src/gromacs/gmxlib/nonbonded/nb_kernel_adress_c/nb_kernel121_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_kernel121_adress.h"
  28
  29
  30
  31 /*
  32  * Gromacs nonbonded kernel nb_kernel121_adress_cg
  33  * Coulomb interaction:     Normal Coulomb
  34  * VdW interaction:         Buckingham
  35  * water optimization:      SPC/TIP3P - other atoms
  36  * Calculate forces:        yes
  37  */
  38 void nb_kernel121_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         jq;
  80     real         qq,vcoul,vctot;
  81     int           nti;
  82     int           tj;
  83     real         rinvsix;
  84     real         Vvdw6,Vvdwtot;
  85     real         Vvdwexp,br;
  86     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
  87     real         ix2,iy2,iz2,fix2,fiy2,fiz2;
  88     real         ix3,iy3,iz3,fix3,fiy3,fiz3;
  89     real         jx1,jy1,jz1,fjx1,fjy1,fjz1;
  90     real         dx11,dy11,dz11,rsq11,rinv11;
  91     real         dx21,dy21,dz21,rsq21,rinv21;
  92     real         dx31,dy31,dz31,rsq31,rinv31;
  93     real         qO,qH;
  94     real         c6,cexp1,cexp2;
  95     real         weight_cg1, weight_cg2, weight_product;
  96     real         hybscal;
  97
  98     nri              = *p_nri;
  99     ntype            = *p_ntype;
 100     nthreads         = *p_nthreads;
 101     facel            = *p_facel;
 102     krf              = *p_krf;
 103     crf              = *p_crf;
 104     tabscale         = *p_tabscale;
 105     ii               = iinr[0];
 106     qO               = facel*charge[ii];
 107     qH               = facel*charge[ii+1];
 108     nti              = 3*ntype*type[ii];
 109
 110     nouter           = 0;
 111     ninner           = 0;
 112
 113     do
 114     {
 115         #ifdef GMX_THREAD_SHM_FDECOMP
 116         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 117         nn0              = *count;
 118         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 119         *count           = nn1;
 120         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 121         if(nn1>nri) nn1=nri;
 122         #else
 123         nn0 = 0;
 124         nn1 = nri;
 125         #endif
 126
 127         for(n=nn0; (n<nn1); n++)
 128         {
 129             is3              = 3*shift[n];
 130             shX              = shiftvec[is3];
 131             shY              = shiftvec[is3+1];
 132             shZ              = shiftvec[is3+2];
 133             nj0              = jindex[n];
 134             nj1              = jindex[n+1];
 135             ii               = iinr[n];
 136             ii3              = 3*ii;
 137             ix1              = shX + pos[ii3+0];
 138             iy1              = shY + pos[ii3+1];
 139             iz1              = shZ + pos[ii3+2];
 140             ix2              = shX + pos[ii3+3];
 141             iy2              = shY + pos[ii3+4];
 142             iz2              = shZ + pos[ii3+5];
 143             ix3              = shX + pos[ii3+6];
 144             iy3              = shY + pos[ii3+7];
 145             iz3              = shZ + pos[ii3+8];
 146             weight_cg1       = wf[ii];
 147             vctot            = 0;
 148             Vvdwtot          = 0;
 149             fix1             = 0;
 150             fiy1             = 0;
 151             fiz1             = 0;
 152             fix2             = 0;
 153             fiy2             = 0;
 154             fiz2             = 0;
 155             fix3             = 0;
 156             fiy3             = 0;
 157             fiz3             = 0;
 158
 159             for(k=nj0; (k<nj1); k++)
 160             {
 161                 jnr              = jjnr[k];
 162                 weight_cg2       = wf[jnr];
 163                 weight_product   = weight_cg1*weight_cg2;
 164                 if (weight_product < ALMOST_ZERO) {
 165                        hybscal = 1.0;
 166                 }
 167                 else if (weight_product >= ALMOST_ONE)
 168                 {
 169                   /* force is zero, skip this molecule */
 170                        continue;
 171                 }
