src/gmxlib/nonbonded/nb_kernel_c/nb_kernel_template_c.pre

   1 /* ## */
   2 /* ## This file is part of the GROMACS molecular simulation package. */
   3 /* ## */
   4 /* ## Copyright (c) 2012, by the GROMACS development team, led by */
   5 /* ## David van der Spoel, Berk Hess, Erik Lindahl, and including many */
   6 /* ## others, as listed in the AUTHORS file in the top-level source */
   7 /* ## 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 /* #if 0 */
  36 #error This file must be processed with the GROMACS pre-preprocessor
  37 /* #endif */
  38 /* #if INCLUDE_HEADER */
  39 #ifdef HAVE_CONFIG_H
  40 #include <config.h>
  41 #endif
  42
  43 #include <math.h>
  44
  45 #include "../nb_kernel.h"
  46 #include "types/simple.h"
  47 #include "vec.h"
  48 #include "nrnb.h"
  49 /* #endif */
  50
  51 /* ## List of variables set by the generating script:                                    */
  52 /* ##                                                                                    */
  53 /* ## Setttings that apply to the entire kernel:                                         */
  54 /* ## KERNEL_ELEC:           String, choice for electrostatic interactions               */
  55 /* ## KERNEL_VDW:            String, choice for van der Waals interactions               */
  56 /* ## KERNEL_NAME:           String, name of this kernel                                 */
  57 /* ## KERNEL_VF:             String telling if we calculate potential, force, or both    */
  58 /* ## GEOMETRY_I/GEOMETRY_J: String, name of each geometry, e.g. 'Water3' or '1Particle' */
  59 /* ##                                                                                    */
  60 /* ## Setttings that apply to particles in the outer (I) or inner (J) loops:             */
  61 /* ## PARTICLES_I[]/         Arrays with lists of i/j particles to use in kernel. It is  */
  62 /* ## PARTICLES_J[]:         just [0] for particle geometry, but can be longer for water */
  63 /* ## PARTICLES_ELEC_I[]/    Arrays with lists of i/j particle that have electrostatics  */
  64 /* ## PARTICLES_ELEC_J[]:    interactions that should be calculated in this kernel.      */
  65 /* ## PARTICLES_VDW_I[]/     Arrays with the list of i/j particle that have VdW          */
  66 /* ## PARTICLES_VDW_J[]:     interactions that should be calculated in this kernel.      */
  67 /* ##                                                                                    */
  68 /* ## Setttings for pairs of interactions (e.g. 2nd i particle against 1st j particle)   */
  69 /* ## PAIRS_IJ[]:            Array with (i,j) tuples of pairs for which interactions     */
  70 /* ##                        should be calculated in this kernel. Zero-charge particles  */
  71 /* ##                        do not have interactions with particles without vdw, and    */
  72 /* ##                        Vdw-only interactions are not evaluated in a no-vdw-kernel. */
  73 /* ## INTERACTION_FLAGS[][]: 2D matrix, dimension e.g. 3*3 for water-water interactions. */
  74 /* ##                        For each i-j pair, the element [I][J] is a list of strings  */
  75 /* ##                        defining properties/flags of this interaction. Examples     */
  76 /* ##                        include 'electrostatics'/'vdw' if that type of interaction  */
  77 /* ##                        should be evaluated, 'rsq'/'rinv'/'rinvsq' if those values  */
  78 /* ##                        are needed, and 'exactcutoff' or 'shift','switch' to        */
  79 /* ##                        decide if the force/potential should be modified. This way  */
  80 /* ##                        we only calculate values absolutely needed for each case.   */
  81
  82 /* ## Calculate the size and offset for (merged/interleaved) table data */
  83
  84 /* #if ('CubicSplineTable' in [KERNEL_ELEC,KERNEL_VDW]) or KERNEL_VF=='PotentialAndForce' */
  85 /*     #define TABLE_POINT_SIZE 4 */
  86 /* #else */
  87 /*     #define TABLE_POINT_SIZE 2 */
  88 /* #endif */
  89
  90 /* #if 'Table' in KERNEL_ELEC and 'Table' in KERNEL_VDW */
  91 /*     #define TABLE_NINTERACTIONS 3 */
  92 /*     #define TABLE_VDW_OFFSET TABLE_POINT_SIZE */
  93 /* #elif 'Table' in KERNEL_ELEC */
  94 /*     #define TABLE_NINTERACTIONS 1 */
  95 /* #elif 'Table' in KERNEL_VDW */
  96 /*     #define TABLE_NINTERACTIONS 2 */
  97 /*     #define TABLE_VDW_OFFSET 0 */
  98 /* #else */
  99 /*     #define TABLE_NINTERACTIONS 0 */
 100 /* #endif */
 101
 102 /* #if 'Buckingham' in KERNEL_VDW */
 103 /*   #define NVDWPARAM 3 */
 104 /* #else */
 105 /*   #define NVDWPARAM 2 */
 106 /* #endif */
 107
 108 /*
 109  * GROMACS nonbonded kernel:   {KERNEL_NAME}
 110  * Electrostatics interaction: {KERNEL_ELEC}
 111  * VdW interaction:            {KERNEL_VDW}
 112  * Geometry:                   {GEOMETRY_I}-{GEOMETRY_J}
 113  * Calculate force/pot:        {KERNEL_VF}
 114  */
 115 void
 116 {KERNEL_NAME}
 117                     (t_nblist * gmx_restrict                nlist,
 118                      rvec * gmx_restrict                    xx,
 119                      rvec * gmx_restrict                    ff,
 120                      t_forcerec * gmx_restrict              fr,
 121                      t_mdatoms * gmx_restrict               mdatoms,
 122                      nb_kernel_data_t * gmx_restrict        kernel_data,
 123                      t_nrnb * gmx_restrict                  nrnb)
