Merge release-5-0 into master
[alexxy/gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_c / nb_kernel_ElecEw_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 "gmxpre.h"
39
40 #include "config.h"
41
42 #include <math.h>
43
44 #include "../nb_kernel.h"
45 #include "gromacs/legacyheaders/types/simple.h"
46 #include "gromacs/math/vec.h"
47 #include "gromacs/legacyheaders/nrnb.h"
48
49 /*
50  * Gromacs nonbonded kernel:   nb_kernel_ElecEw_VdwNone_GeomP1P1_VF_c
51  * Electrostatics interaction: Ewald
52  * VdW interaction:            None
53  * Geometry:                   Particle-Particle
54  * Calculate force/pot:        PotentialAndForce
55  */
56 void
57 nb_kernel_ElecEw_VdwNone_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)
65 {
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;
72     int              vdwioffset0;
73     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
74     int              vdwjidx0;
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;
78     real             *charge;
79     int              ewitab;
80     real             ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
81     real             *ewtab;
82
83     x                = xx[0];
84     f                = ff[0];
85
86     nri              = nlist->nri;
87     iinr             = nlist->iinr;
88     jindex           = nlist->jindex;
89     jjnr             = nlist->jjnr;
90     shiftidx         = nlist->shift;
91     gid              = nlist->gid;
92     shiftvec         = fr->shift_vec[0];
93     fshift           = fr->fshift[0];
94     facel            = fr->epsfac;
95     charge           = mdatoms->chargeA;
96
97     sh_ewald         = fr->ic->sh_ewald;
98     ewtab            = fr->ic->tabq_coul_FDV0;
99     ewtabscale       = fr->ic->tabq_scale;
100     ewtabhalfspace   = 0.5/ewtabscale;
101
102     outeriter        = 0;
103     inneriter        = 0;
104
105     /* Start outer loop over neighborlists */
106     for(iidx=0; iidx<nri; iidx++)
107     {
108         /* Load shift vector for this list */
109         i_shift_offset   = DIM*shiftidx[iidx];
110         shX              = shiftvec[i_shift_offset+XX];
111         shY              = shiftvec[i_shift_offset+YY];
112         shZ              = shiftvec[i_shift_offset+ZZ];
113
114         /* Load limits for loop over neighbors */
115         j_index_start    = jindex[iidx];
116         j_index_end      = jindex[iidx+1];
117
118         /* Get outer coordinate index */
119         inr              = iinr[iidx];
120         i_coord_offset   = DIM*inr;
121
122         /* Load i particle coords and add shift vector */
123         ix0              = shX + x[i_coord_offset+DIM*0+XX];
124         iy0              = shY + x[i_coord_offset+DIM*0+YY];
125         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
126
127         fix0             = 0.0;
128         fiy0             = 0.0;
129         fiz0             = 0.0;
130
131         /* Load parameters for i particles */
132         iq0              = facel*charge[inr+0];
133
134         /* Reset potential sums */
135         velecsum         = 0.0;
136
137         /* Start inner kernel loop */
138         for(jidx=j_index_start; jidx<j_index_end; jidx++)
139         {
140             /* Get j neighbor index, and coordinate index */
141             jnr              = jjnr[jidx];
142             j_coord_offset   = DIM*jnr;
143
144             /* load j atom coordinates */
145             jx0              = x[j_coord_offset+DIM*0+XX];
146             jy0              = x[j_coord_offset+DIM*0+YY];
147             jz0              = x[j_coord_offset+DIM*0+ZZ];
148
149             /* Calculate displacement vector */
150             dx00             = ix0 - jx0;
151             dy00             = iy0 - jy0;
152             dz00             = iz0 - jz0;
153
154             /* Calculate squared distance and things based on it */
155             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
156
157             rinv00           = gmx_invsqrt(rsq00);
158
159             rinvsq00         = rinv00*rinv00;
160
161             /* Load parameters for j particles */
162             jq0              = charge[jnr+0];
163
164             /**************************
165              * CALCULATE INTERACTIONS *
166              **************************/
167
168             r00              = rsq00*rinv00;
169
170             qq00             = iq0*jq0;
171
172             /* EWALD ELECTROSTATICS */
173
174             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
175             ewrt             = r00*ewtabscale;
176             ewitab           = ewrt;
177             eweps            = ewrt-ewitab;
178             ewitab           = 4*ewitab;
179             felec            = ewtab[ewitab]+eweps*ewtab[ewitab+1];
180             velec            = qq00*(rinv00-(ewtab[ewitab+2]-ewtabhalfspace*eweps*(ewtab[ewitab]+felec)));
181             felec            = qq00*rinv00*(rinvsq00-felec);
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 41 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*41);
232 }
233 /*
234  * Gromacs nonbonded kernel:   nb_kernel_ElecEw_VdwNone_GeomP1P1_F_c
