src/mdlib/ns.c

   1 /* -*- mode: c; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4; c-file-style: "stroustrup"; -*-
   2  *
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  *          GROningen MAchine for Chemical Simulations
   9  *
  10  *                        VERSION 3.2.0
  11  * Written by David van der Spoel, Erik Lindahl, Berk Hess, and others.
  12  * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
  13  * Copyright (c) 2001-2004, The GROMACS development team,
  14  * check out http://www.gromacs.org for more information.
  15
  16  * This program is free software; you can redistribute it and/or
  17  * modify it under the terms of the GNU General Public License
  18  * as published by the Free Software Foundation; either version 2
  19  * of the License, or (at your option) any later version.
  20  *
  21  * If you want to redistribute modifications, please consider that
  22  * scientific software is very special. Version control is crucial -
  23  * bugs must be traceable. We will be happy to consider code for
  24  * inclusion in the official distribution, but derived work must not
  25  * be called official GROMACS. Details are found in the README & COPYING
  26  * files - if they are missing, get the official version at www.gromacs.org.
  27  *
  28  * To help us fund GROMACS development, we humbly ask that you cite
  29  * the papers on the package - you can find them in the top README file.
  30  *
  31  * For more info, check our website at http://www.gromacs.org
  32  *
  33  * And Hey:
  34  * GROwing Monsters And Cloning Shrimps
  35  */
  36 #ifdef HAVE_CONFIG_H
  37 #include <config.h>
  38 #endif
  39
  40 #ifdef GMX_THREAD_SHM_FDECOMP
  41 #include <pthread.h>
  42 #endif
  43
  44 #include <math.h>
  45 #include <string.h>
  46 #include "sysstuff.h"
  47 #include "smalloc.h"
  48 #include "macros.h"
  49 #include "maths.h"
  50 #include "vec.h"
  51 #include "network.h"
  52 #include "nsgrid.h"
  53 #include "force.h"
  54 #include "nonbonded.h"
  55 #include "ns.h"
  56 #include "pbc.h"
  57 #include "names.h"
  58 #include "gmx_fatal.h"
  59 #include "nrnb.h"
  60 #include "txtdump.h"
  61 #include "mtop_util.h"
  62
  63 #include "domdec.h"
  64
  65
  66 /* Uncomment the define below to use the generic charge group - charge group
  67  * inner loops (in src/gmxlib/nonbonded/nb_generic_cg.c).
  68  * All intra charge-group interaction should be excluded.
  69  * There should be no inter charge group exclusions.
  70  * There can be no perturbed LJ types or charges.
  71  * mdrun currently does NOT check for this.
  72  */
  73 /* #define GMX_CG_INNERLOOP */
  74
  75
  76 /*
  77  *    E X C L U S I O N   H A N D L I N G
  78  */
  79
  80 #ifdef DEBUG
  81 static void SETEXCL_(t_excl e[],atom_id i,atom_id j)
  82 {   e[j] = e[j] | (1<<i); }
  83 static void RMEXCL_(t_excl e[],atom_id i,atom_id j)
  84 { e[j]=e[j] & ~(1<<i); }
  85 static bool ISEXCL_(t_excl e[],atom_id i,atom_id j)
  86 { return (bool)(e[j] & (1<<i)); }
  87 static bool NOTEXCL_(t_excl e[],atom_id i,atom_id j)
  88 {  return !(ISEXCL(e,i,j)); }
  89 #else
  90 #define SETEXCL(e,i,j) (e)[((atom_id) (j))] |= (1<<((atom_id) (i)))
  91 #define RMEXCL(e,i,j)  (e)[((atom_id) (j))] &= (~(1<<((atom_id) (i))))
  92 #define ISEXCL(e,i,j)  (bool) ((e)[((atom_id) (j))] & (1<<((atom_id) (i))))
  93 #define NOTEXCL(e,i,j) !(ISEXCL(e,i,j))
  94 #endif
  95
  96 /************************************************
  97  *
  98  *  U T I L I T I E S    F O R    N S
  99  *
 100  ************************************************/
 101
 102 static void reallocate_nblist(t_nblist *nl)
 103 {
 104     if (gmx_debug_at)
 105     {
 106         fprintf(debug,"reallocating neigborlist il_code=%d, maxnri=%d\n",
 107                 nl->il_code,nl->maxnri);
 108     }
 109     srenew(nl->iinr,   nl->maxnri);
 110     if (nl->enlist == enlistCG_CG)
 111     {
 112         srenew(nl->iinr_end,nl->maxnri);
 113     }
 114     srenew(nl->gid,    nl->maxnri);
 115     srenew(nl->shift,  nl->maxnri);
 116     srenew(nl->jindex, nl->maxnri+1);
 117 }
 118
 119 /* ivdw/icoul are used to determine the type of interaction, so we
 120  * can set an innerloop index here. The obvious choice for this would have
 121  * been the vdwtype/coultype values in the forcerecord, but unfortunately
 122  * those types are braindead - for instance both Buckingham and normal
 123  * Lennard-Jones use the same value (evdwCUT), and a separate boolean variable
 124  * to determine which interaction is used. There is further no special value
 125  * for 'no interaction'. For backward compatibility with old TPR files we won't
 126  * change this in the 3.x series, so when calling this routine you should use:
 127  *
 128  * icoul=0 no coulomb interaction
 129  * icoul=1 cutoff standard coulomb
 130  * icoul=2 reaction-field coulomb
 131  * icoul=3 tabulated coulomb
 132  *
 133  * ivdw=0 no vdw interaction
 134  * ivdw=1 standard L-J interaction
 135  * ivdw=2 Buckingham
 136  * ivdw=3 tabulated vdw.
 137  *
 138  * Kind of ugly, but it works.
 139  */
 140 static void init_nblist(t_nblist *nl_sr,t_nblist *nl_lr,
 141                         int maxsr,int maxlr,
 142                         int ivdw, int icoul,
 143                         bool bfree, int enlist)
 144 {
 145     t_nblist *nl;
 146     int      homenr;
 147     int      i,nn;
 148
 149     int inloop[20] =
 150     {
 151         eNR_NBKERNEL_NONE,
 152         eNR_NBKERNEL010,
 153         eNR_NBKERNEL020,
 154         eNR_NBKERNEL030,
 155         eNR_NBKERNEL100,
 156         eNR_NBKERNEL110,
 157         eNR_NBKERNEL120,
 158         eNR_NBKERNEL130,
 159         eNR_NBKERNEL200,
 160         eNR_NBKERNEL210,
 161         eNR_NBKERNEL220,
 162         eNR_NBKERNEL230,
 163         eNR_NBKERNEL300,
 164         eNR_NBKERNEL310,
 165         eNR_NBKERNEL320,
 166         eNR_NBKERNEL330,
 167         eNR_NBKERNEL400,
 168         eNR_NBKERNEL410,
 169         eNR_NBKERNEL_NONE,
 170         eNR_NBKERNEL430
 171     };
 172
 173     for(i=0; (i<2); i++)
 174     {
 175         nl     = (i == 0) ? nl_sr : nl_lr;
 176         homenr = (i == 0) ? maxsr : maxlr;
 177
 178         /* Set coul/vdw in neighborlist, and for the normal loops we determine
 179          * an index of which one to call.
 180          */
 181         nl->ivdw  = ivdw;
 182         nl->icoul = icoul;
 183         nl->free_energy = bfree;
 184
 185         if (bfree)
 186         {
 187             nl->enlist  = enlistATOM_ATOM;
 188             nl->il_code = eNR_NBKERNEL_FREE_ENERGY;
 189         }
 190         else
 191         {
 192             nl->enlist = enlist;
 193
 194             nn = inloop[4*icoul + ivdw];
 195
 196             /* solvent loops follow directly after the corresponding
 197             * ordinary loops, in the order:
 198             *
 199             * SPC, SPC-SPC, TIP4p, TIP4p-TIP4p
 200             *
 201             */
 202             switch (enlist) {
 203             case enlistATOM_ATOM:
 204 #ifdef GMX_CG_INNERLOOP
 205                 nl->enlist = enlistCG_CG;
 206 #endif
 207                 break;
 208             case enlistSPC_ATOM:     nn += 1; break;
 209             case enlistSPC_SPC:      nn += 2; break;
 210             case enlistTIP4P_ATOM:   nn += 3; break;
 211             case enlistTIP4P_TIP4P:  nn += 4; break;
 212             }
 213
 214             nl->il_code = nn;
 215         }
 216
 217         if (debug)
 218             fprintf(debug,"Initiating neighbourlist type %d for %s interactions,\nwith %d SR, %d LR atoms.\n",
 219                     nl->il_code,ENLISTTYPE(enlist),maxsr,maxlr);
 220
 221         /* maxnri is influenced by the number of shifts (maximum is 8)
 222          * and the number of energy groups.
 223          * If it is not enough, nl memory will be reallocated during the run.
 224          * 4 seems to be a reasonable factor, which only causes reallocation
 225          * during runs with tiny and many energygroups.
 226          */
 227         nl->maxnri      = homenr*4;
 228         nl->maxnrj      = 0;
 229         nl->maxlen      = 0;
 230         nl->nri         = -1;
 231         nl->nrj         = 0;
 232         nl->iinr        = NULL;
 233         nl->gid         = NULL;
 234         nl->shift       = NULL;
 235         nl->jindex      = NULL;
 236         reallocate_nblist(nl);
 237         nl->jindex[0] = 0;
 238 #ifdef GMX_THREAD_SHM_FDECOMP
 239         nl->counter = 0;
 240         snew(nl->mtx,1);
 241         pthread_mutex_init(nl->mtx,NULL);
 242 #endif
 243     }
 244 }
 245
 246 void init_neighbor_list(FILE *log,t_forcerec *fr,int homenr)
 247 {
 248    /* Make maxlr tunable! (does not seem to be a big difference though)
 249     * This parameter determines the number of i particles in a long range
 250     * neighbourlist. Too few means many function calls, too many means
 251     * cache trashing.
 252     */
 253    int maxsr,maxsr_wat,maxlr,maxlr_wat;
 254    int icoul,icoulf,ivdw;
 255    int solvent;
 256    int enlist_w,enlist_ww;
 257    int i;
 258    t_nblists *nbl;
 259
 260    /* maxsr     = homenr-fr->nWatMol*3; */
 261    maxsr     = homenr;
 262
 263    if (maxsr < 0)
 264    {
 265      gmx_fatal(FARGS,"%s, %d: Negative number of short range atoms.\n"
 266                  "Call your Gromacs dealer for assistance.",__FILE__,__LINE__);
 267    }
 268    /* This is just for initial allocation, so we do not reallocate
 269     * all the nlist arrays many times in a row.
 270     * The numbers seem very accurate, but they are uncritical.
