src/gromacs/gmxlib/nonbonded/nb_kernel_c/nb_kernel_ElecCoul_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 "config.h"
  39
  40 #include <math.h>
  41
  42 #include "../nb_kernel.h"
  43 #include "gromacs/legacyheaders/types/simple.h"
  44 #include "gromacs/math/vec.h"
  45 #include "gromacs/legacyheaders/nrnb.h"
  46
  47 /*
  48  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwLJ_GeomP1P1_VF_c
  49  * Electrostatics interaction: Coulomb
  50  * VdW interaction:            LennardJones
  51  * Geometry:                   Particle-Particle
  52  * Calculate force/pot:        PotentialAndForce
  53  */
  54 void
  55 nb_kernel_ElecCoul_VdwLJ_GeomP1P1_VF_c
  56                     (t_nblist                    * gmx_restrict       nlist,
  57                      rvec                        * gmx_restrict          xx,
  58                      rvec                        * gmx_restrict          ff,
  59                      t_forcerec                  * gmx_restrict          fr,
  60                      t_mdatoms                   * gmx_restrict     mdatoms,
  61                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  62                      t_nrnb                      * gmx_restrict        nrnb)
  63 {
  64     int              i_shift_offset,i_coord_offset,j_coord_offset;
  65     int              j_index_start,j_index_end;
  66     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
  67     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
  68     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  69     real             *shiftvec,*fshift,*x,*f;
  70     int              vdwioffset0;
  71     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  72     int              vdwjidx0;
  73     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  74     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
  75     real             velec,felec,velecsum,facel,crf,krf,krf2;
  76     real             *charge;
  77     int              nvdwtype;
  78     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
  79     int              *vdwtype;
  80     real             *vdwparam;
  81
  82     x                = xx[0];
  83     f                = ff[0];
  84
  85     nri              = nlist->nri;
  86     iinr             = nlist->iinr;
  87     jindex           = nlist->jindex;
  88     jjnr             = nlist->jjnr;
  89     shiftidx         = nlist->shift;
  90     gid              = nlist->gid;
  91     shiftvec         = fr->shift_vec[0];
  92     fshift           = fr->fshift[0];
  93     facel            = fr->epsfac;
  94     charge           = mdatoms->chargeA;
  95     nvdwtype         = fr->ntype;
  96     vdwparam         = fr->nbfp;
  97     vdwtype          = mdatoms->typeA;
  98
  99     outeriter        = 0;
 100     inneriter        = 0;
 101
 102     /* Start outer loop over neighborlists */
 103     for(iidx=0; iidx<nri; iidx++)
 104     {
 105         /* Load shift vector for this list */
 106         i_shift_offset   = DIM*shiftidx[iidx];
 107         shX              = shiftvec[i_shift_offset+XX];
 108         shY              = shiftvec[i_shift_offset+YY];
 109         shZ              = shiftvec[i_shift_offset+ZZ];
 110
 111         /* Load limits for loop over neighbors */
 112         j_index_start    = jindex[iidx];
 113         j_index_end      = jindex[iidx+1];
 114
 115         /* Get outer coordinate index */
 116         inr              = iinr[iidx];
 117         i_coord_offset   = DIM*inr;
 118
 119         /* Load i particle coords and add shift vector */
 120         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 121         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 122         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 123
 124         fix0             = 0.0;
 125         fiy0             = 0.0;
 126         fiz0             = 0.0;
 127
 128         /* Load parameters for i particles */
 129         iq0              = facel*charge[inr+0];
 130         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 131
 132         /* Reset potential sums */
 133         velecsum         = 0.0;
 134         vvdwsum          = 0.0;
 135
 136         /* Start inner kernel loop */
 137         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 138         {
 139             /* Get j neighbor index, and coordinate index */
 140             jnr              = jjnr[jidx];
 141             j_coord_offset   = DIM*jnr;
 142
 143             /* load j atom coordinates */
 144             jx0              = x[j_coord_offset+DIM*0+XX];
 145             jy0              = x[j_coord_offset+DIM*0+YY];
 146             jz0              = x[j_coord_offset+DIM*0+ZZ];
 147
 148             /* Calculate displacement vector */
 149             dx00             = ix0 - jx0;
 150             dy00             = iy0 - jy0;
 151             dz00             = iz0 - jz0;
 152
 153             /* Calculate squared distance and things based on it */
 154             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 155
 156             rinv00           = gmx_invsqrt(rsq00);
