src/gromacs/gmxlib/nonbonded/nb_kernel_adress_c/nb_kernel330_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_kernel330_adress.h"
  28
  29
  30
  31 /*
  32  * Gromacs nonbonded kernel nb_kernel330_adress_cg
  33  * Coulomb interaction:     Tabulated
  34  * VdW interaction:         Tabulated
  35  * water optimization:      No
  36  * Calculate forces:        yes
  37  */
  38 void nb_kernel330_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         iq;
  79     real         qq,vcoul,vctot;
  80     int           nti;
  81     int           tj;
  82     real         Vvdw6,Vvdwtot;
  83     real         Vvdw12;
  84     real         r,rt,eps,eps2;
  85     int           n0,nnn;
  86     real         Y,F,Geps,Heps2,Fp,VV;
  87     real         FF;
  88     real         fijC;
  89     real         fijD,fijR;
  90     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
  91     real         jx1,jy1,jz1;
  92     real         dx11,dy11,dz11,rsq11,rinv11;
  93     real         c6,c12;
  94     real         weight_cg1, weight_cg2, weight_product;
  95     real         hybscal;
  96
  97     nri              = *p_nri;
  98     ntype            = *p_ntype;
  99     nthreads         = *p_nthreads;
 100     facel            = *p_facel;
 101     krf              = *p_krf;
 102     crf              = *p_crf;
 103     tabscale         = *p_tabscale;
 104     nouter           = 0;
 105     ninner           = 0;
 106
 107     do
 108     {
 109         #ifdef GMX_THREAD_SHM_FDECOMP
 110         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 111         nn0              = *count;
 112         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 113         *count           = nn1;
 114         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 115         if(nn1>nri) nn1=nri;
 116         #else
 117         nn0 = 0;
 118         nn1 = nri;
 119         #endif
 120
 121         for(n=nn0; (n<nn1); n++)
 122         {
 123             is3              = 3*shift[n];
 124             shX              = shiftvec[is3];
 125             shY              = shiftvec[is3+1];
 126             shZ              = shiftvec[is3+2];
 127             nj0              = jindex[n];
 128             nj1              = jindex[n+1];
 129             ii               = iinr[n];
 130             ii3              = 3*ii;
 131             ix1              = shX + pos[ii3+0];
 132             iy1              = shY + pos[ii3+1];
 133             iz1              = shZ + pos[ii3+2];
 134             iq               = facel*charge[ii];
 135             nti              = 2*ntype*type[ii];
 136             weight_cg1       = wf[ii];
 137             vctot            = 0;
 138             Vvdwtot          = 0;
 139             fix1             = 0;
 140             fiy1             = 0;
 141             fiz1             = 0;
 142
 143             for(k=nj0; (k<nj1); k++)
 144             {
 145                 jnr              = jjnr[k];
 146                 weight_cg2       = wf[jnr];
 147                 weight_product   = weight_cg1*weight_cg2;
 148                 if (weight_product < ALMOST_ZERO) {
 149                        hybscal = 1.0;
 150                 }
 151                 else if (weight_product >= ALMOST_ONE)
 152                 {
 153                   /* force is zero, skip this molecule */
 154                        continue;
 155                 }
 156                 else
 157                 {
 158                    hybscal = 1.0 - weight_product;
 159                 }
 160                 j3               = 3*jnr;
 161                 jx1              = pos[j3+0];
 162                 jy1              = pos[j3+1];
 163                 jz1              = pos[j3+2];
 164                 dx11             = ix1 - jx1;
 165                 dy11             = iy1 - jy1;
 166                 dz11             = iz1 - jz1;
 167                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 168                 rinv11           = 1.0/sqrt(rsq11);
 169                 qq               = iq*charge[jnr];
 170                 tj               = nti+2*type[jnr];
 171                 c6               = vdwparam[tj];
 172                 c12              = vdwparam[tj+1];
