src/gromacs/mdlib/force.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
   5  * Copyright (c) 2001-2004, The GROMACS development team.
   6  * Copyright (c) 2013,2014, 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.
  10  *
  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.
  15  *
  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.
  20  *
  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.
  25  *
  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.
  33  *
  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.
  36  */
  37 #include "config.h"
  38
  39 #include <assert.h>
  40 #include <math.h>
  41 #include <string.h>
  42
  43 #include "typedefs.h"
  44 #include "macros.h"
  45 #include "force.h"
  46 #include "nonbonded.h"
  47 #include "names.h"
  48 #include "network.h"
  49 #include "ns.h"
  50 #include "nrnb.h"
  51 #include "bondf.h"
  52 #include "txtdump.h"
  53 #include "coulomb.h"
  54 #include "pme.h"
  55 #include "mdrun.h"
  56 #include "domdec.h"
  57 #include "qmmm.h"
  58 #include "gmx_omp_nthreads.h"
  59
  60 #include "gromacs/legacyheaders/types/commrec.h"
  61 #include "gromacs/math/vec.h"
  62 #include "gromacs/pbcutil/ishift.h"
  63 #include "gromacs/pbcutil/mshift.h"
  64 #include "gromacs/pbcutil/pbc.h"
  65 #include "gromacs/timing/wallcycle.h"
  66 #include "gromacs/utility/fatalerror.h"
  67 #include "gromacs/utility/smalloc.h"
  68
  69 void ns(FILE              *fp,
  70         t_forcerec        *fr,
  71         matrix             box,
  72         gmx_groups_t      *groups,
  73         gmx_localtop_t    *top,
  74         t_mdatoms         *md,
  75         t_commrec         *cr,
  76         t_nrnb            *nrnb,
  77         gmx_bool           bFillGrid,
  78         gmx_bool           bDoLongRangeNS)
  79 {
  80     char   *ptr;
  81     int     nsearch;
  82
  83
  84     if (!fr->ns.nblist_initialized)
  85     {
  86         init_neighbor_list(fp, fr, md->homenr);
  87     }
  88
  89     if (fr->bTwinRange)
  90     {
  91         fr->nlr = 0;
  92     }
  93
  94     nsearch = search_neighbours(fp, fr, box, top, groups, cr, nrnb, md,
  95                                 bFillGrid, bDoLongRangeNS);
  96     if (debug)
  97     {
  98         fprintf(debug, "nsearch = %d\n", nsearch);
  99     }
 100
 101     /* Check whether we have to do dynamic load balancing */
 102     /*if ((nsb->nstDlb > 0) && (mod(step,nsb->nstDlb) == 0))
 103        count_nb(cr,nsb,&(top->blocks[ebCGS]),nns,fr->nlr,
 104        &(top->idef),opts->ngener);
 105      */
 106     if (fr->ns.dump_nl > 0)
 107     {
 108         dump_nblist(fp, cr, fr, fr->ns.dump_nl);
 109     }
 110 }
 111
 112 static void reduce_thread_forces(int n, rvec *f,
 113                                  tensor vir_q, tensor vir_lj,
 114                                  real *Vcorr_q, real *Vcorr_lj,
 115                                  real *dvdl_q, real *dvdl_lj,
 116                                  int nthreads, f_thread_t *f_t)
 117 {
 118     int t, i;
 119
 120     /* This reduction can run over any number of threads */
 121 #pragma omp parallel for num_threads(gmx_omp_nthreads_get(emntBonded)) private(t) schedule(static)
 122     for (i = 0; i < n; i++)
 123     {
 124         for (t = 1; t < nthreads; t++)
 125         {
 126             rvec_inc(f[i], f_t[t].f[i]);
 127         }
 128     }
 129     for (t = 1; t < nthreads; t++)
 130     {
 131         *Vcorr_q  += f_t[t].Vcorr_q;
 132         *Vcorr_lj += f_t[t].Vcorr_lj;
 133         *dvdl_q   += f_t[t].dvdl[efptCOUL];
 134         *dvdl_lj  += f_t[t].dvdl[efptVDW];
 135         m_add(vir_q, f_t[t].vir_q, vir_q);
 136         m_add(vir_lj, f_t[t].vir_lj, vir_lj);
 137     }
 138 }
 139
 140 void gmx_print_sepdvdl(FILE *fplog, const char *s, real v, real dvdlambda)
 141 {
 142     fprintf(fplog, "  %-30s V %12.5e  dVdl %12.5e\n", s, v, dvdlambda);
 143 }
 144
 145 void do_force_lowlevel(FILE       *fplog,   gmx_int64_t step,
 146                        t_forcerec *fr,      t_inputrec *ir,
 147                        t_idef     *idef,    t_commrec  *cr,
 148                        t_nrnb     *nrnb,    gmx_wallcycle_t wcycle,
 149                        t_mdatoms  *md,
 150                        rvec       x[],      history_t  *hist,
 151                        rvec       f[],
 152                        rvec       f_longrange[],
 153                        gmx_enerdata_t *enerd,
 154                        t_fcdata   *fcd,
 155                        gmx_localtop_t *top,