 271     */
 272    maxsr_wat = min(fr->nWatMol,(homenr+2)/3);
 273    if (fr->bTwinRange)
 274    {
 275        maxlr     = 50;
 276        maxlr_wat = min(maxsr_wat,maxlr);
 277    }
 278    else
 279    {
 280      maxlr = maxlr_wat = 0;
 281    }
 282
 283    /* Determine the values for icoul/ivdw. */
 284    /* Start with GB */
 285    if(fr->bGB)
 286    {
 287        icoul=4;
 288    }
 289    else if (fr->bcoultab)
 290    {
 291        icoul = 3;
 292    }
 293    else if (EEL_RF(fr->eeltype))
 294    {
 295        icoul = 2;
 296    }
 297    else
 298    {
 299        icoul = 1;
 300    }
 301
 302    if (fr->bvdwtab)
 303    {
 304        ivdw = 3;
 305    }
 306    else if (fr->bBHAM)
 307    {
 308        ivdw = 2;
 309    }
 310    else
 311    {
 312        ivdw = 1;
 313    }
 314
 315    if (fr->solvent_opt == esolTIP4P) {
 316        enlist_w  = enlistTIP4P_ATOM;
 317        enlist_ww = enlistTIP4P_TIP4P;
 318    } else {
 319        enlist_w  = enlistSPC_ATOM;
 320        enlist_ww = enlistSPC_SPC;
 321    }
 322
 323    for(i=0; i<fr->nnblists; i++)
 324    {
 325        nbl = &(fr->nblists[i]);
 326        init_nblist(&nbl->nlist_sr[eNL_VDWQQ],&nbl->nlist_lr[eNL_VDWQQ],
 327                    maxsr,maxlr,ivdw,icoul,FALSE,enlistATOM_ATOM);
 328        init_nblist(&nbl->nlist_sr[eNL_VDW],&nbl->nlist_lr[eNL_VDW],
 329                    maxsr,maxlr,ivdw,0,FALSE,enlistATOM_ATOM);
 330        init_nblist(&nbl->nlist_sr[eNL_QQ],&nbl->nlist_lr[eNL_QQ],
 331                    maxsr,maxlr,0,icoul,FALSE,enlistATOM_ATOM);
 332        init_nblist(&nbl->nlist_sr[eNL_VDWQQ_WATER],&nbl->nlist_lr[eNL_VDWQQ_WATER],
 333                    maxsr_wat,maxlr_wat,ivdw,icoul, FALSE,enlist_w);
 334        init_nblist(&nbl->nlist_sr[eNL_QQ_WATER],&nbl->nlist_lr[eNL_QQ_WATER],
 335                    maxsr_wat,maxlr_wat,0,icoul, FALSE,enlist_w);
 336        init_nblist(&nbl->nlist_sr[eNL_VDWQQ_WATERWATER],&nbl->nlist_lr[eNL_VDWQQ_WATERWATER],
 337                    maxsr_wat,maxlr_wat,ivdw,icoul, FALSE,enlist_ww);
 338        init_nblist(&nbl->nlist_sr[eNL_QQ_WATERWATER],&nbl->nlist_lr[eNL_QQ_WATERWATER],
 339                    maxsr_wat,maxlr_wat,0,icoul, FALSE,enlist_ww);
 340
 341        if (fr->efep != efepNO)
 342        {
 343            if (fr->bEwald)
 344            {
 345                icoulf = 5;
 346            }
 347            else
 348            {
 349                icoulf = icoul;
 350            }
 351
 352            init_nblist(&nbl->nlist_sr[eNL_VDWQQ_FREE],&nbl->nlist_lr[eNL_VDWQQ_FREE],
 353                        maxsr,maxlr,ivdw,icoulf,TRUE,enlistATOM_ATOM);
 354            init_nblist(&nbl->nlist_sr[eNL_VDW_FREE],&nbl->nlist_lr[eNL_VDW_FREE],
 355                        maxsr,maxlr,ivdw,0,TRUE,enlistATOM_ATOM);
 356            init_nblist(&nbl->nlist_sr[eNL_QQ_FREE],&nbl->nlist_lr[eNL_QQ_FREE],
 357                        maxsr,maxlr,0,icoulf,TRUE,enlistATOM_ATOM);
 358        }
 359    }
 360    /* QMMM MM list */
 361    if (fr->bQMMM && fr->qr->QMMMscheme != eQMMMschemeoniom)
 362    {
 363        init_nblist(&fr->QMMMlist_sr,&fr->QMMMlist_lr,
 364                    maxsr,maxlr,0,icoul,FALSE,enlistATOM_ATOM);
 365    }
 366
 367    fr->ns.nblist_initialized=TRUE;
 368 }
 369
 370  static void reset_nblist(t_nblist *nl)
 371  {
 372      nl->nri       = -1;
 373      nl->nrj       = 0;
 374      nl->maxlen    = 0;
 375      if (nl->jindex)
 376      {
 377          nl->jindex[0] = 0;
 378      }
 379 }
 380
 381 static void reset_neighbor_list(t_forcerec *fr,bool bLR,int nls,int eNL)
 382 {
 383     int n,i;
 384
 385     if (bLR)
 386     {
 387         reset_nblist(&(fr->nblists[nls].nlist_lr[eNL]));
 388     }
 389     else
 390     {
 391         for(n=0; n<fr->nnblists; n++)
 392         {
 393             for(i=0; i<eNL_NR; i++)
 394             {
 395                 reset_nblist(&(fr->nblists[n].nlist_sr[i]));
 396             }
 397         }
 398         if (fr->bQMMM)
 399         {
 400             /* only reset the short-range nblist */
 401             reset_nblist(&(fr->QMMMlist_sr));
 402         }
 403     }
 404 }
 405
 406
 407
 408
 409 static inline void new_i_nblist(t_nblist *nlist,
 410                                 bool bLR,atom_id i_atom,int shift,int gid)
 411 {
 412     int    i,k,nri,nshift;
 413
 414     nri = nlist->nri;
 415
 416     /* Check whether we have to increase the i counter */
 417     if ((nri == -1) ||
 418         (nlist->iinr[nri]  != i_atom) ||
 419         (nlist->shift[nri] != shift) ||
 420         (nlist->gid[nri]   != gid))
 421     {
 422         /* This is something else. Now see if any entries have
 423          * been added in the list of the previous atom.
 424          */
 425         if ((nri == -1) ||
 426             ((nlist->jindex[nri+1] > nlist->jindex[nri]) &&
 427              (nlist->gid[nri] != -1)))
 428         {
 429             /* If so increase the counter */
 430             nlist->nri++;
 431             nri++;
 432             if (nlist->nri >= nlist->maxnri)
 433             {
 434                 nlist->maxnri += over_alloc_large(nlist->nri);
 435                 reallocate_nblist(nlist);
 436             }
 437         }
 438         /* Set the number of neighbours and the atom number */
 439         nlist->jindex[nri+1] = nlist->jindex[nri];
 440         nlist->iinr[nri]     = i_atom;
 441         nlist->gid[nri]      = gid;
 442         nlist->shift[nri]    = shift;
 443     }
 444 }
 445
 446 static inline void close_i_nblist(t_nblist *nlist)
 447 {
 448     int nri = nlist->nri;
 449     int len;
 450
 451     if (nri >= 0)
 452     {
 453         nlist->jindex[nri+1] = nlist->nrj;
 454
 455         len=nlist->nrj -  nlist->jindex[nri];
 456
 457         /* nlist length for water i molecules is treated statically
 458          * in the innerloops
 459          */
 460         if (len > nlist->maxlen)
 461         {
 462             nlist->maxlen = len;
 463         }
 464     }
 465 }
 466
 467 static inline void close_nblist(t_nblist *nlist)
 468 {
 469     /* Only close this nblist when it has been initialized */
 470     if (nlist->jindex)
 471     {
 472         nlist->nri++;
 473     }
 474 }
 475
 476 static inline void close_neighbor_list(t_forcerec *fr,bool bLR,int nls,int eNL,
 477                                        bool bMakeQMMMnblist)
 478 {
 479     int n,i;
 480
 481     if (bMakeQMMMnblist) {
 482         if (bLR)
 483         {
 484             close_nblist(&(fr->QMMMlist_lr));
 485         }
 486         else
 487         {
 488             close_nblist(&(fr->QMMMlist_sr));
 489         }
 490     }
 491     else
 492     {
 493         if (bLR)
 494         {
 495             close_nblist(&(fr->nblists[nls].nlist_lr[eNL]));
 496         }
 497         else
 498         {
 499             for(n=0; n<fr->nnblists; n++)
 500             {
 501                 for(i=0; (i<eNL_NR); i++)
 502                 {
 503                     close_nblist(&(fr->nblists[n].nlist_sr[i]));
 504                 }
 505             }
 506         }
 507     }
 508 }
 509
 510 static inline void add_j_to_nblist(t_nblist *nlist,atom_id j_atom,bool bLR)
 511 {
 512     int nrj=nlist->nrj;
 513
 514     if (nlist->nrj >= nlist->maxnrj)
 515     {
 516         nlist->maxnrj = over_alloc_small(nlist->nrj + 1);
 517         if (gmx_debug_at)
 518             fprintf(debug,"Increasing %s nblist %s j size to %d\n",
 519                     bLR ? "LR" : "SR",nrnb_str(nlist->il_code),nlist->maxnrj);
 520
 521         srenew(nlist->jjnr,nlist->maxnrj);
 522     }
 523
 524     nlist->jjnr[nrj] = j_atom;
 525     nlist->nrj ++;
 526 }
 527
 528 static inline void add_j_to_nblist_cg(t_nblist *nlist,
 529                                       atom_id j_start,int j_end,bool bLR)
 530 {
 531     int nrj=nlist->nrj;
 532
 533     if (nlist->nrj >= nlist->maxnrj)
 534     {
 535         nlist->maxnrj = over_alloc_small(nlist->nrj + 1);
 536         if (gmx_debug_at)
 537             fprintf(debug,"Increasing %s nblist %s j size to %d\n",
 538                     bLR ? "LR" : "SR",nrnb_str(nlist->il_code),nlist->maxnrj);
 539
 540         srenew(nlist->jjnr    ,nlist->maxnrj);
 541         srenew(nlist->jjnr_end,nlist->maxnrj);
 542     }
 543
 544     nlist->jjnr[nrj]     = j_start;
 545     nlist->jjnr_end[nrj] = j_end;
 546     nlist->nrj ++;
 547 }
 548
 549 #ifndef GMX_CG_INNERLOOP
 550 static inline void
 551 put_in_list(bool              bHaveVdW[],
 552             int               ngid,
 553             t_mdatoms *       md,
 554             int               icg,
 555             int               jgid,
 556             int               nj,
 557             atom_id           jjcg[],
 558             atom_id           index[],
 559             t_excl            bExcl[],
 560             int               shift,
 561             t_forcerec *      fr,
 562             bool              bLR,
 563             bool              bDoVdW,
 564             bool              bDoCoul,
 565             bool              bQMMM)
 566 {
 567     /* The a[] index has been removed,
 568      * to put it back in i_atom should be a[i0] and jj should be a[jj].
 569      */
 570     t_nblist *   vdwc;
 571     t_nblist *   vdw;
 572     t_nblist *   coul;
 573     t_nblist *   vdwc_free  = NULL;
 574     t_nblist *   vdw_free   = NULL;
 575     t_nblist *   coul_free  = NULL;
 576     t_nblist *   vdwc_ww    = NULL;
 577     t_nblist *   coul_ww    = NULL;
 578
 579     int             i,j,jcg,igid,gid,nbl_ind,ind_ij;
 580     atom_id   jj,jj0,jj1,i_atom;
 581     int       i0,nicg,len;
 582
 583     int       *cginfo;
 584     int       *type,*typeB;
 585     real      *charge,*chargeB;
 586     real      qi,qiB,qq,rlj;
 587     bool      bFreeEnergy,bFree,bFreeJ,bNotEx,*bPert;
 588     bool      bDoVdW_i,bDoCoul_i,bDoCoul_i_sol;
 589     int       iwater,jwater;
 590     t_nblist  *nlist;
 591
 592     /* Copy some pointers */
 593     cginfo  = fr->cginfo;
 594     charge  = md->chargeA;
 595     chargeB = md->chargeB;
 596     type    = md->typeA;
 597     typeB   = md->typeB;
 598     bPert   = md->bPerturbed;
 599
 600     /* Get atom range */
 601     i0     = index[icg];
 602     nicg   = index[icg+1]-i0;
 603
 604     /* Get the i charge group info */
 605     igid   = GET_CGINFO_GID(cginfo[icg]);
 606     iwater = GET_CGINFO_SOLOPT(cginfo[icg]);
 607
 608     bFreeEnergy = FALSE;
 609     if (md->nPerturbed)
 610     {
 611         /* Check if any of the particles involved are perturbed.
 612          * If not we can do the cheaper normal put_in_list
 613          * and use more solvent optimization.
