src/gromacs/gmxlib/nonbonded/nb_kernel_c/nb_kernel_ElecRF_VdwLJ_GeomP1P1_c.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS c kernel generator.
  37  */
  38 #include "gmxpre.h"
  39
  40 #include "config.h"
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "gromacs/legacyheaders/types/simple.h"
  46 #include "gromacs/math/vec.h"
  47 #include "gromacs/legacyheaders/nrnb.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwLJ_GeomP1P1_VF_c
  51  * Electrostatics interaction: ReactionField
  52  * VdW interaction:            LennardJones
  53  * Geometry:                   Particle-Particle
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecRF_VdwLJ_GeomP1P1_VF_c
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      t_forcerec                  * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     int              i_shift_offset,i_coord_offset,j_coord_offset;
  67     int              j_index_start,j_index_end;
  68     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
  69     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
  70     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  71     real             *shiftvec,*fshift,*x,*f;
  72     int              vdwioffset0;
  73     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  74     int              vdwjidx0;
  75     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  76     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
  77     real             velec,felec,velecsum,facel,crf,krf,krf2;
  78     real             *charge;
  79     int              nvdwtype;
  80     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
  81     int              *vdwtype;
  82     real             *vdwparam;
  83
  84     x                = xx[0];
  85     f                = ff[0];
  86
  87     nri              = nlist->nri;
  88     iinr             = nlist->iinr;
  89     jindex           = nlist->jindex;
  90     jjnr             = nlist->jjnr;
  91     shiftidx         = nlist->shift;
  92     gid              = nlist->gid;
  93     shiftvec         = fr->shift_vec[0];
  94     fshift           = fr->fshift[0];
  95     facel            = fr->epsfac;
  96     charge           = mdatoms->chargeA;
  97     krf              = fr->ic->k_rf;
  98     krf2             = krf*2.0;
  99     crf              = fr->ic->c_rf;
 100     nvdwtype         = fr->ntype;
 101     vdwparam         = fr->nbfp;
 102     vdwtype          = mdatoms->typeA;
 103
 104     outeriter        = 0;
 105     inneriter        = 0;
 106
 107     /* Start outer loop over neighborlists */
 108     for(iidx=0; iidx<nri; iidx++)
 109     {
 110         /* Load shift vector for this list */
 111         i_shift_offset   = DIM*shiftidx[iidx];
 112         shX              = shiftvec[i_shift_offset+XX];
 113         shY              = shiftvec[i_shift_offset+YY];
 114         shZ              = shiftvec[i_shift_offset+ZZ];
 115
 116         /* Load limits for loop over neighbors */
 117         j_index_start    = jindex[iidx];
 118         j_index_end      = jindex[iidx+1];
 119
 120         /* Get outer coordinate index */
 121         inr              = iinr[iidx];
 122         i_coord_offset   = DIM*inr;
 123
 124         /* Load i particle coords and add shift vector */
 125         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 126         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 127         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 128
 129         fix0             = 0.0;
 130         fiy0             = 0.0;
 131         fiz0             = 0.0;
 132
 133         /* Load parameters for i particles */
 134         iq0              = facel*charge[inr+0];
 135         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 136
 137         /* Reset potential sums */
 138         velecsum         = 0.0;
 139         vvdwsum          = 0.0;
 140
 141         /* Start inner kernel loop */
 142         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 143         {
 144             /* Get j neighbor index, and coordinate index */
 145             jnr              = jjnr[jidx];
 146             j_coord_offset   = DIM*jnr;
 147
 148             /* load j atom coordinates */
 149             jx0              = x[j_coord_offset+DIM*0+XX];
 150             jy0              = x[j_coord_offset+DIM*0+YY];
 151             jz0              = x[j_coord_offset+DIM*0+ZZ];
 152
 153             /* Calculate displacement vector */
 154             dx00             = ix0 - jx0;
 155             dy00             = iy0 - jy0;
 156             dz00             = iz0 - jz0;
 157
 158             /* Calculate squared distance and things based on it */
 159             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 160
 161             rinv00           = gmx_invsqrt(rsq00);
 162
 163             rinvsq00         = rinv00*rinv00;
 164
 165             /* Load parameters for j particles */
 166             jq0              = charge[jnr+0];
 167             vdwjidx0         = 2*vdwtype[jnr+0];
 168