 156                        gmx_genborn_t *born,
 157                        t_atomtypes *atype,
 158                        gmx_bool       bBornRadii,
 159                        matrix     box,
 160                        t_lambda   *fepvals,
 161                        real       *lambda,
 162                        t_graph    *graph,
 163                        t_blocka   *excl,
 164                        rvec       mu_tot[],
 165                        int        flags,
 166                        float      *cycles_pme)
 167 {
 168     int         i, j;
 169     int         donb_flags;
 170     gmx_bool    bDoEpot, bSepDVDL, bSB;
 171     int         pme_flags;
 172     matrix      boxs;
 173     rvec        box_size;
 174     t_pbc       pbc;
 175     char        buf[22];
 176     double      clam_i, vlam_i;
 177     real        dvdl_dum[efptNR], dvdl_nb[efptNR], lam_i[efptNR];
 178     real        dvdl_q, dvdl_lj;
 179
 180 #ifdef GMX_MPI
 181     double  t0 = 0.0, t1, t2, t3; /* time measurement for coarse load balancing */
 182 #endif
 183
 184 #define PRINT_SEPDVDL(s, v, dvdlambda) if (bSepDVDL) { gmx_print_sepdvdl(fplog, s, v, dvdlambda); }
 185
 186     set_pbc(&pbc, fr->ePBC, box);
 187
 188     /* reset free energy components */
 189     for (i = 0; i < efptNR; i++)
 190     {
 191         dvdl_nb[i]  = 0;
 192         dvdl_dum[i] = 0;
 193     }
 194
 195     /* Reset box */
 196     for (i = 0; (i < DIM); i++)
 197     {
 198         box_size[i] = box[i][i];
 199     }
 200
 201     bSepDVDL = (fr->bSepDVDL && do_per_step(step, ir->nstlog));
 202     debug_gmx();
 203
 204     /* do QMMM first if requested */
 205     if (fr->bQMMM)
 206     {
 207         enerd->term[F_EQM] = calculate_QMMM(cr, x, f, fr);
 208     }
 209
 210     if (bSepDVDL)
 211     {
 212         fprintf(fplog, "Step %s: non-bonded V and dVdl for rank %d:\n",
 213                 gmx_step_str(step, buf), cr->nodeid);
 214     }
 215
 216     /* Call the short range functions all in one go. */
 217
 218 #ifdef GMX_MPI
 219     /*#define TAKETIME ((cr->npmenodes) && (fr->timesteps < 12))*/
 220 #define TAKETIME FALSE
 221     if (TAKETIME)
 222     {
 223         MPI_Barrier(cr->mpi_comm_mygroup);
 224         t0 = MPI_Wtime();
 225     }
 226 #endif
 227
 228     if (ir->nwall)
 229     {
 230         /* foreign lambda component for walls */
 231         real dvdl_walls = do_walls(ir, fr, box, md, x, f, lambda[efptVDW],
 232                                    enerd->grpp.ener[egLJSR], nrnb);
 233         PRINT_SEPDVDL("Walls", 0.0, dvdl_walls);
 234         enerd->dvdl_lin[efptVDW] += dvdl_walls;
 235     }
 236
 237     /* If doing GB, reset dvda and calculate the Born radii */
 238     if (ir->implicit_solvent)
 239     {
 240         wallcycle_sub_start(wcycle, ewcsNONBONDED);
 241
 242         for (i = 0; i < born->nr; i++)
 243         {
 244             fr->dvda[i] = 0;
 245         }
 246
 247         if (bBornRadii)
 248         {
 249             calc_gb_rad(cr, fr, ir, top, x, &(fr->gblist), born, md, nrnb);
 250         }
 251
 252         wallcycle_sub_stop(wcycle, ewcsNONBONDED);
 253     }
 254
 255     where();
 256     /* We only do non-bonded calculation with group scheme here, the verlet
 257      * calls are done from do_force_cutsVERLET(). */
 258     if (fr->cutoff_scheme == ecutsGROUP && (flags & GMX_FORCE_NONBONDED))
 259     {
 260         donb_flags = 0;
 261         /* Add short-range interactions */
 262         donb_flags |= GMX_NONBONDED_DO_SR;
 263
 264         /* Currently all group scheme kernels always calculate (shift-)forces */
 265         if (flags & GMX_FORCE_FORCES)
 266         {
 267             donb_flags |= GMX_NONBONDED_DO_FORCE;
 268         }
 269         if (flags & GMX_FORCE_VIRIAL)
 270         {
 271             donb_flags |= GMX_NONBONDED_DO_SHIFTFORCE;
 272         }
 273         if (flags & GMX_FORCE_ENERGY)
 274         {
 275             donb_flags |= GMX_NONBONDED_DO_POTENTIAL;
 276         }
 277         if (flags & GMX_FORCE_DO_LR)
 278         {
 279             donb_flags |= GMX_NONBONDED_DO_LR;
 280         }
 281
 282         wallcycle_sub_start(wcycle, ewcsNONBONDED);
 283         do_nonbonded(fr, x, f, f_longrange, md, excl,
 284                      &enerd->grpp, nrnb,
 285                      lambda, dvdl_nb, -1, -1, donb_flags);
 286
 287         /* If we do foreign lambda and we have soft-core interactions
 288          * we have to recalculate the (non-linear) energies contributions.