 173                 r                = rsq11*rinv11;
 174                 rt               = r*tabscale;
 175                 n0               = rt;
 176                 eps              = rt-n0;
 177                 eps2             = eps*eps;
 178                 nnn              = 12*n0;
 179                 Y                = VFtab[nnn];
 180                 F                = VFtab[nnn+1];
 181                 Geps             = eps*VFtab[nnn+2];
 182                 Heps2            = eps2*VFtab[nnn+3];
 183                 Fp               = F+Geps+Heps2;
 184                 VV               = Y+eps*Fp;
 185                 FF               = Fp+Geps+2.0*Heps2;
 186                 vcoul            = qq*VV;
 187                 fijC             = qq*FF;
 188                 vctot            = vctot + vcoul;
 189                 nnn              = nnn+4;
 190                 Y                = VFtab[nnn];
 191                 F                = VFtab[nnn+1];
 192                 Geps             = eps*VFtab[nnn+2];
 193                 Heps2            = eps2*VFtab[nnn+3];
 194                 Fp               = F+Geps+Heps2;
 195                 VV               = Y+eps*Fp;
 196                 FF               = Fp+Geps+2.0*Heps2;
 197                 Vvdw6            = c6*VV;
 198                 fijD             = c6*FF;
 199                 nnn              = nnn+4;
 200                 Y                = VFtab[nnn];
 201                 F                = VFtab[nnn+1];
 202                 Geps             = eps*VFtab[nnn+2];
 203                 Heps2            = eps2*VFtab[nnn+3];
 204                 Fp               = F+Geps+Heps2;
 205                 VV               = Y+eps*Fp;
 206                 FF               = Fp+Geps+2.0*Heps2;
 207                 Vvdw12           = c12*VV;
 208                 fijR             = c12*FF;
 209                 Vvdwtot          = Vvdwtot+ Vvdw6 + Vvdw12;
 210                 fscal            = -((fijC+fijD+fijR)*tabscale)*rinv11;
 211                 fscal *= hybscal;
 212                 tx               = fscal*dx11;
 213                 ty               = fscal*dy11;
 214                 tz               = fscal*dz11;
 215                 fix1             = fix1 + tx;
 216                 fiy1             = fiy1 + ty;
 217                 fiz1             = fiz1 + tz;
 218                 faction[j3+0]    = faction[j3+0] - tx;
 219                 faction[j3+1]    = faction[j3+1] - ty;
 220                 faction[j3+2]    = faction[j3+2] - tz;
 221             }
 222
 223             faction[ii3+0]   = faction[ii3+0] + fix1;
 224             faction[ii3+1]   = faction[ii3+1] + fiy1;
 225             faction[ii3+2]   = faction[ii3+2] + fiz1;
 226             fshift[is3]      = fshift[is3]+fix1;
 227             fshift[is3+1]    = fshift[is3+1]+fiy1;
 228             fshift[is3+2]    = fshift[is3+2]+fiz1;
 229             ggid             = gid[n];
 230             Vc[ggid]         = Vc[ggid] + vctot;
 231             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 232             ninner           = ninner + nj1 - nj0;
 233         }
 234
 235         nouter           = nouter + nn1 - nn0;
 236     }
 237     while (nn1<nri);
 238
 239     *outeriter       = nouter;
 240     *inneriter       = ninner;
 241 }
 242
 243
 244
 245
 246
 247 /*
 248  * Gromacs nonbonded kernel nb_kernel330_adress_ex
 249  * Coulomb interaction:     Tabulated
 250  * VdW interaction:         Tabulated
 251  * water optimization:      No
 252  * Calculate forces:        yes
 253  */
 254 void nb_kernel330_adress_ex(
 255                     int *           p_nri,
 256                     int *           iinr,
 257                     int *           jindex,
 258                     int *           jjnr,
 259                     int *           shift,
 260                     real *         shiftvec,
 261                     real *         fshift,
 262                     int *           gid,
 263                     real *         pos,
 264                     real *         faction,
 265                     real *         charge,
 266                     real *         p_facel,
 267                     real *         p_krf,
