OCL: Make variables const
[gromacs.git] / src / gromacs / mdlib / wall.cpp
blob04bbd147113662aae0216221a086c567ca19c9af
1 /*
2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
5 * Copyright (c) 2001-2008, The GROMACS development team.
6 * Copyright (c) 2013,2014,2015,2016,2017,2018, by the GROMACS development team, led by
7 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
8 * and including many others, as listed in the AUTHORS file in the
9 * top-level source directory and at http://www.gromacs.org.
11 * GROMACS is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU Lesser General Public License
13 * as published by the Free Software Foundation; either version 2.1
14 * of the License, or (at your option) any later version.
16 * GROMACS is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 * Lesser General Public License for more details.
21 * You should have received a copy of the GNU Lesser General Public
22 * License along with GROMACS; if not, see
23 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
24 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
26 * If you want to redistribute modifications to GROMACS, please
27 * consider that scientific software is very special. Version
28 * control is crucial - bugs must be traceable. We will be happy to
29 * consider code for inclusion in the official distribution, but
30 * derived work must not be called official GROMACS. Details are found
31 * in the README & COPYING files - if they are missing, get the
32 * official version at http://www.gromacs.org.
34 * To help us fund GROMACS development, we humbly ask that you cite
35 * the research papers on the package. Check out http://www.gromacs.org.
38 #include "gmxpre.h"
40 #include "wall.h"
42 #include <cstring>
44 #include <algorithm>
46 #include "gromacs/fileio/filetypes.h"
47 #include "gromacs/gmxlib/nrnb.h"
48 #include "gromacs/math/utilities.h"
49 #include "gromacs/math/vec.h"
50 #include "gromacs/mdtypes/forceoutput.h"
51 #include "gromacs/mdtypes/inputrec.h"
52 #include "gromacs/mdtypes/md_enums.h"
53 #include "gromacs/mdtypes/mdatom.h"
54 #include "gromacs/mdtypes/nblist.h"
55 #include "gromacs/tables/forcetable.h"
56 #include "gromacs/topology/topology.h"
57 #include "gromacs/utility/cstringutil.h"
58 #include "gromacs/utility/fatalerror.h"
59 #include "gromacs/utility/smalloc.h"
61 void make_wall_tables(FILE *fplog,
62 const t_inputrec *ir, const char *tabfn,
63 const gmx_groups_t *groups,
64 t_forcerec *fr)
66 int negp_pp;
67 int *nm_ind;
68 char buf[STRLEN];
70 negp_pp = ir->opts.ngener - ir->nwall;
71 nm_ind = groups->grps[egcENER].nm_ind;
73 if (fplog)
75 fprintf(fplog, "Reading user tables for %d energy groups with %d walls\n",
76 negp_pp, ir->nwall);
79 snew(fr->wall_tab, ir->nwall);
80 for (int w = 0; w < ir->nwall; w++)
82 snew(fr->wall_tab[w], negp_pp);
83 for (int egp = 0; egp < negp_pp; egp++)
85 /* If the energy group pair is excluded, we don't need a table */
86 if (!(fr->egp_flags[egp*ir->opts.ngener+negp_pp+w] & EGP_EXCL))
88 sprintf(buf, "%s", tabfn);
89 sprintf(buf + strlen(tabfn) - strlen(ftp2ext(efXVG)) - 1, "_%s_%s.%s",
90 *groups->grpname[nm_ind[egp]],
91 *groups->grpname[nm_ind[negp_pp+w]],
92 ftp2ext(efXVG));
93 fr->wall_tab[w][egp] = make_tables(fplog, fr->ic, buf, 0,
94 GMX_MAKETABLES_FORCEUSER);
96 /* Since wall have no charge, we can compress the table */
97 for (int i = 0; i <= fr->wall_tab[w][egp]->n; i++)
99 for (int j = 0; j < 8; j++)
101 fr->wall_tab[w][egp]->data[8*i+j] =
102 fr->wall_tab[w][egp]->data[12*i+4+j];
110 [[ noreturn ]] static void wall_error(int a, const rvec *x, real r)
112 gmx_fatal(FARGS,
113 "An atom is beyond the wall: coordinates %f %f %f, distance %f\n"
114 "You might want to use the mdp option wall_r_linpot",
115 x[a][XX], x[a][YY], x[a][ZZ], r);
118 static void tableForce(real r,
119 const t_forcetable &tab,
120 real Cd,
121 real Cr,
122 real *V,
123 real *F)
125 const real tabscale = tab.scale;
126 const real *VFtab = tab.data;
128 real rt = r*tabscale;
129 int n0 = static_cast<int>(rt);
