Remainder(contract, A, B, C, J, t, t_mode, add, u, v, 8, 8);  in CeedTensorContract_Avx_Remainder_8_8()
263 …CeedTransposeMode t_mode, const CeedInt add, const CeedScalar *restrict u, CeedScalar *restrict v)…  in CeedTensorContract_Avx_Single_4_8()  argument
264   return CeedTensorContract_Avx_Single(contract, A, B, C, J, t, t_mode, add, u, v, 4, 8);  in CeedTensorContract_Avx_Single_4_8()
271 …CeedTransposeMode t_mode, const CeedInt add, const CeedScalar *restrict u, CeedScalar *restrict v)…  in CeedTensorContractApply_Avx()  argument
280     CeedTensorContract_Avx_Single_4_8(contract, A, B, C, J, t, t_mode, true, u, v);  in CeedTensorContractApply_Avx()
283 …if (C >= blk_size) CeedTensorContract_Avx_Blocked_4_8(contract, A, B, C, J, t, t_mode, true, u, v);  in CeedTensorContractApply_Avx()
285 …if (C % blk_size) CeedTensorContract_Avx_Remainder_8_8(contract, A, B, C, J, t, t_mode, true, u, v…  in CeedTensorContractApply_Avx()