 289          */
 290         if (fepvals->n_lambda > 0 && (flags & GMX_FORCE_DHDL) && fepvals->sc_alpha != 0)
 291         {
 292             for (i = 0; i < enerd->n_lambda; i++)
 293             {
 294                 for (j = 0; j < efptNR; j++)
 295                 {
 296                     lam_i[j] = (i == 0 ? lambda[j] : fepvals->all_lambda[j][i-1]);
 297                 }
 298                 reset_foreign_enerdata(enerd);
 299                 do_nonbonded(fr, x, f, f_longrange, md, excl,
 300                              &(enerd->foreign_grpp), nrnb,
 301                              lam_i, dvdl_dum, -1, -1,
 302                              (donb_flags & ~GMX_NONBONDED_DO_FORCE) | GMX_NONBONDED_DO_FOREIGNLAMBDA);
 303                 sum_epot(&(enerd->foreign_grpp), enerd->foreign_term);
 304                 enerd->enerpart_lambda[i] += enerd->foreign_term[F_EPOT];
 305             }
 306         }
 307         wallcycle_sub_stop(wcycle, ewcsNONBONDED);
 308         where();
 309     }
 310
 311     /* If we are doing GB, calculate bonded forces and apply corrections
 312      * to the solvation forces */
 313     /* MRS: Eventually, many need to include free energy contribution here! */
 314     if (ir->implicit_solvent)
 315     {
 316         wallcycle_sub_start(wcycle, ewcsBONDED);
 317         calc_gb_forces(cr, md, born, top, x, f, fr, idef,
 318                        ir->gb_algorithm, ir->sa_algorithm, nrnb, &pbc, graph, enerd);
 319         wallcycle_sub_stop(wcycle, ewcsBONDED);
 320     }
 321
 322 #ifdef GMX_MPI
 323     if (TAKETIME)
 324     {
 325         t1          = MPI_Wtime();
 326         fr->t_fnbf += t1-t0;
 327     }
 328 #endif
 329
 330     if (fepvals->sc_alpha != 0)
 331     {
 332         enerd->dvdl_nonlin[efptVDW] += dvdl_nb[efptVDW];
 333     }
 334     else
 335     {
 336         enerd->dvdl_lin[efptVDW] += dvdl_nb[efptVDW];
 337     }
 338
 339     if (fepvals->sc_alpha != 0)
 340
 341     /* even though coulomb part is linear, we already added it, beacuse we
 342        need to go through the vdw calculation anyway */
 343     {
 344         enerd->dvdl_nonlin[efptCOUL] += dvdl_nb[efptCOUL];
 345     }
 346     else
 347     {
 348         enerd->dvdl_lin[efptCOUL] += dvdl_nb[efptCOUL];
 349     }
 350
 351     if (bSepDVDL)
 352     {
 353         real V_short_range    = 0;
 354         real dvdl_short_range = 0;
 355
 356         for (i = 0; i < enerd->grpp.nener; i++)
 357         {
 358             V_short_range +=
 359                 (fr->bBHAM ?
 360                  enerd->grpp.ener[egBHAMSR][i] :
 361                  enerd->grpp.ener[egLJSR][i])
 362                 + enerd->grpp.ener[egCOULSR][i] + enerd->grpp.ener[egGB][i];
 363         }
 364         dvdl_short_range = dvdl_nb[efptVDW] + dvdl_nb[efptCOUL];
 365         PRINT_SEPDVDL("VdW and Coulomb SR particle-p.",
 366                       V_short_range,
 367                       dvdl_short_range);
 368     }
 369     debug_gmx();
 370
 371
 372     if (debug)
 373     {
 374         pr_rvecs(debug, 0, "fshift after SR", fr->fshift, SHIFTS);
 375     }
 376
 377     /* Shift the coordinates. Must be done before bonded forces and PPPM,
 378      * but is also necessary for SHAKE and update, therefore it can NOT
 379      * go when no bonded forces have to be evaluated.
 380      */
 381
 382     /* Here sometimes we would not need to shift with NBFonly,
 383      * but we do so anyhow for consistency of the returned coordinates.
 384      */
 385     if (graph)
 386     {
 387         shift_self(graph, box, x);
 388         if (TRICLINIC(box))
 389         {
 390             inc_nrnb(nrnb, eNR_SHIFTX, 2*graph->nnodes);
 391         }
 392         else
 393         {
 394             inc_nrnb(nrnb, eNR_SHIFTX, graph->nnodes);
 395         }
 396     }
 397     /* Check whether we need to do bondeds or correct for exclusions */
 398     if (fr->bMolPBC &&
 399         ((flags & GMX_FORCE_BONDED)
 400          || EEL_RF(fr->eeltype) || EEL_FULL(fr->eeltype) || EVDW_PME(fr->vdwtype)))
 401     {
 402         /* Since all atoms are in the rectangular or triclinic unit-cell,
 403          * only single box vector shifts (2 in x) are required.
 404          */
 405         set_pbc_dd(&pbc, fr->ePBC, cr->dd, TRUE, box);
 406     }
 407     debug_gmx();
 408
 409     if (flags & GMX_FORCE_BONDED)
 410     {
 411         wallcycle_sub_start(wcycle, ewcsBONDED);
 412         calc_bonds(fplog, cr->ms,
 413                    idef, x, hist, f, fr, &pbc, graph, enerd, nrnb, lambda, md, fcd,
 414                    DOMAINDECOMP(cr) ? cr->dd->gatindex : NULL, atype, born,
 415                    flags,
 416                    fr->bSepDVDL && do_per_step(step, ir->nstlog), step);
 417
 418         /* Check if we have to determine energy differences
 419          * at foreign lambda's.
