src/gromacs/gmxlib/nonbonded/nb_kernel_c/nb_kernel_ElecNone_VdwCSTab_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_ElecNone_VdwCSTab_GeomP1P1_VF_c
  51  * Electrostatics interaction: None
  52  * VdW interaction:            CubicSplineTable
  53  * Geometry:                   Particle-Particle
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecNone_VdwCSTab_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     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     int              vfitab;
  82     real             rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
  83     real             *vftab;
  84
  85     x                = xx[0];
  86     f                = ff[0];
  87
  88     nri              = nlist->nri;
  89     iinr             = nlist->iinr;
  90     jindex           = nlist->jindex;
  91     jjnr             = nlist->jjnr;
  92     shiftidx         = nlist->shift;
  93     gid              = nlist->gid;
  94     shiftvec         = fr->shift_vec[0];
  95     fshift           = fr->fshift[0];
  96     nvdwtype         = fr->ntype;
  97     vdwparam         = fr->nbfp;
  98     vdwtype          = mdatoms->typeA;
  99
 100     vftab            = kernel_data->table_vdw->data;
 101     vftabscale       = kernel_data->table_vdw->scale;
 102
 103     outeriter        = 0;
 104     inneriter        = 0;
 105
 106     /* Start outer loop over neighborlists */
 107     for(iidx=0; iidx<nri; iidx++)
 108     {
 109         /* Load shift vector for this list */
 110         i_shift_offset   = DIM*shiftidx[iidx];
 111         shX              = shiftvec[i_shift_offset+XX];
 112         shY              = shiftvec[i_shift_offset+YY];
 113         shZ              = shiftvec[i_shift_offset+ZZ];
 114
 115         /* Load limits for loop over neighbors */
 116         j_index_start    = jindex[iidx];
 117         j_index_end      = jindex[iidx+1];
 118
 119         /* Get outer coordinate index */
 120         inr              = iinr[iidx];
 121         i_coord_offset   = DIM*inr;
 122
 123         /* Load i particle coords and add shift vector */
 124         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 125         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 126         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 127
 128         fix0             = 0.0;
 129         fiy0             = 0.0;
 130         fiz0             = 0.0;
 131
 132         /* Load parameters for i particles */
 133         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 134
 135         /* Reset potential sums */
 136         vvdwsum          = 0.0;
 137
 138         /* Start inner kernel loop */
 139         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 140         {
 141             /* Get j neighbor index, and coordinate index */
 142             jnr              = jjnr[jidx];
 143             j_coord_offset   = DIM*jnr;
 144
 145             /* load j atom coordinates */
 146             jx0              = x[j_coord_offset+DIM*0+XX];
 147             jy0              = x[j_coord_offset+DIM*0+YY];
 148             jz0              = x[j_coord_offset+DIM*0+ZZ];
 149
 150             /* Calculate displacement vector */
 151             dx00             = ix0 - jx0;
 152             dy00             = iy0 - jy0;
 153             dz00             = iz0 - jz0;
 154
 155             /* Calculate squared distance and things based on it */
 156             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 157
 158             rinv00           = gmx_invsqrt(rsq00);
 159
 160             /* Load parameters for j particles */
 161             vdwjidx0         = 2*vdwtype[jnr+0];
 162
 163             /**************************
 164              * CALCULATE INTERACTIONS *
 165              **************************/
 166
 167             r00              = rsq00*rinv00;
 168
 169             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 170             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 171
 172             /* Calculate table index by multiplying r with table scale and truncate to integer */
 173             rt               = r00*vftabscale;
 174             vfitab           = rt;
 175             vfeps            = rt-vfitab;
 176             vfitab           = 2*4*vfitab;
 177
 178             /* CUBIC SPLINE TABLE DISPERSION */
 179             vfitab          += 0;
 180             Y                = vftab[vfitab];
 181             F                = vftab[vfitab+1];
 182             Geps             = vfeps*vftab[vfitab+2];
 183             Heps2            = vfeps*vfeps*vftab[vfitab+3];
 184             Fp               = F+Geps+Heps2;
 185             VV               = Y+vfeps*Fp;
 186             vvdw6            = c6_00*VV;
 187             FF               = Fp+Geps+2.0*Heps2;
 188             fvdw6            = c6_00*FF;
 189
 190             /* CUBIC SPLINE TABLE REPULSION */
 191             Y                = vftab[vfitab+4];
 192             F                = vftab[vfitab+5];
 193             Geps             = vfeps*vftab[vfitab+6];
