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.
42 #include "../nb_kernel.h"
43 #include "types/simple.h"
44 #include "gromacs/math/vec.h"
48 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW4P1_VF_c
49 * Electrostatics interaction: CubicSplineTable
50 * VdW interaction: LennardJones
51 * Geometry: Water4-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecCSTab_VdwLJ_GeomW4P1_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)
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;
71 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
73 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
75 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
77 real ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
79 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
80 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
81 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
82 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
83 real dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
84 real velec,felec,velecsum,facel,crf,krf,krf2;
87 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
91 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
99 jindex = nlist->jindex;
101 shiftidx = nlist->shift;
103 shiftvec = fr->shift_vec[0];
104 fshift = fr->fshift[0];
106 charge = mdatoms->chargeA;
107 nvdwtype = fr->ntype;
109 vdwtype = mdatoms->typeA;
111 vftab = kernel_data->table_elec->data;
112 vftabscale = kernel_data->table_elec->scale;
114 /* Setup water-specific parameters */
115 inr = nlist->iinr[0];
116 iq1 = facel*charge[inr+1];
117 iq2 = facel*charge[inr+2];
118 iq3 = facel*charge[inr+3];
119 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
124 /* Start outer loop over neighborlists */
125 for(iidx=0; iidx<nri; iidx++)
127 /* Load shift vector for this list */
128 i_shift_offset = DIM*shiftidx[iidx];
129 shX = shiftvec[i_shift_offset+XX];
130 shY = shiftvec[i_shift_offset+YY];
131 shZ = shiftvec[i_shift_offset+ZZ];
133 /* Load limits for loop over neighbors */
134 j_index_start = jindex[iidx];
135 j_index_end = jindex[iidx+1];
137 /* Get outer coordinate index */
139 i_coord_offset = DIM*inr;
141 /* Load i particle coords and add shift vector */
142 ix0 = shX + x[i_coord_offset+DIM*0+XX];
143 iy0 = shY + x[i_coord_offset+DIM*0+YY];
144 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
145 ix1 = shX + x[i_coord_offset+DIM*1+XX];
146 iy1 = shY + x[i_coord_offset+DIM*1+YY];
147 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
148 ix2 = shX + x[i_coord_offset+DIM*2+XX];
149 iy2 = shY + x[i_coord_offset+DIM*2+YY];
150 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
151 ix3 = shX + x[i_coord_offset+DIM*3+XX];
152 iy3 = shY + x[i_coord_offset+DIM*3+YY];
153 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
168 /* Reset potential sums */
172 /* Start inner kernel loop */
173 for(jidx=j_index_start; jidx<j_index_end; jidx++)
175 /* Get j neighbor index, and coordinate index */
177 j_coord_offset = DIM*jnr;
179 /* load j atom coordinates */
180 jx0 = x[j_coord_offset+DIM*0+XX];
181 jy0 = x[j_coord_offset+DIM*0+YY];
182 jz0 = x[j_coord_offset+DIM*0+ZZ];
184 /* Calculate displacement vector */
198 /* Calculate squared distance and things based on it */
199 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
200 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
201 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
202 rsq30 = dx30*dx30+dy30*dy30+dz30*dz30;
204 rinv10 = gmx_invsqrt(rsq10);
205 rinv20 = gmx_invsqrt(rsq20);
206 rinv30 = gmx_invsqrt(rsq30);
208 rinvsq00 = 1.0/rsq00;
