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_VdwCSTab_GeomW4P1_VF_c
49 * Electrostatics interaction: CubicSplineTable
50 * VdW interaction: CubicSplineTable
51 * Geometry: Water4-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecCSTab_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_elec_vdw->data;
112 vftabscale = kernel_data->table_elec_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 /* Load parameters for j particles */
211 vdwjidx0 = 2*vdwtype[jnr+0];
213 /**************************
214 * CALCULATE INTERACTIONS *
215 **************************/
219 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
220 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
222 /* Calculate table index by multiplying r with table scale and truncate to integer */
228 /* CUBIC SPLINE TABLE DISPERSION */
232 Geps = vfeps*vftab[vfitab+2];
233 Heps2 = vfeps*vfeps*vftab[vfitab+3];
237 FF = Fp+Geps+2.0*Heps2;
240 /* CUBIC SPLINE TABLE REPULSION */
243 Geps = vfeps*vftab[vfitab+6];
244 Heps2 = vfeps*vfeps*vftab[vfitab+7];
248 FF = Fp+Geps+2.0*Heps2;
251 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
253 /* Update potential sums from outer loop */
258 /* Calculate temporary vectorial force */
263 /* Update vectorial force */
267 f[j_coord_offset+DIM*0+XX] -= tx;
268 f[j_coord_offset+DIM*0+YY] -= ty;
269 f[j_coord_offset+DIM*0+ZZ] -= tz;
271 /**************************
272 * CALCULATE INTERACTIONS *
273 **************************/
279 /* Calculate table index by multiplying r with table scale and truncate to integer */
285 /* CUBIC SPLINE TABLE ELECTROSTATICS */
288 Geps = vfeps*vftab[vfitab+2];
289 Heps2 = vfeps*vfeps*vftab[vfitab+3];
293 FF = Fp+Geps+2.0*Heps2;
294 felec = -qq10*FF*vftabscale*rinv10;
296 /* Update potential sums from outer loop */
301 /* Calculate temporary vectorial force */
306 /* Update vectorial force */
310 f[j_coord_offset+DIM*0+XX] -= tx;
311 f[j_coord_offset+DIM*0+YY] -= ty;
312 f[j_coord_offset+DIM*0+ZZ] -= tz;
314 /**************************
315 * CALCULATE INTERACTIONS *
316 **************************/
322 /* Calculate table index by multiplying r with table scale and truncate to integer */
328 /* CUBIC SPLINE TABLE ELECTROSTATICS */
331 Geps = vfeps*vftab[vfitab+2];
332 Heps2 = vfeps*vfeps*vftab[vfitab+3];
336 FF = Fp+Geps+2.0*Heps2;
337 felec = -qq20*FF*vftabscale*rinv20;
339 /* Update potential sums from outer loop */
344 /* Calculate temporary vectorial force */
349 /* Update vectorial force */
353 f[j_coord_offset+DIM*0+XX] -= tx;
354 f[j_coord_offset+DIM*0+YY] -= ty;
355 f[j_coord_offset+DIM*0+ZZ] -= tz;
357 /**************************
358 * CALCULATE INTERACTIONS *
359 **************************/
365 /* Calculate table index by multiplying r with table scale and truncate to integer */
371 /* CUBIC SPLINE TABLE ELECTROSTATICS */
374 Geps = vfeps*vftab[vfitab+2];
375 Heps2 = vfeps*vfeps*vftab[vfitab+3];
379 FF = Fp+Geps+2.0*Heps2;
380 felec = -qq30*FF*vftabscale*rinv30;
382 /* Update potential sums from outer loop */
387 /* Calculate temporary vectorial force */
392 /* Update vectorial force */
396 f[j_coord_offset+DIM*0+XX] -= tx;
397 f[j_coord_offset+DIM*0+YY] -= ty;
398 f[j_coord_offset+DIM*0+ZZ] -= tz;
400 /* Inner loop uses 181 flops */
402 /* End of innermost loop */
405 f[i_coord_offset+DIM*0+XX] += fix0;
406 f[i_coord_offset+DIM*0+YY] += fiy0;
407 f[i_coord_offset+DIM*0+ZZ] += fiz0;
411 f[i_coord_offset+DIM*1+XX] += fix1;
412 f[i_coord_offset+DIM*1+YY] += fiy1;
413 f[i_coord_offset+DIM*1+ZZ] += fiz1;
417 f[i_coord_offset+DIM*2+XX] += fix2;
418 f[i_coord_offset+DIM*2+YY] += fiy2;
419 f[i_coord_offset+DIM*2+ZZ] += fiz2;
423 f[i_coord_offset+DIM*3+XX] += fix3;
424 f[i_coord_offset+DIM*3+YY] += fiy3;
425 f[i_coord_offset+DIM*3+ZZ] += fiz3;
429 fshift[i_shift_offset+XX] += tx;
430 fshift[i_shift_offset+YY] += ty;
431 fshift[i_shift_offset+ZZ] += tz;
434 /* Update potential energies */
435 kernel_data->energygrp_elec[ggid] += velecsum;
436 kernel_data->energygrp_vdw[ggid] += vvdwsum;
438 /* Increment number of inner iterations */
439 inneriter += j_index_end - j_index_start;
441 /* Outer loop uses 41 flops */
444 /* Increment number of outer iterations */
447 /* Update outer/inner flops */
449 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*41 + inneriter*181);
452 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_F_c
453 * Electrostatics interaction: CubicSplineTable
454 * VdW interaction: CubicSplineTable
455 * Geometry: Water4-Particle
456 * Calculate force/pot: Force
459 nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_F_c
460 (t_nblist * gmx_restrict nlist,
461 rvec * gmx_restrict xx,
462 rvec * gmx_restrict ff,
463 t_forcerec * gmx_restrict fr,
464 t_mdatoms * gmx_restrict mdatoms,
465 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
466 t_nrnb * gmx_restrict nrnb)
468 int i_shift_offset,i_coord_offset,j_coord_offset;
469 int j_index_start,j_index_end;
470 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
