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 "gromacs/legacyheaders/types/simple.h"
44 #include "gromacs/math/vec.h"
45 #include "gromacs/legacyheaders/nrnb.h"
48 * Gromacs nonbonded kernel: nb_kernel_ElecRFCut_VdwCSTab_GeomP1P1_VF_c
49 * Electrostatics interaction: ReactionField
50 * VdW interaction: CubicSplineTable
51 * Geometry: Particle-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecRFCut_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;
90 jindex = nlist->jindex;
92 shiftidx = nlist->shift;
94 shiftvec = fr->shift_vec[0];
95 fshift = fr->fshift[0];
97 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 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
109 rcutoff = fr->rcoulomb;
110 rcutoff2 = rcutoff*rcutoff;
115 /* Start outer loop over neighborlists */
116 for(iidx=0; iidx<nri; iidx++)
118 /* Load shift vector for this list */
119 i_shift_offset = DIM*shiftidx[iidx];
120 shX = shiftvec[i_shift_offset+XX];
121 shY = shiftvec[i_shift_offset+YY];
122 shZ = shiftvec[i_shift_offset+ZZ];
124 /* Load limits for loop over neighbors */
125 j_index_start = jindex[iidx];
126 j_index_end = jindex[iidx+1];
128 /* Get outer coordinate index */
130 i_coord_offset = DIM*inr;
132 /* Load i particle coords and add shift vector */
133 ix0 = shX + x[i_coord_offset+DIM*0+XX];
134 iy0 = shY + x[i_coord_offset+DIM*0+YY];
135 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
141 /* Load parameters for i particles */
142 iq0 = facel*charge[inr+0];
143 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
145 /* Reset potential sums */
149 /* Start inner kernel loop */
150 for(jidx=j_index_start; jidx<j_index_end; jidx++)
152 /* Get j neighbor index, and coordinate index */
154 j_coord_offset = DIM*jnr;
156 /* load j atom coordinates */
157 jx0 = x[j_coord_offset+DIM*0+XX];
158 jy0 = x[j_coord_offset+DIM*0+YY];
159 jz0 = x[j_coord_offset+DIM*0+ZZ];
161 /* Calculate displacement vector */
166 /* Calculate squared distance and things based on it */
167 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
169 rinv00 = gmx_invsqrt(rsq00);
171 rinvsq00 = rinv00*rinv00;
173 /* Load parameters for j particles */
175 vdwjidx0 = 2*vdwtype[jnr+0];
177 /**************************
178 * CALCULATE INTERACTIONS *
179 **************************/
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 /* REACTION-FIELD ELECTROSTATICS */
197 velec = qq00*(rinv00+krf*rsq00-crf);
198 felec = qq00*(rinv00*rinvsq00-krf2);
200 /* CUBIC SPLINE TABLE DISPERSION */
204 Geps = vfeps*vftab[vfitab+2];
205 Heps2 = vfeps*vfeps*vftab[vfitab+3];
209 FF = Fp+Geps+2.0*Heps2;
212 /* CUBIC SPLINE TABLE REPULSION */
215 Geps = vfeps*vftab[vfitab+6];
216 Heps2 = vfeps*vfeps*vftab[vfitab+7];
220 FF = Fp+Geps+2.0*Heps2;
223 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
225 /* Update potential sums from outer loop */
231 /* Calculate temporary vectorial force */
236 /* Update vectorial force */
240 f[j_coord_offset+DIM*0+XX] -= tx;
241 f[j_coord_offset+DIM*0+YY] -= ty;
242 f[j_coord_offset+DIM*0+ZZ] -= tz;
246 /* Inner loop uses 66 flops */
248 /* End of innermost loop */
251 f[i_coord_offset+DIM*0+XX] += fix0;
252 f[i_coord_offset+DIM*0+YY] += fiy0;
253 f[i_coord_offset+DIM*0+ZZ] += fiz0;
257 fshift[i_shift_offset+XX] += tx;
258 fshift[i_shift_offset+YY] += ty;
259 fshift[i_shift_offset+ZZ] += tz;
262 /* Update potential energies */
263 kernel_data->energygrp_elec[ggid] += velecsum;
264 kernel_data->energygrp_vdw[ggid] += vvdwsum;
266 /* Increment number of inner iterations */
267 inneriter += j_index_end - j_index_start;
269 /* Outer loop uses 15 flops */
272 /* Increment number of outer iterations */
275 /* Update outer/inner flops */
277 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*66);
280 * Gromacs nonbonded kernel: nb_kernel_ElecRFCut_VdwCSTab_GeomP1P1_F_c
281 * Electrostatics interaction: ReactionField
282 * VdW interaction: CubicSplineTable
283 * Geometry: Particle-Particle
284 * Calculate force/pot: Force
287 nb_kernel_ElecRFCut_VdwCSTab_GeomP1P1_F_c
288 (t_nblist * gmx_restrict nlist,
289 rvec * gmx_restrict xx,
290 rvec * gmx_restrict ff,
