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_ElecCoul_VdwCSTab_GeomW4P1_VF_c
49 * Electrostatics interaction: Coulomb
50 * VdW interaction: CubicSplineTable
51 * Geometry: Water4-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecCoul_VdwCSTab_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_vdw->data;
112 vftabscale = kernel_data->table_vdw->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 rinv00 = gmx_invsqrt(rsq00);
205 rinv10 = gmx_invsqrt(rsq10);
206 rinv20 = gmx_invsqrt(rsq20);
207 rinv30 = gmx_invsqrt(rsq30);
209 rinvsq10 = rinv10*rinv10;
210 rinvsq20 = rinv20*rinv20;
211 rinvsq30 = rinv30*rinv30;
213 /* Load parameters for j particles */
215 vdwjidx0 = 2*vdwtype[jnr+0];
217 /**************************
218 * CALCULATE INTERACTIONS *
219 **************************/
223 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
224 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
226 /* Calculate table index by multiplying r with table scale and truncate to integer */
232 /* CUBIC SPLINE TABLE DISPERSION */
236 Geps = vfeps*vftab[vfitab+2];
237 Heps2 = vfeps*vfeps*vftab[vfitab+3];
241 FF = Fp+Geps+2.0*Heps2;
244 /* CUBIC SPLINE TABLE REPULSION */
247 Geps = vfeps*vftab[vfitab+6];
248 Heps2 = vfeps*vfeps*vftab[vfitab+7];
252 FF = Fp+Geps+2.0*Heps2;
255 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
257 /* Update potential sums from outer loop */
262 /* Calculate temporary vectorial force */
267 /* Update vectorial force */
271 f[j_coord_offset+DIM*0+XX] -= tx;
272 f[j_coord_offset+DIM*0+YY] -= ty;
273 f[j_coord_offset+DIM*0+ZZ] -= tz;
275 /**************************
276 * CALCULATE INTERACTIONS *
277 **************************/
281 /* COULOMB ELECTROSTATICS */
283 felec = velec*rinvsq10;
285 /* Update potential sums from outer loop */
290 /* Calculate temporary vectorial force */
295 /* Update vectorial force */
299 f[j_coord_offset+DIM*0+XX] -= tx;
300 f[j_coord_offset+DIM*0+YY] -= ty;
301 f[j_coord_offset+DIM*0+ZZ] -= tz;
303 /**************************
304 * CALCULATE INTERACTIONS *
305 **************************/
309 /* COULOMB ELECTROSTATICS */
311 felec = velec*rinvsq20;
313 /* Update potential sums from outer loop */
318 /* Calculate temporary vectorial force */
323 /* Update vectorial force */
327 f[j_coord_offset+DIM*0+XX] -= tx;
328 f[j_coord_offset+DIM*0+YY] -= ty;
329 f[j_coord_offset+DIM*0+ZZ] -= tz;
331 /**************************
332 * CALCULATE INTERACTIONS *
333 **************************/
337 /* COULOMB ELECTROSTATICS */
339 felec = velec*rinvsq30;
341 /* Update potential sums from outer loop */
346 /* Calculate temporary vectorial force */
351 /* Update vectorial force */
355 f[j_coord_offset+DIM*0+XX] -= tx;
356 f[j_coord_offset+DIM*0+YY] -= ty;
357 f[j_coord_offset+DIM*0+ZZ] -= tz;
359 /* Inner loop uses 139 flops */
361 /* End of innermost loop */
364 f[i_coord_offset+DIM*0+XX] += fix0;
365 f[i_coord_offset+DIM*0+YY] += fiy0;
366 f[i_coord_offset+DIM*0+ZZ] += fiz0;
370 f[i_coord_offset+DIM*1+XX] += fix1;
371 f[i_coord_offset+DIM*1+YY] += fiy1;
372 f[i_coord_offset+DIM*1+ZZ] += fiz1;
376 f[i_coord_offset+DIM*2+XX] += fix2;
377 f[i_coord_offset+DIM*2+YY] += fiy2;
378 f[i_coord_offset+DIM*2+ZZ] += fiz2;
382 f[i_coord_offset+DIM*3+XX] += fix3;
383 f[i_coord_offset+DIM*3+YY] += fiy3;
384 f[i_coord_offset+DIM*3+ZZ] += fiz3;
388 fshift[i_shift_offset+XX] += tx;
389 fshift[i_shift_offset+YY] += ty;
390 fshift[i_shift_offset+ZZ] += tz;
393 /* Update potential energies */
394 kernel_data->energygrp_elec[ggid] += velecsum;
395 kernel_data->energygrp_vdw[ggid] += vvdwsum;