 124 {
 125     /* ## Not all variables are used for all kernels, but any optimizing compiler fixes that, */
 126     /* ## so there is no point in going to extremes to exclude variables that are not needed. */
 127     int              i_shift_offset,i_coord_offset,j_coord_offset;
 128     int              j_index_start,j_index_end;
 129     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
 130     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
 131     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 132     real             *shiftvec,*fshift,*x,*f;
 133     /* #for I in PARTICLES_I */
 134     int              vdwioffset{I};
 135     real             ix{I},iy{I},iz{I},fix{I},fiy{I},fiz{I},iq{I},isai{I};
 136     /* #endfor */
 137     /* #for J in PARTICLES_J */
 138     int              vdwjidx{J};
 139     real             jx{J},jy{J},jz{J},fjx{J},fjy{J},fjz{J},jq{J},isaj{J};
 140     /* #endfor */
 141     /* #for I,J in PAIRS_IJ */
 142     real             dx{I}{J},dy{I}{J},dz{I}{J},rsq{I}{J},rinv{I}{J},rinvsq{I}{J},r{I}{J},qq{I}{J},c6_{I}{J},c12_{I}{J},cexp1_{I}{J},cexp2_{I}{J};
 143     /* #endfor */
 144     /* #if KERNEL_ELEC != 'None' */
 145     real             velec,felec,velecsum,facel,crf,krf,krf2;
 146     real             *charge;
 147     /* #endif */
 148     /* #if 'GeneralizedBorn' in KERNEL_ELEC */
 149     int              gbitab;
 150     real             vgb,fgb,vgbsum,dvdasum,gbscale,gbtabscale,isaprod,gbqqfactor,gbinvepsdiff,dvdaj,gbeps,dvdatmp;
 151     real             *invsqrta,*dvda,*gbtab;
 152     /* #endif */
 153     /* #if KERNEL_VDW != 'None' */
 154     int              nvdwtype;
 155     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
 156     int              *vdwtype;
 157     real             *vdwparam;
 158     /* #endif */
 159     /* #if 'Table' in KERNEL_ELEC or 'GeneralizedBorn' in KERNEL_ELEC or 'Table' in KERNEL_VDW */
 160     int              vfitab;
 161     real             rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
 162     real             *vftab;
 163     /* #endif */
 164     /* #if 'Ewald' in KERNEL_ELEC */
 165     int              ewitab;
 166     real             ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
 167     real             *ewtab;
 168     /* #endif */
 169     /* #if 'PotentialSwitch' in [KERNEL_MOD_ELEC,KERNEL_MOD_VDW] */
 170     real             rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
 171     /* #endif */
 172
 173     x                = xx[0];
 174     f                = ff[0];
 175
 176     nri              = nlist->nri;
 177     iinr             = nlist->iinr;
 178     jindex           = nlist->jindex;
 179     jjnr             = nlist->jjnr;
 180     shiftidx         = nlist->shift;
 181     gid              = nlist->gid;
 182     shiftvec         = fr->shift_vec[0];
 183     fshift           = fr->fshift[0];
 184     /* #if KERNEL_ELEC != 'None' */
 185     facel            = fr->epsfac;
 186     charge           = mdatoms->chargeA;
 187     /*     #if 'ReactionField' in KERNEL_ELEC */
 188     krf              = fr->ic->k_rf;
 189     krf2             = krf*2.0;
 190     crf              = fr->ic->c_rf;
 191     /*     #endif */
 192     /* #endif */
 193     /* #if KERNEL_VDW != 'None' */
 194     nvdwtype         = fr->ntype;
 195     vdwparam         = fr->nbfp;
 196     vdwtype          = mdatoms->typeA;
 197     /* #endif */
 198
 199     /* #if 'Table' in KERNEL_ELEC and 'Table' in KERNEL_VDW */
 200     vftab            = kernel_data->table_elec_vdw->data;
 201     vftabscale       = kernel_data->table_elec_vdw->scale;
 202     /* #elif 'Table' in KERNEL_ELEC */
 203     vftab            = kernel_data->table_elec->data;
 204     vftabscale       = kernel_data->table_elec->scale;
 205     /* #elif 'Table' in KERNEL_VDW */
 206     vftab            = kernel_data->table_vdw->data;
 207     vftabscale       = kernel_data->table_vdw->scale;
 208     /* #endif */
 209
 210     /* #if 'Ewald' in KERNEL_ELEC */
 211     sh_ewald         = fr->ic->sh_ewald;
 212     /*     #if KERNEL_VF=='Force' and KERNEL_MOD_ELEC!='PotentialSwitch' */
 213     ewtab            = fr->ic->tabq_coul_F;
 214     ewtabscale       = fr->ic->tabq_scale;
 215     ewtabhalfspace   = 0.5/ewtabscale;
 216     /*     #else */
 217     ewtab            = fr->ic->tabq_coul_FDV0;
 218     ewtabscale       = fr->ic->tabq_scale;
 219     ewtabhalfspace   = 0.5/ewtabscale;
 220      /*     #endif */
 221     /* #endif */
 222
 223     /* #if KERNEL_ELEC=='GeneralizedBorn' */
 224     invsqrta         = fr->invsqrta;
 225     dvda             = fr->dvda;
 226     gbtabscale       = fr->gbtab.scale;
 227     gbtab            = fr->gbtab.data;
 228     gbinvepsdiff     = (1.0/fr->epsilon_r) - (1.0/fr->gb_epsilon_solvent);
 229     /* #endif */
 230
 231     /* #if 'Water' in GEOMETRY_I */
 232     /* Setup water-specific parameters */
 233     inr              = nlist->iinr[0];
 234     /*     #for I in PARTICLES_ELEC_I */
 235     iq{I}              = facel*charge[inr+{I}];
 236     /*     #endfor */
 237     /*     #for I in PARTICLES_VDW_I */
 238     vdwioffset{I}      = {NVDWPARAM}*nvdwtype*vdwtype[inr+{I}];
 239     /*     #endfor */
 240     /* #endif */
 241
 242     /* #if 'Water' in GEOMETRY_J */
 243     /*     #for J in PARTICLES_ELEC_J */
 244     jq{J}              = charge[inr+{J}];
 245     /*     #endfor */
 246     /*     #for J in PARTICLES_VDW_J */
