Introduce gmxpre.h for truly global definitions
[alexxy/gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_c / nb_kernel_ElecNone_VdwLJEwSh_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_ElecNone_VdwLJEwSh_GeomP1P1_VF_c
51  * Electrostatics interaction: None
52  * VdW interaction:            LJEwald
53  * Geometry:                   Particle-Particle
54  * Calculate force/pot:        PotentialAndForce
55  */
56 void
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)
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     int              nvdwtype;
78     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
79     int              *vdwtype;
80     real             *vdwparam;
81     real             c6grid_00;
82     real             ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
83     real             *vdwgridparam;
84
85     x                = xx[0];
86     f                = ff[0];
87
88     nri              = nlist->nri;
89     iinr             = nlist->iinr;
90     jindex           = nlist->jindex;
91     jjnr             = nlist->jjnr;
92     shiftidx         = nlist->shift;
93     gid              = nlist->gid;
94     shiftvec         = fr->shift_vec[0];
95     fshift           = fr->fshift[0];
96     nvdwtype         = fr->ntype;
97     vdwparam         = fr->nbfp;
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;
104
105     rcutoff          = fr->rvdw;
106     rcutoff2         = rcutoff*rcutoff;
107
108     sh_vdw_invrcut6  = fr->ic->sh_invrc6;
109     rvdw             = fr->rvdw;
110
111     outeriter        = 0;
112     inneriter        = 0;
113
114     /* Start outer loop over neighborlists */
115     for(iidx=0; iidx<nri; iidx++)
116     {
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];
122
123         /* Load limits for loop over neighbors */
124         j_index_start    = jindex[iidx];
125         j_index_end      = jindex[iidx+1];
126
127         /* Get outer coordinate index */
128         inr              = iinr[iidx];
129         i_coord_offset   = DIM*inr;
130
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];
135
136         fix0             = 0.0;
137         fiy0             = 0.0;
138         fiz0             = 0.0;
139
140         /* Load parameters for i particles */
141         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
142
143         /* Reset potential sums */
144         vvdwsum          = 0.0;
145
146         /* Start inner kernel loop */
147         for(jidx=j_index_start; jidx<j_index_end; jidx++)
148         {
149             /* Get j neighbor index, and coordinate index */
150             jnr              = jjnr[jidx];
151             j_coord_offset   = DIM*jnr;
152
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];
157
158             /* Calculate displacement vector */
159             dx00             = ix0 - jx0;
160             dy00             = iy0 - jy0;
161             dz00             = iz0 - jz0;
162
163             /* Calculate squared distance and things based on it */
164             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
165
166             rinv00           = gmx_invsqrt(rsq00);
167
168             rinvsq00         = rinv00*rinv00;
169
170             /* Load parameters for j particles */
171             vdwjidx0         = 2*vdwtype[jnr+0];
172
173             /**************************
174              * CALCULATE INTERACTIONS *
175              **************************/
176
177             if (rsq00<rcutoff2)
178             {
179
180             r00              = rsq00*rinv00;
181
182             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
183             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
184             c6grid_00        = vdwgridparam[vdwioffset0+vdwjidx0];
185
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;
194
195             /* Update potential sums from outer loop */
196             vvdwsum         += vvdw;
197
198             fscal            = fvdw;
199
200             /* Calculate temporary vectorial force */
201             tx               = fscal*dx00;
202             ty               = fscal*dy00;
203             tz               = fscal*dz00;
204
205             /* Update vectorial force */
206             fix0            += tx;
207             fiy0            += ty;
208             fiz0            += tz;
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;
212
213             }
214
215             /* Inner loop uses 55 flops */
216         }
217         /* End of innermost loop */
218
219         tx = ty = tz = 0;
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;
223         tx                         += fix0;
224         ty                         += fiy0;
225         tz                         += fiz0;
226         fshift[i_shift_offset+XX]  += tx;
227         fshift[i_shift_offset+YY]  += ty;
228         fshift[i_shift_offset+ZZ]  += tz;
229
230         ggid                        = gid[iidx];
231         /* Update potential energies */
232         kernel_data->energygrp_vdw[ggid] += vvdwsum;
233
234         /* Increment number of inner iterations */
235         inneriter                  += j_index_end - j_index_start;
236
237         /* Outer loop uses 13 flops */
238     }
239
