Introduce gmxpre.h for truly global definitions
[alexxy/gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_c / nb_kernel_ElecCoul_VdwLJ_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_ElecCoul_VdwLJ_GeomP1P1_VF_c
51  * Electrostatics interaction: Coulomb
52  * VdW interaction:            LennardJones
53  * Geometry:                   Particle-Particle
54  * Calculate force/pot:        PotentialAndForce
55  */
56 void
57 nb_kernel_ElecCoul_VdwLJ_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              nvdwtype;
80     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
81     int              *vdwtype;
82     real             *vdwparam;
83
84     x                = xx[0];
85     f                = ff[0];
86
87     nri              = nlist->nri;
88     iinr             = nlist->iinr;
89     jindex           = nlist->jindex;
90     jjnr             = nlist->jjnr;
91     shiftidx         = nlist->shift;
92     gid              = nlist->gid;
93     shiftvec         = fr->shift_vec[0];
94     fshift           = fr->fshift[0];
95     facel            = fr->epsfac;
96     charge           = mdatoms->chargeA;
97     nvdwtype         = fr->ntype;
98     vdwparam         = fr->nbfp;
99     vdwtype          = mdatoms->typeA;
100
101     outeriter        = 0;
102     inneriter        = 0;
103
104     /* Start outer loop over neighborlists */
105     for(iidx=0; iidx<nri; iidx++)
106     {
107         /* Load shift vector for this list */
108         i_shift_offset   = DIM*shiftidx[iidx];
109         shX              = shiftvec[i_shift_offset+XX];
110         shY              = shiftvec[i_shift_offset+YY];
111         shZ              = shiftvec[i_shift_offset+ZZ];
112
113         /* Load limits for loop over neighbors */
114         j_index_start    = jindex[iidx];
115         j_index_end      = jindex[iidx+1];
116
117         /* Get outer coordinate index */
118         inr              = iinr[iidx];
119         i_coord_offset   = DIM*inr;
120
121         /* Load i particle coords and add shift vector */
122         ix0              = shX + x[i_coord_offset+DIM*0+XX];
123         iy0              = shY + x[i_coord_offset+DIM*0+YY];
124         iz0              = shZ + x[i_coord_offset+DIM*0+ZZ];
125
126         fix0             = 0.0;
127         fiy0             = 0.0;
128         fiz0             = 0.0;
129
130         /* Load parameters for i particles */
131         iq0              = facel*charge[inr+0];
132         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
133
134         /* Reset potential sums */
135         velecsum         = 0.0;
136         vvdwsum          = 0.0;
137
138         /* Start inner kernel loop */
139         for(jidx=j_index_start; jidx<j_index_end; jidx++)
140         {
141             /* Get j neighbor index, and coordinate index */
142             jnr              = jjnr[jidx];
143             j_coord_offset   = DIM*jnr;
144
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];
149
150             /* Calculate displacement vector */
151             dx00             = ix0 - jx0;
152             dy00             = iy0 - jy0;
153             dz00             = iz0 - jz0;
154
155             /* Calculate squared distance and things based on it */
156             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
157
158             rinv00           = gmx_invsqrt(rsq00);
159
160             rinvsq00         = rinv00*rinv00;
161
162             /* Load parameters for j particles */
163             jq0              = charge[jnr+0];
164             vdwjidx0         = 2*vdwtype[jnr+0];
165
166             /**************************
167              * CALCULATE INTERACTIONS *
168              **************************/
169
170             qq00             = iq0*jq0;
171             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
172             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
173
174             /* COULOMB ELECTROSTATICS */
175             velec            = qq00*rinv00;
176             felec            = velec*rinvsq00;
177
178             /* LENNARD-JONES DISPERSION/REPULSION */
179
180             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
181             vvdw6            = c6_00*rinvsix;
182             vvdw12           = c12_00*rinvsix*rinvsix;
183             vvdw             = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
184             fvdw             = (vvdw12-vvdw6)*rinvsq00;
185
186             /* Update potential sums from outer loop */
187             velecsum        += velec;
188             vvdwsum         += vvdw;
189
190             fscal            = felec+fvdw;
191
192             /* Calculate temporary vectorial force */
193             tx               = fscal*dx00;
194             ty               = fscal*dy00;
195             tz               = fscal*dz00;
196
197             /* Update vectorial force */
198             fix0            += tx;
199             fiy0            += ty;
200             fiz0            += tz;
201             f[j_coord_offset+DIM*0+XX] -= tx;
202             f[j_coord_offset+DIM*0+YY] -= ty;
203             f[j_coord_offset+DIM*0+ZZ] -= tz;
204
205             /* Inner loop uses 40 flops */
206         }
207         /* End of innermost loop */
208
209         tx = ty = tz = 0;
210         f[i_coord_offset+DIM*0+XX] += fix0;
211         f[i_coord_offset+DIM*0+YY] += fiy0;
212         f[i_coord_offset+DIM*0+ZZ] += fiz0;
213         tx                         += fix0;
214         ty                         += fiy0;
215         tz                         += fiz0;
216         fshift[i_shift_offset+XX]  += tx;
217         fshift[i_shift_offset+YY]  += ty;
218         fshift[i_shift_offset+ZZ]  += tz;
219
220         ggid                        = gid[iidx];
221         /* Update potential energies */
222         kernel_data->energygrp_elec[ggid] += velecsum;
223         kernel_data->energygrp_vdw[ggid] += vvdwsum;
224
225         /* Increment number of inner iterations */
226         inneriter                  += j_index_end - j_index_start;
227
228         /* Outer loop uses 15 flops */
229     }
230
231     /* Increment number of outer iterations */
232     outeriter        += nri;
233
234     /* Update outer/inner flops */
235
236     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*40);
237 }
238 /*
239  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwLJ_GeomP1P1_F_c
240  * Electrostatics interaction: Coulomb