235  * Electrostatics interaction: Ewald
236  * VdW interaction:            None
237  * Geometry:                   Particle-Particle
238  * Calculate force/pot:        Force
239  */
240 void
241 nb_kernel_ElecEw_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              ewitab;
264     real             ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace;
265     real             *ewtab;
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     sh_ewald         = fr->ic->sh_ewald;
282     ewtab            = fr->ic->tabq_coul_F;
283     ewtabscale       = fr->ic->tabq_scale;
284     ewtabhalfspace   = 0.5/ewtabscale;
285
286     outeriter        = 0;
287     inneriter        = 0;
288
289     /* Start outer loop over neighborlists */
290     for(iidx=0; iidx<nri; iidx++)
291     {
292         /* Load shift vector for this list */
293         i_shift_offset   = DIM*shiftidx[iidx];
294         shX              = shiftvec[i_shift_offset+XX];
295         shY              = shiftvec[i_shift_offset+YY];
296         shZ              = shiftvec[i_shift_offset+ZZ];
297
298         /* Load limits for loop over neighbors */
299         j_index_start    = jindex[iidx];
300         j_index_end      = jindex[iidx+1];
301
302         /* Get outer coordinate index */
303         inr              = iinr[iidx];
304         i_coord_offset   = DIM*inr;
305
306         /* Load i particle coords and add shift vector */
307         ix0              = shX + x[i_coord_offset+DIM*0+XX];
308         iy0              = shY + x[i_coord_offset+DIM*0+YY];
309         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
310
311         fix0             = 0.0;
312         fiy0             = 0.0;
313         fiz0             = 0.0;
314
315         /* Load parameters for i particles */
316         iq0              = facel*charge[inr+0];
317
318         /* Start inner kernel loop */
319         for(jidx=j_index_start; jidx<j_index_end; jidx++)
320         {
321             /* Get j neighbor index, and coordinate index */
322             jnr              = jjnr[jidx];
323             j_coord_offset   = DIM*jnr;
324
325             /* load j atom coordinates */
326             jx0              = x[j_coord_offset+DIM*0+XX];
327             jy0              = x[j_coord_offset+DIM*0+YY];
328             jz0              = x[j_coord_offset+DIM*0+ZZ];
329
330             /* Calculate displacement vector */
331             dx00             = ix0 - jx0;
332             dy00             = iy0 - jy0;
333             dz00             = iz0 - jz0;
334
335             /* Calculate squared distance and things based on it */
336             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
337
338             rinv00           = gmx_invsqrt(rsq00);
339
340             rinvsq00         = rinv00*rinv00;
341
342             /* Load parameters for j particles */
343             jq0              = charge[jnr+0];
344
345             /**************************
346              * CALCULATE INTERACTIONS *
347              **************************/
348
349             r00              = rsq00*rinv00;
350
351             qq00             = iq0*jq0;
352
353             /* EWALD ELECTROSTATICS */
354
355             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
356             ewrt             = r00*ewtabscale;
357             ewitab           = ewrt;
358             eweps            = ewrt-ewitab;
359             felec            = (1.0-eweps)*ewtab[ewitab]+eweps*ewtab[ewitab+1];
360             felec            = qq00*rinv00*(rinvsq00-felec);
361
362             fscal            = felec;
363
364             /* Calculate temporary vectorial force */
365             tx               = fscal*dx00;
366             ty               = fscal*dy00;
367             tz               = fscal*dz00;
368
369             /* Update vectorial force */
370             fix0            += tx;
371             fiy0            += ty;
372             fiz0            += tz;
373             f[j_coord_offset+DIM*0+XX] -= tx;
374             f[j_coord_offset+DIM*0+YY] -= ty;
375             f[j_coord_offset+DIM*0+ZZ] -= tz;
376
377             /* Inner loop uses 34 flops */
378         }
379         /* End of innermost loop */
380
381         tx = ty = tz = 0;
382         f[i_coord_offset+DIM*0+XX] += fix0;
383         f[i_coord_offset+DIM*0+YY] += fiy0;
384         f[i_coord_offset+DIM*0+ZZ] += fiz0;
385         tx                         += fix0;
386         ty                         += fiy0;
387         tz                         += fiz0;
388         fshift[i_shift_offset+XX]  += tx;
389         fshift[i_shift_offset+YY]  += ty;
390         fshift[i_shift_offset+ZZ]  += tz;
391
392         /* Increment number of inner iterations */
393         inneriter                  += j_index_end - j_index_start;
394
395         /* Outer loop uses 13 flops */
396     }
397
398     /* Increment number of outer iterations */
399     outeriter        += nri;
400
401     /* Update outer/inner flops */
402
403     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_F,outeriter*13 + inneriter*34);
404 }