2 * This file is part of the GROMACS molecular simulation package.
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.
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.
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.
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.
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.
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.
36 * Note: this file was generated by the GROMACS c kernel generator.
42 #include "../nb_kernel.h"
43 #include "types/simple.h"
44 #include "gromacs/math/vec.h"
48 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomP1P1_VF_c
49 * Electrostatics interaction: Ewald
50 * VdW interaction: CubicSplineTable
51 * Geometry: Particle-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecEw_VdwCSTab_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)
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;
71 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
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;
78 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
82 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
85 real ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
93 jindex = nlist->jindex;
95 shiftidx = nlist->shift;
97 shiftvec = fr->shift_vec[0];
98 fshift = fr->fshift[0];
100 charge = mdatoms->chargeA;
101 nvdwtype = fr->ntype;
103 vdwtype = mdatoms->typeA;
105 vftab = kernel_data->table_vdw->data;
106 vftabscale = kernel_data->table_vdw->scale;
108 sh_ewald = fr->ic->sh_ewald;
109 ewtab = fr->ic->tabq_coul_FDV0;
110 ewtabscale = fr->ic->tabq_scale;
111 ewtabhalfspace = 0.5/ewtabscale;
116 /* Start outer loop over neighborlists */
117 for(iidx=0; iidx<nri; iidx++)
119 /* Load shift vector for this list */
120 i_shift_offset = DIM*shiftidx[iidx];
121 shX = shiftvec[i_shift_offset+XX];
122 shY = shiftvec[i_shift_offset+YY];
123 shZ = shiftvec[i_shift_offset+ZZ];
125 /* Load limits for loop over neighbors */
126 j_index_start = jindex[iidx];
127 j_index_end = jindex[iidx+1];
129 /* Get outer coordinate index */
131 i_coord_offset = DIM*inr;
133 /* Load i particle coords and add shift vector */
134 ix0 = shX + x[i_coord_offset+DIM*0+XX];
135 iy0 = shY + x[i_coord_offset+DIM*0+YY];
136 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
142 /* Load parameters for i particles */
143 iq0 = facel*charge[inr+0];
144 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
146 /* Reset potential sums */
150 /* Start inner kernel loop */
151 for(jidx=j_index_start; jidx<j_index_end; jidx++)
153 /* Get j neighbor index, and coordinate index */
155 j_coord_offset = DIM*jnr;
157 /* load j atom coordinates */
158 jx0 = x[j_coord_offset+DIM*0+XX];
159 jy0 = x[j_coord_offset+DIM*0+YY];
160 jz0 = x[j_coord_offset+DIM*0+ZZ];
162 /* Calculate displacement vector */
167 /* Calculate squared distance and things based on it */
168 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
170 rinv00 = gmx_invsqrt(rsq00);
172 rinvsq00 = rinv00*rinv00;
174 /* Load parameters for j particles */
176 vdwjidx0 = 2*vdwtype[jnr+0];
178 /**************************
179 * CALCULATE INTERACTIONS *
180 **************************/
185 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
186 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
188 /* Calculate table index by multiplying r with table scale and truncate to integer */
194 /* EWALD ELECTROSTATICS */
196 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
197 ewrt = r00*ewtabscale;
201 felec = ewtab[ewitab]+eweps*ewtab[ewitab+1];
202 velec = qq00*(rinv00-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
203 felec = qq00*rinv00*(rinvsq00-felec);
205 /* CUBIC SPLINE TABLE DISPERSION */
209 Geps = vfeps*vftab[vfitab+2];
210 Heps2 = vfeps*vfeps*vftab[vfitab+3];
214 FF = Fp+Geps+2.0*Heps2;
217 /* CUBIC SPLINE TABLE REPULSION */
220 Geps = vfeps*vftab[vfitab+6];
221 Heps2 = vfeps*vfeps*vftab[vfitab+7];
225 FF = Fp+Geps+2.0*Heps2;
228 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
230 /* Update potential sums from outer loop */
236 /* Calculate temporary vectorial force */
241 /* Update vectorial force */
245 f[j_coord_offset+DIM*0+XX] -= tx;
246 f[j_coord_offset+DIM*0+YY] -= ty;
247 f[j_coord_offset+DIM*0+ZZ] -= tz;
249 /* Inner loop uses 74 flops */
251 /* End of innermost loop */
254 f[i_coord_offset+DIM*0+XX] += fix0;
255 f[i_coord_offset+DIM*0+YY] += fiy0;
256 f[i_coord_offset+DIM*0+ZZ] += fiz0;
260 fshift[i_shift_offset+XX] += tx;
261 fshift[i_shift_offset+YY] += ty;
262 fshift[i_shift_offset+ZZ] += tz;
265 /* Update potential energies */
266 kernel_data->energygrp_elec[ggid] += velecsum;
267 kernel_data->energygrp_vdw[ggid] += vvdwsum;
269 /* Increment number of inner iterations */
270 inneriter += j_index_end - j_index_start;
272 /* Outer loop uses 15 flops */