 247     vdwjidx{J}         = {NVDWPARAM}*vdwtype[inr+{J}];
 248     /*     #endfor */
 249     /*     #for I,J in PAIRS_IJ */
 250     /*         #if 'electrostatics' in INTERACTION_FLAGS[I][J] */
 251     qq{I}{J}             = iq{I}*jq{J};
 252     /*         #endif */
 253     /*         #if 'vdw' in INTERACTION_FLAGS[I][J] */
 254     /*             #if 'Buckingham' in KERNEL_VDW */
 255     c6_{I}{J}            = vdwparam[vdwioffset{I}+vdwjidx{J}];
 256     cexp1_{I}{J}         = vdwparam[vdwioffset{I}+vdwjidx{J}+1];
 257     cexp2_{I}{J}         = vdwparam[vdwioffset{I}+vdwjidx{J}+2];
 258     /*             #else */
 259     c6_{I}{J}            = vdwparam[vdwioffset{I}+vdwjidx{J}];
 260     c12_{I}{J}           = vdwparam[vdwioffset{I}+vdwjidx{J}+1];
 261     /*             #endif */
 262     /*         #endif */
 263     /*     #endfor */
 264     /* #endif */
 265
 266     /* #if KERNEL_MOD_ELEC!='None' or KERNEL_MOD_VDW!='None' */
 267     /*     #if KERNEL_ELEC!='None' */
 268     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 269     rcutoff          = fr->rcoulomb;
 270     /*     #else */
 271     rcutoff          = fr->rvdw;
 272     /*     #endif */
 273     rcutoff2         = rcutoff*rcutoff;
 274     /* #endif */
 275
 276     /* #if KERNEL_MOD_VDW=='PotentialShift' */
 277     sh_vdw_invrcut6  = fr->ic->sh_invrc6;
 278     rvdw             = fr->rvdw;
 279     /* #endif */
 280
 281     /* #if 'PotentialSwitch' in [KERNEL_MOD_ELEC,KERNEL_MOD_VDW] */
 282     /*     #if KERNEL_MOD_ELEC=='PotentialSwitch'  */
 283     rswitch          = fr->rcoulomb_switch;
 284     /*     #else */
 285     rswitch          = fr->rvdw_switch;
 286     /*     #endif */
 287     /* Setup switch parameters */
 288     d                = rcutoff-rswitch;
 289     swV3             = -10.0/(d*d*d);
 290     swV4             =  15.0/(d*d*d*d);
 291     swV5             =  -6.0/(d*d*d*d*d);
 292     /*     #if 'Force' in KERNEL_VF */
 293     swF2             = -30.0/(d*d*d);
 294     swF3             =  60.0/(d*d*d*d);
 295     swF4             = -30.0/(d*d*d*d*d);
 296     /*     #endif */
 297     /* #endif */
 298
 299     /* ## Keep track of the floating point operations we issue for reporting! */
 300     /* #define OUTERFLOPS 0 */
 301     /* #define INNERFLOPS 0 */
 302     outeriter        = 0;
 303     inneriter        = 0;
 304
 305     /* Start outer loop over neighborlists */
 306     for(iidx=0; iidx<nri; iidx++)
 307     {
 308         /* Load shift vector for this list */
 309         i_shift_offset   = DIM*shiftidx[iidx];
 310         shX              = shiftvec[i_shift_offset+XX];
 311         shY              = shiftvec[i_shift_offset+YY];
 312         shZ              = shiftvec[i_shift_offset+ZZ];
 313
 314         /* Load limits for loop over neighbors */
 315         j_index_start    = jindex[iidx];
 316         j_index_end      = jindex[iidx+1];
 317
 318         /* Get outer coordinate index */
 319         inr              = iinr[iidx];
 320         i_coord_offset   = DIM*inr;
 321
 322         /* Load i particle coords and add shift vector */
 323         /* ## Loop over i particles, but only include ones that we use - skip e.g. vdw-only sites for elec-only kernel */
 324         /*     #for I in PARTICLES_I */
 325         ix{I}              = shX + x[i_coord_offset+DIM*{I}+XX];
 326         iy{I}              = shY + x[i_coord_offset+DIM*{I}+YY];
 327         iz{I}              = shZ + x[i_coord_offset+DIM*{I}+ZZ];
 328         /*     #define OUTERFLOPS OUTERFLOPS+3 */
 329         /* #endfor */
 330
 331         /* #if 'Force' in KERNEL_VF */
 332         /*     #for I in PARTICLES_I */
 333         fix{I}             = 0.0;
 334         fiy{I}             = 0.0;
 335         fiz{I}             = 0.0;
 336         /*     #endfor */
 337         /* #endif */
 338
 339         /* ## For water we already preloaded parameters at the start of the kernel */
 340         /* #if not 'Water' in GEOMETRY_I */
 341         /* Load parameters for i particles */
 342         /*     #for I in PARTICLES_ELEC_I */
 343         iq{I}              = facel*charge[inr+{I}];
 344         /*         #define OUTERFLOPS OUTERFLOPS+1 */
 345         /*         #if KERNEL_ELEC=='GeneralizedBorn' */
 346         isai{I}            = invsqrta[inr+{I}];
 347         /*         #endif */
 348         /*     #endfor */
 349         /*     #for I in PARTICLES_VDW_I */
 350         vdwioffset{I}      = {NVDWPARAM}*nvdwtype*vdwtype[inr+{I}];
 351         /*     #endfor */
 352         /* #endif */
 353
 354         /* #if 'Potential' in KERNEL_VF */
 355         /* Reset potential sums */
 356         /*     #if KERNEL_ELEC != 'None' */
 357         velecsum         = 0.0;
 358         /*     #endif */
 359         /*     #if 'GeneralizedBorn' in KERNEL_ELEC */
 360         vgbsum           = 0.0;
 361         /*     #endif */
 362         /*     #if KERNEL_VDW != 'None' */
 363         vvdwsum          = 0.0;
 364         /*     #endif */
 365         /* #endif */
 366         /*     #if 'GeneralizedBorn' in KERNEL_ELEC and 'Force' in KERNEL_VF */
 367         dvdasum          = 0.0;
 368         /*     #endif */
 369
 370         /* Start inner kernel loop */
 371         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 372         {
 373             /* Get j neighbor index, and coordinate index */
 374             jnr              = jjnr[jidx];
