2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013,2014,2015,2016,2018, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source 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.
36 #ifndef _nbnxn_consts_h
37 #define _nbnxn_consts_h
39 /* With CPU kernels the i-cluster size is always 4 atoms.
40 * With x86 SIMD the j-cluster size can be 2, 4 or 8, otherwise 4.
42 #define NBNXN_CPU_CLUSTER_I_SIZE 4
44 #define NBNXN_CPU_CLUSTER_I_SIZE_2LOG 2
46 // Lower limit for square interaction distances in nonbonded kernels.
47 // For smaller values we will overflow when calculating r^-1 or r^-12, but
48 // to keep it simple we always apply the limit from the tougher r^-12 condition.
50 // Some double precision SIMD architectures use single precision in the first
51 // step, so although the double precision criterion would allow smaller rsq,
52 // we need to stay in single precision with some margin for the N-R iterations.
53 #define NBNXN_MIN_RSQ 1.0e-36
55 // The worst intermediate value we might evaluate is r^-12, which
56 // means we should ensure r^2 stays above pow(GMX_FLOAT_MAX,-1.0/6.0)*1.01 (some margin)
57 #define NBNXN_MIN_RSQ 3.82e-07f // r > 6.2e-4
61 /* Cluster-pair Interaction masks for 4xN and 2xNN kernels.
62 * Bit i*CJ_SIZE + j tells if atom i and j interact.
64 /* All interaction mask is the same for all kernels */
65 #define NBNXN_INTERACTION_MASK_ALL 0xffffffffU
66 /* 4x4 kernel diagonal mask */
67 #define NBNXN_INTERACTION_MASK_DIAG 0x08ceU
68 /* 4x2 kernel diagonal masks */
69 #define NBNXN_INTERACTION_MASK_DIAG_J2_0 0x0002U
70 #define NBNXN_INTERACTION_MASK_DIAG_J2_1 0x002fU
71 /* 4x8 kernel diagonal masks */
72 #define NBNXN_INTERACTION_MASK_DIAG_J8_0 0xf0f8fcfeU
73 #define NBNXN_INTERACTION_MASK_DIAG_J8_1 0x0080c0e0U
75 /* The number of clusters in a super-cluster, used for GPU */
76 #define c_nbnxnGpuNumClusterPerSupercluster 8
78 /* With GPU kernels we group cluster pairs in 4 to optimize memory usage
79 * of integers containing 32 bits.
81 #define c_nbnxnGpuJgroupSize (32/c_nbnxnGpuNumClusterPerSupercluster)