210 /* Load parameters for j particles */
212 vdwjidx0 = 2*vdwtype[jnr+0];
214 /**************************
215 * CALCULATE INTERACTIONS *
216 **************************/
218 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
219 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
221 /* LENNARD-JONES DISPERSION/REPULSION */
223 rinvsix = rinvsq00*rinvsq00*rinvsq00;
224 vvdw6 = c6_00*rinvsix;
225 vvdw12 = c12_00*rinvsix*rinvsix;
226 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
227 fvdw = (vvdw12-vvdw6)*rinvsq00;
229 /* Update potential sums from outer loop */
234 /* Calculate temporary vectorial force */
239 /* Update vectorial force */
243 f[j_coord_offset+DIM*0+XX] -= tx;
244 f[j_coord_offset+DIM*0+YY] -= ty;
245 f[j_coord_offset+DIM*0+ZZ] -= tz;
247 /**************************
248 * CALCULATE INTERACTIONS *
249 **************************/
255 /* Calculate table index by multiplying r with table scale and truncate to integer */
261 /* CUBIC SPLINE TABLE ELECTROSTATICS */
264 Geps = vfeps*vftab[vfitab+2];
265 Heps2 = vfeps*vfeps*vftab[vfitab+3];
269 FF = Fp+Geps+2.0*Heps2;
270 felec = -qq10*FF*vftabscale*rinv10;
272 /* Update potential sums from outer loop */
277 /* Calculate temporary vectorial force */
282 /* Update vectorial force */
286 f[j_coord_offset+DIM*0+XX] -= tx;
287 f[j_coord_offset+DIM*0+YY] -= ty;
288 f[j_coord_offset+DIM*0+ZZ] -= tz;
290 /**************************
291 * CALCULATE INTERACTIONS *
292 **************************/
298 /* Calculate table index by multiplying r with table scale and truncate to integer */
304 /* CUBIC SPLINE TABLE ELECTROSTATICS */
307 Geps = vfeps*vftab[vfitab+2];
308 Heps2 = vfeps*vfeps*vftab[vfitab+3];
312 FF = Fp+Geps+2.0*Heps2;
313 felec = -qq20*FF*vftabscale*rinv20;
315 /* Update potential sums from outer loop */
320 /* Calculate temporary vectorial force */
325 /* Update vectorial force */
329 f[j_coord_offset+DIM*0+XX] -= tx;
330 f[j_coord_offset+DIM*0+YY] -= ty;
331 f[j_coord_offset+DIM*0+ZZ] -= tz;
333 /**************************
334 * CALCULATE INTERACTIONS *
335 **************************/
341 /* Calculate table index by multiplying r with table scale and truncate to integer */
347 /* CUBIC SPLINE TABLE ELECTROSTATICS */
350 Geps = vfeps*vftab[vfitab+2];
351 Heps2 = vfeps*vfeps*vftab[vfitab+3];
355 FF = Fp+Geps+2.0*Heps2;
356 felec = -qq30*FF*vftabscale*rinv30;
358 /* Update potential sums from outer loop */
363 /* Calculate temporary vectorial force */
368 /* Update vectorial force */
372 f[j_coord_offset+DIM*0+XX] -= tx;
373 f[j_coord_offset+DIM*0+YY] -= ty;
374 f[j_coord_offset+DIM*0+ZZ] -= tz;
376 /* Inner loop uses 158 flops */
378 /* End of innermost loop */
381 f[i_coord_offset+DIM*0+XX] += fix0;
382 f[i_coord_offset+DIM*0+YY] += fiy0;
383 f[i_coord_offset+DIM*0+ZZ] += fiz0;
387 f[i_coord_offset+DIM*1+XX] += fix1;
388 f[i_coord_offset+DIM*1+YY] += fiy1;
389 f[i_coord_offset+DIM*1+ZZ] += fiz1;
393 f[i_coord_offset+DIM*2+XX] += fix2;
394 f[i_coord_offset+DIM*2+YY] += fiy2;
395 f[i_coord_offset+DIM*2+ZZ] += fiz2;
399 f[i_coord_offset+DIM*3+XX] += fix3;
400 f[i_coord_offset+DIM*3+YY] += fiy3;
401 f[i_coord_offset+DIM*3+ZZ] += fiz3;
405 fshift[i_shift_offset+XX] += tx;
406 fshift[i_shift_offset+YY] += ty;
407 fshift[i_shift_offset+ZZ] += tz;
410 /* Update potential energies */
411 kernel_data->energygrp_elec[ggid] += velecsum;
412 kernel_data->energygrp_vdw[ggid] += vvdwsum;
414 /* Increment number of inner iterations */
415 inneriter += j_index_end - j_index_start;
417 /* Outer loop uses 41 flops */
420 /* Increment number of outer iterations */
423 /* Update outer/inner flops */