 157
 158             rinvsq00         = rinv00*rinv00;
 159
 160             /* Load parameters for j particles */
 161             jq0              = charge[jnr+0];
 162             vdwjidx0         = 2*vdwtype[jnr+0];
 163
 164             /**************************
 165              * CALCULATE INTERACTIONS *
 166              **************************/
 167
 168             qq00             = iq0*jq0;
 169             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 170             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 171
 172             /* COULOMB ELECTROSTATICS */
 173             velec            = qq00*rinv00;
 174             felec            = velec*rinvsq00;
 175
 176             /* LENNARD-JONES DISPERSION/REPULSION */
 177
 178             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
 179             vvdw6            = c6_00*rinvsix;
 180             vvdw12           = c12_00*rinvsix*rinvsix;
 181             vvdw             = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
 182             fvdw             = (vvdw12-vvdw6)*rinvsq00;
 183
 184             /* Update potential sums from outer loop */
 185             velecsum        += velec;
 186             vvdwsum         += vvdw;
 187
 188             fscal            = felec+fvdw;
 189
 190             /* Calculate temporary vectorial force */
 191             tx               = fscal*dx00;
 192             ty               = fscal*dy00;
 193             tz               = fscal*dz00;
 194
 195             /* Update vectorial force */
 196             fix0            += tx;
 197             fiy0            += ty;
 198             fiz0            += tz;
 199             f[j_coord_offset+DIM*0+XX] -= tx;
 200             f[j_coord_offset+DIM*0+YY] -= ty;
 201             f[j_coord_offset+DIM*0+ZZ] -= tz;
 202
 203             /* Inner loop uses 40 flops */
 204         }
 205         /* End of innermost loop */
 206
 207         tx = ty = tz = 0;
 208         f[i_coord_offset+DIM*0+XX] += fix0;
 209         f[i_coord_offset+DIM*0+YY] += fiy0;
 210         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 211         tx                         += fix0;
 212         ty                         += fiy0;
 213         tz                         += fiz0;
 214         fshift[i_shift_offset+XX]  += tx;
 215         fshift[i_shift_offset+YY]  += ty;
 216         fshift[i_shift_offset+ZZ]  += tz;
 217
 218         ggid                        = gid[iidx];
 219         /* Update potential energies */
 220         kernel_data->energygrp_elec[ggid] += velecsum;
 221         kernel_data->energygrp_vdw[ggid] += vvdwsum;
 222
 223         /* Increment number of inner iterations */
 224         inneriter                  += j_index_end - j_index_start;
 225
 226         /* Outer loop uses 15 flops */
 227     }
 228
 229     /* Increment number of outer iterations */
 230     outeriter        += nri;
 231
 232     /* Update outer/inner flops */
 233
 234     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*40);
 235 }
 236 /*
 237  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwLJ_GeomP1P1_F_c
 238  * Electrostatics interaction: Coulomb
 239  * VdW interaction:            LennardJones
 240  * Geometry:                   Particle-Particle
 241  * Calculate force/pot:        Force
 242  */
 243 void
 244 nb_kernel_ElecCoul_VdwLJ_GeomP1P1_F_c
 245                     (t_nblist                    * gmx_restrict       nlist,
 246                      rvec                        * gmx_restrict          xx,
 247                      rvec                        * gmx_restrict          ff,
 248                      t_forcerec                  * gmx_restrict          fr,
 249                      t_mdatoms                   * gmx_restrict     mdatoms,
 250                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 251                      t_nrnb                      * gmx_restrict        nrnb)
 252 {
 253     int              i_shift_offset,i_coord_offset,j_coord_offset;
 254     int              j_index_start,j_index_end;
 255     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
 256     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
 257     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 258     real             *shiftvec,*fshift,*x,*f;
 259     int              vdwioffset0;
 260     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 261     int              vdwjidx0;
 262     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 263     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
 264     real             velec,felec,velecsum,facel,crf,krf,krf2;
 265     real             *charge;
 266     int              nvdwtype;
 267     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
 268     int              *vdwtype;
 269     real             *vdwparam;
 270
 271     x                = xx[0];
 272     f                = ff[0];
 273
 274     nri              = nlist->nri;
 275     iinr             = nlist->iinr;
 276     jindex           = nlist->jindex;
 277     jjnr             = nlist->jjnr;