 375             j_coord_offset   = DIM*jnr;
 376
 377             /* load j atom coordinates */
 378             /* #for J in PARTICLES_J */
 379             jx{J}              = x[j_coord_offset+DIM*{J}+XX];
 380             jy{J}              = x[j_coord_offset+DIM*{J}+YY];
 381             jz{J}              = x[j_coord_offset+DIM*{J}+ZZ];
 382             /* #endfor */
 383
 384             /* Calculate displacement vector */
 385             /* #for I,J in PAIRS_IJ */
 386             dx{I}{J}             = ix{I} - jx{J};
 387             dy{I}{J}             = iy{I} - jy{J};
 388             dz{I}{J}             = iz{I} - jz{J};
 389             /*     #define INNERFLOPS INNERFLOPS+3 */
 390             /* #endfor */
 391
 392             /* Calculate squared distance and things based on it */
 393             /* #for I,J in PAIRS_IJ */
 394             rsq{I}{J}            = dx{I}{J}*dx{I}{J}+dy{I}{J}*dy{I}{J}+dz{I}{J}*dz{I}{J};
 395             /*     #define INNERFLOPS INNERFLOPS+5 */
 396             /* #endfor */
 397
 398             /* #for I,J in PAIRS_IJ */
 399             /*     #if 'rinv' in INTERACTION_FLAGS[I][J] */
 400             rinv{I}{J}           = gmx_invsqrt(rsq{I}{J});
 401             /*         #define INNERFLOPS INNERFLOPS+5 */
 402             /*     #endif */
 403             /* #endfor */
 404
 405             /* #for I,J in PAIRS_IJ */
 406             /*     #if 'rinvsq' in INTERACTION_FLAGS[I][J] */
 407             /*         # if 'rinv' not in INTERACTION_FLAGS[I][J] */
 408             rinvsq{I}{J}         = 1.0/rsq{I}{J};
 409             /*             #define INNERFLOPS INNERFLOPS+4 */
 410             /*         #else */
 411             rinvsq{I}{J}         = rinv{I}{J}*rinv{I}{J};
 412             /*             #define INNERFLOPS INNERFLOPS+1 */
 413             /*         #endif */
 414             /*     #endif */
 415             /* #endfor */
 416
 417             /* #if not 'Water' in GEOMETRY_J */
 418             /* Load parameters for j particles */
 419             /*     #for J in PARTICLES_ELEC_J */
 420             jq{J}              = charge[jnr+{J}];
 421             /*         #if KERNEL_ELEC=='GeneralizedBorn' */
 422             isaj{J}           = invsqrta[jnr+{J}];
 423             /*         #endif */
 424             /*     #endfor */
 425             /*     #for J in PARTICLES_VDW_J */
 426             vdwjidx{J}         = {NVDWPARAM}*vdwtype[jnr+{J}];
 427             /*     #endfor */
 428             /* #endif */
 429
 430             /* #for I,J in PAIRS_IJ */
 431
 432             /**************************
 433              * CALCULATE INTERACTIONS *
 434              **************************/
 435
 436             /*     ## Note special check for TIP4P-TIP4P. Since we are cutting of all hydrogen interactions we also cut the LJ-only O-O interaction */
 437             /*     #if 'exactcutoff' in INTERACTION_FLAGS[I][J] or (GEOMETRY_I=='Water4' and GEOMETRY_J=='Water4' and 'exactcutoff' in INTERACTION_FLAGS[1][1]) */
 438             /*         ## We always calculate rinv/rinvsq above to enable pipelineing in compilers (performance tested on x86) */
 439             if (rsq{I}{J}<rcutoff2)
 440             {
 441                 /*     #if 0    ## this and the next two lines is a hack to maintain auto-indentation in template file */
 442             }
 443             /*         #endif */
 444             /*     #endif */
 445
 446             /*     #if 'r' in INTERACTION_FLAGS[I][J] */
 447             r{I}{J}              = rsq{I}{J}*rinv{I}{J};
 448             /*         #define INNERFLOPS INNERFLOPS+1 */
 449             /*     #endif */
 450
 451             /*     ## For water geometries we already loaded parameters at the start of the kernel */
 452             /*     #if not 'Water' in GEOMETRY_J */
 453             /*         #if 'electrostatics' in INTERACTION_FLAGS[I][J] */
 454             qq{I}{J}             = iq{I}*jq{J};
 455             /*             #define INNERFLOPS INNERFLOPS+1 */
 456             /*         #endif */
 457             /*         #if 'vdw' in INTERACTION_FLAGS[I][J] */
 458             /*             #if KERNEL_VDW=='Buckingham' */
 459             c6_{I}{J}            = vdwparam[vdwioffset{I}+vdwjidx{J}];
 460             cexp1_{I}{J}         = vdwparam[vdwioffset{I}+vdwjidx{J}+1];
 461             cexp2_{I}{J}         = vdwparam[vdwioffset{I}+vdwjidx{J}+2];
 462             /*             #else */
 463             c6_{I}{J}            = vdwparam[vdwioffset{I}+vdwjidx{J}];
 464             c12_{I}{J}           = vdwparam[vdwioffset{I}+vdwjidx{J}+1];
 465             /*             #endif */
 466             /*         #endif */
 467             /*     #endif */
 468
 469             /*     #if 'table' in INTERACTION_FLAGS[I][J] */
 470             /* Calculate table index by multiplying r with table scale and truncate to integer */
 471             rt               = r{I}{J}*vftabscale;
 472             vfitab           = rt;
 473             vfeps            = rt-vfitab;
 474             vfitab           = {TABLE_NINTERACTIONS}*{TABLE_POINT_SIZE}*vfitab;
 475             /*         #define INNERFLOPS INNERFLOPS+2 */
 476             /*     #endif */
 477
 478             /*     ## ELECTROSTATIC INTERACTIONS */
 479             /*     #if 'electrostatics' in INTERACTION_FLAGS[I][J] */
 480
 481             /*         #if KERNEL_ELEC=='Coulomb' */