 268                     real *         p_crf,
 269                     real *         Vc,
 270                     int *           type,
 271                     int *           p_ntype,
 272                     real *         vdwparam,
 273                     real *         Vvdw,
 274                     real *         p_tabscale,
 275                     real *         VFtab,
 276                     real *         invsqrta,
 277                     real *         dvda,
 278                     real *         p_gbtabscale,
 279                     real *         GBtab,
 280                     int *           p_nthreads,
 281                     int *           count,
 282                     void *          mtx,
 283                     int *           outeriter,
 284                     int *           inneriter,
 285                     real           force_cap,
 286                     real *         wf)
 287 {
 288     int           nri,ntype,nthreads;
 289     real         facel,krf,crf,tabscale,gbtabscale;
 290     int           n,ii,is3,ii3,k,nj0,nj1,jnr,j3,ggid;
 291     int           nn0,nn1,nouter,ninner;
 292     real         shX,shY,shZ;
 293     real         fscal,tx,ty,tz;
 294     real         iq;
 295     real         qq,vcoul,vctot;
 296     int           nti;
 297     int           tj;
 298     real         Vvdw6,Vvdwtot;
 299     real         Vvdw12;
 300     real         r,rt,eps,eps2;
 301     int           n0,nnn;
 302     real         Y,F,Geps,Heps2,Fp,VV;
 303     real         FF;
 304     real         fijC;
 305     real         fijD,fijR;
 306     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
 307     real         jx1,jy1,jz1;
 308     real         dx11,dy11,dz11,rsq11,rinv11;
 309     real         c6,c12;
 310     real         weight_cg1, weight_cg2, weight_product;
 311     real         hybscal;
 312
 313     nri              = *p_nri;
 314     ntype            = *p_ntype;
 315     nthreads         = *p_nthreads;
 316     facel            = *p_facel;
 317     krf              = *p_krf;
 318     crf              = *p_crf;
 319     tabscale         = *p_tabscale;
 320     nouter           = 0;
 321     ninner           = 0;
 322
 323     do
 324     {
 325         #ifdef GMX_THREAD_SHM_FDECOMP
 326         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 327         nn0              = *count;
 328         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 329         *count           = nn1;
 330         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 331         if(nn1>nri) nn1=nri;
 332         #else
 333         nn0 = 0;
 334         nn1 = nri;
 335         #endif
 336
 337         for(n=nn0; (n<nn1); n++)
 338         {
 339             is3              = 3*shift[n];
 340             shX              = shiftvec[is3];
 341             shY              = shiftvec[is3+1];
 342             shZ              = shiftvec[is3+2];
 343             nj0              = jindex[n];
 344             nj1              = jindex[n+1];
 345             ii               = iinr[n];
 346             ii3              = 3*ii;
 347             ix1              = shX + pos[ii3+0];
 348             iy1              = shY + pos[ii3+1];
 349             iz1              = shZ + pos[ii3+2];
 350             iq               = facel*charge[ii];
 351             nti              = 2*ntype*type[ii];
 352             weight_cg1       = wf[ii];
 353             vctot            = 0;
 354             Vvdwtot          = 0;
 355             fix1             = 0;
 356             fiy1             = 0;
 357             fiz1             = 0;
 358
 359             for(k=nj0; (k<nj1); k++)
 360             {
 361                 jnr              = jjnr[k];
 362                 weight_cg2       = wf[jnr];
 363                 weight_product   = weight_cg1*weight_cg2;
 364                 if (weight_product < ALMOST_ZERO) {
 365                 /* force is zero, skip this molecule */
 366                  continue;
 367                 }
 368                 else if (weight_product >= ALMOST_ONE)
 369                 {
 370                        hybscal = 1.0;
 371                 }
 372                 else
 373                 {
