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"
50 * Gromacs nonbonded kernel: nb_kernel_ElecNone_VdwLJEwSh_GeomP1P1_VF_c
51 * Electrostatics interaction: None
52 * VdW interaction: LJEwald
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecNone_VdwLJEwSh_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;
78 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
82 real ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
90 jindex = nlist->jindex;
92 shiftidx = nlist->shift;
94 shiftvec = fr->shift_vec[0];
95 fshift = fr->fshift[0];
98 vdwtype = mdatoms->typeA;
99 vdwgridparam = fr->ljpme_c6grid;
100 ewclj = fr->ewaldcoeff_lj;
101 sh_lj_ewald = fr->ic->sh_lj_ewald;
102 ewclj2 = ewclj*ewclj;
103 ewclj6 = ewclj2*ewclj2*ewclj2;
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 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
143 /* Reset potential sums */
146 /* Start inner kernel loop */
147 for(jidx=j_index_start; jidx<j_index_end; jidx++)
149 /* Get j neighbor index, and coordinate index */
151 j_coord_offset = DIM*jnr;
153 /* load j atom coordinates */
154 jx0 = x[j_coord_offset+DIM*0+XX];
155 jy0 = x[j_coord_offset+DIM*0+YY];
156 jz0 = x[j_coord_offset+DIM*0+ZZ];
158 /* Calculate displacement vector */
163 /* Calculate squared distance and things based on it */
164 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
166 rinv00 = gmx_invsqrt(rsq00);
168 rinvsq00 = rinv00*rinv00;
170 /* Load parameters for j particles */
171 vdwjidx0 = 2*vdwtype[jnr+0];
173 /**************************
174 * CALCULATE INTERACTIONS *
175 **************************/
182 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
183 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
184 c6grid_00 = vdwgridparam[vdwioffset0+vdwjidx0];
186 rinvsix = rinvsq00*rinvsq00*rinvsq00;
187 ewcljrsq = ewclj2*rsq00;
188 exponent = exp(-ewcljrsq);
189 poly = exponent*(1.0 + ewcljrsq + ewcljrsq*ewcljrsq*0.5);
190 vvdw6 = (c6_00-c6grid_00*(1.0-poly))*rinvsix;
191 vvdw12 = c12_00*rinvsix*rinvsix;
192 vvdw = (vvdw12 - c12_00*sh_vdw_invrcut6*sh_vdw_invrcut6)*(1.0/12.0) - (vvdw6 - c6_00*sh_vdw_invrcut6 - c6grid_00*sh_lj_ewald)*(1.0/6.0);
193 fvdw = (vvdw12 - vvdw6 - c6grid_00*(1.0/6.0)*exponent*ewclj6)*rinvsq00;
195 /* Update potential sums from outer loop */
200 /* Calculate temporary vectorial force */
205 /* Update vectorial force */
209 f[j_coord_offset+DIM*0+XX] -= tx;
210 f[j_coord_offset+DIM*0+YY] -= ty;
211 f[j_coord_offset+DIM*0+ZZ] -= tz;
215 /* Inner loop uses 55 flops */
217 /* End of innermost loop */
220 f[i_coord_offset+DIM*0+XX] += fix0;
221 f[i_coord_offset+DIM*0+YY] += fiy0;
222 f[i_coord_offset+DIM*0+ZZ] += fiz0;
226 fshift[i_shift_offset+XX] += tx;
227 fshift[i_shift_offset+YY] += ty;
228 fshift[i_shift_offset+ZZ] += tz;
231 /* Update potential energies */
232 kernel_data->energygrp_vdw[ggid] += vvdwsum;
234 /* Increment number of inner iterations */
235 inneriter += j_index_end - j_index_start;
237 /* Outer loop uses 13 flops */
240 /* Increment number of outer iterations */
243 /* Update outer/inner flops */
245 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*13 + inneriter*55);
248 * Gromacs nonbonded kernel: nb_kernel_ElecNone_VdwLJEwSh_GeomP1P1_F_c
249 * Electrostatics interaction: None
250 * VdW interaction: LJEwald
251 * Geometry: Particle-Particle
252 * Calculate force/pot: Force
255 nb_kernel_ElecNone_VdwLJEwSh_GeomP1P1_F_c
256 (t_nblist * gmx_restrict nlist,
257 rvec * gmx_restrict xx,
258 rvec * gmx_restrict ff,
259 t_forcerec * gmx_restrict fr,
260 t_mdatoms * gmx_restrict mdatoms,
261 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
262 t_nrnb * gmx_restrict nrnb)