 614          */
 615         for(i=0; i<nicg; i++)
 616         {
 617             bFreeEnergy |= bPert[i0+i];
 618         }
 619         /* Loop over the j charge groups */
 620         for(j=0; (j<nj && !bFreeEnergy); j++)
 621         {
 622             jcg = jjcg[j];
 623             jj0 = index[jcg];
 624             jj1 = index[jcg+1];
 625             /* Finally loop over the atoms in the j-charge group */
 626             for(jj=jj0; jj<jj1; jj++)
 627             {
 628                 bFreeEnergy |= bPert[jj];
 629             }
 630         }
 631     }
 632
 633     /* Unpack pointers to neighbourlist structs */
 634     if (fr->nnblists == 1)
 635     {
 636         nbl_ind = 0;
 637     }
 638     else
 639     {
 640         nbl_ind = fr->gid2nblists[GID(igid,jgid,ngid)];
 641     }
 642     if (bLR)
 643     {
 644         nlist = fr->nblists[nbl_ind].nlist_lr;
 645     }
 646     else
 647     {
 648         nlist = fr->nblists[nbl_ind].nlist_sr;
 649     }
 650
 651     if (iwater != esolNO)
 652     {
 653         vdwc = &nlist[eNL_VDWQQ_WATER];
 654         vdw  = &nlist[eNL_VDW];
 655         coul = &nlist[eNL_QQ_WATER];
 656 #ifndef DISABLE_WATERWATER_NLIST
 657         vdwc_ww = &nlist[eNL_VDWQQ_WATERWATER];
 658         coul_ww = &nlist[eNL_QQ_WATERWATER];
 659 #endif
 660     }
 661     else if (bQMMM)
 662     {
 663         vdwc = NULL;
 664         vdw = NULL;
 665         coul = (bLR) ? &fr->QMMMlist_lr : &fr->QMMMlist_sr;
 666     }
 667     else
 668     {
 669         vdwc = &nlist[eNL_VDWQQ];
 670         vdw  = &nlist[eNL_VDW];
 671         coul = &nlist[eNL_QQ];
 672     }
 673
 674     if (!bFreeEnergy)
 675     {
 676         if (iwater != esolNO)
 677         {
 678             /* Loop over the atoms in the i charge group */
 679             i_atom  = i0;
 680             gid     = GID(igid,jgid,ngid);
 681             /* Create new i_atom for each energy group */
 682             if (bDoCoul && bDoVdW)
 683             {
 684                 new_i_nblist(vdwc,bLR,i_atom,shift,gid);
 685 #ifndef DISABLE_WATERWATER_NLIST
 686                 new_i_nblist(vdwc_ww,bLR,i_atom,shift,gid);
 687 #endif
 688             }
 689             if (bDoVdW)
 690             {
 691                 new_i_nblist(vdw,bLR,i_atom,shift,gid);
 692             }
 693             if (bDoCoul)
 694             {
 695                 new_i_nblist(coul,bLR,i_atom,shift,gid);
 696 #ifndef DISABLE_WATERWATER_NLIST
 697                 new_i_nblist(coul_ww,bLR,i_atom,shift,gid);
 698 #endif
 699             }
 700           /* Loop over the j charge groups */
 701             for(j=0; (j<nj); j++)
 702             {
 703                 jcg=jjcg[j];
 704
 705                 if (jcg == icg)
 706                 {
 707                     continue;
 708                 }
 709
 710                 jj0 = index[jcg];
 711                 jwater = GET_CGINFO_SOLOPT(cginfo[jcg]);
 712
 713                 if (iwater == esolSPC && jwater == esolSPC)
 714                 {
 715                     /* Interaction between two SPC molecules */
 716                     if (!bDoCoul)
 717                     {
 718                         /* VdW only - only first atoms in each water interact */
 719                         add_j_to_nblist(vdw,jj0,bLR);
 720                     }
 721                     else
 722                     {
 723 #ifdef DISABLE_WATERWATER_NLIST
 724                         /* Add entries for the three atoms - only do VdW if we need to */
 725                         if (!bDoVdW)
 726                         {
 727                             add_j_to_nblist(coul,jj0,bLR);
 728                         }
 729                         else
 730                         {
 731                             add_j_to_nblist(vdwc,jj0,bLR);
 732                         }
 733                         add_j_to_nblist(coul,jj0+1,bLR);
 734                         add_j_to_nblist(coul,jj0+2,bLR);
 735 #else
 736                         /* One entry for the entire water-water interaction */
 737                         if (!bDoVdW)
 738                         {
 739                             add_j_to_nblist(coul_ww,jj0,bLR);
 740                         }
 741                         else
 742                         {
 743                             add_j_to_nblist(vdwc_ww,jj0,bLR);
 744                         }
 745 #endif
 746                     }
 747                 }
 748                 else if (iwater == esolTIP4P && jwater == esolTIP4P)
 749                 {
 750                     /* Interaction between two TIP4p molecules */
 751                     if (!bDoCoul)
 752                     {
 753                         /* VdW only - only first atoms in each water interact */
 754                         add_j_to_nblist(vdw,jj0,bLR);
 755                     }
 756                     else
 757                     {
 758 #ifdef DISABLE_WATERWATER_NLIST
 759                         /* Add entries for the four atoms - only do VdW if we need to */
 760                         if (bDoVdW)
 761                         {
 762                             add_j_to_nblist(vdw,jj0,bLR);
 763                         }
 764                         add_j_to_nblist(coul,jj0+1,bLR);
 765                         add_j_to_nblist(coul,jj0+2,bLR);
 766                         add_j_to_nblist(coul,jj0+3,bLR);
 767 #else
 768                         /* One entry for the entire water-water interaction */
 769                         if (!bDoVdW)
 770                         {
 771                             add_j_to_nblist(coul_ww,jj0,bLR);
 772                         }
 773                         else
 774                         {
 775                             add_j_to_nblist(vdwc_ww,jj0,bLR);
 776                         }
 777 #endif
 778                     }
 779                 }
 780                 else
 781                 {
 782                     /* j charge group is not water, but i is.
 783                      * Add entries to the water-other_atom lists; the geometry of the water
 784                      * molecule doesn't matter - that is taken care of in the nonbonded kernel,
 785                      * so we don't care if it is SPC or TIP4P...
 786                      */
 787
 788                     jj1 = index[jcg+1];
 789
 790                     if (!bDoVdW)
 791                     {
 792                         for(jj=jj0; (jj<jj1); jj++)
 793                         {
 794                             if (charge[jj] != 0)
 795                             {
 796                                 add_j_to_nblist(coul,jj,bLR);
 797                             }
 798                         }
 799                     }
 800                     else if (!bDoCoul)
 801                     {
 802                         for(jj=jj0; (jj<jj1); jj++)
 803                         {
 804                             if (bHaveVdW[type[jj]])
 805                             {
 806                                 add_j_to_nblist(vdw,jj,bLR);
 807                             }
 808                         }
 809                     }
 810                     else
 811                     {
 812                         /* _charge_ _groups_ interact with both coulomb and LJ */
 813                         /* Check which atoms we should add to the lists!       */
 814                         for(jj=jj0; (jj<jj1); jj++)
 815                         {
 816                             if (bHaveVdW[type[jj]])
 817                             {
 818                                 if (charge[jj] != 0)
 819                                 {
 820                                     add_j_to_nblist(vdwc,jj,bLR);
 821                                 }
 822                                 else
 823                                 {
 824                                     add_j_to_nblist(vdw,jj,bLR);
 825                                 }
 826                             }
 827                             else if (charge[jj] != 0)
 828                             {
 829                                 add_j_to_nblist(coul,jj,bLR);
 830                             }
 831                         }
 832                     }
 833                 }
 834             }
 835             close_i_nblist(vdw);
 836             close_i_nblist(coul);
 837             close_i_nblist(vdwc);
 838 #ifndef DISABLE_WATERWATER_NLIST
 839             close_i_nblist(coul_ww);
 840             close_i_nblist(vdwc_ww);
 841 #endif
 842         }
 843         else if (bQMMM)
 844         {
 845             /* QMMM atoms as i charge groups */
 846             if (!bLR)
 847             {
 848                 /* Loop over atoms in the ith charge group */
 849                 for (i=0;i<nicg;i++)
 850                 {
 851                     i_atom = i0+i;
 852                     gid    = GID(igid,jgid,ngid);
 853                     /* Create new i_atom for each energy group */
 854                     new_i_nblist(coul,bLR,i_atom,shift,gid);
 855
 856                     /* Loop over the j charge groups */
 857                     for (j=0;j<nj;j++)
 858                     {
 859                         jcg=jjcg[j];
 860
 861                         /* Charge groups cannot have QM and MM atoms simultaneously */
 862                         if (jcg!=icg)
 863                         {
 864                             jj0 = index[jcg];
 865                             jj1 = index[jcg+1];
 866                             /* Finally loop over the atoms in the j-charge group */
 867                             for(jj=jj0; jj<jj1; jj++)
 868                             {
 869                                 bNotEx = NOTEXCL(bExcl,i,jj);
 870                                 if(bNotEx)
 871                                     add_j_to_nblist(coul,jj,bLR);
 872                             }
 873                         }
 874                     }
 875                     close_i_nblist(coul);
 876                 }
 877             }
 878         }
 879         else
 880         {
 881             /* no solvent as i charge group */
 882             /* Loop over the atoms in the i charge group */
 883             for(i=0; i<nicg; i++)
 884             {
 885                 i_atom  = i0+i;
 886                 gid     = GID(igid,jgid,ngid);
 887                 qi      = charge[i_atom];
 888
 889                 /* Create new i_atom for each energy group */
 890                 if (bDoVdW && bDoCoul)
 891                 {
 892                     new_i_nblist(vdwc,bLR,i_atom,shift,gid);
 893                 }
 894                 if (bDoVdW)
 895                 {
 896                     new_i_nblist(vdw,bLR,i_atom,shift,gid);
 897                 }
 898                 if (bDoCoul)
 899                 {
 900                     new_i_nblist(coul,bLR,i_atom,shift,gid);
 901                 }
 902                 bDoVdW_i  = (bDoVdW  && bHaveVdW[type[i_atom]]);
 903                 bDoCoul_i = (bDoCoul && qi!=0);
 904
 905                 if (bDoVdW_i || bDoCoul_i)
 906                 {
 907                     /* Loop over the j charge groups */
 908                     for(j=0; (j<nj); j++)
 909                     {
 910                         jcg=jjcg[j];
 911
 912                         /* Check for large charge groups */
 913                         if (jcg == icg)
 914                         {
 915                             jj0 = i0 + i + 1;
 916                         }
 917                         else
 918                         {
 919                             jj0 = index[jcg];
 920                         }
 921
 922                         jj1=index[jcg+1];
 923                         /* Finally loop over the atoms in the j-charge group */
 924                         for(jj=jj0; jj<jj1; jj++)
 925                         {
 926                             bNotEx = NOTEXCL(bExcl,i,jj);
 927
 928                             if (bNotEx)
 929                             {
 930                                 if (!bDoVdW_i)
 931                                 {
 932                                     if (charge[jj] != 0)
 933                                     {
 934                                         add_j_to_nblist(coul,jj,bLR);
 935                                     }
 936                                 }
 937                                 else if (!bDoCoul_i)
 938                                 {
 939                                     if (bHaveVdW[type[jj]])
 940                                     {
 941                                         add_j_to_nblist(vdw,jj,bLR);
 942                                     }
 943                                 }
 944                                 else
 945                                 {
 946                                     if (bHaveVdW[type[jj]])
 947                                     {
 948                                         if (charge[jj] != 0)
 949                                         {
 950                                             add_j_to_nblist(vdwc,jj,bLR);
 951                                         }
 952                                         else
 953                                         {
 954                                             add_j_to_nblist(vdw,jj,bLR);
 955                                         }
 956                                     }
 957                                     else if (charge[jj] != 0)
 958                                     {
 959                                         add_j_to_nblist(coul,jj,bLR);
 960                                     }
 961                                 }
 962                             }
 963                         }
 964                     }
 965                 }
 966                 close_i_nblist(vdw);
 967                 close_i_nblist(coul);
 968                 close_i_nblist(vdwc);
 969             }
 970         }
 971     }
 972     else
 973     {
 974         /* we are doing free energy */
 975         vdwc_free = &nlist[eNL_VDWQQ_FREE];
 976         vdw_free  = &nlist[eNL_VDW_FREE];
 977         coul_free = &nlist[eNL_QQ_FREE];
 978         /* Loop over the atoms in the i charge group */
 979         for(i=0; i<nicg; i++)
 980         {
 981             i_atom  = i0+i;
 982             gid     = GID(igid,jgid,ngid);
 983             qi      = charge[i_atom];
 984             qiB     = chargeB[i_atom];
 985
 986             /* Create new i_atom for each energy group */
 987             if (bDoVdW && bDoCoul)
 988                 new_i_nblist(vdwc,bLR,i_atom,shift,gid);
 989             if (bDoVdW)
 990                 new_i_nblist(vdw,bLR,i_atom,shift,gid);
 991             if (bDoCoul)
 992                 new_i_nblist(coul,bLR,i_atom,shift,gid);
 993
 994             new_i_nblist(vdw_free,bLR,i_atom,shift,gid);
 995             new_i_nblist(coul_free,bLR,i_atom,shift,gid);
 996             new_i_nblist(vdwc_free,bLR,i_atom,shift,gid);
 997
 998             bDoVdW_i  = (bDoVdW  &&
 999                          (bHaveVdW[type[i_atom]] || bHaveVdW[typeB[i_atom]]));
1000             bDoCoul_i = (bDoCoul && (qi!=0 || qiB!=0));
1001             /* For TIP4P the first atom does not have a charge,
1002              * but the last three do. So we should still put an atom
1003              * without LJ but with charge in the water-atom neighborlist
1004              * for a TIP4p i charge group.
1005              * For SPC type water the first atom has LJ and charge,
1006              * so there is no such problem.
