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_ElecEw_VdwLJ_GeomP1P1_VF_c
49 * Electrostatics interaction: Ewald
50 * VdW interaction: LennardJones
51 * Geometry: Particle-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecEw_VdwLJ_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 ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
90 jindex = nlist->jindex;
92 shiftidx = nlist->shift;
94 shiftvec = fr->shift_vec[0];
95 fshift = fr->fshift[0];
97 charge = mdatoms->chargeA;
100 vdwtype = mdatoms->typeA;
102 sh_ewald = fr->ic->sh_ewald;
103 ewtab = fr->ic->tabq_coul_FDV0;
104 ewtabscale = fr->ic->tabq_scale;
105 ewtabhalfspace = 0.5/ewtabscale;
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 /* EWALD ELECTROSTATICS */
184 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
185 ewrt = r00*ewtabscale;
189 felec = ewtab[ewitab]+eweps*ewtab[ewitab+1];
190 velec = qq00*(rinv00-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
191 felec = qq00*rinv00*(rinvsq00-felec);
193 /* LENNARD-JONES DISPERSION/REPULSION */
195 rinvsix = rinvsq00*rinvsq00*rinvsq00;
196 vvdw6 = c6_00*rinvsix;
197 vvdw12 = c12_00*rinvsix*rinvsix;
198 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
199 fvdw = (vvdw12-vvdw6)*rinvsq00;
201 /* Update potential sums from outer loop */
207 /* Calculate temporary vectorial force */
212 /* Update vectorial force */
216 f[j_coord_offset+DIM*0+XX] -= tx;
217 f[j_coord_offset+DIM*0+YY] -= ty;
218 f[j_coord_offset+DIM*0+ZZ] -= tz;
220 /* Inner loop uses 53 flops */
222 /* End of innermost loop */
225 f[i_coord_offset+DIM*0+XX] += fix0;
226 f[i_coord_offset+DIM*0+YY] += fiy0;
227 f[i_coord_offset+DIM*0+ZZ] += fiz0;
231 fshift[i_shift_offset+XX] += tx;
232 fshift[i_shift_offset+YY] += ty;
233 fshift[i_shift_offset+ZZ] += tz;
236 /* Update potential energies */
237 kernel_data->energygrp_elec[ggid] += velecsum;
238 kernel_data->energygrp_vdw[ggid] += vvdwsum;
240 /* Increment number of inner iterations */
241 inneriter += j_index_end - j_index_start;
243 /* Outer loop uses 15 flops */
246 /* Increment number of outer iterations */
249 /* Update outer/inner flops */
251 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*53);
254 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwLJ_GeomP1P1_F_c
255 * Electrostatics interaction: Ewald
256 * VdW interaction: LennardJones
257 * Geometry: Particle-Particle
258 * Calculate force/pot: Force
261 nb_kernel_ElecEw_VdwLJ_GeomP1P1_F_c
262 (t_nblist * gmx_restrict nlist,
263 rvec * gmx_restrict xx,
264 rvec * gmx_restrict ff,
265 t_forcerec * gmx_restrict fr,
266 t_mdatoms * gmx_restrict mdatoms,
267 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
268 t_nrnb * gmx_restrict nrnb)
270 int i_shift_offset,i_coord_offset,j_coord_offset;
271 int j_index_start,j_index_end;
272 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
273 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
274 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
275 real *shiftvec,*fshift,*x,*f;
277 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
279 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
280 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
281 real velec,felec,velecsum,facel,crf,krf,krf2;
284 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
288 real ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
296 jindex = nlist->jindex;
298 shiftidx = nlist->shift;
300 shiftvec = fr->shift_vec[0];
301 fshift = fr->fshift[0];
303 charge = mdatoms->chargeA;
304 nvdwtype = fr->ntype;
306 vdwtype = mdatoms->typeA;
308 sh_ewald = fr->ic->sh_ewald;
309 ewtab = fr->ic->tabq_coul_F;
310 ewtabscale = fr->ic->tabq_scale;
311 ewtabhalfspace = 0.5/ewtabscale;
316 /* Start outer loop over neighborlists */
317 for(iidx=0; iidx<nri; iidx++)
319 /* Load shift vector for this list */
320 i_shift_offset = DIM*shiftidx[iidx];
321 shX = shiftvec[i_shift_offset+XX];
322 shY = shiftvec[i_shift_offset+YY];
323 shZ = shiftvec[i_shift_offset+ZZ];
325 /* Load limits for loop over neighbors */
326 j_index_start = jindex[iidx];
327 j_index_end = jindex[iidx+1];
329 /* Get outer coordinate index */
331 i_coord_offset = DIM*inr;
333 /* Load i particle coords and add shift vector */
334 ix0 = shX + x[i_coord_offset+DIM*0+XX];
335 iy0 = shY + x[i_coord_offset+DIM*0+YY];
336 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
342 /* Load parameters for i particles */
343 iq0 = facel*charge[inr+0];
344 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
346 /* Start inner kernel loop */
347 for(jidx=j_index_start; jidx<j_index_end; jidx++)
349 /* Get j neighbor index, and coordinate index */
351 j_coord_offset = DIM*jnr;
353 /* load j atom coordinates */
354 jx0 = x[j_coord_offset+DIM*0+XX];
355 jy0 = x[j_coord_offset+DIM*0+YY];
356 jz0 = x[j_coord_offset+DIM*0+ZZ];
358 /* Calculate displacement vector */
363 /* Calculate squared distance and things based on it */
364 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
366 rinv00 = gmx_invsqrt(rsq00);
368 rinvsq00 = rinv00*rinv00;
370 /* Load parameters for j particles */
372 vdwjidx0 = 2*vdwtype[jnr+0];
374 /**************************
375 * CALCULATE INTERACTIONS *
376 **************************/
381 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
382 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
384 /* EWALD ELECTROSTATICS */
386 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
387 ewrt = r00*ewtabscale;
390 felec = (1.0-eweps)*ewtab[ewitab]+eweps*ewtab[ewitab+1];
391 felec = qq00*rinv00*(rinvsq00-felec);
393 /* LENNARD-JONES DISPERSION/REPULSION */
395 rinvsix = rinvsq00*rinvsq00*rinvsq00;
396 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
400 /* Calculate temporary vectorial force */
405 /* Update vectorial force */
409 f[j_coord_offset+DIM*0+XX] -= tx;
410 f[j_coord_offset+DIM*0+YY] -= ty;
411 f[j_coord_offset+DIM*0+ZZ] -= tz;
413 /* Inner loop uses 41 flops */
415 /* End of innermost loop */
418 f[i_coord_offset+DIM*0+XX] += fix0;
419 f[i_coord_offset+DIM*0+YY] += fiy0;
420 f[i_coord_offset+DIM*0+ZZ] += fiz0;
424 fshift[i_shift_offset+XX] += tx;
425 fshift[i_shift_offset+YY] += ty;
426 fshift[i_shift_offset+ZZ] += tz;
428 /* Increment number of inner iterations */
429 inneriter += j_index_end - j_index_start;
431 /* Outer loop uses 13 flops */
434 /* Increment number of outer iterations */
437 /* Update outer/inner flops */
439 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*41);