 194             Heps2            = vfeps*vfeps*vftab[vfitab+7];
 195             Fp               = F+Geps+Heps2;
 196             VV               = Y+vfeps*Fp;
 197             vvdw12           = c12_00*VV;
 198             FF               = Fp+Geps+2.0*Heps2;
 199             fvdw12           = c12_00*FF;
 200             vvdw             = vvdw12+vvdw6;
 201             fvdw             = -(fvdw6+fvdw12)*vftabscale*rinv00;
 202
 203             /* Update potential sums from outer loop */
 204             vvdwsum         += vvdw;
 205
 206             fscal            = fvdw;
 207
 208             /* Calculate temporary vectorial force */
 209             tx               = fscal*dx00;
 210             ty               = fscal*dy00;
 211             tz               = fscal*dz00;
 212
 213             /* Update vectorial force */
 214             fix0            += tx;
 215             fiy0            += ty;
 216             fiz0            += tz;
 217             f[j_coord_offset+DIM*0+XX] -= tx;
 218             f[j_coord_offset+DIM*0+YY] -= ty;
 219             f[j_coord_offset+DIM*0+ZZ] -= tz;
 220
 221             /* Inner loop uses 55 flops */
 222         }
 223         /* End of innermost loop */
 224
 225         tx = ty = tz = 0;
 226         f[i_coord_offset+DIM*0+XX] += fix0;
 227         f[i_coord_offset+DIM*0+YY] += fiy0;
 228         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 229         tx                         += fix0;
 230         ty                         += fiy0;
 231         tz                         += fiz0;
 232         fshift[i_shift_offset+XX]  += tx;
 233         fshift[i_shift_offset+YY]  += ty;
 234         fshift[i_shift_offset+ZZ]  += tz;
 235
 236         ggid                        = gid[iidx];
 237         /* Update potential energies */
 238         kernel_data->energygrp_vdw[ggid] += vvdwsum;
 239
 240         /* Increment number of inner iterations */
 241         inneriter                  += j_index_end - j_index_start;
 242
 243         /* Outer loop uses 13 flops */
 244     }
 245
 246     /* Increment number of outer iterations */
 247     outeriter        += nri;
 248
 249     /* Update outer/inner flops */
 250
 251     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*13 + inneriter*55);
 252 }
 253 /*
 254  * Gromacs nonbonded kernel:   nb_kernel_ElecNone_VdwCSTab_GeomP1P1_F_c
 255  * Electrostatics interaction: None
 256  * VdW interaction:            CubicSplineTable
 257  * Geometry:                   Particle-Particle
 258  * Calculate force/pot:        Force
 259  */
 260 void
 261 nb_kernel_ElecNone_VdwCSTab_GeomP1P1_F_c
 262                     (t_nblist                    * gmx_restrict       nlist,
 263                      rvec                        * gmx_restrict          xx,
 264                      rvec                        * gmx_restrict          ff,
 265                      t_forcerec                  * gmx_restrict          fr,
 266                      t_mdatoms                   * gmx_restrict     mdatoms,
 267                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 268                      t_nrnb                      * gmx_restrict        nrnb)
 269 {
 270     int              i_shift_offset,i_coord_offset,j_coord_offset;
 271     int              j_index_start,j_index_end;
 272     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
 273     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
 274     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 275     real             *shiftvec,*fshift,*x,*f;
 276     int              vdwioffset0;
 277     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 278     int              vdwjidx0;
 279     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 280     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
 281     int              nvdwtype;
 282     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
 283     int              *vdwtype;
 284     real             *vdwparam;
 285     int              vfitab;
 286     real             rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
 287     real             *vftab;
 288
 289     x                = xx[0];
 290     f                = ff[0];
 291
 292     nri              = nlist->nri;
 293     iinr             = nlist->iinr;
 294     jindex           = nlist->jindex;
 295     jjnr             = nlist->jjnr;
 296     shiftidx         = nlist->shift;
 297     gid              = nlist->gid;
 298     shiftvec         = fr->shift_vec[0];
 299     fshift           = fr->fshift[0];
 300     nvdwtype         = fr->ntype;
 301     vdwparam         = fr->nbfp;
 302     vdwtype          = mdatoms->typeA;
 303
 304     vftab            = kernel_data->table_vdw->data;
 305     vftabscale       = kernel_data->table_vdw->scale;
 306
 307     outeriter        = 0;
 308     inneriter        = 0;
 309
 310     /* Start outer loop over neighborlists */
 311     for(iidx=0; iidx<nri; iidx++)
 312     {
 313         /* Load shift vector for this list */
 314         i_shift_offset   = DIM*shiftidx[iidx];