275 /* Increment number of outer iterations */
278 /* Update outer/inner flops */
280 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*74);
283 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomP1P1_F_c
284 * Electrostatics interaction: Ewald
285 * VdW interaction: CubicSplineTable
286 * Geometry: Particle-Particle
287 * Calculate force/pot: Force
290 nb_kernel_ElecEw_VdwCSTab_GeomP1P1_F_c
291 (t_nblist * gmx_restrict nlist,
292 rvec * gmx_restrict xx,
293 rvec * gmx_restrict ff,
294 t_forcerec * gmx_restrict fr,
295 t_mdatoms * gmx_restrict mdatoms,
296 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
297 t_nrnb * gmx_restrict nrnb)
299 int i_shift_offset,i_coord_offset,j_coord_offset;
300 int j_index_start,j_index_end;
301 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
302 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
303 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
304 real *shiftvec,*fshift,*x,*f;
306 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
308 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
309 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
310 real velec,felec,velecsum,facel,crf,krf,krf2;
313 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
317 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
320 real ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
328 jindex = nlist->jindex;
330 shiftidx = nlist->shift;
332 shiftvec = fr->shift_vec[0];
333 fshift = fr->fshift[0];
335 charge = mdatoms->chargeA;
336 nvdwtype = fr->ntype;
338 vdwtype = mdatoms->typeA;
340 vftab = kernel_data->table_vdw->data;
341 vftabscale = kernel_data->table_vdw->scale;
343 sh_ewald = fr->ic->sh_ewald;
344 ewtab = fr->ic->tabq_coul_F;
345 ewtabscale = fr->ic->tabq_scale;
346 ewtabhalfspace = 0.5/ewtabscale;
351 /* Start outer loop over neighborlists */
352 for(iidx=0; iidx<nri; iidx++)
354 /* Load shift vector for this list */
355 i_shift_offset = DIM*shiftidx[iidx];
356 shX = shiftvec[i_shift_offset+XX];
357 shY = shiftvec[i_shift_offset+YY];
358 shZ = shiftvec[i_shift_offset+ZZ];
360 /* Load limits for loop over neighbors */
361 j_index_start = jindex[iidx];
362 j_index_end = jindex[iidx+1];
364 /* Get outer coordinate index */
366 i_coord_offset = DIM*inr;
368 /* Load i particle coords and add shift vector */
369 ix0 = shX + x[i_coord_offset+DIM*0+XX];
370 iy0 = shY + x[i_coord_offset+DIM*0+YY];
371 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
377 /* Load parameters for i particles */
378 iq0 = facel*charge[inr+0];
379 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
381 /* Start inner kernel loop */
382 for(jidx=j_index_start; jidx<j_index_end; jidx++)
384 /* Get j neighbor index, and coordinate index */
386 j_coord_offset = DIM*jnr;
388 /* load j atom coordinates */
389 jx0 = x[j_coord_offset+DIM*0+XX];
390 jy0 = x[j_coord_offset+DIM*0+YY];
391 jz0 = x[j_coord_offset+DIM*0+ZZ];
393 /* Calculate displacement vector */
398 /* Calculate squared distance and things based on it */
399 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
401 rinv00 = gmx_invsqrt(rsq00);
403 rinvsq00 = rinv00*rinv00;
405 /* Load parameters for j particles */
407 vdwjidx0 = 2*vdwtype[jnr+0];
409 /**************************
410 * CALCULATE INTERACTIONS *
411 **************************/
416 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
417 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
419 /* Calculate table index by multiplying r with table scale and truncate to integer */
425 /* EWALD ELECTROSTATICS */
427 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
428 ewrt = r00*ewtabscale;
431 felec = (1.0-eweps)*ewtab[ewitab]+eweps*ewtab[ewitab+1];
432 felec = qq00*rinv00*(rinvsq00-felec);
434 /* CUBIC SPLINE TABLE DISPERSION */
437 Geps = vfeps*vftab[vfitab+2];
438 Heps2 = vfeps*vfeps*vftab[vfitab+3];
440 FF = Fp+Geps+2.0*Heps2;
443 /* CUBIC SPLINE TABLE REPULSION */
445 Geps = vfeps*vftab[vfitab+6];
446 Heps2 = vfeps*vfeps*vftab[vfitab+7];
448 FF = Fp+Geps+2.0*Heps2;
450 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
454 /* Calculate temporary vectorial force */
459 /* Update vectorial force */
463 f[j_coord_offset+DIM*0+XX] -= tx;
464 f[j_coord_offset+DIM*0+YY] -= ty;
465 f[j_coord_offset+DIM*0+ZZ] -= tz;
467 /* Inner loop uses 59 flops */
469 /* End of innermost loop */
472 f[i_coord_offset+DIM*0+XX] += fix0;
473 f[i_coord_offset+DIM*0+YY] += fiy0;
474 f[i_coord_offset+DIM*0+ZZ] += fiz0;
478 fshift[i_shift_offset+XX] += tx;
479 fshift[i_shift_offset+YY] += ty;
480 fshift[i_shift_offset+ZZ] += tz;
482 /* Increment number of inner iterations */
483 inneriter += j_index_end - j_index_start;
485 /* Outer loop uses 13 flops */
488 /* Increment number of outer iterations */
491 /* Update outer/inner flops */
493 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*59);