 482
 483             /* COULOMB ELECTROSTATICS */
 484             velec            = qq{I}{J}*rinv{I}{J};
 485             /*             #define INNERFLOPS INNERFLOPS+1 */
 486             /*             #if 'Force' in KERNEL_VF */
 487             felec            = velec*rinvsq{I}{J};
 488             /*                 #define INNERFLOPS INNERFLOPS+2 */
 489             /*             #endif */
 490
 491             /*         #elif KERNEL_ELEC=='ReactionField' */
 492
 493             /* REACTION-FIELD ELECTROSTATICS */
 494             /*             #if 'Potential' in KERNEL_VF */
 495             velec            = qq{I}{J}*(rinv{I}{J}+krf*rsq{I}{J}-crf);
 496             /*                 #define INNERFLOPS INNERFLOPS+4 */
 497             /*             #endif */
 498             /*             #if 'Force' in KERNEL_VF */
 499             felec            = qq{I}{J}*(rinv{I}{J}*rinvsq{I}{J}-krf2);
 500             /*                 #define INNERFLOPS INNERFLOPS+3 */
 501             /*             #endif */
 502
 503             /*         #elif KERNEL_ELEC=='GeneralizedBorn' */
 504
 505             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 506             isaprod          = isai{I}*isaj{J};
 507             gbqqfactor       = isaprod*(-qq{I}{J})*gbinvepsdiff;
 508             gbscale          = isaprod*gbtabscale;
 509             dvdaj            = dvda[jnr+{J}];
 510             /*             #define INNERFLOPS INNERFLOPS+5 */
 511
 512             /* Calculate generalized born table index - this is a separate table from the normal one,
 513              * but we use the same procedure by multiplying r with scale and truncating to integer.
 514              */
 515             rt               = r{I}{J}*gbscale;
 516             gbitab           = rt;
 517             gbeps            = rt-gbitab;
 518             gbitab           = 4*gbitab;
 519
 520             Y                = gbtab[gbitab];
 521             F                = gbtab[gbitab+1];
 522             Geps             = gbeps*gbtab[gbitab+2];
 523             Heps2            = gbeps*gbeps*gbtab[gbitab+3];
 524             Fp               = F+Geps+Heps2;
 525             VV               = Y+gbeps*Fp;
 526             vgb              = gbqqfactor*VV;
 527             /*             #define INNERFLOPS INNERFLOPS+10 */
 528
 529             /*             #if 'Force' in KERNEL_VF */
 530             FF               = Fp+Geps+2.0*Heps2;
 531             fgb              = gbqqfactor*FF*gbscale;
 532             dvdatmp          = -0.5*(vgb+fgb*r{I}{J});
 533             dvdasum          = dvdasum + dvdatmp;
 534             dvda[jnr]        = dvdaj+dvdatmp*isaj{J}*isaj{J};
 535             /*                 #define INNERFLOPS INNERFLOPS+13 */
 536             /*             #endif */
 537             velec            = qq{I}{J}*rinv{I}{J};
 538             /*                 #define INNERFLOPS INNERFLOPS+1 */
 539             /*             #if 'Force' in KERNEL_VF */
 540             felec            = (velec*rinv{I}{J}-fgb)*rinv{I}{J};
 541             /*                 #define INNERFLOPS INNERFLOPS+3 */
 542             /*             #endif */
 543
 544             /*         #elif KERNEL_ELEC=='Ewald' */
 545             /* EWALD ELECTROSTATICS */
 546
 547             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 548             ewrt             = r{I}{J}*ewtabscale;
 549             ewitab           = ewrt;
 550             eweps            = ewrt-ewitab;
 551             /*             #define INNERFLOPS INNERFLOPS+2 */
 552             /*             #if 'Potential' in KERNEL_VF or KERNEL_MOD_ELEC=='PotentialSwitch' */
 553             ewitab           = 4*ewitab;
 554             felec            = ewtab[ewitab]+eweps*ewtab[ewitab+1];
 555             /*                 #define INNERFLOPS INNERFLOPS+4 */
 556             /*                 #if KERNEL_MOD_ELEC=='PotentialShift' */
 557             velec            = qq{I}{J}*((rinv{I}{J}-sh_ewald)-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
 558             /*                     #define INNERFLOPS INNERFLOPS+7 */
 559             /*                 #else */
 560             velec            = qq{I}{J}*(rinv{I}{J}-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
 561             /*                     #define INNERFLOPS INNERFLOPS+6 */
 562             /*                 #endif */
 563             /*                 #if 'Force' in KERNEL_VF */
 564             felec            = qq{I}{J}*rinv{I}{J}*(rinvsq{I}{J}-felec);
 565             /*                      #define INNERFLOPS INNERFLOPS+3 */
 566             /*                 #endif */
 567             /*             #elif KERNEL_VF=='Force' */
 568             felec            = (1.0-eweps)*ewtab[ewitab]+eweps*ewtab[ewitab+1];
 569             felec            = qq{I}{J}*rinv{I}{J}*(rinvsq{I}{J}-felec);
 570             /*                 #define INNERFLOPS INNERFLOPS+7 */
 571             /*             #endif */
 572
 573             /*         #elif KERNEL_ELEC=='CubicSplineTable' */
 574
 575             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 576             /*             #if 'Potential' in KERNEL_VF */
 577             Y                = vftab[vfitab];
 578             /*             #endif */
 579             F                = vftab[vfitab+1];
 580             Geps             = vfeps*vftab[vfitab+2];