425 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*41 + inneriter*158);
428 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW4P1_F_c
429 * Electrostatics interaction: CubicSplineTable
430 * VdW interaction: LennardJones
431 * Geometry: Water4-Particle
432 * Calculate force/pot: Force
435 nb_kernel_ElecCSTab_VdwLJ_GeomW4P1_F_c
436 (t_nblist * gmx_restrict nlist,
437 rvec * gmx_restrict xx,
438 rvec * gmx_restrict ff,
439 t_forcerec * gmx_restrict fr,
440 t_mdatoms * gmx_restrict mdatoms,
441 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
442 t_nrnb * gmx_restrict nrnb)
444 int i_shift_offset,i_coord_offset,j_coord_offset;
445 int j_index_start,j_index_end;
446 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
447 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
448 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
449 real *shiftvec,*fshift,*x,*f;
451 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
453 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
455 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
457 real ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
459 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
460 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
461 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
462 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
463 real dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
464 real velec,felec,velecsum,facel,crf,krf,krf2;
467 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
471 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
479 jindex = nlist->jindex;
481 shiftidx = nlist->shift;
483 shiftvec = fr->shift_vec[0];
484 fshift = fr->fshift[0];
486 charge = mdatoms->chargeA;
487 nvdwtype = fr->ntype;
489 vdwtype = mdatoms->typeA;
491 vftab = kernel_data->table_elec->data;
492 vftabscale = kernel_data->table_elec->scale;
494 /* Setup water-specific parameters */
495 inr = nlist->iinr[0];
496 iq1 = facel*charge[inr+1];
497 iq2 = facel*charge[inr+2];
498 iq3 = facel*charge[inr+3];
499 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
504 /* Start outer loop over neighborlists */
505 for(iidx=0; iidx<nri; iidx++)
507 /* Load shift vector for this list */
508 i_shift_offset = DIM*shiftidx[iidx];
509 shX = shiftvec[i_shift_offset+XX];
510 shY = shiftvec[i_shift_offset+YY];
511 shZ = shiftvec[i_shift_offset+ZZ];
513 /* Load limits for loop over neighbors */
514 j_index_start = jindex[iidx];
515 j_index_end = jindex[iidx+1];
517 /* Get outer coordinate index */
519 i_coord_offset = DIM*inr;
521 /* Load i particle coords and add shift vector */
522 ix0 = shX + x[i_coord_offset+DIM*0+XX];
523 iy0 = shY + x[i_coord_offset+DIM*0+YY];
524 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
525 ix1 = shX + x[i_coord_offset+DIM*1+XX];
526 iy1 = shY + x[i_coord_offset+DIM*1+YY];
527 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
528 ix2 = shX + x[i_coord_offset+DIM*2+XX];
529 iy2 = shY + x[i_coord_offset+DIM*2+YY];
530 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
531 ix3 = shX + x[i_coord_offset+DIM*3+XX];
532 iy3 = shY + x[i_coord_offset+DIM*3+YY];
533 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
548 /* Start inner kernel loop */
549 for(jidx=j_index_start; jidx<j_index_end; jidx++)
551 /* Get j neighbor index, and coordinate index */
553 j_coord_offset = DIM*jnr;
555 /* load j atom coordinates */
556 jx0 = x[j_coord_offset+DIM*0+XX];
557 jy0 = x[j_coord_offset+DIM*0+YY];
558 jz0 = x[j_coord_offset+DIM*0+ZZ];
560 /* Calculate displacement vector */
574 /* Calculate squared distance and things based on it */