 420          */
 421         if (fepvals->n_lambda > 0 && (flags & GMX_FORCE_DHDL) &&
 422             idef->ilsort != ilsortNO_FE)
 423         {
 424             if (idef->ilsort != ilsortFE_SORTED)
 425             {
 426                 gmx_incons("The bonded interactions are not sorted for free energy");
 427             }
 428             for (i = 0; i < enerd->n_lambda; i++)
 429             {
 430                 reset_foreign_enerdata(enerd);
 431                 for (j = 0; j < efptNR; j++)
 432                 {
 433                     lam_i[j] = (i == 0 ? lambda[j] : fepvals->all_lambda[j][i-1]);
 434                 }
 435                 calc_bonds_lambda(fplog, idef, x, fr, &pbc, graph, &(enerd->foreign_grpp), enerd->foreign_term, nrnb, lam_i, md,
 436                                   fcd, DOMAINDECOMP(cr) ? cr->dd->gatindex : NULL);
 437                 sum_epot(&(enerd->foreign_grpp), enerd->foreign_term);
 438                 enerd->enerpart_lambda[i] += enerd->foreign_term[F_EPOT];
 439             }
 440         }
 441         debug_gmx();
 442
 443         wallcycle_sub_stop(wcycle, ewcsBONDED);
 444     }
 445
 446     where();
 447
 448     *cycles_pme = 0;
 449     clear_mat(fr->vir_el_recip);
 450     clear_mat(fr->vir_lj_recip);
 451
 452     /* Do long-range electrostatics and/or LJ-PME, including related short-range
 453      * corrections.
 454      */
 455     if (EEL_FULL(fr->eeltype) || EVDW_PME(fr->vdwtype))
 456     {
 457         real Vlr               = 0, Vcorr = 0;
 458         real dvdl_long_range   = 0;
 459         int  status            = 0;
 460         real Vlr_q             = 0, Vlr_lj = 0, Vcorr_q = 0, Vcorr_lj = 0;
 461         real dvdl_long_range_q = 0, dvdl_long_range_lj = 0;
 462
 463         bSB = (ir->nwall == 2);
 464         if (bSB)
 465         {
 466             copy_mat(box, boxs);
 467             svmul(ir->wall_ewald_zfac, boxs[ZZ], boxs[ZZ]);
 468             box_size[ZZ] *= ir->wall_ewald_zfac;
 469         }
 470
 471         if (EEL_PME_EWALD(fr->eeltype) || EVDW_PME(fr->vdwtype))
 472         {
 473             real dvdl_long_range_correction_q   = 0;
 474             real dvdl_long_range_correction_lj  = 0;
 475             /* With the Verlet scheme exclusion forces are calculated
 476              * in the non-bonded kernel.
 477              */
 478             /* The TPI molecule does not have exclusions with the rest
 479              * of the system and no intra-molecular PME grid
 480              * contributions will be calculated in
 481              * gmx_pme_calc_energy.
 482              */
 483             if ((ir->cutoff_scheme == ecutsGROUP && fr->n_tpi == 0) ||
 484                 ir->ewald_geometry != eewg3D ||
 485                 ir->epsilon_surface != 0)
 486             {
 487                 int nthreads, t;
 488
 489                 wallcycle_sub_start(wcycle, ewcsEWALD_CORRECTION);
 490
 491                 if (fr->n_tpi > 0)
 492                 {
 493                     gmx_fatal(FARGS, "TPI with PME currently only works in a 3D geometry with tin-foil boundary conditions");
 494                 }
 495
 496                 nthreads = gmx_omp_nthreads_get(emntBonded);
 497 #pragma omp parallel for num_threads(nthreads) schedule(static)
 498                 for (t = 0; t < nthreads; t++)
 499                 {
 500                     int     s, e, i;
 501                     rvec   *fnv;
 502                     tensor *vir_q, *vir_lj;
 503                     real   *Vcorrt_q, *Vcorrt_lj, *dvdlt_q, *dvdlt_lj;
 504                     if (t == 0)
 505                     {
 506                         fnv       = fr->f_novirsum;
 507                         vir_q     = &fr->vir_el_recip;
 508                         vir_lj    = &fr->vir_lj_recip;
 509                         Vcorrt_q  = &Vcorr_q;
 510                         Vcorrt_lj = &Vcorr_lj;
 511                         dvdlt_q   = &dvdl_long_range_correction_q;
 512                         dvdlt_lj  = &dvdl_long_range_correction_lj;
 513                     }
 514                     else
 515                     {
 516                         fnv       = fr->f_t[t].f;
 517                         vir_q     = &fr->f_t[t].vir_q;
 518                         vir_lj    = &fr->f_t[t].vir_lj;
 519                         Vcorrt_q  = &fr->f_t[t].Vcorr_q;
 520                         Vcorrt_lj = &fr->f_t[t].Vcorr_lj;
 521                         dvdlt_q   = &fr->f_t[t].dvdl[efptCOUL];
 522                         dvdlt_lj  = &fr->f_t[t].dvdl[efptVDW];
 523                         for (i = 0; i < fr->natoms_force; i++)
 524                         {
 525                             clear_rvec(fnv[i]);
 526                         }
 527                         clear_mat(*vir_q);
 528                         clear_mat(*vir_lj);
 529                     }
 530                     *dvdlt_q  = 0;
 531                     *dvdlt_lj = 0;
 532
 533                     ewald_LRcorrection(fr->excl_load[t], fr->excl_load[t+1],
 534                                        cr, t, fr,
 535                                        md->chargeA,
 536                                        md->nChargePerturbed ? md->chargeB : NULL,
 537                                        md->sqrt_c6A,
 538                                        md->nTypePerturbed ? md->sqrt_c6B : NULL,