 172                 else
 173                 {
 174                    hybscal = 1.0 - weight_product;
 175                 }
 176                 j3               = 3*jnr;
 177                 jx1              = pos[j3+0];
 178                 jy1              = pos[j3+1];
 179                 jz1              = pos[j3+2];
 180                 dx11             = ix1 - jx1;
 181                 dy11             = iy1 - jy1;
 182                 dz11             = iz1 - jz1;
 183                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 184                 dx21             = ix2 - jx1;
 185                 dy21             = iy2 - jy1;
 186                 dz21             = iz2 - jz1;
 187                 rsq21            = dx21*dx21+dy21*dy21+dz21*dz21;
 188                 dx31             = ix3 - jx1;
 189                 dy31             = iy3 - jy1;
 190                 dz31             = iz3 - jz1;
 191                 rsq31            = dx31*dx31+dy31*dy31+dz31*dz31;
 192                 rinv11           = 1.0/sqrt(rsq11);
 193                 rinv21           = 1.0/sqrt(rsq21);
 194                 rinv31           = 1.0/sqrt(rsq31);
 195                 jq               = charge[jnr+0];
 196                 qq               = qO*jq;
 197                 tj               = nti+3*type[jnr];
 198                 c6               = vdwparam[tj];
 199                 cexp1            = vdwparam[tj+1];
 200                 cexp2            = vdwparam[tj+2];
 201                 rinvsq           = rinv11*rinv11;
 202                 vcoul            = qq*rinv11;
 203                 vctot            = vctot+vcoul;
 204                 rinvsix          = rinvsq*rinvsq*rinvsq;
 205                 Vvdw6            = c6*rinvsix;
 206                 br               = cexp2*rsq11*rinv11;
 207                 Vvdwexp          = cexp1*exp(-br);
 208                 Vvdwtot          = Vvdwtot+Vvdwexp-Vvdw6;
 209                 fscal            = (vcoul+br*Vvdwexp-6.0*Vvdw6)*rinvsq;
 210                 fscal *= hybscal;
 211                 tx               = fscal*dx11;
 212                 ty               = fscal*dy11;
 213                 tz               = fscal*dz11;
 214                 fix1             = fix1 + tx;
 215                 fiy1             = fiy1 + ty;
 216                 fiz1             = fiz1 + tz;
 217                 fjx1             = faction[j3+0] - tx;
 218                 fjy1             = faction[j3+1] - ty;
 219                 fjz1             = faction[j3+2] - tz;
 220                 qq               = qH*jq;
 221                 rinvsq           = rinv21*rinv21;
 222                 vcoul            = qq*rinv21;
 223                 vctot            = vctot+vcoul;
 224                 fscal            = (vcoul)*rinvsq;
 225                 fscal *= hybscal;
 226                 tx               = fscal*dx21;
 227                 ty               = fscal*dy21;
 228                 tz               = fscal*dz21;
 229                 fix2             = fix2 + tx;
 230                 fiy2             = fiy2 + ty;
 231                 fiz2             = fiz2 + tz;
 232                 fjx1             = fjx1 - tx;
 233                 fjy1             = fjy1 - ty;
 234                 fjz1             = fjz1 - tz;
 235                 rinvsq           = rinv31*rinv31;
 236                 vcoul            = qq*rinv31;
 237                 vctot            = vctot+vcoul;
 238                 fscal            = (vcoul)*rinvsq;
 239                 fscal *= hybscal;
 240                 tx               = fscal*dx31;
 241                 ty               = fscal*dy31;
 242                 tz               = fscal*dz31;
 243                 fix3             = fix3 + tx;
 244                 fiy3             = fiy3 + ty;
 245                 fiz3             = fiz3 + tz;
 246                 faction[j3+0]    = fjx1 - tx;
 247                 faction[j3+1]    = fjy1 - ty;
 248                 faction[j3+2]    = fjz1 - tz;
 249             }
 250
 251             faction[ii3+0]   = faction[ii3+0] + fix1;
 252             faction[ii3+1]   = faction[ii3+1] + fiy1;
