277ef6c770b3b90893d0befa25dd5ac9051848ea
[alexxy/gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_c / nb_kernel_ElecCSTab_VdwNone_GeomP1P1_c.c
1 /*
2  * This file is part of the GROMACS molecular simulation package.
3  *
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.
8  *
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.
13  *
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.
18  *
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.
23  *
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.
31  *
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.
34  */
35 /*
36  * Note: this file was generated by the GROMACS c kernel generator.
37  */
38 #include "config.h"
39
40 #include <math.h>
41
42 #include "../nb_kernel.h"
43 #include "gromacs/legacyheaders/types/simple.h"
44 #include "gromacs/math/vec.h"
45 #include "gromacs/legacyheaders/nrnb.h"
46
47 /*
48  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwNone_GeomP1P1_VF_c
49  * Electrostatics interaction: CubicSplineTable
50  * VdW interaction:            None
51  * Geometry:                   Particle-Particle
52  * Calculate force/pot:        PotentialAndForce
53  */
54 void
55 nb_kernel_ElecCSTab_VdwNone_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)
63 {
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;
70     int              vdwioffset0;
71     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
72     int              vdwjidx0;
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;
76     real             *charge;
77     int              vfitab;
78     real             rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
79     real             *vftab;
80
81     x                = xx[0];
82     f                = ff[0];
83
84     nri              = nlist->nri;
85     iinr             = nlist->iinr;
86     jindex           = nlist->jindex;
87     jjnr             = nlist->jjnr;
88     shiftidx         = nlist->shift;
89     gid              = nlist->gid;
90     shiftvec         = fr->shift_vec[0];
91     fshift           = fr->fshift[0];
92     facel            = fr->epsfac;
93     charge           = mdatoms->chargeA;
94
95     vftab            = kernel_data->table_elec->data;
96     vftabscale       = kernel_data->table_elec->scale;
97
98     outeriter        = 0;
99     inneriter        = 0;
100
101     /* Start outer loop over neighborlists */
102     for(iidx=0; iidx<nri; iidx++)
103     {
104         /* Load shift vector for this list */
105         i_shift_offset   = DIM*shiftidx[iidx];
106         shX              = shiftvec[i_shift_offset+XX];
107         shY              = shiftvec[i_shift_offset+YY];
108         shZ              = shiftvec[i_shift_offset+ZZ];
109
110         /* Load limits for loop over neighbors */
111         j_index_start    = jindex[iidx];
112         j_index_end      = jindex[iidx+1];
113
114         /* Get outer coordinate index */
115         inr              = iinr[iidx];
116         i_coord_offset   = DIM*inr;
117
118         /* Load i particle coords and add shift vector */
119         ix0              = shX + x[i_coord_offset+DIM*0+XX];
120         iy0              = shY + x[i_coord_offset+DIM*0+YY];
121         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
122
123         fix0             = 0.0;
124         fiy0             = 0.0;
125         fiz0             = 0.0;
126
127         /* Load parameters for i particles */
128         iq0              = facel*charge[inr+0];
129
130         /* Reset potential sums */
131         velecsum         = 0.0;
132
133         /* Start inner kernel loop */
134         for(jidx=j_index_start; jidx<j_index_end; jidx++)
135         {
136             /* Get j neighbor index, and coordinate index */
137             jnr              = jjnr[jidx];
138             j_coord_offset   = DIM*jnr;
139
140             /* load j atom coordinates */
141             jx0              = x[j_coord_offset+DIM*0+XX];
142             jy0              = x[j_coord_offset+DIM*0+YY];
143             jz0              = x[j_coord_offset+DIM*0+ZZ];
144
145             /* Calculate displacement vector */
146             dx00             = ix0 - jx0;
147             dy00             = iy0 - jy0;
148             dz00             = iz0 - jz0;
149
150             /* Calculate squared distance and things based on it */
151             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
152
153             rinv00           = gmx_invsqrt(rsq00);
154
155             /* Load parameters for j particles */
156             jq0              = charge[jnr+0];
157
158             /**************************
159              * CALCULATE INTERACTIONS *
160              **************************/
161
162             r00              = rsq00*rinv00;
163
164             qq00             = iq0*jq0;
165