 539                                        md->sigmaA,
 540                                        md->nTypePerturbed ? md->sigmaB : NULL,
 541                                        md->sigma3A,
 542                                        md->nTypePerturbed ? md->sigma3B : NULL,
 543                                        ir->cutoff_scheme != ecutsVERLET,
 544                                        excl, x, bSB ? boxs : box, mu_tot,
 545                                        ir->ewald_geometry,
 546                                        ir->epsilon_surface,
 547                                        fnv, *vir_q, *vir_lj,
 548                                        Vcorrt_q, Vcorrt_lj,
 549                                        lambda[efptCOUL], lambda[efptVDW],
 550                                        dvdlt_q, dvdlt_lj);
 551                 }
 552                 if (nthreads > 1)
 553                 {
 554                     reduce_thread_forces(fr->natoms_force, fr->f_novirsum,
 555                                          fr->vir_el_recip, fr->vir_lj_recip,
 556                                          &Vcorr_q, &Vcorr_lj,
 557                                          &dvdl_long_range_correction_q,
 558                                          &dvdl_long_range_correction_lj,
 559                                          nthreads, fr->f_t);
 560                 }
 561                 wallcycle_sub_stop(wcycle, ewcsEWALD_CORRECTION);
 562             }
 563
 564             if (EEL_PME_EWALD(fr->eeltype) && fr->n_tpi == 0)
 565             {
 566                 Vcorr_q += ewald_charge_correction(cr, fr, lambda[efptCOUL], box,
 567                                                    &dvdl_long_range_correction_q,
 568                                                    fr->vir_el_recip);
 569             }
 570
 571             PRINT_SEPDVDL("Ewald excl./charge/dip. corr.", Vcorr_q, dvdl_long_range_correction_q);
 572             PRINT_SEPDVDL("Ewald excl. corr. LJ", Vcorr_lj, dvdl_long_range_correction_lj);
 573             enerd->dvdl_lin[efptCOUL] += dvdl_long_range_correction_q;
 574             enerd->dvdl_lin[efptVDW]  += dvdl_long_range_correction_lj;
 575
 576             if ((EEL_PME(fr->eeltype) || EVDW_PME(fr->vdwtype)) && (cr->duty & DUTY_PME))
 577             {
 578                 /* Do reciprocal PME for Coulomb and/or LJ. */
 579                 assert(fr->n_tpi >= 0);
 580                 if (fr->n_tpi == 0 || (flags & GMX_FORCE_STATECHANGED))
 581                 {
 582                     pme_flags = GMX_PME_SPREAD | GMX_PME_SOLVE;
 583                     if (EEL_PME(fr->eeltype))
 584                     {
 585                         pme_flags     |= GMX_PME_DO_COULOMB;
 586                     }
 587                     if (EVDW_PME(fr->vdwtype))
 588                     {
 589                         pme_flags |= GMX_PME_DO_LJ;
 590                     }
 591                     if (flags & GMX_FORCE_FORCES)
 592                     {
 593                         pme_flags |= GMX_PME_CALC_F;
 594                     }
 595                     if (flags & GMX_FORCE_VIRIAL)
 596                     {
 597                         pme_flags |= GMX_PME_CALC_ENER_VIR;
 598                     }
 599                     if (fr->n_tpi > 0)
 600                     {
 601                         /* We don't calculate f, but we do want the potential */
 602                         pme_flags |= GMX_PME_CALC_POT;
 603                     }
 604                     wallcycle_start(wcycle, ewcPMEMESH);
 605                     status = gmx_pme_do(fr->pmedata,
 606                                         0, md->homenr - fr->n_tpi,
 607                                         x, fr->f_novirsum,
 608                                         md->chargeA, md->chargeB,
 609                                         md->sqrt_c6A, md->sqrt_c6B,
 610                                         md->sigmaA, md->sigmaB,
 611                                         bSB ? boxs : box, cr,
 612                                         DOMAINDECOMP(cr) ? dd_pme_maxshift_x(cr->dd) : 0,
 613                                         DOMAINDECOMP(cr) ? dd_pme_maxshift_y(cr->dd) : 0,
 614                                         nrnb, wcycle,
 615                                         fr->vir_el_recip, fr->ewaldcoeff_q,
 616                                         fr->vir_lj_recip, fr->ewaldcoeff_lj,
 617                                         &Vlr_q, &Vlr_lj,
 618                                         lambda[efptCOUL], lambda[efptVDW],
 619                                         &dvdl_long_range_q, &dvdl_long_range_lj, pme_flags);
 620                     *cycles_pme = wallcycle_stop(wcycle, ewcPMEMESH);
 621                     if (status != 0)
 622                     {
 623                         gmx_fatal(FARGS, "Error %d in reciprocal PME routine", status);
 624                     }
 625                     /* We should try to do as little computation after
 626                      * this as possible, because parallel PME synchronizes
 627                      * the nodes, so we want all load imbalance of the
 628                      * rest of the force calculation to be before the PME
 629                      * call.  DD load balancing is done on the whole time
 630                      * of the force call (without PME).