 253             faction[ii3+2]   = faction[ii3+2] + fiz1;
 254             faction[ii3+3]   = faction[ii3+3] + fix2;
 255             faction[ii3+4]   = faction[ii3+4] + fiy2;
 256             faction[ii3+5]   = faction[ii3+5] + fiz2;
 257             faction[ii3+6]   = faction[ii3+6] + fix3;
 258             faction[ii3+7]   = faction[ii3+7] + fiy3;
 259             faction[ii3+8]   = faction[ii3+8] + fiz3;
 260             fshift[is3]      = fshift[is3]+fix1+fix2+fix3;
 261             fshift[is3+1]    = fshift[is3+1]+fiy1+fiy2+fiy3;
 262             fshift[is3+2]    = fshift[is3+2]+fiz1+fiz2+fiz3;
 263             ggid             = gid[n];
 264             Vc[ggid]         = Vc[ggid] + vctot;
 265             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 266             ninner           = ninner + nj1 - nj0;
 267         }
 268
 269         nouter           = nouter + nn1 - nn0;
 270     }
 271     while (nn1<nri);
 272
 273     *outeriter       = nouter;
 274     *inneriter       = ninner;
 275 }
 276
 277
 278
 279
 280
 281 /*
 282  * Gromacs nonbonded kernel nb_kernel121_adress_ex
 283  * Coulomb interaction:     Normal Coulomb
 284  * VdW interaction:         Buckingham
 285  * water optimization:      SPC/TIP3P - other atoms
 286  * Calculate forces:        yes
 287  */
 288 void nb_kernel121_adress_ex(
 289                     int *           p_nri,
 290                     int *           iinr,
 291                     int *           jindex,
 292                     int *           jjnr,
 293                     int *           shift,
 294                     real *         shiftvec,
 295                     real *         fshift,
 296                     int *           gid,
 297                     real *         pos,
 298                     real *         faction,
 299                     real *         charge,
 300                     real *         p_facel,
 301                     real *         p_krf,
 302                     real *         p_crf,
 303                     real *         Vc,
 304                     int *           type,
 305                     int *           p_ntype,
 306                     real *         vdwparam,
 307                     real *         Vvdw,
 308                     real *         p_tabscale,
 309                     real *         VFtab,
 310                     real *         invsqrta,
 311                     real *         dvda,
 312                     real *         p_gbtabscale,
 313                     real *         GBtab,
 314                     int *           p_nthreads,
 315                     int *           count,
 316                     void *          mtx,
 317                     int *           outeriter,
 318                     int *           inneriter,
 319                     real           force_cap,
 320                     real *         wf)
 321 {
 322     int           nri,ntype,nthreads;
 323     real         facel,krf,crf,tabscale,gbtabscale;
 324     int           n,ii,is3,ii3,k,nj0,nj1,jnr,j3,ggid;
 325     int           nn0,nn1,nouter,ninner;
 326     real         shX,shY,shZ;
 327     real         fscal,tx,ty,tz;
 328     real         rinvsq;
 329     real         jq;
 330     real         qq,vcoul,vctot;
 331     int           nti;
 332     int           tj;
 333     real         rinvsix;
 334     real         Vvdw6,Vvdwtot;
 335     real         Vvdwexp,br;
 336     real         ix1,iy1,iz1,fix1,fiy1,fiz1;
 337     real         ix2,iy2,iz2,fix2,fiy2,fiz2;
 338     real         ix3,iy3,iz3,fix3,fiy3,fiz3;
 339     real         jx1,jy1,jz1,fjx1,fjy1,fjz1;
 340     real         dx11,dy11,dz11,rsq11,rinv11;
 341     real         dx21,dy21,dz21,rsq21,rinv21;
 342     real         dx31,dy31,dz31,rsq31,rinv31;
 343     real         qO,qH;
 344     real         c6,cexp1,cexp2;
 345     real         weight_cg1, weight_cg2, weight_product;
 346     real         hybscal;
 347
 348     nri              = *p_nri;
 349     ntype            = *p_ntype;
 350     nthreads         = *p_nthreads;
 351     facel            = *p_facel;
 352     krf              = *p_krf;
 353     crf              = *p_crf;