 169             /**************************
 170              * CALCULATE INTERACTIONS *
 171              **************************/
 172
 173             qq00             = iq0*jq0;
 174             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 175             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 176
 177             /* REACTION-FIELD ELECTROSTATICS */
 178             velec            = qq00*(rinv00+krf*rsq00-crf);
 179             felec            = qq00*(rinv00*rinvsq00-krf2);
 180
 181             /* LENNARD-JONES DISPERSION/REPULSION */
 182
 183             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
 184             vvdw6            = c6_00*rinvsix;
 185             vvdw12           = c12_00*rinvsix*rinvsix;
 186             vvdw             = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
 187             fvdw             = (vvdw12-vvdw6)*rinvsq00;
 188
 189             /* Update potential sums from outer loop */
 190             velecsum        += velec;
 191             vvdwsum         += vvdw;
 192
 193             fscal            = felec+fvdw;
 194
 195             /* Calculate temporary vectorial force */
 196             tx               = fscal*dx00;
 197             ty               = fscal*dy00;
 198             tz               = fscal*dz00;
 199
 200             /* Update vectorial force */
 201             fix0            += tx;
 202             fiy0            += ty;
 203             fiz0            += tz;
 204             f[j_coord_offset+DIM*0+XX] -= tx;
 205             f[j_coord_offset+DIM*0+YY] -= ty;
 206             f[j_coord_offset+DIM*0+ZZ] -= tz;
 207
 208             /* Inner loop uses 44 flops */
 209         }
 210         /* End of innermost loop */
 211
 212         tx = ty = tz = 0;
 213         f[i_coord_offset+DIM*0+XX] += fix0;
 214         f[i_coord_offset+DIM*0+YY] += fiy0;
 215         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 216         tx                         += fix0;
 217         ty                         += fiy0;
 218         tz                         += fiz0;
 219         fshift[i_shift_offset+XX]  += tx;
 220         fshift[i_shift_offset+YY]  += ty;
 221         fshift[i_shift_offset+ZZ]  += tz;
 222
 223         ggid                        = gid[iidx];
 224         /* Update potential energies */
 225         kernel_data->energygrp_elec[ggid] += velecsum;
 226         kernel_data->energygrp_vdw[ggid] += vvdwsum;
 227
 228         /* Increment number of inner iterations */
 229         inneriter                  += j_index_end - j_index_start;
 230
 231         /* Outer loop uses 15 flops */
 232     }
 233
 234     /* Increment number of outer iterations */
 235     outeriter        += nri;
 236
 237     /* Update outer/inner flops */
 238
 239     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*44);
 240 }
 241 /*
 242  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwLJ_GeomP1P1_F_c
 243  * Electrostatics interaction: ReactionField
 244  * VdW interaction:            LennardJones
 245  * Geometry:                   Particle-Particle
 246  * Calculate force/pot:        Force
 247  */
 248 void
 249 nb_kernel_ElecRF_VdwLJ_GeomP1P1_F_c
 250                     (t_nblist                    * gmx_restrict       nlist,
 251                      rvec                        * gmx_restrict          xx,
 252                      rvec                        * gmx_restrict          ff,
 253                      t_forcerec                  * gmx_restrict          fr,
 254                      t_mdatoms                   * gmx_restrict     mdatoms,
 255                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 256                      t_nrnb                      * gmx_restrict        nrnb)
 257 {
 258     int              i_shift_offset,i_coord_offset,j_coord_offset;
 259     int              j_index_start,j_index_end;
 260     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
 261     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
 262     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 263     real             *shiftvec,*fshift,*x,*f;
 264     int              vdwioffset0;
 265     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 266     int              vdwjidx0;
 267     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 268     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
 269     real             velec,felec,velecsum,facel,crf,krf,krf2;
 270     real             *charge;
 271     int              nvdwtype;
 272     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
 273     int              *vdwtype;
 274     real             *vdwparam;
 275
 276     x                = xx[0];
 277     f                = ff[0];
 278
 279     nri              = nlist->nri;
 280     iinr             = nlist->iinr;
 281     jindex           = nlist->jindex;
 282     jjnr             = nlist->jjnr;
 283     shiftidx         = nlist->shift;
 284     gid              = nlist->gid;
 285     shiftvec         = fr->shift_vec[0];
 286     fshift           = fr->fshift[0];