291 t_forcerec * gmx_restrict fr,
292 t_mdatoms * gmx_restrict mdatoms,
293 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
294 t_nrnb * gmx_restrict nrnb)
296 int i_shift_offset,i_coord_offset,j_coord_offset;
297 int j_index_start,j_index_end;
298 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
299 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
300 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
301 real *shiftvec,*fshift,*x,*f;
303 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
305 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
306 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
307 real velec,felec,velecsum,facel,crf,krf,krf2;
310 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
314 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
322 jindex = nlist->jindex;
324 shiftidx = nlist->shift;
326 shiftvec = fr->shift_vec[0];
327 fshift = fr->fshift[0];
329 charge = mdatoms->chargeA;
333 nvdwtype = fr->ntype;
335 vdwtype = mdatoms->typeA;
337 vftab = kernel_data->table_vdw->data;
338 vftabscale = kernel_data->table_vdw->scale;
340 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
341 rcutoff = fr->rcoulomb;
342 rcutoff2 = rcutoff*rcutoff;
347 /* Start outer loop over neighborlists */
348 for(iidx=0; iidx<nri; iidx++)
350 /* Load shift vector for this list */
351 i_shift_offset = DIM*shiftidx[iidx];
352 shX = shiftvec[i_shift_offset+XX];
353 shY = shiftvec[i_shift_offset+YY];
354 shZ = shiftvec[i_shift_offset+ZZ];
356 /* Load limits for loop over neighbors */
357 j_index_start = jindex[iidx];
358 j_index_end = jindex[iidx+1];
360 /* Get outer coordinate index */
362 i_coord_offset = DIM*inr;
364 /* Load i particle coords and add shift vector */
365 ix0 = shX + x[i_coord_offset+DIM*0+XX];
366 iy0 = shY + x[i_coord_offset+DIM*0+YY];
367 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
373 /* Load parameters for i particles */
374 iq0 = facel*charge[inr+0];
375 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
377 /* Start inner kernel loop */
378 for(jidx=j_index_start; jidx<j_index_end; jidx++)
380 /* Get j neighbor index, and coordinate index */
382 j_coord_offset = DIM*jnr;
384 /* load j atom coordinates */
385 jx0 = x[j_coord_offset+DIM*0+XX];
386 jy0 = x[j_coord_offset+DIM*0+YY];
387 jz0 = x[j_coord_offset+DIM*0+ZZ];
389 /* Calculate displacement vector */
394 /* Calculate squared distance and things based on it */
395 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
397 rinv00 = gmx_invsqrt(rsq00);
399 rinvsq00 = rinv00*rinv00;
401 /* Load parameters for j particles */
403 vdwjidx0 = 2*vdwtype[jnr+0];
405 /**************************
406 * CALCULATE INTERACTIONS *
407 **************************/
415 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
416 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
418 /* Calculate table index by multiplying r with table scale and truncate to integer */
424 /* REACTION-FIELD ELECTROSTATICS */
425 felec = qq00*(rinv00*rinvsq00-krf2);
427 /* CUBIC SPLINE TABLE DISPERSION */
430 Geps = vfeps*vftab[vfitab+2];
431 Heps2 = vfeps*vfeps*vftab[vfitab+3];
433 FF = Fp+Geps+2.0*Heps2;
436 /* CUBIC SPLINE TABLE REPULSION */
438 Geps = vfeps*vftab[vfitab+6];
439 Heps2 = vfeps*vfeps*vftab[vfitab+7];
441 FF = Fp+Geps+2.0*Heps2;
443 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
447 /* Calculate temporary vectorial force */
452 /* Update vectorial force */
456 f[j_coord_offset+DIM*0+XX] -= tx;
457 f[j_coord_offset+DIM*0+YY] -= ty;
458 f[j_coord_offset+DIM*0+ZZ] -= tz;
462 /* Inner loop uses 53 flops */
464 /* End of innermost loop */
467 f[i_coord_offset+DIM*0+XX] += fix0;
468 f[i_coord_offset+DIM*0+YY] += fiy0;
469 f[i_coord_offset+DIM*0+ZZ] += fiz0;
473 fshift[i_shift_offset+XX] += tx;
474 fshift[i_shift_offset+YY] += ty;
475 fshift[i_shift_offset+ZZ] += tz;
477 /* Increment number of inner iterations */
478 inneriter += j_index_end - j_index_start;
480 /* Outer loop uses 13 flops */
483 /* Increment number of outer iterations */
486 /* Update outer/inner flops */
488 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*53);