 354     tabscale         = *p_tabscale;
 355     ii               = iinr[0];
 356     qO               = facel*charge[ii];
 357     qH               = facel*charge[ii+1];
 358     nti              = 3*ntype*type[ii];
 359
 360     nouter           = 0;
 361     ninner           = 0;
 362
 363     do
 364     {
 365         #ifdef GMX_THREAD_SHM_FDECOMP
 366         tMPI_Thread_mutex_lock((tMPI_Thread_mutex_t *)mtx);
 367         nn0              = *count;
 368         nn1              = nn0+(nri-nn0)/(2*nthreads)+10;
 369         *count           = nn1;
 370         tMPI_Thread_mutex_unlock((tMPI_Thread_mutex_t *)mtx);
 371         if(nn1>nri) nn1=nri;
 372         #else
 373         nn0 = 0;
 374         nn1 = nri;
 375         #endif
 376
 377         for(n=nn0; (n<nn1); n++)
 378         {
 379             is3              = 3*shift[n];
 380             shX              = shiftvec[is3];
 381             shY              = shiftvec[is3+1];
 382             shZ              = shiftvec[is3+2];
 383             nj0              = jindex[n];
 384             nj1              = jindex[n+1];
 385             ii               = iinr[n];
 386             ii3              = 3*ii;
 387             ix1              = shX + pos[ii3+0];
 388             iy1              = shY + pos[ii3+1];
 389             iz1              = shZ + pos[ii3+2];
 390             ix2              = shX + pos[ii3+3];
 391             iy2              = shY + pos[ii3+4];
 392             iz2              = shZ + pos[ii3+5];
 393             ix3              = shX + pos[ii3+6];
 394             iy3              = shY + pos[ii3+7];
 395             iz3              = shZ + pos[ii3+8];
 396             weight_cg1       = wf[ii];
 397             vctot            = 0;
 398             Vvdwtot          = 0;
 399             fix1             = 0;
 400             fiy1             = 0;
 401             fiz1             = 0;
 402             fix2             = 0;
 403             fiy2             = 0;
 404             fiz2             = 0;
 405             fix3             = 0;
 406             fiy3             = 0;
 407             fiz3             = 0;
 408
 409             for(k=nj0; (k<nj1); k++)
 410             {
 411                 jnr              = jjnr[k];
 412                 weight_cg2       = wf[jnr];
 413                 weight_product   = weight_cg1*weight_cg2;
 414                 if (weight_product < ALMOST_ZERO) {
 415                 /* force is zero, skip this molecule */
 416                  continue;
 417                 }
 418                 else if (weight_product >= ALMOST_ONE)
 419                 {
 420                        hybscal = 1.0;
 421                 }
 422                 else
 423                 {
 424                    hybscal = weight_product;
 425                 }
 426                 j3               = 3*jnr;
 427                 jx1              = pos[j3+0];
 428                 jy1              = pos[j3+1];
 429                 jz1              = pos[j3+2];
 430                 dx11             = ix1 - jx1;
 431                 dy11             = iy1 - jy1;
 432                 dz11             = iz1 - jz1;
 433                 rsq11            = dx11*dx11+dy11*dy11+dz11*dz11;
 434                 dx21             = ix2 - jx1;
 435                 dy21             = iy2 - jy1;
 436                 dz21             = iz2 - jz1;
 437                 rsq21            = dx21*dx21+dy21*dy21+dz21*dz21;
 438                 dx31             = ix3 - jx1;
 439                 dy31             = iy3 - jy1;
 440                 dz31             = iz3 - jz1;
 441                 rsq31            = dx31*dx31+dy31*dy31+dz31*dz31;
 442                 rinv11           = 1.0/sqrt(rsq11);
 443                 rinv21           = 1.0/sqrt(rsq21);
 444                 rinv31           = 1.0/sqrt(rsq31);
 445                 jq               = charge[jnr+0];
 446                 qq               = qO*jq;
 447                 tj               = nti+3*type[jnr];
 448                 c6               = vdwparam[tj];
 449                 cexp1            = vdwparam[tj+1];