 315         shX              = shiftvec[i_shift_offset+XX];
 316         shY              = shiftvec[i_shift_offset+YY];
 317         shZ              = shiftvec[i_shift_offset+ZZ];
 318
 319         /* Load limits for loop over neighbors */
 320         j_index_start    = jindex[iidx];
 321         j_index_end      = jindex[iidx+1];
 322
 323         /* Get outer coordinate index */
 324         inr              = iinr[iidx];
 325         i_coord_offset   = DIM*inr;
 326
 327         /* Load i particle coords and add shift vector */
 328         ix0              = shX + x[i_coord_offset+DIM*0+XX];
 329         iy0              = shY + x[i_coord_offset+DIM*0+YY];
 330         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
 331
 332         fix0             = 0.0;
 333         fiy0             = 0.0;
 334         fiz0             = 0.0;
 335
 336         /* Load parameters for i particles */
 337         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 338
 339         /* Start inner kernel loop */
 340         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 341         {
 342             /* Get j neighbor index, and coordinate index */
 343             jnr              = jjnr[jidx];
 344             j_coord_offset   = DIM*jnr;
 345
 346             /* load j atom coordinates */
 347             jx0              = x[j_coord_offset+DIM*0+XX];
 348             jy0              = x[j_coord_offset+DIM*0+YY];
 349             jz0              = x[j_coord_offset+DIM*0+ZZ];
 350
 351             /* Calculate displacement vector */
 352             dx00             = ix0 - jx0;
 353             dy00             = iy0 - jy0;
 354             dz00             = iz0 - jz0;
 355
 356             /* Calculate squared distance and things based on it */
 357             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
 358
 359             rinv00           = gmx_invsqrt(rsq00);
 360
 361             /* Load parameters for j particles */
 362             vdwjidx0         = 2*vdwtype[jnr+0];
 363
 364             /**************************
 365              * CALCULATE INTERACTIONS *
 366              **************************/
 367
 368             r00              = rsq00*rinv00;
 369
 370             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
 371             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
 372
 373             /* Calculate table index by multiplying r with table scale and truncate to integer */
 374             rt               = r00*vftabscale;
 375             vfitab           = rt;
 376             vfeps            = rt-vfitab;
 377             vfitab           = 2*4*vfitab;
 378
 379             /* CUBIC SPLINE TABLE DISPERSION */
 380             vfitab          += 0;
 381             F                = vftab[vfitab+1];
 382             Geps             = vfeps*vftab[vfitab+2];
 383             Heps2            = vfeps*vfeps*vftab[vfitab+3];
 384             Fp               = F+Geps+Heps2;
 385             FF               = Fp+Geps+2.0*Heps2;
 386             fvdw6            = c6_00*FF;
 387
 388             /* CUBIC SPLINE TABLE REPULSION */
 389             F                = vftab[vfitab+5];
 390             Geps             = vfeps*vftab[vfitab+6];
 391             Heps2            = vfeps*vfeps*vftab[vfitab+7];
 392             Fp               = F+Geps+Heps2;
 393             FF               = Fp+Geps+2.0*Heps2;
 394             fvdw12           = c12_00*FF;
 395             fvdw             = -(fvdw6+fvdw12)*vftabscale*rinv00;
 396
 397             fscal            = fvdw;
 398
 399             /* Calculate temporary vectorial force */
 400             tx               = fscal*dx00;
 401             ty               = fscal*dy00;
 402             tz               = fscal*dz00;
 403
 404             /* Update vectorial force */
 405             fix0            += tx;
 406             fiy0            += ty;
 407             fiz0            += tz;
 408             f[j_coord_offset+DIM*0+XX] -= tx;
 409             f[j_coord_offset+DIM*0+YY] -= ty;
 410             f[j_coord_offset+DIM*0+ZZ] -= tz;
 411
 412             /* Inner loop uses 47 flops */
 413         }
 414         /* End of innermost loop */
 415
 416         tx = ty = tz = 0;
 417         f[i_coord_offset+DIM*0+XX] += fix0;
 418         f[i_coord_offset+DIM*0+YY] += fiy0;
 419         f[i_coord_offset+DIM*0+ZZ] += fiz0;
 420         tx                         += fix0;
 421         ty                         += fiy0;
 422         tz                         += fiz0;
 423         fshift[i_shift_offset+XX]  += tx;
 424         fshift[i_shift_offset+YY]  += ty;
 425         fshift[i_shift_offset+ZZ]  += tz;
 426
 427         /* Increment number of inner iterations */
 428         inneriter                  += j_index_end - j_index_start;
 429
 430         /* Outer loop uses 12 flops */
 431     }
 432
 433     /* Increment number of outer iterations */
 434     outeriter        += nri;
 435
 436     /* Update outer/inner flops */
 437
 438     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*12 + inneriter*47);
 439 }