 631                      */
 632                 }
 633                 if (fr->n_tpi > 0)
 634                 {
 635                     if (EVDW_PME(ir->vdwtype))
 636                     {
 637
 638                         gmx_fatal(FARGS, "Test particle insertion not implemented with LJ-PME");
 639                     }
 640                     /* Determine the PME grid energy of the test molecule
 641                      * with the PME grid potential of the other charges.
 642                      */
 643                     gmx_pme_calc_energy(fr->pmedata, fr->n_tpi,
 644                                         x + md->homenr - fr->n_tpi,
 645                                         md->chargeA + md->homenr - fr->n_tpi,
 646                                         &Vlr_q);
 647                 }
 648                 PRINT_SEPDVDL("PME mesh", Vlr_q + Vlr_lj, dvdl_long_range_q+dvdl_long_range_lj);
 649             }
 650         }
 651
 652         if (!EEL_PME(fr->eeltype) && EEL_PME_EWALD(fr->eeltype))
 653         {
 654             Vlr_q = do_ewald(ir, x, fr->f_novirsum,
 655                              md->chargeA, md->chargeB,
 656                              box_size, cr, md->homenr,
 657                              fr->vir_el_recip, fr->ewaldcoeff_q,
 658                              lambda[efptCOUL], &dvdl_long_range_q, fr->ewald_table);
 659             PRINT_SEPDVDL("Ewald long-range", Vlr_q, dvdl_long_range_q);
 660         }
 661
 662         /* Note that with separate PME nodes we get the real energies later */
 663         enerd->dvdl_lin[efptCOUL] += dvdl_long_range_q;
 664         enerd->dvdl_lin[efptVDW]  += dvdl_long_range_lj;
 665         enerd->term[F_COUL_RECIP]  = Vlr_q + Vcorr_q;
 666         enerd->term[F_LJ_RECIP]    = Vlr_lj + Vcorr_lj;
 667         if (debug)
 668         {
 669             fprintf(debug, "Vlr_q = %g, Vcorr_q = %g, Vlr_corr_q = %g\n",
 670                     Vlr_q, Vcorr_q, enerd->term[F_COUL_RECIP]);
 671             pr_rvecs(debug, 0, "vir_el_recip after corr", fr->vir_el_recip, DIM);
 672             pr_rvecs(debug, 0, "fshift after LR Corrections", fr->fshift, SHIFTS);
 673             fprintf(debug, "Vlr_lj: %g, Vcorr_lj = %g, Vlr_corr_lj = %g\n",
 674                     Vlr_lj, Vcorr_lj, enerd->term[F_LJ_RECIP]);
 675             pr_rvecs(debug, 0, "vir_lj_recip after corr", fr->vir_lj_recip, DIM);
 676         }
 677     }
 678     else
 679     {
 680         /* Is there a reaction-field exclusion correction needed? */
 681         if (EEL_RF(fr->eeltype) && eelRF_NEC != fr->eeltype)
 682         {
 683             /* With the Verlet scheme, exclusion forces are calculated
 684              * in the non-bonded kernel.
 685              */
 686             if (ir->cutoff_scheme != ecutsVERLET)
 687             {
 688                 real dvdl_rf_excl      = 0;
 689                 enerd->term[F_RF_EXCL] =
 690                     RF_excl_correction(fr, graph, md, excl, x, f,
 691                                        fr->fshift, &pbc, lambda[efptCOUL], &dvdl_rf_excl);
 692
 693                 enerd->dvdl_lin[efptCOUL] += dvdl_rf_excl;
 694                 PRINT_SEPDVDL("RF exclusion correction",
 695                               enerd->term[F_RF_EXCL], dvdl_rf_excl);
 696             }
 697         }
 698     }
 699     where();
 700     debug_gmx();
 701
 702     if (debug)
 703     {
 704         print_nrnb(debug, nrnb);
 705     }
 706     debug_gmx();
 707
 708 #ifdef GMX_MPI
 709     if (TAKETIME)
 710     {
 711         t2 = MPI_Wtime();
 712         MPI_Barrier(cr->mpi_comm_mygroup);
 713         t3          = MPI_Wtime();
 714         fr->t_wait += t3-t2;
 715         if (fr->timesteps == 11)
 716         {
 717             fprintf(stderr, "* PP load balancing info: rank %d, step %s, rel wait time=%3.0f%% , load string value: %7.2f\n",
 718                     cr->nodeid, gmx_step_str(fr->timesteps, buf),
 719                     100*fr->t_wait/(fr->t_wait+fr->t_fnbf),
 720                     (fr->t_fnbf+fr->t_wait)/fr->t_fnbf);
 721         }
 722         fr->timesteps++;
 723     }
 724 #endif
 725
 726     if (debug)
 727     {
 728         pr_rvecs(debug, 0, "fshift after bondeds", fr->fshift, SHIFTS);
 729     }
 730
 731 }
 732
 733 void init_enerdata(int ngener, int n_lambda, gmx_enerdata_t *enerd)
 734 {
 735     int i, n2;
 736
 737     for (i = 0; i < F_NRE; i++)
 738     {
 739         enerd->term[i]         = 0;
 740         enerd->foreign_term[i] = 0;
 741     }
 742
 743
 744     for (i = 0; i < efptNR; i++)
 745     {
 746         enerd->dvdl_lin[i]     = 0;