 581             Heps2            = vfeps*vfeps*vftab[vfitab+3];
 582             Fp               = F+Geps+Heps2;
 583             /*             #define INNERFLOPS INNERFLOPS+5 */
 584             /*             #if 'Potential' in KERNEL_VF */
 585             VV               = Y+vfeps*Fp;
 586             velec            = qq{I}{J}*VV;
 587             /*                 #define INNERFLOPS INNERFLOPS+3 */
 588             /*             #endif */
 589             /*             #if 'Force' in KERNEL_VF */
 590             FF               = Fp+Geps+2.0*Heps2;
 591             felec            = -qq{I}{J}*FF*vftabscale*rinv{I}{J};
 592             /*                 #define INNERFLOPS INNERFLOPS+7 */
 593             /*             #endif */
 594             /*         #endif */
 595             /*         ## End of check for electrostatics interaction forms */
 596             /*     #endif */
 597             /*     ## END OF ELECTROSTATIC INTERACTION CHECK FOR PAIR I-J */
 598
 599             /*     #if 'vdw' in INTERACTION_FLAGS[I][J] */
 600
 601             /*         #if KERNEL_VDW=='LennardJones' */
 602
 603             /* LENNARD-JONES DISPERSION/REPULSION */
 604
 605             rinvsix          = rinvsq{I}{J}*rinvsq{I}{J}*rinvsq{I}{J};
 606             /*             #define INNERFLOPS INNERFLOPS+2 */
 607             /*             #if 'Potential' in KERNEL_VF or KERNEL_MOD_VDW=='PotentialSwitch' */
 608             vvdw6            = c6_{I}{J}*rinvsix;
 609             vvdw12           = c12_{I}{J}*rinvsix*rinvsix;
 610             /*                 #define INNERFLOPS INNERFLOPS+3 */
 611             /*                 #if KERNEL_MOD_VDW=='PotentialShift' */
 612             vvdw             = (vvdw12 - c12_{I}{J}*sh_vdw_invrcut6*sh_vdw_invrcut6)*(1.0/12.0) - (vvdw6 - c6_{I}{J}*sh_vdw_invrcut6)*(1.0/6.0);
 613             /*                     #define INNERFLOPS INNERFLOPS+8 */
 614             /*                 #else */
 615             vvdw             = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
 616             /*                     #define INNERFLOPS INNERFLOPS+3 */
 617             /*                 #endif */
 618             /*                 ## Check for force inside potential check, i.e. this means we already did the potential part */
 619             /*                 #if 'Force' in KERNEL_VF */
 620             fvdw             = (vvdw12-vvdw6)*rinvsq{I}{J};
 621             /*                     #define INNERFLOPS INNERFLOPS+2 */
 622             /*                 #endif */
 623             /*             #elif KERNEL_VF=='Force' */
 624             /*                 ## Force-only LennardJones makes it possible to save 1 flop (they do add up...) */
 625             fvdw             = (c12_{I}{J}*rinvsix-c6_{I}{J})*rinvsix*rinvsq{I}{J};
 626             /*                 #define INNERFLOPS INNERFLOPS+4 */
 627             /*             #endif */
 628
 629             /*         #elif KERNEL_VDW=='Buckingham' */
 630
 631             /* BUCKINGHAM DISPERSION/REPULSION */
 632             rinvsix          = rinvsq{I}{J}*rinvsq{I}{J}*rinvsq{I}{J};
 633             vvdw6            = c6_{I}{J}*rinvsix;
 634             br               = cexp2_{I}{J}*r{I}{J};
 635             vvdwexp          = cexp1_{I}{J}*exp(-br);
 636             /*             ## Estimate exp() to 25 flops */
 637             /*             #define INNERFLOPS INNERFLOPS+31 */
 638             /*             #if 'Potential' in KERNEL_VF or KERNEL_MOD_VDW=='PotentialSwitch'  */
 639             /*                 #if KERNEL_MOD_VDW=='PotentialShift' */
 640             vvdw             = (vvdwexp-cexp1_{I}{J}*exp(-cexp2_{I}{J}*rvdw)) - (vvdw6 - c6_{I}{J}*sh_vdw_invrcut6)*(1.0/6.0);
 641             /*                     #define INNERFLOPS INNERFLOPS+33 */
 642             /*                  #else */
 643             vvdw             = vvdwexp - vvdw6*(1.0/6.0);
 644             /*                     #define INNERFLOPS INNERFLOPS+2 */
 645             /*                 #endif */
 646             /*             #endif */
 647             /*             #if 'Force' in KERNEL_VF */
 648             fvdw             = (br*vvdwexp-vvdw6)*rinvsq{I}{J};
 649             /*                 #define INNERFLOPS INNERFLOPS+3 */
 650             /*             #endif */
 651
 652             /*         #elif KERNEL_VDW=='CubicSplineTable' */
 653
 654             /* CUBIC SPLINE TABLE DISPERSION */
 655             vfitab          += {TABLE_VDW_OFFSET};
 656             /*             #if 'Potential' in KERNEL_VF */
 657             Y                = vftab[vfitab];
 658             /*             #endif */
 659             F                = vftab[vfitab+1];
 660             Geps             = vfeps*vftab[vfitab+2];
 661             Heps2            = vfeps*vfeps*vftab[vfitab+3];
 662             Fp               = F+Geps+Heps2;
 663             /*             #define INNERFLOPS INNERFLOPS+5 */
 664             /*             #if 'Potential' in KERNEL_VF */
 665             VV               = Y+vfeps*Fp;
 666             vvdw6            = c6_{I}{J}*VV;
 667             /*                 #define INNERFLOPS INNERFLOPS+3 */
 668             /*             #endif */
 669             /*             #if 'Force' in KERNEL_VF */
 670             FF               = Fp+Geps+2.0*Heps2;
 671             fvdw6            = c6_{I}{J}*FF;
