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.
44 #include "../nb_kernel.h"
45 #include "gromacs/legacyheaders/types/simple.h"
46 #include "gromacs/math/vec.h"
47 #include "gromacs/legacyheaders/nrnb.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomP1P1_VF_c
51 * Electrostatics interaction: CubicSplineTable
52 * VdW interaction: LennardJones
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecCSTab_VdwLJ_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;
92 jindex = nlist->jindex;
94 shiftidx = nlist->shift;
96 shiftvec = fr->shift_vec[0];
97 fshift = fr->fshift[0];
99 charge = mdatoms->chargeA;
100 nvdwtype = fr->ntype;
102 vdwtype = mdatoms->typeA;
104 vftab = kernel_data->table_elec->data;
105 vftabscale = kernel_data->table_elec->scale;
110 /* Start outer loop over neighborlists */
111 for(iidx=0; iidx<nri; iidx++)
113 /* Load shift vector for this list */
114 i_shift_offset = DIM*shiftidx[iidx];
115 shX = shiftvec[i_shift_offset+XX];
116 shY = shiftvec[i_shift_offset+YY];
117 shZ = shiftvec[i_shift_offset+ZZ];
119 /* Load limits for loop over neighbors */
120 j_index_start = jindex[iidx];
121 j_index_end = jindex[iidx+1];
123 /* Get outer coordinate index */
125 i_coord_offset = DIM*inr;
127 /* Load i particle coords and add shift vector */
128 ix0 = shX + x[i_coord_offset+DIM*0+XX];
129 iy0 = shY + x[i_coord_offset+DIM*0+YY];
130 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
136 /* Load parameters for i particles */
137 iq0 = facel*charge[inr+0];
138 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
140 /* Reset potential sums */
144 /* Start inner kernel loop */
145 for(jidx=j_index_start; jidx<j_index_end; jidx++)
147 /* Get j neighbor index, and coordinate index */
149 j_coord_offset = DIM*jnr;
151 /* load j atom coordinates */
152 jx0 = x[j_coord_offset+DIM*0+XX];
153 jy0 = x[j_coord_offset+DIM*0+YY];
154 jz0 = x[j_coord_offset+DIM*0+ZZ];
156 /* Calculate displacement vector */
161 /* Calculate squared distance and things based on it */
162 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
164 rinv00 = gmx_invsqrt(rsq00);
166 rinvsq00 = rinv00*rinv00;
168 /* Load parameters for j particles */
170 vdwjidx0 = 2*vdwtype[jnr+0];
172 /**************************
173 * CALCULATE INTERACTIONS *
174 **************************/
179 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
180 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
182 /* Calculate table index by multiplying r with table scale and truncate to integer */
188 /* CUBIC SPLINE TABLE ELECTROSTATICS */
191 Geps = vfeps*vftab[vfitab+2];
192 Heps2 = vfeps*vfeps*vftab[vfitab+3];
196 FF = Fp+Geps+2.0*Heps2;
197 felec = -qq00*FF*vftabscale*rinv00;
199 /* LENNARD-JONES DISPERSION/REPULSION */
201 rinvsix = rinvsq00*rinvsq00*rinvsq00;
202 vvdw6 = c6_00*rinvsix;
203 vvdw12 = c12_00*rinvsix*rinvsix;
204 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
205 fvdw = (vvdw12-vvdw6)*rinvsq00;
207 /* Update potential sums from outer loop */
213 /* Calculate temporary vectorial force */
218 /* Update vectorial force */
222 f[j_coord_offset+DIM*0+XX] -= tx;
223 f[j_coord_offset+DIM*0+YY] -= ty;
224 f[j_coord_offset+DIM*0+ZZ] -= tz;
226 /* Inner loop uses 55 flops */
228 /* End of innermost loop */
231 f[i_coord_offset+DIM*0+XX] += fix0;
232 f[i_coord_offset+DIM*0+YY] += fiy0;
233 f[i_coord_offset+DIM*0+ZZ] += fiz0;
237 fshift[i_shift_offset+XX] += tx;
238 fshift[i_shift_offset+YY] += ty;
239 fshift[i_shift_offset+ZZ] += tz;
242 /* Update potential energies */
243 kernel_data->energygrp_elec[ggid] += velecsum;
244 kernel_data->energygrp_vdw[ggid] += vvdwsum;
246 /* Increment number of inner iterations */
247 inneriter += j_index_end - j_index_start;
249 /* Outer loop uses 15 flops */
252 /* Increment number of outer iterations */
255 /* Update outer/inner flops */
257 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*55);
260 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomP1P1_F_c
261 * Electrostatics interaction: CubicSplineTable
262 * VdW interaction: LennardJones
263 * Geometry: Particle-Particle
