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 "types/simple.h"
46 #include "gromacs/math/vec.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecRFCut_VdwLJSh_GeomP1P1_VF_c
51 * Electrostatics interaction: ReactionField
52 * VdW interaction: LennardJones
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecRFCut_VdwLJSh_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;
89 jindex = nlist->jindex;
91 shiftidx = nlist->shift;
93 shiftvec = fr->shift_vec[0];
94 fshift = fr->fshift[0];
96 charge = mdatoms->chargeA;
100 nvdwtype = fr->ntype;
102 vdwtype = mdatoms->typeA;
104 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
105 rcutoff = fr->rcoulomb;
106 rcutoff2 = rcutoff*rcutoff;
108 sh_vdw_invrcut6 = fr->ic->sh_invrc6;
114 /* Start outer loop over neighborlists */
115 for(iidx=0; iidx<nri; iidx++)
117 /* Load shift vector for this list */
118 i_shift_offset = DIM*shiftidx[iidx];
119 shX = shiftvec[i_shift_offset+XX];
120 shY = shiftvec[i_shift_offset+YY];
121 shZ = shiftvec[i_shift_offset+ZZ];
123 /* Load limits for loop over neighbors */
124 j_index_start = jindex[iidx];
125 j_index_end = jindex[iidx+1];
127 /* Get outer coordinate index */
129 i_coord_offset = DIM*inr;
131 /* Load i particle coords and add shift vector */
132 ix0 = shX + x[i_coord_offset+DIM*0+XX];
133 iy0 = shY + x[i_coord_offset+DIM*0+YY];
134 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
140 /* Load parameters for i particles */
141 iq0 = facel*charge[inr+0];
142 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
144 /* Reset potential sums */
148 /* Start inner kernel loop */
149 for(jidx=j_index_start; jidx<j_index_end; jidx++)
151 /* Get j neighbor index, and coordinate index */
153 j_coord_offset = DIM*jnr;
155 /* load j atom coordinates */
156 jx0 = x[j_coord_offset+DIM*0+XX];
157 jy0 = x[j_coord_offset+DIM*0+YY];
158 jz0 = x[j_coord_offset+DIM*0+ZZ];
160 /* Calculate displacement vector */
165 /* Calculate squared distance and things based on it */
166 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
168 rinv00 = gmx_invsqrt(rsq00);
170 rinvsq00 = rinv00*rinv00;
172 /* Load parameters for j particles */
174 vdwjidx0 = 2*vdwtype[jnr+0];
176 /**************************
177 * CALCULATE INTERACTIONS *
178 **************************/
184 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
185 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
187 /* REACTION-FIELD ELECTROSTATICS */
188 velec = qq00*(rinv00+krf*rsq00-crf);
189 felec = qq00*(rinv00*rinvsq00-krf2);
191 /* LENNARD-JONES DISPERSION/REPULSION */
193 rinvsix = rinvsq00*rinvsq00*rinvsq00;
194 vvdw6 = c6_00*rinvsix;
195 vvdw12 = c12_00*rinvsix*rinvsix;
196 vvdw = (vvdw12 - c12_00*sh_vdw_invrcut6*sh_vdw_invrcut6)*(1.0/12.0) - (vvdw6 - c6_00*sh_vdw_invrcut6)*(1.0/6.0);
197 fvdw = (vvdw12-vvdw6)*rinvsq00;
199 /* Update potential sums from outer loop */
205 /* Calculate temporary vectorial force */
210 /* Update vectorial force */
214 f[j_coord_offset+DIM*0+XX] -= tx;
215 f[j_coord_offset+DIM*0+YY] -= ty;
216 f[j_coord_offset+DIM*0+ZZ] -= tz;
220 /* Inner loop uses 49 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*49);
254 * Gromacs nonbonded kernel: nb_kernel_ElecRFCut_VdwLJSh_GeomP1P1_F_c
255 * Electrostatics interaction: ReactionField
256 * VdW interaction: LennardJones
257 * Geometry: Particle-Particle
258 * Calculate force/pot: Force
261 nb_kernel_ElecRFCut_VdwLJSh_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;
293 jindex = nlist->jindex;
295 shiftidx = nlist->shift;
297 shiftvec = fr->shift_vec[0];
298 fshift = fr->fshift[0];
300 charge = mdatoms->chargeA;
304 nvdwtype = fr->ntype;
306 vdwtype = mdatoms->typeA;
308 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
309 rcutoff = fr->rcoulomb;
310 rcutoff2 = rcutoff*rcutoff;
312 sh_vdw_invrcut6 = fr->ic->sh_invrc6;
318 /* Start outer loop over neighborlists */
319 for(iidx=0; iidx<nri; iidx++)
321 /* Load shift vector for this list */
322 i_shift_offset = DIM*shiftidx[iidx];
323 shX = shiftvec[i_shift_offset+XX];
324 shY = shiftvec[i_shift_offset+YY];
325 shZ = shiftvec[i_shift_offset+ZZ];
327 /* Load limits for loop over neighbors */
328 j_index_start = jindex[iidx];
329 j_index_end = jindex[iidx+1];
331 /* Get outer coordinate index */
333 i_coord_offset = DIM*inr;
335 /* Load i particle coords and add shift vector */
336 ix0 = shX + x[i_coord_offset+DIM*0+XX];
337 iy0 = shY + x[i_coord_offset+DIM*0+YY];
338 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
344 /* Load parameters for i particles */
345 iq0 = facel*charge[inr+0];
346 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
348 /* Start inner kernel loop */
349 for(jidx=j_index_start; jidx<j_index_end; jidx++)
351 /* Get j neighbor index, and coordinate index */
353 j_coord_offset = DIM*jnr;
355 /* load j atom coordinates */
356 jx0 = x[j_coord_offset+DIM*0+XX];
357 jy0 = x[j_coord_offset+DIM*0+YY];
358 jz0 = x[j_coord_offset+DIM*0+ZZ];
360 /* Calculate displacement vector */
365 /* Calculate squared distance and things based on it */
366 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
368 rinv00 = gmx_invsqrt(rsq00);
370 rinvsq00 = rinv00*rinv00;
372 /* Load parameters for j particles */
374 vdwjidx0 = 2*vdwtype[jnr+0];
376 /**************************
377 * CALCULATE INTERACTIONS *
378 **************************/
384 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
385 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
387 /* REACTION-FIELD ELECTROSTATICS */
388 felec = qq00*(rinv00*rinvsq00-krf2);
390 /* LENNARD-JONES DISPERSION/REPULSION */
392 rinvsix = rinvsq00*rinvsq00*rinvsq00;
393 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
397 /* Calculate temporary vectorial force */
402 /* Update vectorial force */
406 f[j_coord_offset+DIM*0+XX] -= tx;
407 f[j_coord_offset+DIM*0+YY] -= ty;
408 f[j_coord_offset+DIM*0+ZZ] -= tz;
412 /* Inner loop uses 34 flops */
414 /* End of innermost loop */
417 f[i_coord_offset+DIM*0+XX] += fix0;
418 f[i_coord_offset+DIM*0+YY] += fiy0;
419 f[i_coord_offset+DIM*0+ZZ] += fiz0;
423 fshift[i_shift_offset+XX] += tx;
424 fshift[i_shift_offset+YY] += ty;
425 fshift[i_shift_offset+ZZ] += tz;
427 /* Increment number of inner iterations */
428 inneriter += j_index_end - j_index_start;
430 /* Outer loop uses 13 flops */
433 /* Increment number of outer iterations */
436 /* Update outer/inner flops */
438 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*34);