1007              */
1008             if (iwater == esolNO)
1009             {
1010                 bDoCoul_i_sol = bDoCoul_i;
1011             }
1012             else
1013             {
1014                 bDoCoul_i_sol = bDoCoul;
1015             }
1016
1017             if (bDoVdW_i || bDoCoul_i_sol)
1018             {
1019                 /* Loop over the j charge groups */
1020                 for(j=0; (j<nj); j++)
1021                 {
1022                     jcg=jjcg[j];
1023
1024                     /* Check for large charge groups */
1025                     if (jcg == icg)
1026                     {
1027                         jj0 = i0 + i + 1;
1028                     }
1029                     else
1030                     {
1031                         jj0 = index[jcg];
1032                     }
1033
1034                     jj1=index[jcg+1];
1035                     /* Finally loop over the atoms in the j-charge group */
1036                     bFree = bPert[i_atom];
1037                     for(jj=jj0; (jj<jj1); jj++)
1038                     {
1039                         bFreeJ = bFree || bPert[jj];
1040                         /* Complicated if, because the water H's should also
1041                          * see perturbed j-particles
1042                          */
1043                         if (iwater==esolNO || i==0 || bFreeJ)
1044                         {
1045                             bNotEx = NOTEXCL(bExcl,i,jj);
1046
1047                             if (bNotEx)
1048                             {
1049                                 if (bFreeJ)
1050                                 {
1051                                     if (!bDoVdW_i)
1052                                     {
1053                                         if (charge[jj]!=0 || chargeB[jj]!=0)
1054                                         {
1055                                             add_j_to_nblist(coul_free,jj,bLR);
1056                                         }
1057                                     }
1058                                     else if (!bDoCoul_i)
1059                                     {
1060                                         if (bHaveVdW[type[jj]] || bHaveVdW[typeB[jj]])
1061                                         {
1062                                             add_j_to_nblist(vdw_free,jj,bLR);
1063                                         }
1064                                     }
1065                                     else
1066                                     {
1067                                         if (bHaveVdW[type[jj]] || bHaveVdW[typeB[jj]])
1068                                         {
1069                                             if (charge[jj]!=0 || chargeB[jj]!=0)
1070                                             {
1071                                                 add_j_to_nblist(vdwc_free,jj,bLR);
1072                                             }
1073                                             else
1074                                             {
1075                                                 add_j_to_nblist(vdw_free,jj,bLR);
1076                                             }
1077                                         }
1078                                         else if (charge[jj]!=0 || chargeB[jj]!=0)
1079                                             add_j_to_nblist(coul_free,jj,bLR);
1080                                     }
1081                                 }
1082                                 else if (!bDoVdW_i)
1083                                 {
1084                                     /* This is done whether or not bWater is set */
1085                                     if (charge[jj] != 0)
1086                                     {
1087                                         add_j_to_nblist(coul,jj,bLR);
1088                                     }
1089                                 }
1090                                 else if (!bDoCoul_i_sol)
1091                                 {
1092                                     if (bHaveVdW[type[jj]])
1093                                     {
1094                                         add_j_to_nblist(vdw,jj,bLR);
1095                                     }
1096                                 }
1097                                 else
1098                                 {
1099                                     if (bHaveVdW[type[jj]])
1100                                     {
1101                                         if (charge[jj] != 0)
1102                                         {
1103                                             add_j_to_nblist(vdwc,jj,bLR);
1104                                         }
1105                                         else
1106                                         {
1107                                             add_j_to_nblist(vdw,jj,bLR);
1108                                         }
1109                                     }
1110                                     else if (charge[jj] != 0)
1111                                     {
1112                                         add_j_to_nblist(coul,jj,bLR);
1113                                     }
1114                                 }
1115                             }
1116                         }
1117                     }
1118                 }
1119             }
1120             close_i_nblist(vdw);
1121             close_i_nblist(coul);
1122             close_i_nblist(vdwc);
1123             close_i_nblist(vdw_free);
1124             close_i_nblist(coul_free);
1125             close_i_nblist(vdwc_free);
1126         }
1127     }
1128 }
1129 #endif
1130
1131 #ifdef GMX_CG_INNERLOOP
1132 static inline void
1133 put_in_list(bool              bHaveVdW[],
1134             int               ngid,
1135             t_mdatoms *       md,
1136             int               icg,
1137             int               jgid,
1138             int               nj,
1139             atom_id           jjcg[],
1140             atom_id           index[],
1141             t_excl            bExcl[],
1142             int               shift,
1143             t_forcerec *      fr,
1144             bool              bLR,
1145             bool              bDoVdW,
1146             bool              bDoCoul,
1147             bool              bQMMM)
1148 {
1149     int          igid,gid,nbl_ind;
1150     t_nblist *   vdwc;
1151     int          j,jcg;
1152
1153     igid = GET_CGINFO_GID(fr->cginfo[icg]);
1154     gid  = GID(igid,jgid,ngid);
1155
1156     /* Unpack pointers to neighbourlist structs */
1157     if (fr->nnblists == 1)
1158     {
1159         nbl_ind = 0;
1160     }
1161     else
1162     {
1163         nbl_ind = fr->gid2nblists[gid];
1164     }
1165     if (bLR)
1166     {
1167         vdwc = &fr->nblists[nbl_ind].nlist_lr[eNL_VDWQQ];
1168     }
1169     else
1170     {
1171         vdwc = &fr->nblists[nbl_ind].nlist_sr[eNL_VDWQQ];
1172     }
1173
1174     /* Make a new neighbor list for charge group icg.
1175      * Currently simply one neighbor list is made with LJ and Coulomb.
1176      * If required, zerp interactions could be removed here
1177      * or in the force loop.
1178      */
1179     new_i_nblist(vdwc,bLR,index[icg],shift,gid);
1180     vdwc->iinr_end[vdwc->nri] = index[icg+1];
1181
1182     for(j=0; (j<nj); j++)
1183     {
1184         jcg = jjcg[j];
1185         if (jcg != icg)
1186         {
1187             /* Here we simply add the j charge group jcg to the list
1188              * without checking exclusions, LJ interactions or charges.
1189              */
1190             add_j_to_nblist_cg(vdwc,index[jcg],index[jcg+1],bLR);
1191         }
1192     }
1193
1194     close_i_nblist(vdwc);
1195 }
1196 #endif
1197
1198 static void setexcl(atom_id start,atom_id end,t_blocka *excl,bool b,
1199                     t_excl bexcl[])
1200 {
1201     atom_id i,k;
1202
1203     if (b)
1204     {
1205         for(i=start; i<end; i++)
1206         {
1207             for(k=excl->index[i]; k<excl->index[i+1]; k++)
1208             {
1209                 SETEXCL(bexcl,i-start,excl->a[k]);
1210             }
1211         }
1212     }
1213     else
1214     {
1215         for(i=start; i<end; i++)
1216         {
1217             for(k=excl->index[i]; k<excl->index[i+1]; k++)
1218             {
1219                 RMEXCL(bexcl,i-start,excl->a[k]);
1220             }
1221         }
1222     }
1223 }
1224
1225 int calc_naaj(int icg,int cgtot)
1226 {
1227     int naaj;
1228
1229     if ((cgtot % 2) == 1)
1230     {
1231         /* Odd number of charge groups, easy */
1232         naaj = 1 + (cgtot/2);
1233     }
1234     else if ((cgtot % 4) == 0)
1235     {
1236     /* Multiple of four is hard */
1237         if (icg < cgtot/2)
1238         {
1239             if ((icg % 2) == 0)
1240             {
1241                 naaj=1+(cgtot/2);
1242             }
1243             else
1244             {
1245                 naaj=cgtot/2;
1246             }
1247         }
1248         else
1249         {
1250             if ((icg % 2) == 1)
1251             {
1252                 naaj=1+(cgtot/2);
1253             }
1254             else
1255             {
1256                 naaj=cgtot/2;
1257             }
1258         }
1259     }
1260     else
1261     {
1262         /* cgtot/2 = odd */
1263         if ((icg % 2) == 0)
1264         {
1265             naaj=1+(cgtot/2);
1266         }
1267         else
1268         {
1269             naaj=cgtot/2;
1270         }
1271     }
1272 #ifdef DEBUG
1273     fprintf(log,"naaj=%d\n",naaj);
1274 #endif
1275
1276     return naaj;
1277 }
1278
1279 /************************************************
1280  *
1281  *  S I M P L E      C O R E     S T U F F
1282  *
1283  ************************************************/
1284
1285 static real calc_image_tric(rvec xi,rvec xj,matrix box,
1286                             rvec b_inv,int *shift)
1287 {
1288     /* This code assumes that the cut-off is smaller than
1289      * a half times the smallest diagonal element of the box.
1290      */
1291     const real h25=2.5;
1292     real dx,dy,dz;
1293     real r2;
1294     int  tx,ty,tz;
1295
1296     /* Compute diff vector */
1297     dz = xj[ZZ] - xi[ZZ];
1298     dy = xj[YY] - xi[YY];
1299     dx = xj[XX] - xi[XX];
1300
1301   /* Perform NINT operation, using trunc operation, therefore
1302    * we first add 2.5 then subtract 2 again
1303    */
1304     tz = dz*b_inv[ZZ] + h25;
1305     tz -= 2;
1306     dz -= tz*box[ZZ][ZZ];
1307     dy -= tz*box[ZZ][YY];
1308     dx -= tz*box[ZZ][XX];
1309
1310     ty = dy*b_inv[YY] + h25;
1311     ty -= 2;
1312     dy -= ty*box[YY][YY];
1313     dx -= ty*box[YY][XX];
1314
1315     tx = dx*b_inv[XX]+h25;
1316     tx -= 2;
1317     dx -= tx*box[XX][XX];
1318
1319     /* Distance squared */
1320     r2 = (dx*dx) + (dy*dy) + (dz*dz);
1321
1322     *shift = XYZ2IS(tx,ty,tz);
1323
1324     return r2;
1325 }
1326
1327 static real calc_image_rect(rvec xi,rvec xj,rvec box_size,
1328                             rvec b_inv,int *shift)
1329 {
1330     const real h15=1.5;
1331     real ddx,ddy,ddz;
1332     real dx,dy,dz;
1333     real r2;
1334     int  tx,ty,tz;
1335
1336     /* Compute diff vector */
1337     dx = xj[XX] - xi[XX];
1338     dy = xj[YY] - xi[YY];
1339     dz = xj[ZZ] - xi[ZZ];
1340
1341     /* Perform NINT operation, using trunc operation, therefore
1342      * we first add 1.5 then subtract 1 again
1343      */
1344     tx = dx*b_inv[XX] + h15;
1345     ty = dy*b_inv[YY] + h15;
1346     tz = dz*b_inv[ZZ] + h15;
1347     tx--;
1348     ty--;
1349     tz--;
1350
1351     /* Correct diff vector for translation */
1352     ddx = tx*box_size[XX] - dx;
1353     ddy = ty*box_size[YY] - dy;
1354     ddz = tz*box_size[ZZ] - dz;
1355
1356     /* Distance squared */
1357     r2 = (ddx*ddx) + (ddy*ddy) + (ddz*ddz);
1358
1359     *shift = XYZ2IS(tx,ty,tz);
1360
1361     return r2;
1362 }
1363
1364 static void add_simple(t_ns_buf *nsbuf,int nrj,atom_id cg_j,
1365                        bool bHaveVdW[],int ngid,t_mdatoms *md,
1366                        int icg,int jgid,t_block *cgs,t_excl bexcl[],
1367                        int shift,t_forcerec *fr)
1368 {
1369     if (nsbuf->nj + nrj > MAX_CG)
1370     {
1371         put_in_list(bHaveVdW,ngid,md,icg,jgid,nsbuf->ncg,nsbuf->jcg,
1372                     cgs->index,bexcl,shift,fr,FALSE,TRUE,TRUE,FALSE);
1373         /* Reset buffer contents */
1374         nsbuf->ncg = nsbuf->nj = 0;
1375     }
1376     nsbuf->jcg[nsbuf->ncg++] = cg_j;
1377     nsbuf->nj += nrj;
1378 }
1379
1380 static void ns_inner_tric(rvec x[],int icg,int *i_egp_flags,
1381                           int njcg,atom_id jcg[],
1382                           matrix box,rvec b_inv,real rcut2,
1383                           t_block *cgs,t_ns_buf **ns_buf,
1384                           bool bHaveVdW[],int ngid,t_mdatoms *md,
1385                           t_excl bexcl[],t_forcerec *fr)
1386 {
1387     int      shift;
1388     int      j,nrj,jgid;
1389     int      *cginfo=fr->cginfo;
1390     atom_id  cg_j,*cgindex;
1391     t_ns_buf *nsbuf;
1392
1393     cgindex = cgs->index;
1394     shift   = CENTRAL;
1395     for(j=0; (j<njcg); j++)
1396     {
1397         cg_j   = jcg[j];
1398         nrj    = cgindex[cg_j+1]-cgindex[cg_j];
1399         if (calc_image_tric(x[icg],x[cg_j],box,b_inv,&shift) < rcut2)
1400         {
1401             jgid  = GET_CGINFO_GID(cginfo[cg_j]);
1402             if (!(i_egp_flags[jgid] & EGP_EXCL))
