2 * This file is part of the GROMACS molecular simulation package.
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.
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 "gromacs/legacyheaders/types/simple.h"
46 #include "gromacs/math/vec.h"
47 #include "gromacs/legacyheaders/nrnb.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_VF_c
51 * Electrostatics interaction: CubicSplineTable
52 * VdW interaction: CubicSplineTable
53 * Geometry: Particle-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecCSTab_VdwCSTab_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;
77 real velec,felec,velecsum,facel,crf,krf,krf2;
80 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
84 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
92 jindex = nlist->jindex;
94 shiftidx = nlist->shift;
96 shiftvec = fr->shift_vec[0];
97 fshift = fr->fshift[0];
99 charge = mdatoms->chargeA;
100 nvdwtype = fr->ntype;
102 vdwtype = mdatoms->typeA;
104 vftab = kernel_data->table_elec_vdw->data;
105 vftabscale = kernel_data->table_elec_vdw->scale;
110 /* Start outer loop over neighborlists */
111 for(iidx=0; iidx<nri; iidx++)
113 /* Load shift vector for this list */
114 i_shift_offset = DIM*shiftidx[iidx];
115 shX = shiftvec[i_shift_offset+XX];
116 shY = shiftvec[i_shift_offset+YY];
117 shZ = shiftvec[i_shift_offset+ZZ];
119 /* Load limits for loop over neighbors */
120 j_index_start = jindex[iidx];
121 j_index_end = jindex[iidx+1];
123 /* Get outer coordinate index */
125 i_coord_offset = DIM*inr;
127 /* Load i particle coords and add shift vector */
128 ix0 = shX + x[i_coord_offset+DIM*0+XX];
129 iy0 = shY + x[i_coord_offset+DIM*0+YY];
130 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
136 /* Load parameters for i particles */
137 iq0 = facel*charge[inr+0];
138 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
140 /* Reset potential sums */
144 /* Start inner kernel loop */
145 for(jidx=j_index_start; jidx<j_index_end; jidx++)
147 /* Get j neighbor index, and coordinate index */
149 j_coord_offset = DIM*jnr;
151 /* load j atom coordinates */
152 jx0 = x[j_coord_offset+DIM*0+XX];
153 jy0 = x[j_coord_offset+DIM*0+YY];
154 jz0 = x[j_coord_offset+DIM*0+ZZ];
156 /* Calculate displacement vector */
161 /* Calculate squared distance and things based on it */
162 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
164 rinv00 = gmx_invsqrt(rsq00);
166 /* Load parameters for j particles */
168 vdwjidx0 = 2*vdwtype[jnr+0];
170 /**************************
171 * CALCULATE INTERACTIONS *
172 **************************/
177 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
178 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
180 /* Calculate table index by multiplying r with table scale and truncate to integer */
186 /* CUBIC SPLINE TABLE ELECTROSTATICS */
189 Geps = vfeps*vftab[vfitab+2];
190 Heps2 = vfeps*vfeps*vftab[vfitab+3];
194 FF = Fp+Geps+2.0*Heps2;
195 felec = -qq00*FF*vftabscale*rinv00;
197 /* CUBIC SPLINE TABLE DISPERSION */
201 Geps = vfeps*vftab[vfitab+2];
202 Heps2 = vfeps*vfeps*vftab[vfitab+3];
206 FF = Fp+Geps+2.0*Heps2;
209 /* CUBIC SPLINE TABLE REPULSION */
212 Geps = vfeps*vftab[vfitab+6];
213 Heps2 = vfeps*vfeps*vftab[vfitab+7];
217 FF = Fp+Geps+2.0*Heps2;
220 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
222 /* Update potential sums from outer loop */
228 /* Calculate temporary vectorial force */
233 /* Update vectorial force */
237 f[j_coord_offset+DIM*0+XX] -= tx;
238 f[j_coord_offset+DIM*0+YY] -= ty;
239 f[j_coord_offset+DIM*0+ZZ] -= tz;
241 /* Inner loop uses 73 flops */
243 /* End of innermost loop */
246 f[i_coord_offset+DIM*0+XX] += fix0;
247 f[i_coord_offset+DIM*0+YY] += fiy0;
248 f[i_coord_offset+DIM*0+ZZ] += fiz0;
252 fshift[i_shift_offset+XX] += tx;
253 fshift[i_shift_offset+YY] += ty;
254 fshift[i_shift_offset+ZZ] += tz;
257 /* Update potential energies */
258 kernel_data->energygrp_elec[ggid] += velecsum;
259 kernel_data->energygrp_vdw[ggid] += vvdwsum;
261 /* Increment number of inner iterations */
262 inneriter += j_index_end - j_index_start;
264 /* Outer loop uses 15 flops */
267 /* Increment number of outer iterations */
270 /* Update outer/inner flops */
272 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*15 + inneriter*73);
275 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_F_c
276 * Electrostatics interaction: CubicSplineTable