 747         enerd->dvdl_nonlin[i]  = 0;
 748     }
 749
 750     n2 = ngener*ngener;
 751     if (debug)
 752     {
 753         fprintf(debug, "Creating %d sized group matrix for energies\n", n2);
 754     }
 755     enerd->grpp.nener         = n2;
 756     enerd->foreign_grpp.nener = n2;
 757     for (i = 0; (i < egNR); i++)
 758     {
 759         snew(enerd->grpp.ener[i], n2);
 760         snew(enerd->foreign_grpp.ener[i], n2);
 761     }
 762
 763     if (n_lambda)
 764     {
 765         enerd->n_lambda = 1 + n_lambda;
 766         snew(enerd->enerpart_lambda, enerd->n_lambda);
 767     }
 768     else
 769     {
 770         enerd->n_lambda = 0;
 771     }
 772 }
 773
 774 void destroy_enerdata(gmx_enerdata_t *enerd)
 775 {
 776     int i;
 777
 778     for (i = 0; (i < egNR); i++)
 779     {
 780         sfree(enerd->grpp.ener[i]);
 781     }
 782
 783     for (i = 0; (i < egNR); i++)
 784     {
 785         sfree(enerd->foreign_grpp.ener[i]);
 786     }
 787
 788     if (enerd->n_lambda)
 789     {
 790         sfree(enerd->enerpart_lambda);
 791     }
 792 }
 793
 794 static real sum_v(int n, real v[])
 795 {
 796     real t;
 797     int  i;
 798
 799     t = 0.0;
 800     for (i = 0; (i < n); i++)
 801     {
 802         t = t + v[i];
 803     }
 804
 805     return t;
 806 }
 807
 808 void sum_epot(gmx_grppairener_t *grpp, real *epot)
 809 {
 810     int i;
 811
 812     /* Accumulate energies */
 813     epot[F_COUL_SR]  = sum_v(grpp->nener, grpp->ener[egCOULSR]);
 814     epot[F_LJ]       = sum_v(grpp->nener, grpp->ener[egLJSR]);
 815     epot[F_LJ14]     = sum_v(grpp->nener, grpp->ener[egLJ14]);
 816     epot[F_COUL14]   = sum_v(grpp->nener, grpp->ener[egCOUL14]);
 817     epot[F_COUL_LR]  = sum_v(grpp->nener, grpp->ener[egCOULLR]);
 818     epot[F_LJ_LR]    = sum_v(grpp->nener, grpp->ener[egLJLR]);
 819     /* We have already added 1-2,1-3, and 1-4 terms to F_GBPOL */
 820     epot[F_GBPOL]   += sum_v(grpp->nener, grpp->ener[egGB]);
 821
 822 /* lattice part of LR doesnt belong to any group
 823  * and has been added earlier
 824  */
 825     epot[F_BHAM]     = sum_v(grpp->nener, grpp->ener[egBHAMSR]);
 826     epot[F_BHAM_LR]  = sum_v(grpp->nener, grpp->ener[egBHAMLR]);
 827
 828     epot[F_EPOT] = 0;
 829     for (i = 0; (i < F_EPOT); i++)
 830     {
 831         if (i != F_DISRESVIOL && i != F_ORIRESDEV)
 832         {
 833             epot[F_EPOT] += epot[i];
 834         }
 835     }
 836 }
 837
 838 void sum_dhdl(gmx_enerdata_t *enerd, real *lambda, t_lambda *fepvals)
 839 {
 840     int    i, j, index;
 841     double dlam;
 842
 843     enerd->dvdl_lin[efptVDW] += enerd->term[F_DVDL_VDW];  /* include dispersion correction */
 844     enerd->term[F_DVDL]       = 0.0;
 845     for (i = 0; i < efptNR; i++)
 846     {
 847         if (fepvals->separate_dvdl[i])
 848         {
 849             /* could this be done more readably/compactly? */
 850             switch (i)
 851             {
 852                 case (efptMASS):
 853                     index = F_DKDL;
 854                     break;
 855                 case (efptCOUL):
 856                     index = F_DVDL_COUL;
 857                     break;
 858                 case (efptVDW):
 859                     index = F_DVDL_VDW;
 860                     break;
 861                 case (efptBONDED):
 862                     index = F_DVDL_BONDED;
 863                     break;
 864                 case (efptRESTRAINT):
 865                     index = F_DVDL_RESTRAINT;
 866                     break;
 867                 default:
 868                     index = F_DVDL;
 869                     break;
 870             }
 871             enerd->term[index] = enerd->dvdl_lin[i] + enerd->dvdl_nonlin[i];
 872             if (debug)
 873             {
 874                 fprintf(debug, "dvdl-%s[%2d]: %f: non-linear %f + linear %f\n",
 875                         efpt_names[i], i, enerd->term[index], enerd->dvdl_nonlin[i], enerd->dvdl_lin[i]);
 876             }
 877         }
 878         else
 879         {
 880             enerd->term[F_DVDL] += enerd->dvdl_lin[i] + enerd->dvdl_nonlin[i];
 881             if (debug)
 882             {
 883                 fprintf(debug, "dvd-%sl[%2d]: %f: non-linear %f + linear %f\n",
 884                         efpt_names[0], i, enerd->term[F_DVDL], enerd->dvdl_nonlin[i], enerd->dvdl_lin[i]);
 885             }
 886         }
 887     }
 888
 889     /* Notes on the foreign lambda free energy difference evaluation:
 890      * Adding the potential and ekin terms that depend linearly on lambda
 891      * as delta lam * dvdl to the energy differences is exact.