 450                 cexp2            = vdwparam[tj+2];
 451                 rinvsq           = rinv11*rinv11;
 452                 vcoul            = qq*rinv11;
 453                 vctot            = vctot+vcoul;
 454                 rinvsix          = rinvsq*rinvsq*rinvsq;
 455                 Vvdw6            = c6*rinvsix;
 456                 br               = cexp2*rsq11*rinv11;
 457                 Vvdwexp          = cexp1*exp(-br);
 458                 Vvdwtot          = Vvdwtot+Vvdwexp-Vvdw6;
 459                 fscal            = (vcoul+br*Vvdwexp-6.0*Vvdw6)*rinvsq;
 460                 fscal *= hybscal;
 461                 if(force_cap>0 && (fabs(fscal)> force_cap)){
 462                 fscal=force_cap*fscal/fabs(fscal);
 463                 }
 464                 tx               = fscal*dx11;
 465                 ty               = fscal*dy11;
 466                 tz               = fscal*dz11;
 467                 fix1             = fix1 + tx;
 468                 fiy1             = fiy1 + ty;
 469                 fiz1             = fiz1 + tz;
 470                 fjx1             = faction[j3+0] - tx;
 471                 fjy1             = faction[j3+1] - ty;
 472                 fjz1             = faction[j3+2] - tz;
 473                 qq               = qH*jq;
 474                 rinvsq           = rinv21*rinv21;
 475                 vcoul            = qq*rinv21;
 476                 vctot            = vctot+vcoul;
 477                 fscal            = (vcoul)*rinvsq;
 478                 fscal *= hybscal;
 479                 if(force_cap>0 && (fabs(fscal)> force_cap)){
 480                 fscal=force_cap*fscal/fabs(fscal);
 481                 }
 482                 tx               = fscal*dx21;
 483                 ty               = fscal*dy21;
 484                 tz               = fscal*dz21;
 485                 fix2             = fix2 + tx;
 486                 fiy2             = fiy2 + ty;
 487                 fiz2             = fiz2 + tz;
 488                 fjx1             = fjx1 - tx;
 489                 fjy1             = fjy1 - ty;
 490                 fjz1             = fjz1 - tz;
 491                 rinvsq           = rinv31*rinv31;
 492                 vcoul            = qq*rinv31;
 493                 vctot            = vctot+vcoul;
 494                 fscal            = (vcoul)*rinvsq;
 495                 fscal *= hybscal;
 496                 if(force_cap>0 && (fabs(fscal)> force_cap)){
 497                 fscal=force_cap*fscal/fabs(fscal);
 498                 }
 499                 tx               = fscal*dx31;
 500                 ty               = fscal*dy31;
 501                 tz               = fscal*dz31;
 502                 fix3             = fix3 + tx;
 503                 fiy3             = fiy3 + ty;
 504                 fiz3             = fiz3 + tz;
 505                 faction[j3+0]    = fjx1 - tx;
 506                 faction[j3+1]    = fjy1 - ty;
 507                 faction[j3+2]    = fjz1 - tz;
 508             }
 509
 510             faction[ii3+0]   = faction[ii3+0] + fix1;
 511             faction[ii3+1]   = faction[ii3+1] + fiy1;
 512             faction[ii3+2]   = faction[ii3+2] + fiz1;
 513             faction[ii3+3]   = faction[ii3+3] + fix2;
 514             faction[ii3+4]   = faction[ii3+4] + fiy2;
 515             faction[ii3+5]   = faction[ii3+5] + fiz2;
 516             faction[ii3+6]   = faction[ii3+6] + fix3;
 517             faction[ii3+7]   = faction[ii3+7] + fiy3;
 518             faction[ii3+8]   = faction[ii3+8] + fiz3;
 519             fshift[is3]      = fshift[is3]+fix1+fix2+fix3;
 520             fshift[is3+1]    = fshift[is3+1]+fiy1+fiy2+fiy3;
 521             fshift[is3+2]    = fshift[is3+2]+fiz1+fiz2+fiz3;
 522             ggid             = gid[n];
 523             Vc[ggid]         = Vc[ggid] + vctot;
 524             Vvdw[ggid]       = Vvdw[ggid] + Vvdwtot;
 525             ninner           = ninner + nj1 - nj0;
 526         }
 527
 528         nouter           = nouter + nn1 - nn0;
 529     }
 530     while (nn1<nri);
 531
 532     *outeriter       = nouter;
 533     *inneriter       = ninner;
 534 }
 535
 536