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 "types/simple.h"
44 #include "gromacs/math/vec.h"
48 * Gromacs nonbonded kernel: nb_kernel_ElecNone_VdwLJEw_GeomP1P1_VF_c
49 * Electrostatics interaction: None
50 * VdW interaction: LJEwald
51 * Geometry: Particle-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecNone_VdwLJEw_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;
76 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
80 real ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
88 jindex = nlist->jindex;
90 shiftidx = nlist->shift;
92 shiftvec = fr->shift_vec[0];
93 fshift = fr->fshift[0];
96 vdwtype = mdatoms->typeA;
97 vdwgridparam = fr->ljpme_c6grid;
98 ewclj = fr->ewaldcoeff_lj;
99 sh_lj_ewald = fr->ic->sh_lj_ewald;
100 ewclj2 = ewclj*ewclj;
101 ewclj6 = ewclj2*ewclj2*ewclj2;
106 /* Start outer loop over neighborlists */
107 for(iidx=0; iidx<nri; iidx++)
109 /* Load shift vector for this list */
110 i_shift_offset = DIM*shiftidx[iidx];
111 shX = shiftvec[i_shift_offset+XX];
112 shY = shiftvec[i_shift_offset+YY];
113 shZ = shiftvec[i_shift_offset+ZZ];
115 /* Load limits for loop over neighbors */
116 j_index_start = jindex[iidx];
117 j_index_end = jindex[iidx+1];
119 /* Get outer coordinate index */
121 i_coord_offset = DIM*inr;
123 /* Load i particle coords and add shift vector */
124 ix0 = shX + x[i_coord_offset+DIM*0+XX];
125 iy0 = shY + x[i_coord_offset+DIM*0+YY];
126 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
132 /* Load parameters for i particles */
133 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
135 /* Reset potential sums */
138 /* Start inner kernel loop */
139 for(jidx=j_index_start; jidx<j_index_end; jidx++)
141 /* Get j neighbor index, and coordinate index */
143 j_coord_offset = DIM*jnr;
145 /* load j atom coordinates */
146 jx0 = x[j_coord_offset+DIM*0+XX];
147 jy0 = x[j_coord_offset+DIM*0+YY];
148 jz0 = x[j_coord_offset+DIM*0+ZZ];
150 /* Calculate displacement vector */
155 /* Calculate squared distance and things based on it */
156 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
158 rinv00 = gmx_invsqrt(rsq00);
160 rinvsq00 = rinv00*rinv00;
162 /* Load parameters for j particles */
163 vdwjidx0 = 2*vdwtype[jnr+0];
165 /**************************
166 * CALCULATE INTERACTIONS *
167 **************************/
171 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
172 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
173 c6grid_00 = vdwgridparam[vdwioffset0+vdwjidx0];
175 rinvsix = rinvsq00*rinvsq00*rinvsq00;
176 ewcljrsq = ewclj2*rsq00;
177 exponent = exp(-ewcljrsq);
178 poly = exponent*(1.0 + ewcljrsq + ewcljrsq*ewcljrsq*0.5);
179 vvdw6 = (c6_00-c6grid_00*(1.0-poly))*rinvsix;
180 vvdw12 = c12_00*rinvsix*rinvsix;
181 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
182 fvdw = (vvdw12 - vvdw6 - c6grid_00*(1.0/6.0)*exponent*ewclj6)*rinvsq00;
184 /* Update potential sums from outer loop */
189 /* Calculate temporary vectorial force */
194 /* Update vectorial force */
198 f[j_coord_offset+DIM*0+XX] -= tx;
199 f[j_coord_offset+DIM*0+YY] -= ty;
200 f[j_coord_offset+DIM*0+ZZ] -= tz;
202 /* Inner loop uses 49 flops */
204 /* End of innermost loop */
207 f[i_coord_offset+DIM*0+XX] += fix0;
208 f[i_coord_offset+DIM*0+YY] += fiy0;
209 f[i_coord_offset+DIM*0+ZZ] += fiz0;
213 fshift[i_shift_offset+XX] += tx;
214 fshift[i_shift_offset+YY] += ty;
215 fshift[i_shift_offset+ZZ] += tz;
218 /* Update potential energies */
219 kernel_data->energygrp_vdw[ggid] += vvdwsum;
221 /* Increment number of inner iterations */
222 inneriter += j_index_end - j_index_start;
224 /* Outer loop uses 13 flops */
227 /* Increment number of outer iterations */
230 /* Update outer/inner flops */
232 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*13 + inneriter*49);
235 * Gromacs nonbonded kernel: nb_kernel_ElecNone_VdwLJEw_GeomP1P1_F_c
236 * Electrostatics interaction: None
237 * VdW interaction: LJEwald
238 * Geometry: Particle-Particle