 672             /*                 #define INNERFLOPS INNERFLOPS+4 */
 673             /*             #endif */
 674
 675             /* CUBIC SPLINE TABLE REPULSION */
 676             /*             #if 'Potential' in KERNEL_VF */
 677             Y                = vftab[vfitab+4];
 678             /*             #endif */
 679             F                = vftab[vfitab+5];
 680             Geps             = vfeps*vftab[vfitab+6];
 681             Heps2            = vfeps*vfeps*vftab[vfitab+7];
 682             Fp               = F+Geps+Heps2;
 683             /*             #define INNERFLOPS INNERFLOPS+5 */
 684             /*             #if 'Potential' in KERNEL_VF */
 685             VV               = Y+vfeps*Fp;
 686             vvdw12           = c12_{I}{J}*VV;
 687             /*                 #define INNERFLOPS INNERFLOPS+3 */
 688             /*             #endif */
 689             /*             #if 'Force' in KERNEL_VF */
 690             FF               = Fp+Geps+2.0*Heps2;
 691             fvdw12           = c12_{I}{J}*FF;
 692             /*                 #define INNERFLOPS INNERFLOPS+4 */
 693             /*             #endif */
 694             /*             #if 'Potential' in KERNEL_VF */
 695             vvdw             = vvdw12+vvdw6;
 696             /*                 #define INNERFLOPS INNERFLOPS+1 */
 697             /*             #endif */
 698             /*             #if 'Force' in KERNEL_VF */
 699             fvdw             = -(fvdw6+fvdw12)*vftabscale*rinv{I}{J};
 700             /*                 #define INNERFLOPS INNERFLOPS+4 */
 701             /*             #endif */
 702             /*         #endif */
 703             /*         ## End of check for vdw interaction forms */
 704             /*     #endif */
 705             /*     ## END OF VDW INTERACTION CHECK FOR PAIR I-J */
 706
 707             /*     #if 'switch' in INTERACTION_FLAGS[I][J] */
 708             d                = r{I}{J}-rswitch;
 709             d                = (d>0.0) ? d : 0.0;
 710             d2               = d*d;
 711             sw               = 1.0+d2*d*(swV3+d*(swV4+d*swV5));
 712             /*         #define INNERFLOPS INNERFLOPS+9 */
 713
 714             /*         #if 'Force' in KERNEL_VF */
 715             dsw              = d2*(swF2+d*(swF3+d*swF4));
 716             /*             #define INNERFLOPS INNERFLOPS+5 */
 717             /*         #endif */
 718
 719             /* Evaluate switch function */
 720             /*         #if 'Force' in KERNEL_VF */
 721             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 722             /*             #if 'electrostatics' in INTERACTION_FLAGS[I][J] and KERNEL_MOD_ELEC=='PotentialSwitch' */
 723             felec            = felec*sw - rinv{I}{J}*velec*dsw;
 724             /*                 #define INNERFLOPS INNERFLOPS+3 */
 725             /*             #endif */
 726             /*             #if 'vdw' in INTERACTION_FLAGS[I][J] and KERNEL_MOD_VDW=='PotentialSwitch' */
 727             fvdw             = fvdw*sw - rinv{I}{J}*vvdw*dsw;
 728             /*                 #define INNERFLOPS INNERFLOPS+3 */
 729             /*             #endif */
 730             /*         #endif */
 731             /*         #if 'Potential' in KERNEL_VF */
 732             /*             #if 'electrostatics' in INTERACTION_FLAGS[I][J] and KERNEL_MOD_ELEC=='PotentialSwitch' */
 733             velec           *= sw;
 734             /*                 #define INNERFLOPS INNERFLOPS+1 */
 735             /*             #endif */
 736             /*             #if 'vdw' in INTERACTION_FLAGS[I][J] and KERNEL_MOD_VDW=='PotentialSwitch' */
 737             vvdw            *= sw;
 738             /*                 #define INNERFLOPS INNERFLOPS+1 */
 739             /*             #endif */
 740             /*         #endif */
 741             /*     #endif */
 742
 743             /*     #if 'Potential' in KERNEL_VF */
 744             /* Update potential sums from outer loop */
 745             /*         #if 'electrostatics' in INTERACTION_FLAGS[I][J] */
 746             velecsum        += velec;
 747             /*             #define INNERFLOPS INNERFLOPS+1 */
 748             /*             #if KERNEL_ELEC=='GeneralizedBorn' */
 749             vgbsum          += vgb;
 750             /*                 #define INNERFLOPS INNERFLOPS+1 */
 751             /*             #endif */
 752             /*         #endif */
 753             /*         #if 'vdw' in INTERACTION_FLAGS[I][J] */
 754             vvdwsum         += vvdw;
 755             /*             #define INNERFLOPS INNERFLOPS+1 */
 756             /*         #endif */
 757             /*     #endif */
 758
 759             /*     #if 'Force' in KERNEL_VF */
 760
 761             /*         #if 'electrostatics' in INTERACTION_FLAGS[I][J] and 'vdw' in INTERACTION_FLAGS[I][J] */
 762             fscal            = felec+fvdw;
 763             /*             #define INNERFLOPS INNERFLOPS+1 */
 764             /*         #elif 'electrostatics' in INTERACTION_FLAGS[I][J] */
 765             fscal            = felec;
 766             /*         #elif 'vdw' in INTERACTION_FLAGS[I][J] */
 767             fscal            = fvdw;
 768             /*        #endif */
 769
 770             /* Calculate temporary vectorial force */
 771             tx               = fscal*dx{I}{J};
 772             ty               = fscal*dy{I}{J};