264 int i_shift_offset,i_coord_offset,j_coord_offset;
265 int j_index_start,j_index_end;
266 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
267 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
268 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
269 real *shiftvec,*fshift,*x,*f;
271 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
273 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
274 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
276 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
280 real ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
288 jindex = nlist->jindex;
290 shiftidx = nlist->shift;
292 shiftvec = fr->shift_vec[0];
293 fshift = fr->fshift[0];
294 nvdwtype = fr->ntype;
296 vdwtype = mdatoms->typeA;
297 vdwgridparam = fr->ljpme_c6grid;
298 ewclj = fr->ewaldcoeff_lj;
299 sh_lj_ewald = fr->ic->sh_lj_ewald;
300 ewclj2 = ewclj*ewclj;
301 ewclj6 = ewclj2*ewclj2*ewclj2;
304 rcutoff2 = rcutoff*rcutoff;
306 sh_vdw_invrcut6 = fr->ic->sh_invrc6;
312 /* Start outer loop over neighborlists */
313 for(iidx=0; iidx<nri; iidx++)
315 /* Load shift vector for this list */
316 i_shift_offset = DIM*shiftidx[iidx];
317 shX = shiftvec[i_shift_offset+XX];
318 shY = shiftvec[i_shift_offset+YY];
319 shZ = shiftvec[i_shift_offset+ZZ];
321 /* Load limits for loop over neighbors */
322 j_index_start = jindex[iidx];
323 j_index_end = jindex[iidx+1];
325 /* Get outer coordinate index */
327 i_coord_offset = DIM*inr;
329 /* Load i particle coords and add shift vector */
330 ix0 = shX + x[i_coord_offset+DIM*0+XX];
331 iy0 = shY + x[i_coord_offset+DIM*0+YY];
332 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
338 /* Load parameters for i particles */
339 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
341 /* Start inner kernel loop */
342 for(jidx=j_index_start; jidx<j_index_end; jidx++)
344 /* Get j neighbor index, and coordinate index */
346 j_coord_offset = DIM*jnr;
348 /* load j atom coordinates */
349 jx0 = x[j_coord_offset+DIM*0+XX];
350 jy0 = x[j_coord_offset+DIM*0+YY];
351 jz0 = x[j_coord_offset+DIM*0+ZZ];
353 /* Calculate displacement vector */
358 /* Calculate squared distance and things based on it */
359 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
361 rinv00 = gmx_invsqrt(rsq00);
363 rinvsq00 = rinv00*rinv00;
365 /* Load parameters for j particles */
366 vdwjidx0 = 2*vdwtype[jnr+0];
368 /**************************
369 * CALCULATE INTERACTIONS *
370 **************************/
377 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
378 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
379 c6grid_00 = vdwgridparam[vdwioffset0+vdwjidx0];
381 rinvsix = rinvsq00*rinvsq00*rinvsq00;
382 ewcljrsq = ewclj2*rsq00;
383 exponent = exp(-ewcljrsq);
384 poly = exponent*(1.0 + ewcljrsq + ewcljrsq*ewcljrsq*0.5);
385 fvdw = (((c12_00*rinvsix - c6_00 + c6grid_00*(1.0-poly))*rinvsix) - c6grid_00*(1.0/6.0)*exponent*ewclj6)*rinvsq00;
389 /* Calculate temporary vectorial force */
394 /* Update vectorial force */
398 f[j_coord_offset+DIM*0+XX] -= tx;
399 f[j_coord_offset+DIM*0+YY] -= ty;
400 f[j_coord_offset+DIM*0+ZZ] -= tz;
404 /* Inner loop uses 44 flops */
406 /* End of innermost loop */
409 f[i_coord_offset+DIM*0+XX] += fix0;
410 f[i_coord_offset+DIM*0+YY] += fiy0;
411 f[i_coord_offset+DIM*0+ZZ] += fiz0;
415 fshift[i_shift_offset+XX] += tx;
416 fshift[i_shift_offset+YY] += ty;
417 fshift[i_shift_offset+ZZ] += tz;
419 /* Increment number of inner iterations */
420 inneriter += j_index_end - j_index_start;
422 /* Outer loop uses 12 flops */
425 /* Increment number of outer iterations */
428 /* Update outer/inner flops */
430 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*12 + inneriter*44);