239 * Calculate force/pot: Force
242 nb_kernel_ElecNone_VdwLJEw_GeomP1P1_F_c
243 (t_nblist * gmx_restrict nlist,
244 rvec * gmx_restrict xx,
245 rvec * gmx_restrict ff,
246 t_forcerec * gmx_restrict fr,
247 t_mdatoms * gmx_restrict mdatoms,
248 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
249 t_nrnb * gmx_restrict nrnb)
251 int i_shift_offset,i_coord_offset,j_coord_offset;
252 int j_index_start,j_index_end;
253 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
254 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
255 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
256 real *shiftvec,*fshift,*x,*f;
258 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
260 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
261 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
263 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
267 real ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
275 jindex = nlist->jindex;
277 shiftidx = nlist->shift;
279 shiftvec = fr->shift_vec[0];
280 fshift = fr->fshift[0];
281 nvdwtype = fr->ntype;
283 vdwtype = mdatoms->typeA;
284 vdwgridparam = fr->ljpme_c6grid;
285 ewclj = fr->ewaldcoeff_lj;
286 sh_lj_ewald = fr->ic->sh_lj_ewald;
287 ewclj2 = ewclj*ewclj;
288 ewclj6 = ewclj2*ewclj2*ewclj2;
293 /* Start outer loop over neighborlists */
294 for(iidx=0; iidx<nri; iidx++)
296 /* Load shift vector for this list */
297 i_shift_offset = DIM*shiftidx[iidx];
298 shX = shiftvec[i_shift_offset+XX];
299 shY = shiftvec[i_shift_offset+YY];
300 shZ = shiftvec[i_shift_offset+ZZ];
302 /* Load limits for loop over neighbors */
303 j_index_start = jindex[iidx];
304 j_index_end = jindex[iidx+1];
306 /* Get outer coordinate index */
308 i_coord_offset = DIM*inr;
310 /* Load i particle coords and add shift vector */
311 ix0 = shX + x[i_coord_offset+DIM*0+XX];
312 iy0 = shY + x[i_coord_offset+DIM*0+YY];
313 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
319 /* Load parameters for i particles */
320 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
322 /* Start inner kernel loop */
323 for(jidx=j_index_start; jidx<j_index_end; jidx++)
325 /* Get j neighbor index, and coordinate index */
327 j_coord_offset = DIM*jnr;
329 /* load j atom coordinates */
330 jx0 = x[j_coord_offset+DIM*0+XX];
331 jy0 = x[j_coord_offset+DIM*0+YY];
332 jz0 = x[j_coord_offset+DIM*0+ZZ];
334 /* Calculate displacement vector */
339 /* Calculate squared distance and things based on it */
340 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
342 rinv00 = gmx_invsqrt(rsq00);
344 rinvsq00 = rinv00*rinv00;
346 /* Load parameters for j particles */
347 vdwjidx0 = 2*vdwtype[jnr+0];
349 /**************************
350 * CALCULATE INTERACTIONS *
351 **************************/
355 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
356 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
357 c6grid_00 = vdwgridparam[vdwioffset0+vdwjidx0];
359 rinvsix = rinvsq00*rinvsq00*rinvsq00;
360 ewcljrsq = ewclj2*rsq00;
361 exponent = exp(-ewcljrsq);
362 poly = exponent*(1.0 + ewcljrsq + ewcljrsq*ewcljrsq*0.5);
363 fvdw = (((c12_00*rinvsix - c6_00 + c6grid_00*(1.0-poly))*rinvsix) - c6grid_00*(1.0/6.0)*exponent*ewclj6)*rinvsq00;
367 /* Calculate temporary vectorial force */
372 /* Update vectorial force */
376 f[j_coord_offset+DIM*0+XX] -= tx;
377 f[j_coord_offset+DIM*0+YY] -= ty;
378 f[j_coord_offset+DIM*0+ZZ] -= tz;
380 /* Inner loop uses 44 flops */
382 /* End of innermost loop */
385 f[i_coord_offset+DIM*0+XX] += fix0;
386 f[i_coord_offset+DIM*0+YY] += fiy0;
387 f[i_coord_offset+DIM*0+ZZ] += fiz0;
391 fshift[i_shift_offset+XX] += tx;
392 fshift[i_shift_offset+YY] += ty;
393 fshift[i_shift_offset+ZZ] += tz;
395 /* Increment number of inner iterations */
396 inneriter += j_index_end - j_index_start;
398 /* Outer loop uses 12 flops */
401 /* Increment number of outer iterations */
404 /* Update outer/inner flops */
406 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*12 + inneriter*44);