2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013, 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.
44 #include "../nb_kernel.h"
45 #include "types/simple.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomP1P1_VF_c
51 * Electrostatics interaction: Ewald
52 * VdW interaction: CubicSplineTable
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecEw_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)
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;
73 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
75 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
76 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
77 real velec,felec,velecsum,facel,crf,krf,krf2;
80 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
84 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
87 real ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
95 jindex = nlist->jindex;
97 shiftidx = nlist->shift;
99 shiftvec = fr->shift_vec[0];
100 fshift = fr->fshift[0];
102 charge = mdatoms->chargeA;
103 nvdwtype = fr->ntype;
105 vdwtype = mdatoms->typeA;
107 vftab = kernel_data->table_vdw->data;
108 vftabscale = kernel_data->table_vdw->scale;
110 sh_ewald = fr->ic->sh_ewald;
111 ewtab = fr->ic->tabq_coul_FDV0;
112 ewtabscale = fr->ic->tabq_scale;
113 ewtabhalfspace = 0.5/ewtabscale;
118 /* Start outer loop over neighborlists */
119 for(iidx=0; iidx<nri; iidx++)
121 /* Load shift vector for this list */
122 i_shift_offset = DIM*shiftidx[iidx];
123 shX = shiftvec[i_shift_offset+XX];
124 shY = shiftvec[i_shift_offset+YY];
125 shZ = shiftvec[i_shift_offset+ZZ];
127 /* Load limits for loop over neighbors */
128 j_index_start = jindex[iidx];
129 j_index_end = jindex[iidx+1];
131 /* Get outer coordinate index */
133 i_coord_offset = DIM*inr;
135 /* Load i particle coords and add shift vector */
136 ix0 = shX + x[i_coord_offset+DIM*0+XX];
137 iy0 = shY + x[i_coord_offset+DIM*0+YY];
138 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
144 /* Load parameters for i particles */
145 iq0 = facel*charge[inr+0];
146 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
148 /* Reset potential sums */
152 /* Start inner kernel loop */
153 for(jidx=j_index_start; jidx<j_index_end; jidx++)
155 /* Get j neighbor index, and coordinate index */
157 j_coord_offset = DIM*jnr;
159 /* load j atom coordinates */
160 jx0 = x[j_coord_offset+DIM*0+XX];
161 jy0 = x[j_coord_offset+DIM*0+YY];
162 jz0 = x[j_coord_offset+DIM*0+ZZ];
164 /* Calculate displacement vector */
169 /* Calculate squared distance and things based on it */
170 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
172 rinv00 = gmx_invsqrt(rsq00);
174 rinvsq00 = rinv00*rinv00;
176 /* Load parameters for j particles */
178 vdwjidx0 = 2*vdwtype[jnr+0];
180 /**************************
181 * CALCULATE INTERACTIONS *
182 **************************/
187 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
188 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
190 /* Calculate table index by multiplying r with table scale and truncate to integer */
196 /* EWALD ELECTROSTATICS */
198 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
199 ewrt = r00*ewtabscale;
203 felec = ewtab[ewitab]+eweps*ewtab[ewitab+1];
204 velec = qq00*(rinv00-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
205 felec = qq00*rinv00*(rinvsq00-felec);
207 /* CUBIC SPLINE TABLE DISPERSION */
211 Geps = vfeps*vftab[vfitab+2];
212 Heps2 = vfeps*vfeps*vftab[vfitab+3];
216 FF = Fp+Geps+2.0*Heps2;
219 /* CUBIC SPLINE TABLE REPULSION */
222 Geps = vfeps*vftab[vfitab+6];
223 Heps2 = vfeps*vfeps*vftab[vfitab+7];
227 FF = Fp+Geps+2.0*Heps2;
230 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
232 /* Update potential sums from outer loop */
238 /* Calculate temporary vectorial force */
243 /* Update vectorial force */
247 f[j_coord_offset+DIM*0+XX] -= tx;
248 f[j_coord_offset+DIM*0+YY] -= ty;
249 f[j_coord_offset+DIM*0+ZZ] -= tz;
251 /* Inner loop uses 74 flops */
253 /* End of innermost loop */
256 f[i_coord_offset+DIM*0+XX] += fix0;
257 f[i_coord_offset+DIM*0+YY] += fiy0;
258 f[i_coord_offset+DIM*0+ZZ] += fiz0;
262 fshift[i_shift_offset+XX] += tx;
263 fshift[i_shift_offset+YY] += ty;
264 fshift[i_shift_offset+ZZ] += tz;
267 /* Update potential energies */
268 kernel_data->energygrp_elec[ggid] += velecsum;
269 kernel_data->energygrp_vdw[ggid] += vvdwsum;
271 /* Increment number of inner iterations */
272 inneriter += j_index_end - j_index_start;
274 /* Outer loop uses 15 flops */
277 /* Increment number of outer iterations */
280 /* Update outer/inner flops */
282 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*74);