166             /* Calculate table index by multiplying r with table scale and truncate to integer */
167             rt               = r00*vftabscale;
168             vfitab           = rt;
169             vfeps            = rt-vfitab;
170             vfitab           = 1*4*vfitab;
171
172             /* CUBIC SPLINE TABLE ELECTROSTATICS */
173             Y                = vftab[vfitab];
174             F                = vftab[vfitab+1];
175             Geps             = vfeps*vftab[vfitab+2];
176             Heps2            = vfeps*vfeps*vftab[vfitab+3];
177             Fp               = F+Geps+Heps2;
178             VV               = Y+vfeps*Fp;
179             velec            = qq00*VV;
180             FF               = Fp+Geps+2.0*Heps2;
181             felec            = -qq00*FF*vftabscale*rinv00;
182
183             /* Update potential sums from outer loop */
184             velecsum        += velec;
185
186             fscal            = felec;
187
188             /* Calculate temporary vectorial force */
189             tx               = fscal*dx00;
190             ty               = fscal*dy00;
191             tz               = fscal*dz00;
192
193             /* Update vectorial force */
194             fix0            += tx;
195             fiy0            += ty;
196             fiz0            += tz;
197             f[j_coord_offset+DIM*0+XX] -= tx;
198             f[j_coord_offset+DIM*0+YY] -= ty;
199             f[j_coord_offset+DIM*0+ZZ] -= tz;
200
201             /* Inner loop uses 42 flops */
202         }
203         /* End of innermost loop */
204
205         tx = ty = tz = 0;
206         f[i_coord_offset+DIM*0+XX] += fix0;
207         f[i_coord_offset+DIM*0+YY] += fiy0;
208         f[i_coord_offset+DIM*0+ZZ] += fiz0;
209         tx                         += fix0;
210         ty                         += fiy0;
211         tz                         += fiz0;
212         fshift[i_shift_offset+XX]  += tx;
213         fshift[i_shift_offset+YY]  += ty;
214         fshift[i_shift_offset+ZZ]  += tz;
215
216         ggid                        = gid[iidx];
217         /* Update potential energies */
218         kernel_data->energygrp_elec[ggid] += velecsum;
219
220         /* Increment number of inner iterations */
221         inneriter                  += j_index_end - j_index_start;
222
223         /* Outer loop uses 14 flops */
224     }
225
226     /* Increment number of outer iterations */
227     outeriter        += nri;
228
229     /* Update outer/inner flops */
230
231     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VF,outeriter*14 + inneriter*42);
232 }
233 /*
234  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwNone_GeomP1P1_F_c
235  * Electrostatics interaction: CubicSplineTable
236  * VdW interaction:            None
237  * Geometry:                   Particle-Particle
238  * Calculate force/pot:        Force
239  */
240 void
241 nb_kernel_ElecCSTab_VdwNone_GeomP1P1_F_c
242                     (t_nblist                    * gmx_restrict       nlist,
243                      rvec                        * gmx_restrict          xx,
244                      rvec                        * gmx_restrict          ff,
245                      t_forcerec                  * gmx_restrict          fr,
246                      t_mdatoms                   * gmx_restrict     mdatoms,
247                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
248                      t_nrnb                      * gmx_restrict        nrnb)
249 {
250     int              i_shift_offset,i_coord_offset,j_coord_offset;
251     int              j_index_start,j_index_end;
252     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
253     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
254     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
255     real             *shiftvec,*fshift,*x,*f;
256     int              vdwioffset0;
257     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
258     int              vdwjidx0;
259     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
260     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
261     real             velec,felec,velecsum,facel,crf,krf,krf2;
262     real             *charge;
263     int              vfitab;
264     real             rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
265     real             *vftab;
266
267     x                = xx[0];
268     f                = ff[0];
269
270     nri              = nlist->nri;
271     iinr             = nlist->iinr;
272     jindex           = nlist->jindex;
273     jjnr             = nlist->jjnr;
274     shiftidx         = nlist->shift;
275     gid              = nlist->gid;
276     shiftvec         = fr->shift_vec[0];
277     fshift           = fr->fshift[0];
278     facel            = fr->epsfac;
279     charge           = mdatoms->chargeA;