264 * Calculate force/pot: Force
267 nb_kernel_ElecCSTab_VdwLJ_GeomP1P1_F_c
268 (t_nblist * gmx_restrict nlist,
269 rvec * gmx_restrict xx,
270 rvec * gmx_restrict ff,
271 t_forcerec * gmx_restrict fr,
272 t_mdatoms * gmx_restrict mdatoms,
273 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
274 t_nrnb * gmx_restrict nrnb)
276 int i_shift_offset,i_coord_offset,j_coord_offset;
277 int j_index_start,j_index_end;
278 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
279 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
280 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
281 real *shiftvec,*fshift,*x,*f;
283 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
285 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
286 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
287 real velec,felec,velecsum,facel,crf,krf,krf2;
290 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
294 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
302 jindex = nlist->jindex;
304 shiftidx = nlist->shift;
306 shiftvec = fr->shift_vec[0];
307 fshift = fr->fshift[0];
309 charge = mdatoms->chargeA;
310 nvdwtype = fr->ntype;
312 vdwtype = mdatoms->typeA;
314 vftab = kernel_data->table_elec->data;
315 vftabscale = kernel_data->table_elec->scale;
320 /* Start outer loop over neighborlists */
321 for(iidx=0; iidx<nri; iidx++)
323 /* Load shift vector for this list */
324 i_shift_offset = DIM*shiftidx[iidx];
325 shX = shiftvec[i_shift_offset+XX];
326 shY = shiftvec[i_shift_offset+YY];
327 shZ = shiftvec[i_shift_offset+ZZ];
329 /* Load limits for loop over neighbors */
330 j_index_start = jindex[iidx];
331 j_index_end = jindex[iidx+1];
333 /* Get outer coordinate index */
335 i_coord_offset = DIM*inr;
337 /* Load i particle coords and add shift vector */
338 ix0 = shX + x[i_coord_offset+DIM*0+XX];
339 iy0 = shY + x[i_coord_offset+DIM*0+YY];
340 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
346 /* Load parameters for i particles */
347 iq0 = facel*charge[inr+0];
348 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
350 /* Start inner kernel loop */
351 for(jidx=j_index_start; jidx<j_index_end; jidx++)
353 /* Get j neighbor index, and coordinate index */
355 j_coord_offset = DIM*jnr;
357 /* load j atom coordinates */
358 jx0 = x[j_coord_offset+DIM*0+XX];
359 jy0 = x[j_coord_offset+DIM*0+YY];
360 jz0 = x[j_coord_offset+DIM*0+ZZ];
362 /* Calculate displacement vector */
367 /* Calculate squared distance and things based on it */
368 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
370 rinv00 = gmx_invsqrt(rsq00);
372 rinvsq00 = rinv00*rinv00;
374 /* Load parameters for j particles */
376 vdwjidx0 = 2*vdwtype[jnr+0];
378 /**************************
379 * CALCULATE INTERACTIONS *
380 **************************/
385 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
386 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
388 /* Calculate table index by multiplying r with table scale and truncate to integer */
394 /* CUBIC SPLINE TABLE ELECTROSTATICS */
396 Geps = vfeps*vftab[vfitab+2];
397 Heps2 = vfeps*vfeps*vftab[vfitab+3];
399 FF = Fp+Geps+2.0*Heps2;
400 felec = -qq00*FF*vftabscale*rinv00;
402 /* LENNARD-JONES DISPERSION/REPULSION */
404 rinvsix = rinvsq00*rinvsq00*rinvsq00;
405 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
409 /* Calculate temporary vectorial force */
414 /* Update vectorial force */
418 f[j_coord_offset+DIM*0+XX] -= tx;
419 f[j_coord_offset+DIM*0+YY] -= ty;
420 f[j_coord_offset+DIM*0+ZZ] -= tz;
422 /* Inner loop uses 46 flops */
424 /* End of innermost loop */
427 f[i_coord_offset+DIM*0+XX] += fix0;
428 f[i_coord_offset+DIM*0+YY] += fiy0;
429 f[i_coord_offset+DIM*0+ZZ] += fiz0;
433 fshift[i_shift_offset+XX] += tx;
434 fshift[i_shift_offset+YY] += ty;
435 fshift[i_shift_offset+ZZ] += tz;
437 /* Increment number of inner iterations */
438 inneriter += j_index_end - j_index_start;
440 /* Outer loop uses 13 flops */
443 /* Increment number of outer iterations */
446 /* Update outer/inner flops */
448 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*46);