575 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
576 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
577 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
578 rsq30 = dx30*dx30+dy30*dy30+dz30*dz30;
580 rinv10 = gmx_invsqrt(rsq10);
581 rinv20 = gmx_invsqrt(rsq20);
582 rinv30 = gmx_invsqrt(rsq30);
584 rinvsq00 = 1.0/rsq00;
586 /* Load parameters for j particles */
588 vdwjidx0 = 2*vdwtype[jnr+0];
590 /**************************
591 * CALCULATE INTERACTIONS *
592 **************************/
594 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
595 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
597 /* LENNARD-JONES DISPERSION/REPULSION */
599 rinvsix = rinvsq00*rinvsq00*rinvsq00;
600 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
604 /* Calculate temporary vectorial force */
609 /* Update vectorial force */
613 f[j_coord_offset+DIM*0+XX] -= tx;
614 f[j_coord_offset+DIM*0+YY] -= ty;
615 f[j_coord_offset+DIM*0+ZZ] -= tz;
617 /**************************
618 * CALCULATE INTERACTIONS *
619 **************************/
625 /* Calculate table index by multiplying r with table scale and truncate to integer */
631 /* CUBIC SPLINE TABLE ELECTROSTATICS */
633 Geps = vfeps*vftab[vfitab+2];
634 Heps2 = vfeps*vfeps*vftab[vfitab+3];
636 FF = Fp+Geps+2.0*Heps2;
637 felec = -qq10*FF*vftabscale*rinv10;
641 /* Calculate temporary vectorial force */
646 /* Update vectorial force */
650 f[j_coord_offset+DIM*0+XX] -= tx;
651 f[j_coord_offset+DIM*0+YY] -= ty;
652 f[j_coord_offset+DIM*0+ZZ] -= tz;
654 /**************************
655 * CALCULATE INTERACTIONS *
656 **************************/
662 /* Calculate table index by multiplying r with table scale and truncate to integer */
668 /* CUBIC SPLINE TABLE ELECTROSTATICS */
670 Geps = vfeps*vftab[vfitab+2];
671 Heps2 = vfeps*vfeps*vftab[vfitab+3];
673 FF = Fp+Geps+2.0*Heps2;
674 felec = -qq20*FF*vftabscale*rinv20;
678 /* Calculate temporary vectorial force */
683 /* Update vectorial force */
687 f[j_coord_offset+DIM*0+XX] -= tx;
688 f[j_coord_offset+DIM*0+YY] -= ty;
689 f[j_coord_offset+DIM*0+ZZ] -= tz;
691 /**************************
692 * CALCULATE INTERACTIONS *
693 **************************/
699 /* Calculate table index by multiplying r with table scale and truncate to integer */
705 /* CUBIC SPLINE TABLE ELECTROSTATICS */
707 Geps = vfeps*vftab[vfitab+2];
708 Heps2 = vfeps*vfeps*vftab[vfitab+3];
710 FF = Fp+Geps+2.0*Heps2;
711 felec = -qq30*FF*vftabscale*rinv30;
715 /* Calculate temporary vectorial force */
720 /* Update vectorial force */
724 f[j_coord_offset+DIM*0+XX] -= tx;
725 f[j_coord_offset+DIM*0+YY] -= ty;
726 f[j_coord_offset+DIM*0+ZZ] -= tz;
728 /* Inner loop uses 141 flops */
730 /* End of innermost loop */
733 f[i_coord_offset+DIM*0+XX] += fix0;
734 f[i_coord_offset+DIM*0+YY] += fiy0;
735 f[i_coord_offset+DIM*0+ZZ] += fiz0;
739 f[i_coord_offset+DIM*1+XX] += fix1;
740 f[i_coord_offset+DIM*1+YY] += fiy1;
741 f[i_coord_offset+DIM*1+ZZ] += fiz1;
745 f[i_coord_offset+DIM*2+XX] += fix2;
746 f[i_coord_offset+DIM*2+YY] += fiy2;
747 f[i_coord_offset+DIM*2+ZZ] += fiz2;
751 f[i_coord_offset+DIM*3+XX] += fix3;
752 f[i_coord_offset+DIM*3+YY] += fiy3;
753 f[i_coord_offset+DIM*3+ZZ] += fiz3;
757 fshift[i_shift_offset+XX] += tx;
758 fshift[i_shift_offset+YY] += ty;
759 fshift[i_shift_offset+ZZ] += tz;
761 /* Increment number of inner iterations */
762 inneriter += j_index_end - j_index_start;
764 /* Outer loop uses 39 flops */
767 /* Increment number of outer iterations */
770 /* Update outer/inner flops */
772 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*39 + inneriter*141);