130 if (n0 >= tab.n)
132 /* Beyond the table range, set V and F to zero */
133 *V = 0;
134 *F = 0;
136 else
138 real eps = rt - n0;
139 real eps2 = eps*eps;
140 /* Dispersion */
141 int nnn = 8*n0;
142 real Yt = VFtab[nnn];
143 real Ft = VFtab[nnn + 1];
144 real Geps = VFtab[nnn + 2]*eps;
145 real Heps2 = VFtab[nnn + 3]*eps2;
146 real Fp = Ft + Geps + Heps2;
147 real VV = Yt + Fp*eps;
148 real FF = Fp + Geps + 2.0*Heps2;
149 real Vd = 6*Cd*VV;
150 real Fd = 6*Cd*FF;
151 /* Repulsion */
152 nnn = nnn + 4;
153 Yt = VFtab[nnn];
154 Ft = VFtab[nnn+1];
155 Geps = VFtab[nnn+2]*eps;
156 Heps2 = VFtab[nnn+3]*eps2;
157 Fp = Ft + Geps + Heps2;
158 VV = Yt + Fp*eps;
159 FF = Fp + Geps + 2.0*Heps2;
160 real Vr = 12*Cr*VV;
161 real Fr = 12*Cr*FF;
162 *V = Vd + Vr;
163 *F = -(Fd + Fr)*tabscale;
167 real do_walls(const t_inputrec &ir, const t_forcerec &fr,
168 const matrix box, const t_mdatoms &md,
169 const rvec *x, gmx::ForceWithVirial *forceWithVirial,
170 real lambda, real Vlj[], t_nrnb *nrnb)
172 constexpr real sixth = 1.0/6.0;
173 constexpr real twelfth = 1.0/12.0;
175 int ntw[2];
176 real fac_d[2], fac_r[2];
177 const unsigned short *gid = md.cENER;
179 const int nwall = ir.nwall;
180 const int ngid = ir.opts.ngener;
181 const int ntype = fr.ntype;
182 const real *nbfp = fr.nbfp;
183 const int *egp_flags = fr.egp_flags;
185 for (int w = 0; w < nwall; w++)
187 ntw[w] = 2*ntype*ir.wall_atomtype[w];
188 switch (ir.wall_type)
190 case ewt93:
191 fac_d[w] = ir.wall_density[w]*M_PI/6;
192 fac_r[w] = ir.wall_density[w]*M_PI/45;
193 break;
194 case ewt104:
195 fac_d[w] = ir.wall_density[w]*M_PI/2;
196 fac_r[w] = ir.wall_density[w]*M_PI/5;
197 break;
198 default:
199 break;
202 const real wall_z[2] = { 0, box[ZZ][ZZ] };
204 rvec * gmx_restrict f = as_rvec_array(forceWithVirial->force_.data());
206 real dvdlambda = 0;
207 double sumRF = 0;
208 for (int lam = 0; lam < (md.nPerturbed ? 2 : 1); lam++)
210 real lamfac;
211 const int *type;
212 if (md.nPerturbed)
214 if (lam == 0)
216 lamfac = 1 - lambda;
217 type = md.typeA;
219 else
221 lamfac = lambda;
222 type = md.typeB;
225 else
227 lamfac = 1;
228 type = md.typeA;
231 real Vlambda = 0;
232 for (int i = 0; i < md.homenr; i++)
234 for (int w = 0; w < std::min(nwall, 2); w++)
236 /* The wall energy groups are always at the end of the list */
237 const int ggid = gid[i]*ngid + ngid - nwall + w;
238 const int at = type[i];
239 /* nbfp now includes the 6/12 derivative prefactors */
240 const real Cd = nbfp[ntw[w] + 2*at]*sixth;
241 const real Cr = nbfp[ntw[w] + 2*at + 1]*twelfth;
242 if (!((Cd == 0 && Cr == 0) || (egp_flags[ggid] & EGP_EXCL)))
244 real r, mr;
245 if (w == 0)
247 r = x[i][ZZ];
249 else
251 r = wall_z[1] - x[i][ZZ];
253 if (r < ir.wall_r_linpot)
255 mr = ir.wall_r_linpot - r;
256 r = ir.wall_r_linpot;
258 else
260 mr = 0;
262 if (r <= 0)
264 wall_error(i, x, r);
267 real V, F;
268 real r1, r2, r4, Vd, Vr;
269 switch (ir.wall_type)
271 case ewtTABLE:
272 tableForce(r, *fr.wall_tab[w][gid[i]], Cd, Cr, &V, &F);
273 F *= lamfac;
274 break;
275 case ewt93:
276 r1 = 1/r;
277 r2 = r1*r1;
278 r4 = r2*r2;
279 Vd = fac_d[w]*Cd*r2*r1;
280 Vr = fac_r[w]*Cr*r4*r4*r1;
281 V = Vr - Vd;
282 F = lamfac*(9*Vr - 3*Vd)*r1;
283 break;
284 case ewt104:
285 r1 = 1/r;
286 r2 = r1*r1;
287 r4 = r2*r2;
288 Vd = fac_d[w]*Cd*r4;
289 Vr = fac_r[w]*Cr*r4*r4*r2;
290 V = Vr - Vd;
291 F = lamfac*(10*Vr - 4*Vd)*r1;
292 break;
293 case ewt126:
294 r1 = 1/r;
295 r2 = r1*r1;
296 r4 = r2*r2;
297 Vd = Cd*r4*r2;
298 Vr = Cr*r4*r4*r4;
299 V = Vr - Vd;
300 F = lamfac*(12*Vr - 6*Vd)*r1;
301 break;
302 default:
303 V = 0;
304 F = 0;
305 break;
307 if (mr > 0)
309 V += mr*F;
311 sumRF += r*F;
312 if (w == 1)
314 F = -F;
316 Vlj[ggid] += lamfac*V;
317 Vlambda += V;
318 f[i][ZZ] += F;
322 if (md.nPerturbed)
324 dvdlambda += (lam == 0 ? -1 : 1)*Vlambda;
327 inc_nrnb(nrnb, eNR_WALLS, md.homenr);
330 if (forceWithVirial->computeVirial_)
332 rvec virial = { 0, 0, static_cast<real>(-0.5*sumRF) };
333 forceWithVirial->addVirialContribution(virial);
336 return dvdlambda;