src/gromacs/gmxlib/nonbonded/nb_kernel_adress_c/nb_kernel110_c_adress.c

   1 /*
   2  * Copyright (c) Erik Lindahl, David van der Spoel 2003
   3  *
   4  * This file is generated automatically at compile time
   5  * by the program mknb in the Gromacs distribution.
   6  *
   7  * Options used when generation this file:
   8  * Language:         c
   9  * Precision:        single
  10  * Threads:          yes
  11  * Software invsqrt: no
  12  * PowerPC invsqrt:  no
  13  * Prefetch forces:  no
  14  * Adress kernel:  yes
  15  * Comments:         no
  16  */
  17 #ifdef HAVE_CONFIG_H
  18 #include<config.h>
  19 #endif
  20 #ifdef GMX_THREAD_SHM_FDECOMP
  21 #include<thread_mpi.h>
  22 #endif
  23 #define ALMOST_ZERO 1e-30
  24 #define ALMOST_ONE 1-(1e-30)
  25 #include<math.h>
  26
  27 #include "nb_kernel110_adress.h"
  28
  29
  30
  31 /*
  32  * Gromacs nonbonded kernel nb_kernel110_adress_cg
  33  * Coulomb interaction:     Normal Coulomb
  34  * VdW interaction:         Lennard-Jones
  35  * water optimization:      No
  36  * Calculate forces:        yes
  37  */
  38 void nb_kernel110_adress_cg(
  39                     int *           p_nri,
  40                     int *           iinr,
  41                     int *           jindex,
  42                     int *           jjnr,
  43                     int *           shift,
  44                     real *         shiftvec,
  45                     real *         fshift,
  46                     int *           gid,
  47                     real *         pos,
  48                     real *         faction,
  49                     real *         charge,
  50                     real *         p_facel,
  51                     real *         p_krf,
  52                     real *         p_crf,
  53                     real *         Vc,
  54                     int *           type,
  55                     int *           p_ntype,
  56                     real *         vdwparam,
  57                     real *         Vvdw,
  58                     real *         p_tabscale,
  59                     real *         VFtab,
  60                     real *         invsqrta,
  61                     real *         dvda,
  62                     real *         p_gbtabscale,
  63                     real *         GBtab,
  64                     int *           p_nthreads,
  65                     int *           count,
  66                     void *          mtx,
  67                     int *           outeriter,
  68                     int *           inneriter,
  69                     real           force_cap,
  70                     real *         wf)
  71 {
  72     int           nri,ntype,nthreads;
  73     real         facel,krf,crf,tabscale,gbtabscale;
  74     int           n,ii,is3,ii3,k,nj0,nj1,jnr,j3,ggid;
  75     int           nn0,nn1,nouter,ninner;
  76     real         shX,shY,shZ;
  77     real         fscal,tx,ty,tz;
  78     real         rinvsq;
  79     real         iq;
  80     real         qq,vcoul,vctot;
  81     int           nti;
  82     int           tj;
  83     real         rinvsix;
  84     real         Vvdw6,Vvdwtot;
  85     real         Vvdw12;
  86     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
  87     real         jx1,jy1,jz1;
  88     real         dx11,dy11,dz11,rsq11,rinv11;
  89     real         c6,c12;
  90     real         weight_cg1, weight_cg2, weight_product;
  91     real         hybscal;
  92
  93     nri              = *p_nri;
  94     ntype            = *p_ntype;
  95     nthreads         = *p_nthreads;
  96     facel            = *p_facel;
  97     krf              = *p_krf;
  98     crf              = *p_crf;
  99     tabscale         = *p_tabscale;
 100     nouter           = 0;
 101     ninner           = 0;
 102
 103     do
 104     {
 105         #ifdef GMX_THREAD_SHM_FDECOMP
 106         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 107         nn0              = *count;
 108         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 109         *count           = nn1;
 110         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 111         if(nn1>nri) nn1=nri;
 112         #else
 113         nn0 = 0;
 114         nn1 = nri;
 115         #endif
 116
 117         for(n=nn0; (n<nn1); n++)
 118         {
 119             is3              = 3*shift[n];
 120             shX              = shiftvec[is3];
 121             shY              = shiftvec[is3+1];
 122             shZ              = shiftvec[is3+2];
 123             nj0              = jindex[n];
 124             nj1              = jindex[n+1];
 125             ii               = iinr[n];
 126             ii3              = 3*ii;
 127             ix1              = shX + pos[ii3+0];
 128             iy1              = shY + pos[ii3+1];
 129             iz1              = shZ + pos[ii3+2];
 130             iq               = facel*charge[ii];
 131             nti              = 2*ntype*type[ii];
 132             weight_cg1       = wf[ii];
 133             vctot            = 0;
 134             Vvdwtot          = 0;
 135             fix1             = 0;
 136             fiy1             = 0;
 137             fiz1             = 0;
 138
 139             for(k=nj0; (k<nj1); k++)
 140             {
 141                 jnr              = jjnr[k];
 142                 weight_cg2       = wf[jnr];