241  * VdW interaction:            LennardJones
242  * Geometry:                   Particle-Particle
243  * Calculate force/pot:        Force
244  */
245 void
246 nb_kernel_ElecCoul_VdwLJ_GeomP1P1_F_c
247                     (t_nblist                    * gmx_restrict       nlist,
248                      rvec                        * gmx_restrict          xx,
249                      rvec                        * gmx_restrict          ff,
250                      t_forcerec                  * gmx_restrict          fr,
251                      t_mdatoms                   * gmx_restrict     mdatoms,
252                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
253                      t_nrnb                      * gmx_restrict        nrnb)
254 {
255     int              i_shift_offset,i_coord_offset,j_coord_offset;
256     int              j_index_start,j_index_end;
257     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
258     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
259     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
260     real             *shiftvec,*fshift,*x,*f;
261     int              vdwioffset0;
262     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
263     int              vdwjidx0;
264     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
265     real             dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
266     real             velec,felec,velecsum,facel,crf,krf,krf2;
267     real             *charge;
268     int              nvdwtype;
269     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
270     int              *vdwtype;
271     real             *vdwparam;
272
273     x                = xx[0];
274     f                = ff[0];
275
276     nri              = nlist->nri;
277     iinr             = nlist->iinr;
278     jindex           = nlist->jindex;
279     jjnr             = nlist->jjnr;
280     shiftidx         = nlist->shift;
281     gid              = nlist->gid;
282     shiftvec         = fr->shift_vec[0];
283     fshift           = fr->fshift[0];
284     facel            = fr->epsfac;
285     charge           = mdatoms->chargeA;
286     nvdwtype         = fr->ntype;
287     vdwparam         = fr->nbfp;
288     vdwtype          = mdatoms->typeA;
289
290     outeriter        = 0;
291     inneriter        = 0;
292
293     /* Start outer loop over neighborlists */
294     for(iidx=0; iidx<nri; iidx++)
295     {
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];
301
302         /* Load limits for loop over neighbors */
303         j_index_start    = jindex[iidx];
304         j_index_end      = jindex[iidx+1];
305
306         /* Get outer coordinate index */
307         inr              = iinr[iidx];
308         i_coord_offset   = DIM*inr;
309
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];
314
315         fix0             = 0.0;
316         fiy0             = 0.0;
317         fiz0             = 0.0;
318
319         /* Load parameters for i particles */
320         iq0              = facel*charge[inr+0];
321         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
322
323         /* Start inner kernel loop */
324         for(jidx=j_index_start; jidx<j_index_end; jidx++)
325         {
326             /* Get j neighbor index, and coordinate index */
327             jnr              = jjnr[jidx];
328             j_coord_offset   = DIM*jnr;
329
330             /* load j atom coordinates */
331             jx0              = x[j_coord_offset+DIM*0+XX];
332             jy0              = x[j_coord_offset+DIM*0+YY];
333             jz0              = x[j_coord_offset+DIM*0+ZZ];
334
335             /* Calculate displacement vector */
336             dx00             = ix0 - jx0;
337             dy00             = iy0 - jy0;
338             dz00             = iz0 - jz0;
339
340             /* Calculate squared distance and things based on it */
341             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
342
343             rinv00           = gmx_invsqrt(rsq00);
344
345             rinvsq00         = rinv00*rinv00;
346
347             /* Load parameters for j particles */
348             jq0              = charge[jnr+0];
349             vdwjidx0         = 2*vdwtype[jnr+0];
350
351             /**************************
352              * CALCULATE INTERACTIONS *
353              **************************/
354
355             qq00             = iq0*jq0;
356             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
357             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
358
359             /* COULOMB ELECTROSTATICS */
360             velec            = qq00*rinv00;
361             felec            = velec*rinvsq00;
362
363             /* LENNARD-JONES DISPERSION/REPULSION */
364
365             rinvsix          = rinvsq00*rinvsq00*rinvsq00;
366             fvdw             = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
367
368             fscal            = felec+fvdw;
369
370             /* Calculate temporary vectorial force */
371             tx               = fscal*dx00;
372             ty               = fscal*dy00;
373             tz               = fscal*dz00;
374
375             /* Update vectorial force */
376             fix0            += tx;
377             fiy0            += ty;
378             fiz0            += tz;
379             f[j_coord_offset+DIM*0+XX] -= tx;
380             f[j_coord_offset+DIM*0+YY] -= ty;
381             f[j_coord_offset+DIM*0+ZZ] -= tz;
382
383             /* Inner loop uses 34 flops */
384         }
385         /* End of innermost loop */
386
387         tx = ty = tz = 0;
388         f[i_coord_offset+DIM*0+XX] += fix0;
389         f[i_coord_offset+DIM*0+YY] += fiy0;
390         f[i_coord_offset+DIM*0+ZZ] += fiz0;
391         tx                         += fix0;
392         ty                         += fiy0;
393         tz                         += fiz0;
394         fshift[i_shift_offset+XX]  += tx;
395         fshift[i_shift_offset+YY]  += ty;
396         fshift[i_shift_offset+ZZ]  += tz;
397
398         /* Increment number of inner iterations */
399         inneriter                  += j_index_end - j_index_start;
400
401         /* Outer loop uses 13 flops */
402     }
403
404     /* Increment number of outer iterations */
405     outeriter        += nri;
406
407     /* Update outer/inner flops */
408
409     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*34);
410 }