240     /* Increment number of outer iterations */
241     outeriter        += nri;
242
243     /* Update outer/inner flops */
244
245     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*13 + inneriter*55);
246 }
247 /*
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
253  */
254 void
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)
263 {
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;
270     int              vdwioffset0;
271     real             ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
272     int              vdwjidx0;
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;
275     int              nvdwtype;
276     real             rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
277     int              *vdwtype;
278     real             *vdwparam;
279     real             c6grid_00;
280     real             ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,sh_lj_ewald;
281     real             *vdwgridparam;
282
283     x                = xx[0];
284     f                = ff[0];
285
286     nri              = nlist->nri;
287     iinr             = nlist->iinr;
288     jindex           = nlist->jindex;
289     jjnr             = nlist->jjnr;
290     shiftidx         = nlist->shift;
291     gid              = nlist->gid;
292     shiftvec         = fr->shift_vec[0];
293     fshift           = fr->fshift[0];
294     nvdwtype         = fr->ntype;
295     vdwparam         = fr->nbfp;
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;
302
303     rcutoff          = fr->rvdw;
304     rcutoff2         = rcutoff*rcutoff;
305
306     sh_vdw_invrcut6  = fr->ic->sh_invrc6;
307     rvdw             = fr->rvdw;
308
309     outeriter        = 0;
310     inneriter        = 0;
311
312     /* Start outer loop over neighborlists */
313     for(iidx=0; iidx<nri; iidx++)
314     {
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];
320
321         /* Load limits for loop over neighbors */
322         j_index_start    = jindex[iidx];
323         j_index_end      = jindex[iidx+1];
324
325         /* Get outer coordinate index */
326         inr              = iinr[iidx];
327         i_coord_offset   = DIM*inr;
328
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];
333
334         fix0             = 0.0;
335         fiy0             = 0.0;
336         fiz0             = 0.0;
337
338         /* Load parameters for i particles */
339         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
340
341         /* Start inner kernel loop */
342         for(jidx=j_index_start; jidx<j_index_end; jidx++)
343         {
344             /* Get j neighbor index, and coordinate index */
345             jnr              = jjnr[jidx];
346             j_coord_offset   = DIM*jnr;
347
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];
352
353             /* Calculate displacement vector */
354             dx00             = ix0 - jx0;
355             dy00             = iy0 - jy0;
356             dz00             = iz0 - jz0;
357
358             /* Calculate squared distance and things based on it */
359             rsq00            = dx00*dx00+dy00*dy00+dz00*dz00;
360
361             rinv00           = gmx_invsqrt(rsq00);
362
363             rinvsq00         = rinv00*rinv00;
364
365             /* Load parameters for j particles */
366             vdwjidx0         = 2*vdwtype[jnr+0];
367
368             /**************************
369              * CALCULATE INTERACTIONS *
370              **************************/
371
372             if (rsq00<rcutoff2)
373             {
374
375             r00              = rsq00*rinv00;
376
377             c6_00            = vdwparam[vdwioffset0+vdwjidx0];
378             c12_00           = vdwparam[vdwioffset0+vdwjidx0+1];
379             c6grid_00        = vdwgridparam[vdwioffset0+vdwjidx0];
380
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;
386
387             fscal            = fvdw;
388
389             /* Calculate temporary vectorial force */
390             tx               = fscal*dx00;
391             ty               = fscal*dy00;
392             tz               = fscal*dz00;
393
394             /* Update vectorial force */
395             fix0            += tx;
396             fiy0            += ty;
397             fiz0            += tz;
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;
401
402             }
403
404             /* Inner loop uses 44 flops */
405         }
406         /* End of innermost loop */
407
408         tx = ty = tz = 0;
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;
412         tx                         += fix0;
413         ty                         += fiy0;
414         tz                         += fiz0;
415         fshift[i_shift_offset+XX]  += tx;
416         fshift[i_shift_offset+YY]  += ty;
417         fshift[i_shift_offset+ZZ]  += tz;
418
419         /* Increment number of inner iterations */
420         inneriter                  += j_index_end - j_index_start;
421
422         /* Outer loop uses 12 flops */
423     }
424
425     /* Increment number of outer iterations */
426     outeriter        += nri;
427
428     /* Update outer/inner flops */
429
430     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*12 + inneriter*44);
431 }