2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013, 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_VdwLJSw_GeomP1P1_VF_c
51 * Electrostatics interaction: None
52 * VdW interaction: LennardJones
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecNone_VdwLJSw_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;
81 real rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
88 jindex = nlist->jindex;
90 shiftidx = nlist->shift;
92 shiftvec = fr->shift_vec[0];
93 fshift = fr->fshift[0];
96 vdwtype = mdatoms->typeA;
99 rcutoff2 = rcutoff*rcutoff;
101 rswitch = fr->rvdw_switch;
102 /* Setup switch parameters */
104 swV3 = -10.0/(d*d*d);
105 swV4 = 15.0/(d*d*d*d);
106 swV5 = -6.0/(d*d*d*d*d);
107 swF2 = -30.0/(d*d*d);
108 swF3 = 60.0/(d*d*d*d);
109 swF4 = -30.0/(d*d*d*d*d);
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];
185 /* LENNARD-JONES DISPERSION/REPULSION */
187 rinvsix = rinvsq00*rinvsq00*rinvsq00;
188 vvdw6 = c6_00*rinvsix;
189 vvdw12 = c12_00*rinvsix*rinvsix;
190 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
191 fvdw = (vvdw12-vvdw6)*rinvsq00;
194 d = (d>0.0) ? d : 0.0;
196 sw = 1.0+d2*d*(swV3+d*(swV4+d*swV5));
198 dsw = d2*(swF2+d*(swF3+d*swF4));
200 /* Evaluate switch function */
201 /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
202 fvdw = fvdw*sw - rinv00*vvdw*dsw;
205 /* Update potential sums from outer loop */
210 /* Calculate temporary vectorial force */
215 /* Update vectorial force */
219 f[j_coord_offset+DIM*0+XX] -= tx;
220 f[j_coord_offset+DIM*0+YY] -= ty;
221 f[j_coord_offset+DIM*0+ZZ] -= tz;
225 /* Inner loop uses 53 flops */
227 /* End of innermost loop */
230 f[i_coord_offset+DIM*0+XX] += fix0;
231 f[i_coord_offset+DIM*0+YY] += fiy0;
232 f[i_coord_offset+DIM*0+ZZ] += fiz0;
236 fshift[i_shift_offset+XX] += tx;
237 fshift[i_shift_offset+YY] += ty;
238 fshift[i_shift_offset+ZZ] += tz;
241 /* Update potential energies */
242 kernel_data->energygrp_vdw[ggid] += vvdwsum;
244 /* Increment number of inner iterations */
245 inneriter += j_index_end - j_index_start;
247 /* Outer loop uses 13 flops */
250 /* Increment number of outer iterations */
253 /* Update outer/inner flops */
255 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*13 + inneriter*53);
258 * Gromacs nonbonded kernel: nb_kernel_ElecNone_VdwLJSw_GeomP1P1_F_c
259 * Electrostatics interaction: None
260 * VdW interaction: LennardJones
261 * Geometry: Particle-Particle
262 * Calculate force/pot: Force
265 nb_kernel_ElecNone_VdwLJSw_GeomP1P1_F_c
266 (t_nblist * gmx_restrict nlist,
267 rvec * gmx_restrict xx,
268 rvec * gmx_restrict ff,
269 t_forcerec * gmx_restrict fr,
270 t_mdatoms * gmx_restrict mdatoms,
271 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
272 t_nrnb * gmx_restrict nrnb)
274 int i_shift_offset,i_coord_offset,j_coord_offset;
275 int j_index_start,j_index_end;
276 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
277 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
278 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
279 real *shiftvec,*fshift,*x,*f;
281 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
283 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
284 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
286 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