397 /* Increment number of inner iterations */
398 inneriter += j_index_end - j_index_start;
400 /* Outer loop uses 41 flops */
403 /* Increment number of outer iterations */
406 /* Update outer/inner flops */
408 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*41 + inneriter*139);
411 * Gromacs nonbonded kernel: nb_kernel_ElecCoul_VdwCSTab_GeomW4P1_F_c
412 * Electrostatics interaction: Coulomb
413 * VdW interaction: CubicSplineTable
414 * Geometry: Water4-Particle
415 * Calculate force/pot: Force
418 nb_kernel_ElecCoul_VdwCSTab_GeomW4P1_F_c
419 (t_nblist * gmx_restrict nlist,
420 rvec * gmx_restrict xx,
421 rvec * gmx_restrict ff,
422 t_forcerec * gmx_restrict fr,
423 t_mdatoms * gmx_restrict mdatoms,
424 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
425 t_nrnb * gmx_restrict nrnb)
427 int i_shift_offset,i_coord_offset,j_coord_offset;
428 int j_index_start,j_index_end;
429 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
430 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
431 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
432 real *shiftvec,*fshift,*x,*f;
434 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
436 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
438 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
440 real ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
442 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
443 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
444 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
445 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
446 real dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
447 real velec,felec,velecsum,facel,crf,krf,krf2;
450 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
454 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
462 jindex = nlist->jindex;
464 shiftidx = nlist->shift;
466 shiftvec = fr->shift_vec[0];
467 fshift = fr->fshift[0];
469 charge = mdatoms->chargeA;
470 nvdwtype = fr->ntype;
472 vdwtype = mdatoms->typeA;
474 vftab = kernel_data->table_vdw->data;
475 vftabscale = kernel_data->table_vdw->scale;
477 /* Setup water-specific parameters */
478 inr = nlist->iinr[0];
479 iq1 = facel*charge[inr+1];
480 iq2 = facel*charge[inr+2];
481 iq3 = facel*charge[inr+3];
482 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
487 /* Start outer loop over neighborlists */
488 for(iidx=0; iidx<nri; iidx++)
490 /* Load shift vector for this list */
491 i_shift_offset = DIM*shiftidx[iidx];
492 shX = shiftvec[i_shift_offset+XX];
493 shY = shiftvec[i_shift_offset+YY];
494 shZ = shiftvec[i_shift_offset+ZZ];
496 /* Load limits for loop over neighbors */
497 j_index_start = jindex[iidx];
498 j_index_end = jindex[iidx+1];
500 /* Get outer coordinate index */
502 i_coord_offset = DIM*inr;
504 /* Load i particle coords and add shift vector */
505 ix0 = shX + x[i_coord_offset+DIM*0+XX];
506 iy0 = shY + x[i_coord_offset+DIM*0+YY];
507 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
508 ix1 = shX + x[i_coord_offset+DIM*1+XX];
509 iy1 = shY + x[i_coord_offset+DIM*1+YY];
510 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
511 ix2 = shX + x[i_coord_offset+DIM*2+XX];
512 iy2 = shY + x[i_coord_offset+DIM*2+YY];
513 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
514 ix3 = shX + x[i_coord_offset+DIM*3+XX];
515 iy3 = shY + x[i_coord_offset+DIM*3+YY];
516 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
531 /* Start inner kernel loop */
532 for(jidx=j_index_start; jidx<j_index_end; jidx++)
534 /* Get j neighbor index, and coordinate index */
536 j_coord_offset = DIM*jnr;