1403             {
1404                 add_simple(&ns_buf[jgid][shift],nrj,cg_j,
1405                            bHaveVdW,ngid,md,icg,jgid,cgs,bexcl,shift,fr);
1406             }
1407         }
1408     }
1409 }
1410
1411 static void ns_inner_rect(rvec x[],int icg,int *i_egp_flags,
1412                           int njcg,atom_id jcg[],
1413                           bool bBox,rvec box_size,rvec b_inv,real rcut2,
1414                           t_block *cgs,t_ns_buf **ns_buf,
1415                           bool bHaveVdW[],int ngid,t_mdatoms *md,
1416                           t_excl bexcl[],t_forcerec *fr)
1417 {
1418     int      shift;
1419     int      j,nrj,jgid;
1420     int      *cginfo=fr->cginfo;
1421     atom_id  cg_j,*cgindex;
1422     t_ns_buf *nsbuf;
1423
1424     cgindex = cgs->index;
1425     if (bBox)
1426     {
1427         shift = CENTRAL;
1428         for(j=0; (j<njcg); j++)
1429         {
1430             cg_j   = jcg[j];
1431             nrj    = cgindex[cg_j+1]-cgindex[cg_j];
1432             if (calc_image_rect(x[icg],x[cg_j],box_size,b_inv,&shift) < rcut2)
1433             {
1434                 jgid  = GET_CGINFO_GID(cginfo[cg_j]);
1435                 if (!(i_egp_flags[jgid] & EGP_EXCL))
1436                 {
1437                     add_simple(&ns_buf[jgid][shift],nrj,cg_j,
1438                                bHaveVdW,ngid,md,icg,jgid,cgs,bexcl,shift,fr);
1439                 }
1440             }
1441         }
1442     }
1443     else
1444     {
1445         for(j=0; (j<njcg); j++)
1446         {
1447             cg_j   = jcg[j];
1448             nrj    = cgindex[cg_j+1]-cgindex[cg_j];
1449             if ((rcut2 == 0) || (distance2(x[icg],x[cg_j]) < rcut2)) {
1450                 jgid  = GET_CGINFO_GID(cginfo[cg_j]);
1451                 if (!(i_egp_flags[jgid] & EGP_EXCL))
1452                 {
1453                     add_simple(&ns_buf[jgid][CENTRAL],nrj,cg_j,
1454                                bHaveVdW,ngid,md,icg,jgid,cgs,bexcl,CENTRAL,fr);
1455                 }
1456             }
1457         }
1458     }
1459 }
1460
1461 /* ns_simple_core needs to be adapted for QMMM still 2005 */
1462
1463 static int ns_simple_core(t_forcerec *fr,
1464                           gmx_localtop_t *top,
1465                           t_mdatoms *md,
1466                           matrix box,rvec box_size,
1467                           t_excl bexcl[],atom_id *aaj,
1468                           int ngid,t_ns_buf **ns_buf,
1469                           bool bHaveVdW[])
1470 {
1471     int      naaj,k;
1472     real     rlist2;
1473     int      nsearch,icg,jcg,igid,i0,nri,nn;
1474     int      *cginfo;
1475     t_ns_buf *nsbuf;
1476     /* atom_id  *i_atoms; */
1477     t_block  *cgs=&(top->cgs);
1478     t_blocka *excl=&(top->excls);
1479     rvec     b_inv;
1480     int      m;
1481     bool     bBox,bTriclinic;
1482     int      *i_egp_flags;
1483
1484     rlist2 = sqr(fr->rlist);
1485
1486     bBox = (fr->ePBC != epbcNONE);
1487     if (bBox)
1488     {
1489         for(m=0; (m<DIM); m++)
1490         {
1491             b_inv[m] = divide(1.0,box_size[m]);
1492         }
1493         bTriclinic = TRICLINIC(box);
1494     }
1495     else
1496     {
1497         bTriclinic = FALSE;
1498     }
1499
1500     cginfo = fr->cginfo;
1501
1502     nsearch=0;
1503     for (icg=fr->cg0; (icg<fr->hcg); icg++)
1504     {
1505         /*
1506           i0        = cgs->index[icg];
1507           nri       = cgs->index[icg+1]-i0;
1508           i_atoms   = &(cgs->a[i0]);
1509           i_eg_excl = fr->eg_excl + ngid*md->cENER[*i_atoms];
1510           setexcl(nri,i_atoms,excl,TRUE,bexcl);
1511         */
1512         igid = GET_CGINFO_GID(cginfo[icg]);
1513         i_egp_flags = fr->egp_flags + ngid*igid;
1514         setexcl(cgs->index[icg],cgs->index[icg+1],excl,TRUE,bexcl);
1515
1516         naaj=calc_naaj(icg,cgs->nr);
1517         if (bTriclinic)
1518         {
1519             ns_inner_tric(fr->cg_cm,icg,i_egp_flags,naaj,&(aaj[icg]),
1520                           box,b_inv,rlist2,cgs,ns_buf,
1521                           bHaveVdW,ngid,md,bexcl,fr);
1522         }
1523         else
1524         {
1525             ns_inner_rect(fr->cg_cm,icg,i_egp_flags,naaj,&(aaj[icg]),
1526                     bBox,box_size,b_inv,rlist2,cgs,ns_buf,
1527                     bHaveVdW,ngid,md,bexcl,fr);
1528         }
1529         nsearch += naaj;
1530
1531         for(nn=0; (nn<ngid); nn++)
1532         {
1533             for(k=0; (k<SHIFTS); k++)
1534             {
1535                 nsbuf = &(ns_buf[nn][k]);
1536                 if (nsbuf->ncg > 0)
1537                 {
1538                     put_in_list(bHaveVdW,ngid,md,icg,nn,nsbuf->ncg,nsbuf->jcg,
1539                                 cgs->index,bexcl,k,fr,FALSE,TRUE,TRUE,FALSE);
1540                     nsbuf->ncg=nsbuf->nj=0;
1541                 }
1542             }
1543         }
1544         /* setexcl(nri,i_atoms,excl,FALSE,bexcl); */
1545         setexcl(cgs->index[icg],cgs->index[icg+1],excl,FALSE,bexcl);
1546     }
1547     close_neighbor_list(fr,FALSE,-1,-1,FALSE);
1548
1549     return nsearch;
1550 }
1551
1552 /************************************************
1553  *
1554  *    N S 5     G R I D     S T U F F
1555  *
1556  ************************************************/
1557
1558 static inline void get_dx(int Nx,real gridx,real rc2,int xgi,real x,
1559                           int *dx0,int *dx1,real *dcx2)
1560 {
1561     real dcx,tmp;
1562     int  xgi0,xgi1,i;
1563
1564     if (xgi < 0)
1565     {
1566         *dx0 = 0;
1567         xgi0 = -1;
1568         *dx1 = -1;
1569         xgi1 = 0;
1570     }
1571     else if (xgi >= Nx)
1572     {
1573         *dx0 = Nx;
1574         xgi0 = Nx-1;
1575         *dx1 = Nx-1;
1576         xgi1 = Nx;
1577     }
1578     else
1579     {
1580         dcx2[xgi] = 0;
1581         *dx0 = xgi;
1582         xgi0 = xgi-1;
1583         *dx1 = xgi;
1584         xgi1 = xgi+1;
1585     }
1586
1587     for(i=xgi0; i>=0; i--)
1588     {
1589         dcx = (i+1)*gridx-x;
1590         tmp = dcx*dcx;
1591         if (tmp >= rc2)
1592             break;
1593         *dx0 = i;
1594         dcx2[i] = tmp;
1595     }
1596     for(i=xgi1; i<Nx; i++)
1597     {
1598         dcx = i*gridx-x;
1599         tmp = dcx*dcx;
1600         if (tmp >= rc2)
1601         {
1602             break;
1603         }
1604         *dx1 = i;
1605         dcx2[i] = tmp;
1606     }
1607 }
1608
1609 static inline void get_dx_dd(int Nx,real gridx,real rc2,int xgi,real x,
1610                              int ncpddc,int shift_min,int shift_max,
1611                              int *g0,int *g1,real *dcx2)
1612 {
1613     real dcx,tmp;
1614     int  g_min,g_max,shift_home;
1615
1616     if (xgi < 0)
1617     {
1618         g_min = 0;
1619         g_max = Nx - 1;
1620         *g0   = 0;
1621         *g1   = -1;
1622     }
1623     else if (xgi >= Nx)
1624     {
1625         g_min = 0;
1626         g_max = Nx - 1;
1627         *g0   = Nx;
1628         *g1   = Nx - 1;
1629     }
1630     else
1631     {
1632         if (ncpddc == 0)
1633         {
1634             g_min = 0;
1635             g_max = Nx - 1;
1636         }
1637         else
1638         {
1639             if (xgi < ncpddc)
1640             {
1641                 shift_home = 0;
1642             }
1643             else
1644             {
1645                 shift_home = -1;
1646             }
1647             g_min = (shift_min == shift_home ? 0          : ncpddc);
1648             g_max = (shift_max == shift_home ? ncpddc - 1 : Nx - 1);
1649         }
1650         if (shift_min > 0)
1651         {
1652             *g0 = g_min;
1653             *g1 = g_min - 1;
1654         }
1655         else if (shift_max < 0)
1656         {
1657             *g0 = g_max + 1;
1658             *g1 = g_max;
1659         }
1660         else
1661         {
1662             *g0 = xgi;
1663             *g1 = xgi;
1664             dcx2[xgi] = 0;
1665         }
1666     }
1667
1668     while (*g0 > g_min)
1669     {
1670         /* Check one grid cell down */
1671         dcx = ((*g0 - 1) + 1)*gridx - x;
1672         tmp = dcx*dcx;
1673         if (tmp >= rc2)
1674         {
1675             break;
1676         }
1677         (*g0)--;
1678         dcx2[*g0] = tmp;
1679     }
1680
1681     while (*g1 < g_max)
1682     {
1683         /* Check one grid cell up */
1684         dcx = (*g1 + 1)*gridx - x;
1685         tmp = dcx*dcx;
1686         if (tmp >= rc2)
1687         {
1688             break;
1689         }
1690         (*g1)++;
1691         dcx2[*g1] = tmp;
1692     }
1693 }
1694
1695
1696 #define sqr(x) ((x)*(x))
1697 #define calc_dx2(XI,YI,ZI,y) (sqr(XI-y[XX]) + sqr(YI-y[YY]) + sqr(ZI-y[ZZ]))
1698 #define calc_cyl_dx2(XI,YI,y) (sqr(XI-y[XX]) + sqr(YI-y[YY]))
1699 /****************************************************
1700  *
1701  *    F A S T   N E I G H B O R  S E A R C H I N G
1702  *
1703  *    Optimized neighboursearching routine using grid
1704  *    at least 1x1x1, see GROMACS manual
1705  *
1706  ****************************************************/
1707
1708 static void do_longrange(t_commrec *cr,gmx_localtop_t *top,t_forcerec *fr,
1709                          int ngid,t_mdatoms *md,int icg,
1710                          int jgid,int nlr,
1711                          atom_id lr[],t_excl bexcl[],int shift,
1712                          rvec x[],rvec box_size,t_nrnb *nrnb,
1713                          real lambda,real *dvdlambda,
1714                          gmx_grppairener_t *grppener,
1715                          bool bDoVdW,bool bDoCoul,
1716                          bool bEvaluateNow,bool bHaveVdW[],
1717                          bool bDoForces,rvec *f)
1718 {
1719     int n,i;
1720     t_nblist *nl;
1721
1722     for(n=0; n<fr->nnblists; n++)
1723     {
1724         for(i=0; (i<eNL_NR); i++)
1725         {
1726             nl = &fr->nblists[n].nlist_lr[i];
1727             if ((nl->nri > nl->maxnri-32) || bEvaluateNow)
1728             {
1729                 close_neighbor_list(fr,TRUE,n,i,FALSE);
1730                 /* Evaluate the energies and forces */
1731                 do_nonbonded(cr,fr,x,f,md,NULL,
1732                              grppener->ener[fr->bBHAM ? egBHAMLR : egLJLR],
1733                              grppener->ener[egCOULLR],
1734                                                          grppener->ener[egGB],box_size,
1735                              nrnb,lambda,dvdlambda,n,i,
1736                              GMX_DONB_LR | GMX_DONB_FORCES);
1737
1738                 reset_neighbor_list(fr,TRUE,n,i);
1739             }
1740         }
1741     }
1742
1743     if (!bEvaluateNow)
1744     {
1745         /* Put the long range particles in a list */
1746         /* Since do_longrange is never called for QMMM, bQMMM is FALSE */
1747         put_in_list(bHaveVdW,ngid,md,icg,jgid,nlr,lr,top->cgs.index,
1748                     bexcl,shift,fr,TRUE,bDoVdW,bDoCoul,FALSE);
1749     }
1750 }
1751
1752 static int nsgrid_core(FILE *log,t_commrec *cr,t_forcerec *fr,
1753                        matrix box,rvec box_size,int ngid,
1754                        gmx_localtop_t *top,
1755                        t_grid *grid,rvec x[],
1756                        t_excl bexcl[],bool *bExcludeAlleg,
1757                        t_nrnb *nrnb,t_mdatoms *md,
1758                        real lambda,real *dvdlambda,
1759                        gmx_grppairener_t *grppener,
1760                        bool bHaveVdW[],
1761                        bool bDoLongRange,bool bDoForces,rvec *f,
1762                        bool bMakeQMMMnblist)
1763 {
1764     gmx_ns_t *ns;
1765     atom_id **nl_lr_ljc,**nl_lr_one,**nl_sr;
1766     int     *nlr_ljc,*nlr_one,*nsr;
1767     gmx_domdec_t *dd=NULL;
1768     t_block *cgs=&(top->cgs);
1769     int     *cginfo=fr->cginfo;
1770     /* atom_id *i_atoms,*cgsindex=cgs->index; */
1771     ivec    sh0,sh1,shp;
1772     int     cell_x,cell_y,cell_z;
1773     int     d,tx,ty,tz,dx,dy,dz,cj;
1774 #ifdef ALLOW_OFFDIAG_LT_HALFDIAG
1775     int     zsh_ty,zsh_tx,ysh_tx;
1776 #endif
1777     int     dx0,dx1,dy0,dy1,dz0,dz1;
1778     int     Nx,Ny,Nz,shift=-1,j,nrj,nns,nn=-1;
1779     real    gridx,gridy,gridz,grid_x,grid_y,grid_z;
1780     real    *dcx2,*dcy2,*dcz2;
1781     int     zgi,ygi,xgi;
1782     int     cg0,cg1,icg=-1,cgsnr,i0,igid,nri,naaj,max_jcg;
1783     int     jcg0,jcg1,jjcg,cgj0,jgid;
1784     int     *grida,*gridnra,*gridind;
1785     bool    rvdw_lt_rcoul,rcoul_lt_rvdw;
1786     rvec    xi,*cgcm,grid_offset;
1787     real    r2,rs2,rvdw2,rcoul2,rm2,rl2,XI,YI,ZI,dcx,dcy,dcz,tmp1,tmp2;
1788     int     *i_egp_flags;
1789     bool    bDomDec,bTriclinicX,bTriclinicY;
1790     ivec    ncpddc;
1791
1792     ns = &fr->ns;
1793
1794     bDomDec = DOMAINDECOMP(cr);
1795     if (bDomDec)
1796     {
1797         dd = cr->dd;
1798     }
1799
1800     bTriclinicX = ((YY < grid->npbcdim &&
1801                     (!bDomDec || dd->nc[YY]==1) && box[YY][XX] != 0) ||
1802                    (ZZ < grid->npbcdim &&
1803                     (!bDomDec || dd->nc[ZZ]==1) && box[ZZ][XX] != 0));
1804     bTriclinicY =  (ZZ < grid->npbcdim &&
1805                     (!bDomDec || dd->nc[ZZ]==1) && box[ZZ][YY] != 0);
1806
1807     cgsnr    = cgs->nr;
1808     rs2      = sqr(fr->rlist);
1809     if (bDoLongRange && fr->bTwinRange)
1810     {
1811         /* The VdW and elec. LR cut-off's could be different,
1812          * so we can not simply set them to rlistlong.