277 * VdW interaction: CubicSplineTable
278 * Geometry: Particle-Particle
279 * Calculate force/pot: Force
282 nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_F_c
283 (t_nblist * gmx_restrict nlist,
284 rvec * gmx_restrict xx,
285 rvec * gmx_restrict ff,
286 t_forcerec * gmx_restrict fr,
287 t_mdatoms * gmx_restrict mdatoms,
288 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
289 t_nrnb * gmx_restrict nrnb)
291 int i_shift_offset,i_coord_offset,j_coord_offset;
292 int j_index_start,j_index_end;
293 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
294 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
295 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
296 real *shiftvec,*fshift,*x,*f;
298 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
300 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
301 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
302 real velec,felec,velecsum,facel,crf,krf,krf2;
305 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
309 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
317 jindex = nlist->jindex;
319 shiftidx = nlist->shift;
321 shiftvec = fr->shift_vec[0];
322 fshift = fr->fshift[0];
324 charge = mdatoms->chargeA;
325 nvdwtype = fr->ntype;
327 vdwtype = mdatoms->typeA;
329 vftab = kernel_data->table_elec_vdw->data;
330 vftabscale = kernel_data->table_elec_vdw->scale;
335 /* Start outer loop over neighborlists */
336 for(iidx=0; iidx<nri; iidx++)
338 /* Load shift vector for this list */
339 i_shift_offset = DIM*shiftidx[iidx];
340 shX = shiftvec[i_shift_offset+XX];
341 shY = shiftvec[i_shift_offset+YY];
342 shZ = shiftvec[i_shift_offset+ZZ];
344 /* Load limits for loop over neighbors */
345 j_index_start = jindex[iidx];
346 j_index_end = jindex[iidx+1];
348 /* Get outer coordinate index */
350 i_coord_offset = DIM*inr;
352 /* Load i particle coords and add shift vector */
353 ix0 = shX + x[i_coord_offset+DIM*0+XX];
354 iy0 = shY + x[i_coord_offset+DIM*0+YY];
355 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
361 /* Load parameters for i particles */
362 iq0 = facel*charge[inr+0];
363 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
365 /* Start inner kernel loop */
366 for(jidx=j_index_start; jidx<j_index_end; jidx++)
368 /* Get j neighbor index, and coordinate index */
370 j_coord_offset = DIM*jnr;
372 /* load j atom coordinates */
373 jx0 = x[j_coord_offset+DIM*0+XX];
374 jy0 = x[j_coord_offset+DIM*0+YY];
375 jz0 = x[j_coord_offset+DIM*0+ZZ];
377 /* Calculate displacement vector */
382 /* Calculate squared distance and things based on it */
383 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
385 rinv00 = gmx_invsqrt(rsq00);
387 /* Load parameters for j particles */
389 vdwjidx0 = 2*vdwtype[jnr+0];
391 /**************************
392 * CALCULATE INTERACTIONS *
393 **************************/
398 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
399 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
401 /* Calculate table index by multiplying r with table scale and truncate to integer */
407 /* CUBIC SPLINE TABLE ELECTROSTATICS */
409 Geps = vfeps*vftab[vfitab+2];
410 Heps2 = vfeps*vfeps*vftab[vfitab+3];
412 FF = Fp+Geps+2.0*Heps2;
413 felec = -qq00*FF*vftabscale*rinv00;
415 /* CUBIC SPLINE TABLE DISPERSION */
418 Geps = vfeps*vftab[vfitab+2];
419 Heps2 = vfeps*vfeps*vftab[vfitab+3];
421 FF = Fp+Geps+2.0*Heps2;
424 /* CUBIC SPLINE TABLE REPULSION */
426 Geps = vfeps*vftab[vfitab+6];
427 Heps2 = vfeps*vfeps*vftab[vfitab+7];
429 FF = Fp+Geps+2.0*Heps2;
431 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
435 /* Calculate temporary vectorial force */
440 /* Update vectorial force */
444 f[j_coord_offset+DIM*0+XX] -= tx;
445 f[j_coord_offset+DIM*0+YY] -= ty;
446 f[j_coord_offset+DIM*0+ZZ] -= tz;
448 /* Inner loop uses 61 flops */
450 /* End of innermost loop */
453 f[i_coord_offset+DIM*0+XX] += fix0;
454 f[i_coord_offset+DIM*0+YY] += fiy0;
455 f[i_coord_offset+DIM*0+ZZ] += fiz0;
459 fshift[i_shift_offset+XX] += tx;
460 fshift[i_shift_offset+YY] += ty;
461 fshift[i_shift_offset+ZZ] += tz;
463 /* Increment number of inner iterations */
464 inneriter += j_index_end - j_index_start;
466 /* Outer loop uses 13 flops */
469 /* Increment number of outer iterations */
472 /* Update outer/inner flops */
474 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*13 + inneriter*61);