289 real rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
296 jindex = nlist->jindex;
298 shiftidx = nlist->shift;
300 shiftvec = fr->shift_vec[0];
301 fshift = fr->fshift[0];
302 nvdwtype = fr->ntype;
304 vdwtype = mdatoms->typeA;
307 rcutoff2 = rcutoff*rcutoff;
309 rswitch = fr->rvdw_switch;
310 /* Setup switch parameters */
312 swV3 = -10.0/(d*d*d);
313 swV4 = 15.0/(d*d*d*d);
314 swV5 = -6.0/(d*d*d*d*d);
315 swF2 = -30.0/(d*d*d);
316 swF3 = 60.0/(d*d*d*d);
317 swF4 = -30.0/(d*d*d*d*d);
322 /* Start outer loop over neighborlists */
323 for(iidx=0; iidx<nri; iidx++)
325 /* Load shift vector for this list */
326 i_shift_offset = DIM*shiftidx[iidx];
327 shX = shiftvec[i_shift_offset+XX];
328 shY = shiftvec[i_shift_offset+YY];
329 shZ = shiftvec[i_shift_offset+ZZ];
331 /* Load limits for loop over neighbors */
332 j_index_start = jindex[iidx];
333 j_index_end = jindex[iidx+1];
335 /* Get outer coordinate index */
337 i_coord_offset = DIM*inr;
339 /* Load i particle coords and add shift vector */
340 ix0 = shX + x[i_coord_offset+DIM*0+XX];
341 iy0 = shY + x[i_coord_offset+DIM*0+YY];
342 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
348 /* Load parameters for i particles */
349 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
351 /* Start inner kernel loop */
352 for(jidx=j_index_start; jidx<j_index_end; jidx++)
354 /* Get j neighbor index, and coordinate index */
356 j_coord_offset = DIM*jnr;
358 /* load j atom coordinates */
359 jx0 = x[j_coord_offset+DIM*0+XX];
360 jy0 = x[j_coord_offset+DIM*0+YY];
361 jz0 = x[j_coord_offset+DIM*0+ZZ];
363 /* Calculate displacement vector */
368 /* Calculate squared distance and things based on it */
369 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
371 rinv00 = gmx_invsqrt(rsq00);
373 rinvsq00 = rinv00*rinv00;
375 /* Load parameters for j particles */
376 vdwjidx0 = 2*vdwtype[jnr+0];
378 /**************************
379 * CALCULATE INTERACTIONS *
380 **************************/
387 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
388 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
390 /* LENNARD-JONES DISPERSION/REPULSION */
392 rinvsix = rinvsq00*rinvsq00*rinvsq00;
393 vvdw6 = c6_00*rinvsix;
394 vvdw12 = c12_00*rinvsix*rinvsix;
395 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
396 fvdw = (vvdw12-vvdw6)*rinvsq00;
399 d = (d>0.0) ? d : 0.0;
401 sw = 1.0+d2*d*(swV3+d*(swV4+d*swV5));
403 dsw = d2*(swF2+d*(swF3+d*swF4));
405 /* Evaluate switch function */
406 /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
407 fvdw = fvdw*sw - rinv00*vvdw*dsw;
411 /* Calculate temporary vectorial force */
416 /* Update vectorial force */
420 f[j_coord_offset+DIM*0+XX] -= tx;
421 f[j_coord_offset+DIM*0+YY] -= ty;
422 f[j_coord_offset+DIM*0+ZZ] -= tz;
426 /* Inner loop uses 51 flops */
428 /* End of innermost loop */
431 f[i_coord_offset+DIM*0+XX] += fix0;
432 f[i_coord_offset+DIM*0+YY] += fiy0;
433 f[i_coord_offset+DIM*0+ZZ] += fiz0;
437 fshift[i_shift_offset+XX] += tx;
438 fshift[i_shift_offset+YY] += ty;
439 fshift[i_shift_offset+ZZ] += tz;
441 /* Increment number of inner iterations */
442 inneriter += j_index_end - j_index_start;
444 /* Outer loop uses 12 flops */
447 /* Increment number of outer iterations */
450 /* Update outer/inner flops */
452 inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*12 + inneriter*51);