 374                    hybscal = weight_product;
 375                 }
 376                 j3               = 3*jnr;
 377                 jx1              = pos[j3+0];
 378                 jy1              = pos[j3+1];
 379                 jz1              = pos[j3+2];
 380                 dx11             = ix1 - jx1;
 381                 dy11             = iy1 - jy1;
 382                 dz11             = iz1 - jz1;
 383                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 384                 rinv11           = 1.0/sqrt(rsq11);
 385                 qq               = iq*charge[jnr];
 386                 tj               = nti+2*type[jnr];
 387                 c6               = vdwparam[tj];
 388                 c12              = vdwparam[tj+1];
 389                 r                = rsq11*rinv11;
 390                 rt               = r*tabscale;
 391                 n0               = rt;
 392                 eps              = rt-n0;
 393                 eps2             = eps*eps;
 394                 nnn              = 12*n0;
 395                 Y                = VFtab[nnn];
 396                 F                = VFtab[nnn+1];
 397                 Geps             = eps*VFtab[nnn+2];
 398                 Heps2            = eps2*VFtab[nnn+3];
 399                 Fp               = F+Geps+Heps2;
 400                 VV               = Y+eps*Fp;
 401                 FF               = Fp+Geps+2.0*Heps2;
 402                 vcoul            = qq*VV;
 403                 fijC             = qq*FF;
 404                 vctot            = vctot + vcoul;
 405                 nnn              = nnn+4;
 406                 Y                = VFtab[nnn];
 407                 F                = VFtab[nnn+1];
 408                 Geps             = eps*VFtab[nnn+2];
 409                 Heps2            = eps2*VFtab[nnn+3];
 410                 Fp               = F+Geps+Heps2;
 411                 VV               = Y+eps*Fp;
 412                 FF               = Fp+Geps+2.0*Heps2;
 413                 Vvdw6            = c6*VV;
 414                 fijD             = c6*FF;
 415                 nnn              = nnn+4;
 416                 Y                = VFtab[nnn];
 417                 F                = VFtab[nnn+1];
 418                 Geps             = eps*VFtab[nnn+2];
 419                 Heps2            = eps2*VFtab[nnn+3];
 420                 Fp               = F+Geps+Heps2;
 421                 VV               = Y+eps*Fp;
 422                 FF               = Fp+Geps+2.0*Heps2;
 423                 Vvdw12           = c12*VV;
 424                 fijR             = c12*FF;
 425                 Vvdwtot          = Vvdwtot+ Vvdw6 + Vvdw12;
 426                 fscal            = -((fijC+fijD+fijR)*tabscale)*rinv11;
 427                 fscal *= hybscal;
 428                 if(force_cap>0 && (fabs(fscal)> force_cap)){
 429                 fscal=force_cap*fscal/fabs(fscal);
 430                 }
 431                 tx               = fscal*dx11;
 432                 ty               = fscal*dy11;
 433                 tz               = fscal*dz11;
 434                 fix1             = fix1 + tx;
 435                 fiy1             = fiy1 + ty;
 436                 fiz1             = fiz1 + tz;
 437                 faction[j3+0]    = faction[j3+0] - tx;
 438                 faction[j3+1]    = faction[j3+1] - ty;
 439                 faction[j3+2]    = faction[j3+2] - tz;
 440             }
 441
 442             faction[ii3+0]   = faction[ii3+0] + fix1;
 443             faction[ii3+1]   = faction[ii3+1] + fiy1;
 444             faction[ii3+2]   = faction[ii3+2] + fiz1;
 445             fshift[is3]      = fshift[is3]+fix1;
 446             fshift[is3+1]    = fshift[is3+1]+fiy1;
 447             fshift[is3+2]    = fshift[is3+2]+fiz1;
 448             ggid             = gid[n];
 449             Vc[ggid]         = Vc[ggid] + vctot;
 450             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 451             ninner           = ninner + nj1 - nj0;
 452         }
 453
 454         nouter           = nouter + nn1 - nn0;
 455     }
 456     while (nn1<nri);
 457
 458     *outeriter       = nouter;
 459     *inneriter       = ninner;
 460 }
 461
 462