 143                 weight_product   = weight_cg1*weight_cg2;
 144                 if (weight_product < ALMOST_ZERO) {
 145                        hybscal = 1.0;
 146                 }
 147                 else if (weight_product >= ALMOST_ONE)
 148                 {
 149                   /* force is zero, skip this molecule */
 150                        continue;
 151                 }
 152                 else
 153                 {
 154                    hybscal = 1.0 - weight_product;
 155                 }
 156                 j3               = 3*jnr;
 157                 jx1              = pos[j3+0];
 158                 jy1              = pos[j3+1];
 159                 jz1              = pos[j3+2];
 160                 dx11             = ix1 - jx1;
 161                 dy11             = iy1 - jy1;
 162                 dz11             = iz1 - jz1;
 163                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 164                 rinv11           = 1.0/sqrt(rsq11);
 165                 qq               = iq*charge[jnr];
 166                 tj               = nti+2*type[jnr];
 167                 c6               = vdwparam[tj];
 168                 c12              = vdwparam[tj+1];
 169                 rinvsq           = rinv11*rinv11;
 170                 vcoul            = qq*rinv11;
 171                 vctot            = vctot+vcoul;
 172                 rinvsix          = rinvsq*rinvsq*rinvsq;
 173                 Vvdw6            = c6*rinvsix;
 174                 Vvdw12           = c12*rinvsix*rinvsix;
 175                 Vvdwtot          = Vvdwtot+Vvdw12-Vvdw6;
 176                 fscal            = (vcoul+12.0*Vvdw12-6.0*Vvdw6)*rinvsq;
 177                 fscal *= hybscal;
 178                 tx               = fscal*dx11;
 179                 ty               = fscal*dy11;
 180                 tz               = fscal*dz11;
 181                 fix1             = fix1 + tx;
 182                 fiy1             = fiy1 + ty;
 183                 fiz1             = fiz1 + tz;
 184                 faction[j3+0]    = faction[j3+0] - tx;
 185                 faction[j3+1]    = faction[j3+1] - ty;
 186                 faction[j3+2]    = faction[j3+2] - tz;
 187             }
 188
 189             faction[ii3+0]   = faction[ii3+0] + fix1;
 190             faction[ii3+1]   = faction[ii3+1] + fiy1;
 191             faction[ii3+2]   = faction[ii3+2] + fiz1;
 192             fshift[is3]      = fshift[is3]+fix1;
 193             fshift[is3+1]    = fshift[is3+1]+fiy1;
 194             fshift[is3+2]    = fshift[is3+2]+fiz1;
 195             ggid             = gid[n];
 196             Vc[ggid]         = Vc[ggid] + vctot;
 197             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 198             ninner           = ninner + nj1 - nj0;
 199         }
 200
 201         nouter           = nouter + nn1 - nn0;
 202     }
 203     while (nn1<nri);
 204
 205     *outeriter       = nouter;
 206     *inneriter       = ninner;
 207 }
 208
 209
 210
 211
 212
 213 /*
 214  * Gromacs nonbonded kernel nb_kernel110_adress_ex
 215  * Coulomb interaction:     Normal Coulomb
 216  * VdW interaction:         Lennard-Jones
 217  * water optimization:      No
 218  * Calculate forces:        yes
 219  */
 220 void nb_kernel110_adress_ex(
 221                     int *           p_nri,
 222                     int *           iinr,
 223                     int *           jindex,
 224                     int *           jjnr,
 225                     int *           shift,
 226                     real *         shiftvec,
 227                     real *         fshift,
 228                     int *           gid,
 229                     real *         pos,
 230                     real *         faction,
 231                     real *         charge,
 232                     real *         p_facel,
 233                     real *         p_krf,
 234                     real *         p_crf,
 235                     real *         Vc,
 236                     int *           type,
 237                     int *           p_ntype,
 238                     real *         vdwparam,
 239                     real *         Vvdw,
 240                     real *         p_tabscale,
 241                     real *         VFtab,
 242                     real *         invsqrta,
 243                     real *         dvda,
 244                     real *         p_gbtabscale,
 245                     real *         GBtab,
 246                     int *           p_nthreads,
 247                     int *           count,
 248                     void *          mtx,
 249                     int *           outeriter,
 250                     int *           inneriter,
 251                     real           force_cap,
 252                     real *         wf)
 253 {
 254     int           nri,ntype,nthreads;
 255     real         facel,krf,crf,tabscale,gbtabscale;
 256     int           n,ii,is3,ii3,k,nj0,nj1,jnr,j3,ggid;
 257     int           nn0,nn1,nouter,ninner;
 258     real         shX,shY,shZ;
 259     real         fscal,tx,ty,tz;
 260     real         rinvsq;
 261     real         iq;
 262     real         qq,vcoul,vctot;
 263     int           nti;
 264     int           tj;
 265     real         rinvsix;
 266     real         Vvdw6,Vvdwtot;