1813          */
1814         if (EVDW_ZERO_AT_CUTOFF(fr->vdwtype) && fr->rvdw > fr->rlist)
1815         {
1816             rvdw2  = sqr(fr->rlistlong);
1817         }
1818         else
1819         {
1820             rvdw2  = sqr(fr->rvdw);
1821         }
1822         if (EEL_ZERO_AT_CUTOFF(fr->eeltype) && fr->rcoulomb > fr->rlist)
1823         {
1824             rcoul2 = sqr(fr->rlistlong);
1825         }
1826         else
1827         {
1828             rcoul2 = sqr(fr->rcoulomb);
1829         }
1830     }
1831     else
1832     {
1833         /* Workaround for a gcc -O3 or -ffast-math problem */
1834         rvdw2  = rs2;
1835         rcoul2 = rs2;
1836     }
1837     rm2 = min(rvdw2,rcoul2);
1838     rl2 = max(rvdw2,rcoul2);
1839     rvdw_lt_rcoul = (rvdw2 >= rcoul2);
1840     rcoul_lt_rvdw = (rcoul2 >= rvdw2);
1841
1842     if (bMakeQMMMnblist)
1843     {
1844         rm2 = rl2;
1845         rs2 = rl2;
1846     }
1847
1848     if (ns->nl_sr == NULL)
1849     {
1850         /* Short range buffers */
1851         snew(ns->nl_sr,ngid);
1852         /* Counters */
1853         snew(ns->nsr,ngid);
1854         snew(ns->nlr_ljc,ngid);
1855         snew(ns->nlr_one,ngid);
1856
1857         if (rm2 > rs2)
1858         {
1859             /* Long range VdW and Coul buffers */
1860             snew(ns->nl_lr_ljc,ngid);
1861         }
1862         if (rl2 > rm2)
1863         {
1864             /* Long range VdW or Coul only buffers */
1865             snew(ns->nl_lr_one,ngid);
1866         }
1867         for(j=0; (j<ngid); j++) {
1868             snew(ns->nl_sr[j],MAX_CG);
1869             if (rm2 > rs2)
1870             {
1871                 snew(ns->nl_lr_ljc[j],MAX_CG);
1872             }
1873             if (rl2 > rm2)
1874             {
1875                 snew(ns->nl_lr_one[j],MAX_CG);
1876             }
1877         }
1878         if (debug)
1879             fprintf(debug,
1880                     "ns5_core: rs2 = %g, rvdw2 = %g, rcoul2 = %g (nm^2)\n",
1881                     rs2,rvdw2,rcoul2);
1882     }
1883     nl_sr     = ns->nl_sr;
1884     nsr       = ns->nsr;
1885     nl_lr_ljc = ns->nl_lr_ljc;
1886     nl_lr_one = ns->nl_lr_one;
1887     nlr_ljc   = ns->nlr_ljc;
1888     nlr_one   = ns->nlr_one;
1889
1890     /* Unpack arrays */
1891     cgcm    = fr->cg_cm;
1892     Nx      = grid->n[XX];
1893     Ny      = grid->n[YY];
1894     Nz      = grid->n[ZZ];
1895     grida   = grid->a;
1896     gridind = grid->index;
1897     gridnra = grid->nra;
1898     nns     = 0;
1899
1900     gridx      = grid->cell_size[XX];
1901     gridy      = grid->cell_size[YY];
1902     gridz      = grid->cell_size[ZZ];
1903     grid_x     = 1/gridx;
1904     grid_y     = 1/gridy;
1905     grid_z     = 1/gridz;
1906     copy_rvec(grid->cell_offset,grid_offset);
1907     copy_ivec(grid->ncpddc,ncpddc);
1908     dcx2       = grid->dcx2;
1909     dcy2       = grid->dcy2;
1910     dcz2       = grid->dcz2;
1911
1912 #ifdef ALLOW_OFFDIAG_LT_HALFDIAG
1913     zsh_ty = floor(-box[ZZ][YY]/box[YY][YY]+0.5);
1914     zsh_tx = floor(-box[ZZ][XX]/box[XX][XX]+0.5);
1915     ysh_tx = floor(-box[YY][XX]/box[XX][XX]+0.5);
1916     if (zsh_tx!=0 && ysh_tx!=0)
1917     {
1918         /* This could happen due to rounding, when both ratios are 0.5 */
1919         ysh_tx = 0;
1920     }
1921 #endif
1922
1923     debug_gmx();
1924
1925     if (fr->n_tpi)
1926     {
1927         /* We only want a list for the test particle */
1928         cg0 = cgsnr - 1;
1929     }
1930     else
1931     {
1932         cg0 = grid->icg0;
1933     }
1934     cg1 = grid->icg1;
1935
1936     /* Set the shift range */
1937     for(d=0; d<DIM; d++)
1938     {
1939         sh0[d] = -1;
1940         sh1[d] = 1;
1941         /* Check if we need periodicity shifts.
1942          * Without PBC or with domain decomposition we don't need them.
1943          */
1944         if (d >= ePBC2npbcdim(fr->ePBC) || (bDomDec && dd->nc[d] > 1))
1945         {
1946             shp[d] = 0;
1947         }
1948         else
1949         {
1950             if (d == XX &&
1951                 box[XX][XX] - fabs(box[YY][XX]) - fabs(box[ZZ][XX]) < sqrt(rl2))
1952             {
1953                 shp[d] = 2;
1954             }
1955             else
1956             {
1957                 shp[d] = 1;
1958             }
1959         }
1960     }
1961
1962     /* Loop over charge groups */
1963     for(icg=cg0; (icg < cg1); icg++)
1964     {
1965         igid = GET_CGINFO_GID(cginfo[icg]);
1966         /* Skip this charge group if all energy groups are excluded! */
1967         if (bExcludeAlleg[igid])
1968         {
1969             continue;
1970         }
1971
1972         i0   = cgs->index[icg];
1973
1974         if (bMakeQMMMnblist)
1975         {
1976             /* Skip this charge group if it is not a QM atom while making a
1977              * QM/MM neighbourlist
1978              */
1979             if (md->bQM[i0]==FALSE)
1980             {
1981                 continue; /* MM particle, go to next particle */
1982             }
1983
1984             /* Compute the number of charge groups that fall within the control
1985              * of this one (icg)
1986              */
1987             naaj    = calc_naaj(icg,cgsnr);
1988             jcg0    = icg;
1989             jcg1    = icg + naaj;
1990             max_jcg = cgsnr;
1991         }
1992         else
1993         {
1994             /* make a normal neighbourlist */
1995
1996             if (bDomDec)
1997             {
1998                 /* Get the j charge-group and dd cell shift ranges */
1999                 dd_get_ns_ranges(cr->dd,icg,&jcg0,&jcg1,sh0,sh1);
2000                 max_jcg = 0;
2001             }
2002             else
2003             {
2004                 /* Compute the number of charge groups that fall within the control
2005                  * of this one (icg)
2006                  */
2007                 naaj = calc_naaj(icg,cgsnr);
2008                 jcg0 = icg;
2009                 jcg1 = icg + naaj;
2010
2011                 if (fr->n_tpi)
2012                 {
2013                     /* The i-particle is awlways the test particle,
2014                      * so we want all j-particles
2015                      */
2016                     max_jcg = cgsnr - 1;
2017                 }
2018                 else
2019                 {
2020                     max_jcg  = jcg1 - cgsnr;
2021                 }
2022             }
2023         }
2024
2025         i_egp_flags = fr->egp_flags + igid*ngid;
2026
2027         /* Set the exclusions for the atoms in charge group icg using a bitmask */
2028         setexcl(i0,cgs->index[icg+1],&top->excls,TRUE,bexcl);
2029
2030         ci2xyz(grid,icg,&cell_x,&cell_y,&cell_z);
2031
2032         /* Changed iicg to icg, DvdS 990115
2033          * (but see consistency check above, DvdS 990330)
2034          */
2035 #ifdef NS5DB
2036         fprintf(log,"icg=%5d, naaj=%5d, cell %d %d %d\n",
2037                 icg,naaj,cell_x,cell_y,cell_z);
2038 #endif
2039         /* Loop over shift vectors in three dimensions */
2040         for (tz=-shp[ZZ]; tz<=shp[ZZ]; tz++)
2041         {
2042             ZI = cgcm[icg][ZZ]+tz*box[ZZ][ZZ];
2043             /* Calculate range of cells in Z direction that have the shift tz */
2044             zgi = cell_z + tz*Nz;
2045 #define FAST_DD_NS
2046 #ifndef FAST_DD_NS
2047             get_dx(Nz,gridz,rl2,zgi,ZI,&dz0,&dz1,dcz2);
2048 #else
2049             get_dx_dd(Nz,gridz,rl2,zgi,ZI-grid_offset[ZZ],
2050                       ncpddc[ZZ],sh0[ZZ],sh1[ZZ],&dz0,&dz1,dcz2);
2051 #endif
2052             if (dz0 > dz1)
2053             {
2054                 continue;
2055             }
2056             for (ty=-shp[YY]; ty<=shp[YY]; ty++)
2057             {
2058                 YI = cgcm[icg][YY]+ty*box[YY][YY]+tz*box[ZZ][YY];
2059                 /* Calculate range of cells in Y direction that have the shift ty */
2060                 if (bTriclinicY)
2061                 {
2062                     ygi = (int)(Ny + (YI - grid_offset[YY])*grid_y) - Ny;
2063                 }
2064                 else
2065                 {
2066                     ygi = cell_y + ty*Ny;
2067                 }
2068 #ifndef FAST_DD_NS
2069                 get_dx(Ny,gridy,rl2,ygi,YI,&dy0,&dy1,dcy2);
2070 #else
2071                 get_dx_dd(Ny,gridy,rl2,ygi,YI-grid_offset[YY],
2072                           ncpddc[YY],sh0[YY],sh1[YY],&dy0,&dy1,dcy2);
2073 #endif
2074                 if (dy0 > dy1)
2075                 {
2076                     continue;
2077                 }
2078                 for (tx=-shp[XX]; tx<=shp[XX]; tx++)
2079                 {
2080                     XI = cgcm[icg][XX]+tx*box[XX][XX]+ty*box[YY][XX]+tz*box[ZZ][XX];
2081                     /* Calculate range of cells in X direction that have the shift tx */
2082                     if (bTriclinicX)
2083                     {
2084                         xgi = (int)(Nx + (XI - grid_offset[XX])*grid_x) - Nx;
2085                     }
2086                     else
2087                     {
2088                         xgi = cell_x + tx*Nx;
2089                     }
2090 #ifndef FAST_DD_NS
2091                     get_dx(Nx,gridx,rl2,xgi*Nx,XI,&dx0,&dx1,dcx2);
2092 #else
2093                     get_dx_dd(Nx,gridx,rl2,xgi,XI-grid_offset[XX],
2094                               ncpddc[XX],sh0[XX],sh1[XX],&dx0,&dx1,dcx2);
2095 #endif
2096                     if (dx0 > dx1)
2097                     {
2098                         continue;
2099                     }
2100                     /* Get shift vector */
2101                     shift=XYZ2IS(tx,ty,tz);
2102 #ifdef NS5DB
2103                     range_check(shift,0,SHIFTS);
2104 #endif
2105                     for(nn=0; (nn<ngid); nn++)
2106                     {
2107                         nsr[nn]      = 0;
2108                         nlr_ljc[nn]  = 0;
2109                         nlr_one[nn] = 0;
2110                     }
2111 #ifdef NS5DB
2112                     fprintf(log,"shift: %2d, dx0,1: %2d,%2d, dy0,1: %2d,%2d, dz0,1: %2d,%2d\n",
2113                             shift,dx0,dx1,dy0,dy1,dz0,dz1);
2114                     fprintf(log,"cgcm: %8.3f  %8.3f  %8.3f\n",cgcm[icg][XX],
2115                             cgcm[icg][YY],cgcm[icg][ZZ]);
2116                     fprintf(log,"xi:   %8.3f  %8.3f  %8.3f\n",XI,YI,ZI);
2117 #endif
2118                     for (dx=dx0; (dx<=dx1); dx++)
2119                     {
2120                         tmp1 = rl2 - dcx2[dx];
2121                         for (dy=dy0; (dy<=dy1); dy++)
2122                         {
2123                             tmp2 = tmp1 - dcy2[dy];
2124                             if (tmp2 > 0)
2125                             {
2126                                 for (dz=dz0; (dz<=dz1); dz++) {
2127                                     if (tmp2 > dcz2[dz]) {
2128                                         /* Find grid-cell cj in which possible neighbours are */
2129                                         cj   = xyz2ci(Ny,Nz,dx,dy,dz);
2130
2131                                         /* Check out how many cgs (nrj) there in this cell */
2132                                         nrj  = gridnra[cj];
2133
2134                                         /* Find the offset in the cg list */
2135                                         cgj0 = gridind[cj];
2136
2137                                         /* Check if all j's are out of range so we
2138                                          * can skip the whole cell.