 287     facel            = fr->epsfac;
 288     charge           = mdatoms->chargeA;
 289     krf              = fr->ic->k_rf;
 290     krf2             = krf*2.0;
 291     crf              = fr->ic->c_rf;
 292     nvdwtype         = fr->ntype;
 293     vdwparam         = fr->nbfp;
 294     vdwtype          = mdatoms->typeA;
 295
 296     outeriter        = 0;
 297     inneriter        = 0;
 298
 299     /* Start outer loop over neighborlists */
 300     for(iidx=0; iidx<nri; iidx++)
 301     {
 302         /* Load shift vector for this list */
 303         i_shift_offset   = DIM*shiftidx[iidx];
 304         shX              = shiftvec[i_shift_offset+XX];
 305         shY              = shiftvec[i_shift_offset+YY];
 306         shZ              = shiftvec[i_shift_offset+ZZ];
 307
 308         /* Load limits for loop over neighbors */
 309         j_index_start    = jindex[iidx];
 310         j_index_end      = jindex[iidx+1];
 311
 312         /* Get outer coordinate index */
 313         inr              = iinr[iidx];
 314         i_coord_offset   = DIM*inr;
 315
 316         /* Load i particle coords and add shift vector */
 317         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 318         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 319         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 320
 321         fix0             = 0.0;
 322         fiy0             = 0.0;
 323         fiz0             = 0.0;
 324
 325         /* Load parameters for i particles */
 326         iq0              = facel*charge[inr+0];
 327         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 328
 329         /* Start inner kernel loop */
 330         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 331         {
 332             /* Get j neighbor index, and coordinate index */
 333             jnr              = jjnr[jidx];
 334             j_coord_offset   = DIM*jnr;
 335
 336             /* load j atom coordinates */
 337             jx0              = x[j_coord_offset+DIM*0+XX];
 338             jy0              = x[j_coord_offset+DIM*0+YY];
 339             jz0              = x[j_coord_offset+DIM*0+ZZ];
 340
 341             /* Calculate displacement vector */
 342             dx00             = ix0 - jx0;
 343             dy00             = iy0 - jy0;
 344             dz00             = iz0 - jz0;
 345
 346             /* Calculate squared distance and things based on it */
 347             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 348
 349             rinv00           = gmx_invsqrt(rsq00);
 350
 351             rinvsq00         = rinv00*rinv00;
 352
 353             /* Load parameters for j particles */
 354             jq0              = charge[jnr+0];
 355             vdwjidx0         = 2*vdwtype[jnr+0];
 356
 357             /**************************
 358              * CALCULATE INTERACTIONS *
 359              **************************/
 360
 361             qq00             = iq0*jq0;
 362             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 363             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 364
 365             /* REACTION-FIELD ELECTROSTATICS */
 366             felec            = qq00*(rinv00*rinvsq00-krf2);
 367
 368             /* LENNARD-JONES DISPERSION/REPULSION */
 369
 370             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
 371             fvdw             = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
 372
 373             fscal            = felec+fvdw;
 374
 375             /* Calculate temporary vectorial force */
 376             tx               = fscal*dx00;
 377             ty               = fscal*dy00;
 378             tz               = fscal*dz00;
 379
 380             /* Update vectorial force */
 381             fix0            += tx;
 382             fiy0            += ty;
 383             fiz0            += tz;
 384             f[j_coord_offset+DIM*0+XX] -= tx;
 385             f[j_coord_offset+DIM*0+YY] -= ty;
 386             f[j_coord_offset+DIM*0+ZZ] -= tz;
 387
 388             /* Inner loop uses 34 flops */
 389         }
 390         /* End of innermost loop */
 391
 392         tx = ty = tz = 0;
 393         f[i_coord_offset+DIM*0+XX] += fix0;
 394         f[i_coord_offset+DIM*0+YY] += fiy0;
 395         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 396         tx                         += fix0;
 397         ty                         += fiy0;
 398         tz                         += fiz0;
 399         fshift[i_shift_offset+XX]  += tx;
 400         fshift[i_shift_offset+YY]  += ty;
 401         fshift[i_shift_offset+ZZ]  += tz;
 402
 403         /* Increment number of inner iterations */
 404         inneriter                  += j_index_end - j_index_start;
 405
 406         /* Outer loop uses 13 flops */
 407     }
 408
 409     /* Increment number of outer iterations */
 410     outeriter        += nri;
 411
 412     /* Update outer/inner flops */
 413
 414     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*34);
 415 }