285 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomP1P1_F_c
286 * Electrostatics interaction: Ewald
287 * VdW interaction: CubicSplineTable
288 * Geometry: Particle-Particle
289 * Calculate force/pot: Force
292 nb_kernel_ElecEw_VdwCSTab_GeomP1P1_F_c
293 (t_nblist * gmx_restrict nlist,
294 rvec * gmx_restrict xx,
295 rvec * gmx_restrict ff,
296 t_forcerec * gmx_restrict fr,
297 t_mdatoms * gmx_restrict mdatoms,
298 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
299 t_nrnb * gmx_restrict nrnb)
301 int i_shift_offset,i_coord_offset,j_coord_offset;
302 int j_index_start,j_index_end;
303 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
304 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
305 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
306 real *shiftvec,*fshift,*x,*f;
308 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
310 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
311 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
312 real velec,felec,velecsum,facel,crf,krf,krf2;
315 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
319 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
322 real ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
330 jindex = nlist->jindex;
332 shiftidx = nlist->shift;
334 shiftvec = fr->shift_vec[0];
335 fshift = fr->fshift[0];
337 charge = mdatoms->chargeA;
338 nvdwtype = fr->ntype;
340 vdwtype = mdatoms->typeA;
342 vftab = kernel_data->table_vdw->data;
343 vftabscale = kernel_data->table_vdw->scale;
345 sh_ewald = fr->ic->sh_ewald;
346 ewtab = fr->ic->tabq_coul_F;
347 ewtabscale = fr->ic->tabq_scale;
348 ewtabhalfspace = 0.5/ewtabscale;
353 /* Start outer loop over neighborlists */
354 for(iidx=0; iidx<nri; iidx++)
356 /* Load shift vector for this list */
357 i_shift_offset = DIM*shiftidx[iidx];
358 shX = shiftvec[i_shift_offset+XX];
359 shY = shiftvec[i_shift_offset+YY];
360 shZ = shiftvec[i_shift_offset+ZZ];
362 /* Load limits for loop over neighbors */
363 j_index_start = jindex[iidx];
364 j_index_end = jindex[iidx+1];
366 /* Get outer coordinate index */
368 i_coord_offset = DIM*inr;
370 /* Load i particle coords and add shift vector */
371 ix0 = shX + x[i_coord_offset+DIM*0+XX];
372 iy0 = shY + x[i_coord_offset+DIM*0+YY];
373 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
379 /* Load parameters for i particles */
380 iq0 = facel*charge[inr+0];
381 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
383 /* Start inner kernel loop */
384 for(jidx=j_index_start; jidx<j_index_end; jidx++)
386 /* Get j neighbor index, and coordinate index */
388 j_coord_offset = DIM*jnr;
390 /* load j atom coordinates */
391 jx0 = x[j_coord_offset+DIM*0+XX];
392 jy0 = x[j_coord_offset+DIM*0+YY];
393 jz0 = x[j_coord_offset+DIM*0+ZZ];
395 /* Calculate displacement vector */
400 /* Calculate squared distance and things based on it */
401 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
403 rinv00 = gmx_invsqrt(rsq00);
405 rinvsq00 = rinv00*rinv00;
407 /* Load parameters for j particles */
409 vdwjidx0 = 2*vdwtype[jnr+0];
411 /**************************
412 * CALCULATE INTERACTIONS *
413 **************************/
418 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
419 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
421 /* Calculate table index by multiplying r with table scale and truncate to integer */
427 /* EWALD ELECTROSTATICS */
429 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
430 ewrt = r00*ewtabscale;
433 felec = (1.0-eweps)*ewtab[ewitab]+eweps*ewtab[ewitab+1];
434 felec = qq00*rinv00*(rinvsq00-felec);
436 /* CUBIC SPLINE TABLE DISPERSION */
439 Geps = vfeps*vftab[vfitab+2];
440 Heps2 = vfeps*vfeps*vftab[vfitab+3];
442 FF = Fp+Geps+2.0*Heps2;
445 /* CUBIC SPLINE TABLE REPULSION */
447 Geps = vfeps*vftab[vfitab+6];
448 Heps2 = vfeps*vfeps*vftab[vfitab+7];
450 FF = Fp+Geps+2.0*Heps2;
452 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
456 /* Calculate temporary vectorial force */
461 /* Update vectorial force */
465 f[j_coord_offset+DIM*0+XX] -= tx;
466 f[j_coord_offset+DIM*0+YY] -= ty;
467 f[j_coord_offset+DIM*0+ZZ] -= tz;
469 /* Inner loop uses 59 flops */
471 /* End of innermost loop */
474 f[i_coord_offset+DIM*0+XX] += fix0;
475 f[i_coord_offset+DIM*0+YY] += fiy0;
476 f[i_coord_offset+DIM*0+ZZ] += fiz0;
480 fshift[i_shift_offset+XX] += tx;
481 fshift[i_shift_offset+YY] += ty;
482 fshift[i_shift_offset+ZZ] += tz;
484 /* Increment number of inner iterations */
485 inneriter += j_index_end - j_index_start;
487 /* Outer loop uses 13 flops */
490 /* Increment number of outer iterations */
493 /* Update outer/inner flops */
495 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*59);