2139                                          * Should save some time, especially with DD.
2140                                          */
2141                                         if (nrj == 0 ||
2142                                             (grida[cgj0] >= max_jcg &&
2143                                              (grida[cgj0] >= jcg1 || grida[cgj0+nrj-1] < jcg0)))
2144                                         {
2145                                             continue;
2146                                         }
2147
2148                                         /* Loop over cgs */
2149                                         for (j=0; (j<nrj); j++)
2150                                         {
2151                                             jjcg = grida[cgj0+j];
2152
2153                                             /* check whether this guy is in range! */
2154                                             if ((jjcg >= jcg0 && jjcg < jcg1) ||
2155                                                 (jjcg < max_jcg))
2156                                             {
2157                                                 r2=calc_dx2(XI,YI,ZI,cgcm[jjcg]);
2158                                                 if (r2 < rl2) {
2159                                                     /* jgid = gid[cgsatoms[cgsindex[jjcg]]]; */
2160                                                     jgid = GET_CGINFO_GID(cginfo[jjcg]);
2161                                                     /* check energy group exclusions */
2162                                                     if (!(i_egp_flags[jgid] & EGP_EXCL))
2163                                                     {
2164                                                         if (r2 < rs2)
2165                                                         {
2166                                                             if (nsr[jgid] >= MAX_CG)
2167                                                             {
2168                                                                 put_in_list(bHaveVdW,ngid,md,icg,jgid,
2169                                                                             nsr[jgid],nl_sr[jgid],
2170                                                                             cgs->index,/* cgsatoms, */ bexcl,
2171                                                                             shift,fr,FALSE,TRUE,TRUE,
2172                                                                             bMakeQMMMnblist);
2173                                                                 nsr[jgid]=0;
2174                                                             }
2175                                                             nl_sr[jgid][nsr[jgid]++]=jjcg;
2176                                                         }
2177                                                         else if (r2 < rm2)
2178                                                         {
2179                                                             if (nlr_ljc[jgid] >= MAX_CG)
2180                                                             {
2181                                                                 do_longrange(cr,top,fr,ngid,md,icg,jgid,
2182                                                                              nlr_ljc[jgid],
2183                                                                              nl_lr_ljc[jgid],bexcl,shift,x,
2184                                                                              box_size,nrnb,
2185                                                                              lambda,dvdlambda,
2186                                                                              grppener,
2187                                                                              TRUE,TRUE,FALSE,
2188                                                                              bHaveVdW,
2189                                                                              bDoForces,f);
2190                                                                 nlr_ljc[jgid]=0;
2191                                                             }
2192                                                             nl_lr_ljc[jgid][nlr_ljc[jgid]++]=jjcg;
2193                                                         }
2194                                                         else
2195                                                         {
2196                                                             if (nlr_one[jgid] >= MAX_CG) {
2197                                                                 do_longrange(cr,top,fr,ngid,md,icg,jgid,
2198                                                                              nlr_one[jgid],
2199                                                                              nl_lr_one[jgid],bexcl,shift,x,
2200                                                                              box_size,nrnb,
2201                                                                              lambda,dvdlambda,
2202                                                                              grppener,
2203                                                                              rvdw_lt_rcoul,rcoul_lt_rvdw,FALSE,
2204                                                                              bHaveVdW,
2205                                                                              bDoForces,f);
2206                                                                 nlr_one[jgid]=0;
2207                                                             }
2208                                                             nl_lr_one[jgid][nlr_one[jgid]++]=jjcg;
2209                                                         }
2210                                                     }
2211                                                 }
2212                                                 nns++;
2213                                             }
2214                                         }
2215                                     }
2216                                 }
2217                             }
2218                         }
2219                     }
2220                     /* CHECK whether there is anything left in the buffers */
2221                     for(nn=0; (nn<ngid); nn++)
2222                     {
2223                         if (nsr[nn] > 0)
2224                         {
2225                             put_in_list(bHaveVdW,ngid,md,icg,nn,nsr[nn],nl_sr[nn],
2226                                         cgs->index, /* cgsatoms, */ bexcl,
2227                                         shift,fr,FALSE,TRUE,TRUE,bMakeQMMMnblist);
2228                         }
2229
2230                         if (nlr_ljc[nn] > 0)
2231                         {
2232                             do_longrange(cr,top,fr,ngid,md,icg,nn,nlr_ljc[nn],
2233                                          nl_lr_ljc[nn],bexcl,shift,x,box_size,nrnb,
2234                                          lambda,dvdlambda,grppener,TRUE,TRUE,FALSE,
2235                                          bHaveVdW,bDoForces,f);
2236                         }
2237
2238                         if (nlr_one[nn] > 0)
2239                         {
2240                             do_longrange(cr,top,fr,ngid,md,icg,nn,nlr_one[nn],
2241                                          nl_lr_one[nn],bexcl,shift,x,box_size,nrnb,
2242                                          lambda,dvdlambda,grppener,
2243                                          rvdw_lt_rcoul,rcoul_lt_rvdw,FALSE,
2244                                          bHaveVdW,bDoForces,f);
2245                         }
2246                     }
2247                 }
2248             }
2249         }
2250         /* setexcl(nri,i_atoms,&top->atoms.excl,FALSE,bexcl); */
2251         setexcl(cgs->index[icg],cgs->index[icg+1],&top->excls,FALSE,bexcl);
2252     }
2253     /* Perform any left over force calculations */
2254     for (nn=0; (nn<ngid); nn++)
2255     {
2256         if (rm2 > rs2)
2257         {
2258             do_longrange(cr,top,fr,0,md,icg,nn,nlr_ljc[nn],
2259                          nl_lr_ljc[nn],bexcl,shift,x,box_size,nrnb,
2260                          lambda,dvdlambda,grppener,
2261                          TRUE,TRUE,TRUE,bHaveVdW,bDoForces,f);
2262         }
2263         if (rl2 > rm2) {
2264             do_longrange(cr,top,fr,0,md,icg,nn,nlr_one[nn],
2265                          nl_lr_one[nn],bexcl,shift,x,box_size,nrnb,
2266                          lambda,dvdlambda,grppener,
2267                          rvdw_lt_rcoul,rcoul_lt_rvdw,
2268                          TRUE,bHaveVdW,bDoForces,f);
2269         }
2270     }
2271     debug_gmx();
2272
2273     /* Close off short range neighbourlists */
2274     close_neighbor_list(fr,FALSE,-1,-1,bMakeQMMMnblist);
2275
2276     return nns;
2277 }
2278
2279 void ns_realloc_natoms(gmx_ns_t *ns,int natoms)
2280 {
2281     int i;
2282
2283     if (natoms > ns->nra_alloc)
2284     {
2285         ns->nra_alloc = over_alloc_dd(natoms);
2286         srenew(ns->bexcl,ns->nra_alloc);
2287         for(i=0; i<ns->nra_alloc; i++)
2288         {
2289             ns->bexcl[i] = 0;
2290         }
2291     }
2292 }
2293
2294 void init_ns(FILE *fplog,const t_commrec *cr,
2295              gmx_ns_t *ns,t_forcerec *fr,
2296              const gmx_mtop_t *mtop,
2297              matrix box)
2298 {
2299     int  mt,icg,nr_in_cg,maxcg,i,j,jcg,ngid,ncg;
2300     t_block *cgs;
2301     char *ptr;
2302
2303     /* Compute largest charge groups size (# atoms) */
2304     nr_in_cg=1;
2305     for(mt=0; mt<mtop->nmoltype; mt++) {
2306         cgs = &mtop->moltype[mt].cgs;
2307         for (icg=0; (icg < cgs->nr); icg++)
2308         {
2309             nr_in_cg=max(nr_in_cg,(int)(cgs->index[icg+1]-cgs->index[icg]));
2310         }
2311     }
2312
2313     /* Verify whether largest charge group is <= max cg.
2314      * This is determined by the type of the local exclusion type
2315      * Exclusions are stored in bits. (If the type is not large
2316      * enough, enlarge it, unsigned char -> unsigned short -> unsigned long)
2317      */
2318     maxcg = sizeof(t_excl)*8;
2319     if (nr_in_cg > maxcg)
2320     {
2321         gmx_fatal(FARGS,"Max #atoms in a charge group: %d > %d\n",
2322                   nr_in_cg,maxcg);
2323     }
2324
2325     ngid = mtop->groups.grps[egcENER].nr;
2326     snew(ns->bExcludeAlleg,ngid);
2327     for(i=0; i<ngid; i++) {
2328         ns->bExcludeAlleg[i] = TRUE;
2329         for(j=0; j<ngid; j++)
2330         {
2331             if (!(fr->egp_flags[i*ngid+j] & EGP_EXCL))
2332             {
2333                 ns->bExcludeAlleg[i] = FALSE;
2334             }
2335         }
2336     }
2337
2338     if (fr->bGrid) {
2339         /* Grid search */
2340         ns->grid = init_grid(fplog,fr);
2341         /* These lists are allocated in ns5_core */
2342         ns->nl_sr = NULL;
2343     }
2344     else
2345     {
2346         /* Simple search */
2347         snew(ns->ns_buf,ngid);
2348         for(i=0; (i<ngid); i++)
2349         {
2350             snew(ns->ns_buf[i],SHIFTS);
2351         }
2352         ncg = ncg_mtop(mtop);
2353         snew(ns->simple_aaj,2*ncg);
2354         for(jcg=0; (jcg<ncg); jcg++)
2355         {
2356             ns->simple_aaj[jcg]     = jcg;
2357             ns->simple_aaj[jcg+ncg] = jcg;
2358         }
2359     }
2360
2361     /* Create array that determines whether or not atoms have VdW */
2362     snew(ns->bHaveVdW,fr->ntype);
2363     for(i=0; (i<fr->ntype); i++)
2364     {
2365         for(j=0; (j<fr->ntype); j++)
2366         {
2367             ns->bHaveVdW[i] = (ns->bHaveVdW[i] ||
2368                                (fr->bBHAM ?