538 /* load j atom coordinates */
539 jx0 = x[j_coord_offset+DIM*0+XX];
540 jy0 = x[j_coord_offset+DIM*0+YY];
541 jz0 = x[j_coord_offset+DIM*0+ZZ];
543 /* Calculate displacement vector */
557 /* Calculate squared distance and things based on it */
558 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
559 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
560 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
561 rsq30 = dx30*dx30+dy30*dy30+dz30*dz30;
563 rinv00 = gmx_invsqrt(rsq00);
564 rinv10 = gmx_invsqrt(rsq10);
565 rinv20 = gmx_invsqrt(rsq20);
566 rinv30 = gmx_invsqrt(rsq30);
568 rinvsq10 = rinv10*rinv10;
569 rinvsq20 = rinv20*rinv20;
570 rinvsq30 = rinv30*rinv30;
572 /* Load parameters for j particles */
574 vdwjidx0 = 2*vdwtype[jnr+0];
576 /**************************
577 * CALCULATE INTERACTIONS *
578 **************************/
582 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
583 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
585 /* Calculate table index by multiplying r with table scale and truncate to integer */
591 /* CUBIC SPLINE TABLE DISPERSION */
594 Geps = vfeps*vftab[vfitab+2];
595 Heps2 = vfeps*vfeps*vftab[vfitab+3];
597 FF = Fp+Geps+2.0*Heps2;
600 /* CUBIC SPLINE TABLE REPULSION */
602 Geps = vfeps*vftab[vfitab+6];
603 Heps2 = vfeps*vfeps*vftab[vfitab+7];
605 FF = Fp+Geps+2.0*Heps2;
607 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
611 /* Calculate temporary vectorial force */
616 /* Update vectorial force */
620 f[j_coord_offset+DIM*0+XX] -= tx;
621 f[j_coord_offset+DIM*0+YY] -= ty;
622 f[j_coord_offset+DIM*0+ZZ] -= tz;
624 /**************************
625 * CALCULATE INTERACTIONS *
626 **************************/
630 /* COULOMB ELECTROSTATICS */
632 felec = velec*rinvsq10;
636 /* Calculate temporary vectorial force */
641 /* Update vectorial force */
645 f[j_coord_offset+DIM*0+XX] -= tx;
646 f[j_coord_offset+DIM*0+YY] -= ty;
647 f[j_coord_offset+DIM*0+ZZ] -= tz;
649 /**************************
650 * CALCULATE INTERACTIONS *
651 **************************/
655 /* COULOMB ELECTROSTATICS */
657 felec = velec*rinvsq20;
661 /* Calculate temporary vectorial force */
666 /* Update vectorial force */
670 f[j_coord_offset+DIM*0+XX] -= tx;
671 f[j_coord_offset+DIM*0+YY] -= ty;
672 f[j_coord_offset+DIM*0+ZZ] -= tz;
674 /**************************
675 * CALCULATE INTERACTIONS *
676 **************************/
680 /* COULOMB ELECTROSTATICS */
682 felec = velec*rinvsq30;
686 /* Calculate temporary vectorial force */
691 /* Update vectorial force */
695 f[j_coord_offset+DIM*0+XX] -= tx;
696 f[j_coord_offset+DIM*0+YY] -= ty;
697 f[j_coord_offset+DIM*0+ZZ] -= tz;
699 /* Inner loop uses 128 flops */
701 /* End of innermost loop */
704 f[i_coord_offset+DIM*0+XX] += fix0;
705 f[i_coord_offset+DIM*0+YY] += fiy0;
706 f[i_coord_offset+DIM*0+ZZ] += fiz0;
710 f[i_coord_offset+DIM*1+XX] += fix1;
711 f[i_coord_offset+DIM*1+YY] += fiy1;
712 f[i_coord_offset+DIM*1+ZZ] += fiz1;
716 f[i_coord_offset+DIM*2+XX] += fix2;
717 f[i_coord_offset+DIM*2+YY] += fiy2;
718 f[i_coord_offset+DIM*2+ZZ] += fiz2;
722 f[i_coord_offset+DIM*3+XX] += fix3;
723 f[i_coord_offset+DIM*3+YY] += fiy3;
724 f[i_coord_offset+DIM*3+ZZ] += fiz3;
728 fshift[i_shift_offset+XX] += tx;
729 fshift[i_shift_offset+YY] += ty;
730 fshift[i_shift_offset+ZZ] += tz;
732 /* Increment number of inner iterations */
733 inneriter += j_index_end - j_index_start;
735 /* Outer loop uses 39 flops */
738 /* Increment number of outer iterations */
741 /* Update outer/inner flops */
743 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*39 + inneriter*128);