 892      * For the constraints this is not exact, but we have no other option
 893      * without literally changing the lengths and reevaluating the energies at each step.
 894      * (try to remedy this post 4.6 - MRS)
 895      * For the non-bonded LR term we assume that the soft-core (if present)
 896      * no longer affects the energy beyond the short-range cut-off,
 897      * which is a very good approximation (except for exotic settings).
 898      * (investigate how to overcome this post 4.6 - MRS)
 899      */
 900     if (fepvals->separate_dvdl[efptBONDED])
 901     {
 902         enerd->term[F_DVDL_BONDED] += enerd->term[F_DVDL_CONSTR];
 903     }
 904     else
 905     {
 906         enerd->term[F_DVDL] += enerd->term[F_DVDL_CONSTR];
 907     }
 908     enerd->term[F_DVDL_CONSTR] = 0;
 909
 910     for (i = 0; i < fepvals->n_lambda; i++)
 911     {
 912         /* note we are iterating over fepvals here!
 913            For the current lam, dlam = 0 automatically,
 914            so we don't need to add anything to the
 915            enerd->enerpart_lambda[0] */
 916
 917         /* we don't need to worry about dvdl_lin contributions to dE at
 918            current lambda, because the contributions to the current
 919            lambda are automatically zeroed */
 920
 921         for (j = 0; j < efptNR; j++)
 922         {
 923             /* Note that this loop is over all dhdl components, not just the separated ones */
 924             dlam = (fepvals->all_lambda[j][i]-lambda[j]);
 925             enerd->enerpart_lambda[i+1] += dlam*enerd->dvdl_lin[j];
 926             if (debug)
 927             {
 928                 fprintf(debug, "enerdiff lam %g: (%15s), non-linear %f linear %f*%f\n",
 929                         fepvals->all_lambda[j][i], efpt_names[j],
 930                         (enerd->enerpart_lambda[i+1] - enerd->enerpart_lambda[0]),
 931                         dlam, enerd->dvdl_lin[j]);
 932             }
 933         }
 934     }
 935 }
 936
 937
 938 void reset_foreign_enerdata(gmx_enerdata_t *enerd)
 939 {
 940     int  i, j;
 941
 942     /* First reset all foreign energy components.  Foreign energies always called on
 943        neighbor search steps */
 944     for (i = 0; (i < egNR); i++)
 945     {
 946         for (j = 0; (j < enerd->grpp.nener); j++)
 947         {
 948             enerd->foreign_grpp.ener[i][j] = 0.0;
 949         }
 950     }
 951
 952     /* potential energy components */
 953     for (i = 0; (i <= F_EPOT); i++)
 954     {
 955         enerd->foreign_term[i] = 0.0;
 956     }
 957 }
 958
 959 void reset_enerdata(t_forcerec *fr, gmx_bool bNS,
 960                     gmx_enerdata_t *enerd,
 961                     gmx_bool bMaster)
 962 {
 963     gmx_bool bKeepLR;
 964     int      i, j;
 965
 966     /* First reset all energy components, except for the long range terms
 967      * on the master at non neighbor search steps, since the long range
 968      * terms have already been summed at the last neighbor search step.
 969      */
 970     bKeepLR = (fr->bTwinRange && !bNS);
 971     for (i = 0; (i < egNR); i++)
 972     {
 973         if (!(bKeepLR && bMaster && (i == egCOULLR || i == egLJLR)))
 974         {
 975             for (j = 0; (j < enerd->grpp.nener); j++)
 976             {
 977                 enerd->grpp.ener[i][j] = 0.0;
 978             }
 979         }
 980     }
 981     for (i = 0; i < efptNR; i++)
 982     {
 983         enerd->dvdl_lin[i]    = 0.0;
 984         enerd->dvdl_nonlin[i] = 0.0;
 985     }
 986
 987     /* Normal potential energy components */
 988     for (i = 0; (i <= F_EPOT); i++)
 989     {
 990         enerd->term[i] = 0.0;
 991     }
 992     /* Initialize the dVdlambda term with the long range contribution */
 993     /* Initialize the dvdl term with the long range contribution */
 994     enerd->term[F_DVDL]            = 0.0;
 995     enerd->term[F_DVDL_COUL]       = 0.0;
 996     enerd->term[F_DVDL_VDW]        = 0.0;
 997     enerd->term[F_DVDL_BONDED]     = 0.0;
 998     enerd->term[F_DVDL_RESTRAINT]  = 0.0;
 999     enerd->term[F_DKDL]            = 0.0;
1000     if (enerd->n_lambda > 0)
1001     {
1002         for (i = 0; i < enerd->n_lambda; i++)
1003         {
1004             enerd->enerpart_lambda[i] = 0.0;
1005         }
1006     }
1007     /* reset foreign energy data - separate function since we also call it elsewhere */
1008     reset_foreign_enerdata(enerd);
1009 }