280
281     vftab            = kernel_data->table_elec->data;
282     vftabscale       = kernel_data->table_elec->scale;
283
284     outeriter        = 0;
285     inneriter        = 0;
286
287     /* Start outer loop over neighborlists */
288     for(iidx=0; iidx<nri; iidx++)
289     {
290         /* Load shift vector for this list */
291         i_shift_offset   = DIM*shiftidx[iidx];
292         shX              = shiftvec[i_shift_offset+XX];
293         shY              = shiftvec[i_shift_offset+YY];
294         shZ              = shiftvec[i_shift_offset+ZZ];
295
296         /* Load limits for loop over neighbors */
297         j_index_start    = jindex[iidx];
298         j_index_end      = jindex[iidx+1];
299
300         /* Get outer coordinate index */
301         inr              = iinr[iidx];
302         i_coord_offset   = DIM*inr;
303
304         /* Load i particle coords and add shift vector */
305         ix0              = shX + x[i_coord_offset+DIM*0+XX];
306         iy0              = shY + x[i_coord_offset+DIM*0+YY];
307         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
308
309         fix0             = 0.0;
310         fiy0             = 0.0;
311         fiz0             = 0.0;
312
313         /* Load parameters for i particles */
314         iq0              = facel*charge[inr+0];
315
316         /* Start inner kernel loop */
317         for(jidx=j_index_start; jidx<j_index_end; jidx++)
318         {
319             /* Get j neighbor index, and coordinate index */
320             jnr              = jjnr[jidx];
321             j_coord_offset   = DIM*jnr;
322
323             /* load j atom coordinates */
324             jx0              = x[j_coord_offset+DIM*0+XX];
325             jy0              = x[j_coord_offset+DIM*0+YY];
326             jz0              = x[j_coord_offset+DIM*0+ZZ];
327
328             /* Calculate displacement vector */
329             dx00             = ix0 - jx0;
330             dy00             = iy0 - jy0;
331             dz00             = iz0 - jz0;
332
333             /* Calculate squared distance and things based on it */
334             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
335
336             rinv00           = gmx_invsqrt(rsq00);
337
338             /* Load parameters for j particles */
339             jq0              = charge[jnr+0];
340
341             /**************************
342              * CALCULATE INTERACTIONS *
343              **************************/
344
345             r00              = rsq00*rinv00;
346
347             qq00             = iq0*jq0;
348
349             /* Calculate table index by multiplying r with table scale and truncate to integer */
350             rt               = r00*vftabscale;
351             vfitab           = rt;
352             vfeps            = rt-vfitab;
353             vfitab           = 1*4*vfitab;
354
355             /* CUBIC SPLINE TABLE ELECTROSTATICS */
356             F                = vftab[vfitab+1];
357             Geps             = vfeps*vftab[vfitab+2];
358             Heps2            = vfeps*vfeps*vftab[vfitab+3];
359             Fp               = F+Geps+Heps2;
360             FF               = Fp+Geps+2.0*Heps2;
361             felec            = -qq00*FF*vftabscale*rinv00;
362
363             fscal            = felec;
364
365             /* Calculate temporary vectorial force */
366             tx               = fscal*dx00;
367             ty               = fscal*dy00;
368             tz               = fscal*dz00;
369
370             /* Update vectorial force */
371             fix0            += tx;
372             fiy0            += ty;
373             fiz0            += tz;
374             f[j_coord_offset+DIM*0+XX] -= tx;
375             f[j_coord_offset+DIM*0+YY] -= ty;
376             f[j_coord_offset+DIM*0+ZZ] -= tz;
377
378             /* Inner loop uses 38 flops */
379         }
380         /* End of innermost loop */
381
382         tx = ty = tz = 0;
383         f[i_coord_offset+DIM*0+XX] += fix0;
384         f[i_coord_offset+DIM*0+YY] += fiy0;
385         f[i_coord_offset+DIM*0+ZZ] += fiz0;
386         tx                         += fix0;
387         ty                         += fiy0;
388         tz                         += fiz0;
389         fshift[i_shift_offset+XX]  += tx;
390         fshift[i_shift_offset+YY]  += ty;
391         fshift[i_shift_offset+ZZ]  += tz;
392
393         /* Increment number of inner iterations */
394         inneriter                  += j_index_end - j_index_start;
395
396         /* Outer loop uses 13 flops */
397     }
398
399     /* Increment number of outer iterations */
400     outeriter        += nri;
401
402     /* Update outer/inner flops */
403
404     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_F,outeriter*13 + inneriter*38);
405 }