 773             tz               = fscal*dz{I}{J};
 774
 775             /* Update vectorial force */
 776             fix{I}            += tx;
 777             fiy{I}            += ty;
 778             fiz{I}            += tz;
 779             f[j_coord_offset+DIM*{J}+XX] -= tx;
 780             f[j_coord_offset+DIM*{J}+YY] -= ty;
 781             f[j_coord_offset+DIM*{J}+ZZ] -= tz;
 782
 783             /*         #define INNERFLOPS INNERFLOPS+9 */
 784             /*     #endif */
 785
 786             /*     ## Note special check for TIP4P-TIP4P. Since we are cutting of all hydrogen interactions we also cut the LJ-only O-O interaction */
 787             /*     #if 'exactcutoff' in INTERACTION_FLAGS[I][J] or (GEOMETRY_I=='Water4' and GEOMETRY_J=='Water4' and 'exactcutoff' in INTERACTION_FLAGS[1][1]) */
 788             /*         #if 0    ## This and next two lines is a hack to maintain indentation in template file */
 789             {
 790                 /*     #endif */
 791             }
 792             /*     #endif */
 793             /*    ## End of check for the interaction being outside the cutoff */
 794
 795             /* #endfor */
 796             /* ## End of loop over i-j interaction pairs */
 797
 798             /* Inner loop uses {INNERFLOPS} flops */
 799         }
 800         /* End of innermost loop */
 801
 802         /* #if 'Force' in KERNEL_VF */
 803         tx = ty = tz = 0;
 804         /*     #for I in PARTICLES_I */
 805         f[i_coord_offset+DIM*{I}+XX] += fix{I};
 806         f[i_coord_offset+DIM*{I}+YY] += fiy{I};
 807         f[i_coord_offset+DIM*{I}+ZZ] += fiz{I};
 808         tx                         += fix{I};
 809         ty                         += fiy{I};
 810         tz                         += fiz{I};
 811         /*         #define OUTERFLOPS OUTERFLOPS+6 */
 812         /*     #endfor */
 813         fshift[i_shift_offset+XX]  += tx;
 814         fshift[i_shift_offset+YY]  += ty;
 815         fshift[i_shift_offset+ZZ]  += tz;
 816         /*     #define OUTERFLOPS OUTERFLOPS+3 */
 817         /* #endif */
 818
 819         /* #if 'Potential' in KERNEL_VF */
 820         ggid                        = gid[iidx];
 821         /* Update potential energies */
 822         /*     #if KERNEL_ELEC != 'None' */
 823         kernel_data->energygrp_elec[ggid] += velecsum;
 824         /*         #define OUTERFLOPS OUTERFLOPS+1 */
 825         /*     #endif */
 826         /*     #if 'GeneralizedBorn' in KERNEL_ELEC */
 827         kernel_data->energygrp_polarization[ggid] += vgbsum;
 828         /*         #define OUTERFLOPS OUTERFLOPS+1 */
 829         /*     #endif */
 830         /*     #if KERNEL_VDW != 'None' */
 831         kernel_data->energygrp_vdw[ggid] += vvdwsum;
 832         /*         #define OUTERFLOPS OUTERFLOPS+1 */
 833         /*     #endif */
 834         /* #endif */
 835         /*     #if 'GeneralizedBorn' in KERNEL_ELEC and 'Force' in KERNEL_VF */
 836         dvda[inr]                   = dvda[inr] + dvdasum*isai{I}*isai{I};
 837         /*     #endif */
 838
 839         /* Increment number of inner iterations */
 840         inneriter                  += j_index_end - j_index_start;
 841
 842         /* Outer loop uses {OUTERFLOPS} flops */
 843     }
 844
 845     /* Increment number of outer iterations */
 846     outeriter        += nri;
 847
 848     /* Update outer/inner flops */
 849     /* ## NB: This is not important, it just affects the flopcount. However, since our preprocessor is */
 850     /* ## primitive and replaces aggressively even in strings inside these directives, we need to      */
 851     /* ## assemble the main part of the name (containing KERNEL/ELEC/VDW) directly in the source.      */
 852     /* #if GEOMETRY_I == 'Water3'            */
 853     /*     #define ISUFFIX '_W3'             */
 854     /* #elif GEOMETRY_I == 'Water4'          */
 855     /*     #define ISUFFIX '_W4'             */
 856     /* #else                                 */
 857     /*     #define ISUFFIX ''                */
 858     /* #endif                                */
 859     /* #if GEOMETRY_J == 'Water3'            */
 860     /*     #define JSUFFIX 'W3'              */
 861     /* #elif GEOMETRY_J == 'Water4'          */
 862     /*     #define JSUFFIX 'W4'              */
 863     /* #else                                 */
 864     /*     #define JSUFFIX ''                */
 865     /* #endif                                */
 866     /* #if 'PotentialAndForce' in KERNEL_VF  */
 867     /*     #define VFSUFFIX  '_VF'           */
 868     /* #elif 'Potential' in KERNEL_VF        */
 869     /*     #define VFSUFFIX '_V'             */
 870     /* #else                                 */
 871     /*     #define VFSUFFIX '_F'             */
 872     /* #endif                                */
 873
 874     /* #if KERNEL_ELEC != 'None' and KERNEL_VDW != 'None' */
 875     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW{ISUFFIX}{JSUFFIX}{VFSUFFIX},outeriter*{OUTERFLOPS} + inneriter*{INNERFLOPS});
 876     /* #elif KERNEL_ELEC != 'None' */
 877     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC{ISUFFIX}{JSUFFIX}{VFSUFFIX},outeriter*{OUTERFLOPS} + inneriter*{INNERFLOPS});
 878     /* #else */
 879     inc_nrnb(nrnb,eNR_NBKERNEL_VDW{ISUFFIX}{JSUFFIX}{VFSUFFIX},outeriter*{OUTERFLOPS} + inneriter*{INNERFLOPS});
 880     /* #endif  */
 881 }