 278     shiftidx         = nlist->shift;
 279     gid              = nlist->gid;
 280     shiftvec         = fr->shift_vec[0];
 281     fshift           = fr->fshift[0];
 282     facel            = fr->epsfac;
 283     charge           = mdatoms->chargeA;
 284     nvdwtype         = fr->ntype;
 285     vdwparam         = fr->nbfp;
 286     vdwtype          = mdatoms->typeA;
 287
 288     outeriter        = 0;
 289     inneriter        = 0;
 290
 291     /* Start outer loop over neighborlists */
 292     for(iidx=0; iidx<nri; iidx++)
 293     {
 294         /* Load shift vector for this list */
 295         i_shift_offset   = DIM*shiftidx[iidx];
 296         shX              = shiftvec[i_shift_offset+XX];
 297         shY              = shiftvec[i_shift_offset+YY];
 298         shZ              = shiftvec[i_shift_offset+ZZ];
 299
 300         /* Load limits for loop over neighbors */
 301         j_index_start    = jindex[iidx];
 302         j_index_end      = jindex[iidx+1];
 303
 304         /* Get outer coordinate index */
 305         inr              = iinr[iidx];
 306         i_coord_offset   = DIM*inr;
 307
 308         /* Load i particle coords and add shift vector */
 309         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 310         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 311         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 312
 313         fix0             = 0.0;
 314         fiy0             = 0.0;
 315         fiz0             = 0.0;
 316
 317         /* Load parameters for i particles */
 318         iq0              = facel*charge[inr+0];
 319         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 320
 321         /* Start inner kernel loop */
 322         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 323         {
 324             /* Get j neighbor index, and coordinate index */
 325             jnr              = jjnr[jidx];
 326             j_coord_offset   = DIM*jnr;
 327
 328             /* load j atom coordinates */
 329             jx0              = x[j_coord_offset+DIM*0+XX];
 330             jy0              = x[j_coord_offset+DIM*0+YY];
 331             jz0              = x[j_coord_offset+DIM*0+ZZ];
 332
 333             /* Calculate displacement vector */
 334             dx00             = ix0 - jx0;
 335             dy00             = iy0 - jy0;
 336             dz00             = iz0 - jz0;
 337
 338             /* Calculate squared distance and things based on it */
 339             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 340
 341             rinv00           = gmx_invsqrt(rsq00);
 342
 343             rinvsq00         = rinv00*rinv00;
 344
 345             /* Load parameters for j particles */
 346             jq0              = charge[jnr+0];
 347             vdwjidx0         = 2*vdwtype[jnr+0];
 348
 349             /**************************
 350              * CALCULATE INTERACTIONS *
 351              **************************/
 352
 353             qq00             = iq0*jq0;
 354             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 355             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 356
 357             /* COULOMB ELECTROSTATICS */
 358             velec            = qq00*rinv00;
 359             felec            = velec*rinvsq00;
 360
 361             /* LENNARD-JONES DISPERSION/REPULSION */
 362
 363             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
 364             fvdw             = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
 365
 366             fscal            = felec+fvdw;
 367
 368             /* Calculate temporary vectorial force */
 369             tx               = fscal*dx00;
 370             ty               = fscal*dy00;
 371             tz               = fscal*dz00;
 372
 373             /* Update vectorial force */
 374             fix0            += tx;
 375             fiy0            += ty;
 376             fiz0            += tz;
 377             f[j_coord_offset+DIM*0+XX] -= tx;
 378             f[j_coord_offset+DIM*0+YY] -= ty;
 379             f[j_coord_offset+DIM*0+ZZ] -= tz;
 380
 381             /* Inner loop uses 34 flops */
 382         }
 383         /* End of innermost loop */
 384
 385         tx = ty = tz = 0;
 386         f[i_coord_offset+DIM*0+XX] += fix0;
 387         f[i_coord_offset+DIM*0+YY] += fiy0;
 388         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 389         tx                         += fix0;
 390         ty                         += fiy0;
 391         tz                         += fiz0;
 392         fshift[i_shift_offset+XX]  += tx;
 393         fshift[i_shift_offset+YY]  += ty;
 394         fshift[i_shift_offset+ZZ]  += tz;
 395
 396         /* Increment number of inner iterations */
 397         inneriter                  += j_index_end - j_index_start;
 398
 399         /* Outer loop uses 13 flops */
 400     }
 401
 402     /* Increment number of outer iterations */
 403     outeriter        += nri;
 404
 405     /* Update outer/inner flops */
 406
 407     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*34);
 408 }