2369                                 ((BHAMA(fr->nbfp,fr->ntype,i,j) != 0) ||
2370                                  (BHAMB(fr->nbfp,fr->ntype,i,j) != 0) ||
2371                                  (BHAMC(fr->nbfp,fr->ntype,i,j) != 0)) :
2372                                 ((C6(fr->nbfp,fr->ntype,i,j) != 0) ||
2373                                  (C12(fr->nbfp,fr->ntype,i,j) != 0))));
2374         }
2375     }
2376     if (debug)
2377         pr_bvec(debug,0,"bHaveVdW",ns->bHaveVdW,fr->ntype,TRUE);
2378
2379     if (!DOMAINDECOMP(cr))
2380     {
2381         /* This could be reduced with particle decomposition */
2382         ns_realloc_natoms(ns,mtop->natoms);
2383     }
2384
2385     ns->nblist_initialized=FALSE;
2386
2387     /* nbr list debug dump */
2388     {
2389         char *ptr=getenv("GMX_DUMP_NL");
2390         if (ptr)
2391         {
2392             ns->dump_nl=strtol(ptr,NULL,0);
2393             if (fplog)
2394             {
2395                 fprintf(fplog, "GMX_DUMP_NL = %d", ns->dump_nl);
2396             }
2397         }
2398         else
2399         {
2400             ns->dump_nl=0;
2401         }
2402     }
2403 }
2404
2405
2406 int search_neighbours(FILE *log,t_forcerec *fr,
2407                       rvec x[],matrix box,
2408                       gmx_localtop_t *top,
2409                       gmx_groups_t *groups,
2410                       t_commrec *cr,
2411                       t_nrnb *nrnb,t_mdatoms *md,
2412                       real lambda,real *dvdlambda,
2413                       gmx_grppairener_t *grppener,
2414                       bool bFillGrid,
2415                       bool bDoLongRange,
2416                       bool bDoForces,rvec *f)
2417 {
2418     t_block  *cgs=&(top->cgs);
2419     rvec     box_size,grid_x0,grid_x1;
2420     int      i,j,m,ngid;
2421     real     min_size,grid_dens;
2422     int      nsearch;
2423     bool     bGrid;
2424     char     *ptr;
2425     bool     *i_egp_flags;
2426     int      cg_start,cg_end,start,end;
2427     gmx_ns_t *ns;
2428     t_grid   *grid;
2429     gmx_domdec_zones_t *dd_zones;
2430
2431     ns = &fr->ns;
2432
2433     /* Set some local variables */
2434     bGrid = fr->bGrid;
2435     ngid = groups->grps[egcENER].nr;
2436
2437     for(m=0; (m<DIM); m++)
2438     {
2439         box_size[m] = box[m][m];
2440     }
2441
2442     if (fr->ePBC != epbcNONE)
2443     {
2444         if (sqr(fr->rlistlong) >= max_cutoff2(fr->ePBC,box))
2445         {
2446             gmx_fatal(FARGS,"One of the box vectors has become shorter than twice the cut-off length or box_yy-|box_zy| or box_zz has become smaller than the cut-off.");
2447         }
2448         if (!bGrid)
2449         {
2450             min_size = min(box_size[XX],min(box_size[YY],box_size[ZZ]));
2451             if (2*fr->rlistlong >= min_size)
2452                 gmx_fatal(FARGS,"One of the box diagonal elements has become smaller than twice the cut-off length.");
2453         }
2454     }
2455
2456     if (DOMAINDECOMP(cr))
2457     {
2458         ns_realloc_natoms(ns,cgs->index[cgs->nr]);
2459     }
2460     debug_gmx();
2461
2462     /* Reset the neighbourlists */
2463     reset_neighbor_list(fr,FALSE,-1,-1);
2464
2465     if (bGrid && bFillGrid)
2466     {
2467
2468         grid = ns->grid;
2469         if (DOMAINDECOMP(cr))
2470         {
2471             dd_zones = domdec_zones(cr->dd);
2472         }
2473         else
2474         {
2475             dd_zones = NULL;
2476
2477             get_nsgrid_boundaries(grid,NULL,box,NULL,NULL,NULL,
2478                                   cgs->nr,fr->cg_cm,grid_x0,grid_x1,&grid_dens);
2479
2480             grid_first(log,grid,NULL,NULL,fr->ePBC,box,grid_x0,grid_x1,
2481                        fr->rlistlong,grid_dens);
2482         }
2483         debug_gmx();
2484
2485         /* Don't know why this all is... (DvdS 3/99) */
2486 #ifndef SEGV
2487         start = 0;
2488         end   = cgs->nr;
2489 #else
2490         start = fr->cg0;
2491         end   = (cgs->nr+1)/2;
2492 #endif
2493
2494         if (DOMAINDECOMP(cr))
2495         {
2496             end = cgs->nr;
2497             fill_grid(log,dd_zones,grid,end,-1,end,fr->cg_cm);
2498             grid->icg0 = 0;
2499             grid->icg1 = dd_zones->izone[dd_zones->nizone-1].cg1;
2500         }
2501         else
2502         {
2503             fill_grid(log,NULL,grid,cgs->nr,fr->cg0,fr->hcg,fr->cg_cm);
2504             grid->icg0 = fr->cg0;
2505             grid->icg1 = fr->hcg;
2506             debug_gmx();
2507
2508             if (PARTDECOMP(cr))
2509                 mv_grid(cr,grid);
2510             debug_gmx();
2511         }
2512
2513         calc_elemnr(log,grid,start,end,cgs->nr);
2514         calc_ptrs(grid);
2515         grid_last(log,grid,start,end,cgs->nr);
2516
2517         if (gmx_debug_at)
2518         {
2519             check_grid(debug,grid);
2520             print_grid(debug,grid);
2521         }
2522     }
2523     else if (fr->n_tpi)
2524     {
2525         /* Set the grid cell index for the test particle only.
2526          * The cell to cg index is not corrected, but that does not matter.
2527          */
2528         fill_grid(log,NULL,ns->grid,fr->hcg,fr->hcg-1,fr->hcg,fr->cg_cm);
2529     }
2530     debug_gmx();
2531
2532     /* Do the core! */
2533     if (bGrid)
2534     {
2535         grid = ns->grid;
2536         nsearch = nsgrid_core(log,cr,fr,box,box_size,ngid,top,
2537                               grid,x,ns->bexcl,ns->bExcludeAlleg,
2538                               nrnb,md,lambda,dvdlambda,grppener,
2539                               ns->bHaveVdW,
2540                               bDoLongRange,bDoForces,f,
2541                               FALSE);
2542
2543         /* neighbour searching withouth QMMM! QM atoms have zero charge in
2544          * the classical calculation. The charge-charge interaction
2545          * between QM and MM atoms is handled in the QMMM core calculation
2546          * (see QMMM.c). The VDW however, we'd like to compute classically
2547          * and the QM MM atom pairs have just been put in the
2548          * corresponding neighbourlists. in case of QMMM we still need to
2549          * fill a special QMMM neighbourlist that contains all neighbours
2550          * of the QM atoms. If bQMMM is true, this list will now be made:
2551          */
2552         if (fr->bQMMM && fr->qr->QMMMscheme!=eQMMMschemeoniom)
2553         {
2554             nsearch += nsgrid_core(log,cr,fr,box,box_size,ngid,top,
2555                                    grid,x,ns->bexcl,ns->bExcludeAlleg,
2556                                    nrnb,md,lambda,dvdlambda,grppener,
2557                                    ns->bHaveVdW,
2558                                    bDoLongRange,bDoForces,f,
2559                                    TRUE);
2560         }
2561     }
2562     else
2563     {
2564         nsearch = ns_simple_core(fr,top,md,box,box_size,
2565                                  ns->bexcl,ns->simple_aaj,
2566                                  ngid,ns->ns_buf,ns->bHaveVdW);
2567     }
2568     debug_gmx();
2569
2570 #ifdef DEBUG
2571     pr_nsblock(log);
2572 #endif
2573
2574     inc_nrnb(nrnb,eNR_NS,nsearch);
2575     /* inc_nrnb(nrnb,eNR_LR,fr->nlr); */
2576
2577     return nsearch;
2578 }
2579
2580 int natoms_beyond_ns_buffer(t_inputrec *ir,t_forcerec *fr,t_block *cgs,
2581                             matrix scale_tot,rvec *x)
2582 {
2583     int  cg0,cg1,cg,a0,a1,a,i,j;
2584     real rint,hbuf2,scale;
2585     rvec *cg_cm,cgsc;
2586     bool bIsotropic;
2587     int  nBeyond;
2588
2589     nBeyond = 0;
2590
2591     rint = max(ir->rcoulomb,ir->rvdw);
2592     if (ir->rlist < rint)
2593     {
2594         gmx_fatal(FARGS,"The neighbor search buffer has negative size: %f nm",
2595                   ir->rlist - rint);
2596     }
2597     cg_cm = fr->cg_cm;
2598
2599     cg0 = fr->cg0;
2600     cg1 = fr->hcg;
2601
2602     if (!EI_DYNAMICS(ir->eI) || !DYNAMIC_BOX(*ir))
2603     {
2604         hbuf2 = sqr(0.5*(ir->rlist - rint));
2605         for(cg=cg0; cg<cg1; cg++)
2606         {
2607             a0 = cgs->index[cg];
2608             a1 = cgs->index[cg+1];
2609             for(a=a0; a<a1; a++)
2610             {
2611                 if (distance2(cg_cm[cg],x[a]) > hbuf2)
2612                 {
2613                     nBeyond++;
2614                 }
2615             }
2616         }
2617     }
2618     else
2619     {
2620         bIsotropic = TRUE;
2621         scale = scale_tot[0][0];
2622         for(i=1; i<DIM; i++)
2623         {
2624             /* With anisotropic scaling, the original spherical ns volumes become
2625              * ellipsoids. To avoid costly transformations we use the minimum
2626              * eigenvalue of the scaling matrix for determining the buffer size.
2627              * Since the lower half is 0, the eigenvalues are the diagonal elements.
2628              */
2629             scale = min(scale,scale_tot[i][i]);
2630             if (scale_tot[i][i] != scale_tot[i-1][i-1])
2631             {
2632                 bIsotropic = FALSE;
2633             }
2634             for(j=0; j<i; j++)
2635             {
2636                 if (scale_tot[i][j] != 0)
2637                 {
2638                     bIsotropic = FALSE;
2639                 }
2640             }
2641         }
2642         hbuf2 = sqr(0.5*(scale*ir->rlist - rint));
2643         if (bIsotropic)
2644         {
2645             for(cg=cg0; cg<cg1; cg++)
2646             {
2647                 svmul(scale,cg_cm[cg],cgsc);
2648                 a0 = cgs->index[cg];
2649                 a1 = cgs->index[cg+1];
2650                 for(a=a0; a<a1; a++)
2651                 {
2652                     if (distance2(cgsc,x[a]) > hbuf2)
2653                     {
2654                         nBeyond++;
2655                     }
2656                 }
2657             }
2658         }
2659         else
2660         {
2661             /* Anistropic scaling */
2662             for(cg=cg0; cg<cg1; cg++)
2663             {
2664                 /* Since scale_tot contains the transpose of the scaling matrix,
2665                  * we need to multiply with the transpose.
2666                  */
2667                 tmvmul_ur0(scale_tot,cg_cm[cg],cgsc);
2668                 a0 = cgs->index[cg];
2669                 a1 = cgs->index[cg+1];
2670                 for(a=a0; a<a1; a++)
2671                 {
2672                     if (distance2(cgsc,x[a]) > hbuf2)
2673                     {
2674                         nBeyond++;
2675                     }
2676                 }
2677             }
2678         }
2679     }
2680
2681     return nBeyond;
2682 }