 267     real         Vvdw12;
 268     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
 269     real         jx1,jy1,jz1;
 270     real         dx11,dy11,dz11,rsq11,rinv11;
 271     real         c6,c12;
 272     real         weight_cg1, weight_cg2, weight_product;
 273     real         hybscal;
 274
 275     nri              = *p_nri;
 276     ntype            = *p_ntype;
 277     nthreads         = *p_nthreads;
 278     facel            = *p_facel;
 279     krf              = *p_krf;
 280     crf              = *p_crf;
 281     tabscale         = *p_tabscale;
 282     nouter           = 0;
 283     ninner           = 0;
 284
 285     do
 286     {
 287         #ifdef GMX_THREAD_SHM_FDECOMP
 288         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 289         nn0              = *count;
 290         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 291         *count           = nn1;
 292         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 293         if(nn1>nri) nn1=nri;
 294         #else
 295         nn0 = 0;
 296         nn1 = nri;
 297         #endif
 298
 299         for(n=nn0; (n<nn1); n++)
 300         {
 301             is3              = 3*shift[n];
 302             shX              = shiftvec[is3];
 303             shY              = shiftvec[is3+1];
 304             shZ              = shiftvec[is3+2];
 305             nj0              = jindex[n];
 306             nj1              = jindex[n+1];
 307             ii               = iinr[n];
 308             ii3              = 3*ii;
 309             ix1              = shX + pos[ii3+0];
 310             iy1              = shY + pos[ii3+1];
 311             iz1              = shZ + pos[ii3+2];
 312             iq               = facel*charge[ii];
 313             nti              = 2*ntype*type[ii];
 314             weight_cg1       = wf[ii];
 315             vctot            = 0;
 316             Vvdwtot          = 0;
 317             fix1             = 0;
 318             fiy1             = 0;
 319             fiz1             = 0;
 320
 321             for(k=nj0; (k<nj1); k++)
 322             {
 323                 jnr              = jjnr[k];
 324                 weight_cg2       = wf[jnr];
 325                 weight_product   = weight_cg1*weight_cg2;
 326                 if (weight_product < ALMOST_ZERO) {
 327                 /* force is zero, skip this molecule */
 328                  continue;
 329                 }
 330                 else if (weight_product >= ALMOST_ONE)
 331                 {
 332                        hybscal = 1.0;
 333                 }
 334                 else
 335                 {
 336                    hybscal = weight_product;
 337                 }
 338                 j3               = 3*jnr;
 339                 jx1              = pos[j3+0];
 340                 jy1              = pos[j3+1];
 341                 jz1              = pos[j3+2];
 342                 dx11             = ix1 - jx1;
 343                 dy11             = iy1 - jy1;
 344                 dz11             = iz1 - jz1;
 345                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 346                 rinv11           = 1.0/sqrt(rsq11);
 347                 qq               = iq*charge[jnr];
 348                 tj               = nti+2*type[jnr];
 349                 c6               = vdwparam[tj];
 350                 c12              = vdwparam[tj+1];
 351                 rinvsq           = rinv11*rinv11;
 352                 vcoul            = qq*rinv11;
 353                 vctot            = vctot+vcoul;
 354                 rinvsix          = rinvsq*rinvsq*rinvsq;
 355                 Vvdw6            = c6*rinvsix;
 356                 Vvdw12           = c12*rinvsix*rinvsix;
 357                 Vvdwtot          = Vvdwtot+Vvdw12-Vvdw6;
 358                 fscal            = (vcoul+12.0*Vvdw12-6.0*Vvdw6)*rinvsq;
 359                 fscal *= hybscal;
 360                 if(force_cap>0 && (fabs(fscal)> force_cap)){
 361                 fscal=force_cap*fscal/fabs(fscal);
 362                 }
 363                 tx               = fscal*dx11;
 364                 ty               = fscal*dy11;
 365                 tz               = fscal*dz11;
 366                 fix1             = fix1 + tx;
 367                 fiy1             = fiy1 + ty;
 368                 fiz1             = fiz1 + tz;
 369                 faction[j3+0]    = faction[j3+0] - tx;
 370                 faction[j3+1]    = faction[j3+1] - ty;
 371                 faction[j3+2]    = faction[j3+2] - tz;
 372             }
 373
 374             faction[ii3+0]   = faction[ii3+0] + fix1;
 375             faction[ii3+1]   = faction[ii3+1] + fiy1;
 376             faction[ii3+2]   = faction[ii3+2] + fiz1;
 377             fshift[is3]      = fshift[is3]+fix1;
 378             fshift[is3+1]    = fshift[is3+1]+fiy1;
 379             fshift[is3+2]    = fshift[is3+2]+fiz1;
 380             ggid             = gid[n];
 381             Vc[ggid]         = Vc[ggid] + vctot;
 382             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 383             ninner           = ninner + nj1 - nj0;
 384         }
 385
 386         nouter           = nouter + nn1 - nn0;
 387     }
 388     while (nn1<nri);
 389
 390     *outeriter       = nouter;
 391     *inneriter       = ninner;
 392 }
 393
 394