2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013, by the GROMACS development team, led by
5 * David van der Spoel, Berk Hess, Erik Lindahl, and including many
6 * others, as listed in the AUTHORS file in the top-level source
7 * directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
35 #include "gmx_simd_macros.h"
36 #include "gmx_simd_vec.h"
37 #include "../../nbnxn_consts.h"
38 #ifdef CALC_COUL_EWALD
43 #define SUM_SIMD4(x) (x[0]+x[1]+x[2]+x[3])
45 #define UNROLLI NBNXN_CPU_CLUSTER_I_SIZE
46 #define UNROLLJ (GMX_SIMD_WIDTH_HERE/GMX_SIMD_J_UNROLL_SIZE)
48 /* The stride of all the atom data arrays is equal to half the SIMD width */
49 #define STRIDE (GMX_SIMD_WIDTH_HERE/GMX_SIMD_J_UNROLL_SIZE)
51 #if GMX_SIMD_WIDTH_HERE == 8
52 #define SUM_SIMD(x) (x[0]+x[1]+x[2]+x[3]+x[4]+x[5]+x[6]+x[7])
54 #if GMX_SIMD_WIDTH_HERE == 16
55 /* This is getting ridiculous, SIMD horizontal adds would help,
56 * but this is not performance critical (only used to reduce energies)
58 #define SUM_SIMD(x) (x[0]+x[1]+x[2]+x[3]+x[4]+x[5]+x[6]+x[7]+x[8]+x[9]+x[10]+x[11]+x[12]+x[13]+x[14]+x[15])
60 #error "unsupported kernel configuration"
64 #include "../nbnxn_kernel_simd_utils.h"
67 gmx_load_simd_2xnn_interactions(int excl,
68 gmx_exclfilter filter_S0,
69 gmx_exclfilter filter_S2,
70 gmx_mm_pb *interact_S0,
71 gmx_mm_pb *interact_S2)
73 /* Load integer interaction mask */
74 gmx_exclfilter mask_pr_S = gmx_load1_exclfilter(excl);
75 *interact_S0 = gmx_checkbitmask_pb(mask_pr_S, filter_S0);
76 *interact_S2 = gmx_checkbitmask_pb(mask_pr_S, filter_S2);
79 /* All functionality defines are set here, except for:
80 * CALC_ENERGIES, ENERGY_GROUPS which are defined before.
81 * CHECK_EXCLS, which is set just before including the inner loop contents.
82 * The combination rule defines, LJ_COMB_GEOM or LJ_COMB_LB are currently
83 * set before calling the kernel function. We might want to move that
84 * to inside the n-loop and have a different combination rule for different
85 * ci's, as no combination rule gives a 50% performance hit for LJ.
88 /* We always calculate shift forces, because it's cheap anyhow */
89 #define CALC_SHIFTFORCES
91 /* Assumes all LJ parameters are identical */
92 /* #define FIX_LJ_C */