471 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
472 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
473 real *shiftvec,*fshift,*x,*f;
475 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
477 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
479 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
481 real ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
483 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
484 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
485 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
486 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
487 real dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
488 real velec,felec,velecsum,facel,crf,krf,krf2;
491 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
495 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
503 jindex = nlist->jindex;
505 shiftidx = nlist->shift;
507 shiftvec = fr->shift_vec[0];
508 fshift = fr->fshift[0];
510 charge = mdatoms->chargeA;
511 nvdwtype = fr->ntype;
513 vdwtype = mdatoms->typeA;
515 vftab = kernel_data->table_elec_vdw->data;
516 vftabscale = kernel_data->table_elec_vdw->scale;
518 /* Setup water-specific parameters */
519 inr = nlist->iinr[0];
520 iq1 = facel*charge[inr+1];
521 iq2 = facel*charge[inr+2];
522 iq3 = facel*charge[inr+3];
523 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
528 /* Start outer loop over neighborlists */
529 for(iidx=0; iidx<nri; iidx++)
531 /* Load shift vector for this list */
532 i_shift_offset = DIM*shiftidx[iidx];
533 shX = shiftvec[i_shift_offset+XX];
534 shY = shiftvec[i_shift_offset+YY];
535 shZ = shiftvec[i_shift_offset+ZZ];
537 /* Load limits for loop over neighbors */
538 j_index_start = jindex[iidx];
539 j_index_end = jindex[iidx+1];
541 /* Get outer coordinate index */
543 i_coord_offset = DIM*inr;
545 /* Load i particle coords and add shift vector */
546 ix0 = shX + x[i_coord_offset+DIM*0+XX];
547 iy0 = shY + x[i_coord_offset+DIM*0+YY];
548 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
549 ix1 = shX + x[i_coord_offset+DIM*1+XX];
550 iy1 = shY + x[i_coord_offset+DIM*1+YY];
551 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
552 ix2 = shX + x[i_coord_offset+DIM*2+XX];
553 iy2 = shY + x[i_coord_offset+DIM*2+YY];
554 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
555 ix3 = shX + x[i_coord_offset+DIM*3+XX];
556 iy3 = shY + x[i_coord_offset+DIM*3+YY];
557 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
572 /* Start inner kernel loop */
573 for(jidx=j_index_start; jidx<j_index_end; jidx++)
575 /* Get j neighbor index, and coordinate index */
577 j_coord_offset = DIM*jnr;
579 /* load j atom coordinates */
580 jx0 = x[j_coord_offset+DIM*0+XX];
581 jy0 = x[j_coord_offset+DIM*0+YY];
582 jz0 = x[j_coord_offset+DIM*0+ZZ];
584 /* Calculate displacement vector */
598 /* Calculate squared distance and things based on it */
599 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
600 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
601 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
602 rsq30 = dx30*dx30+dy30*dy30+dz30*dz30;
604 rinv00 = gmx_invsqrt(rsq00);
605 rinv10 = gmx_invsqrt(rsq10);
606 rinv20 = gmx_invsqrt(rsq20);
607 rinv30 = gmx_invsqrt(rsq30);
609 /* Load parameters for j particles */
611 vdwjidx0 = 2*vdwtype[jnr+0];
613 /**************************
614 * CALCULATE INTERACTIONS *
615 **************************/
619 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
620 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
622 /* Calculate table index by multiplying r with table scale and truncate to integer */
628 /* CUBIC SPLINE TABLE DISPERSION */
631 Geps = vfeps*vftab[vfitab+2];
632 Heps2 = vfeps*vfeps*vftab[vfitab+3];
634 FF = Fp+Geps+2.0*Heps2;
637 /* CUBIC SPLINE TABLE REPULSION */
639 Geps = vfeps*vftab[vfitab+6];
640 Heps2 = vfeps*vfeps*vftab[vfitab+7];
642 FF = Fp+Geps+2.0*Heps2;
644 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
648 /* Calculate temporary vectorial force */
653 /* Update vectorial force */
657 f[j_coord_offset+DIM*0+XX] -= tx;
658 f[j_coord_offset+DIM*0+YY] -= ty;
659 f[j_coord_offset+DIM*0+ZZ] -= tz;
661 /**************************
662 * CALCULATE INTERACTIONS *
663 **************************/
669 /* Calculate table index by multiplying r with table scale and truncate to integer */
675 /* CUBIC SPLINE TABLE ELECTROSTATICS */
677 Geps = vfeps*vftab[vfitab+2];
678 Heps2 = vfeps*vfeps*vftab[vfitab+3];
680 FF = Fp+Geps+2.0*Heps2;
681 felec = -qq10*FF*vftabscale*rinv10;
685 /* Calculate temporary vectorial force */
690 /* Update vectorial force */
694 f[j_coord_offset+DIM*0+XX] -= tx;
695 f[j_coord_offset+DIM*0+YY] -= ty;
696 f[j_coord_offset+DIM*0+ZZ] -= tz;
698 /**************************
699 * CALCULATE INTERACTIONS *
700 **************************/
706 /* Calculate table index by multiplying r with table scale and truncate to integer */
712 /* CUBIC SPLINE TABLE ELECTROSTATICS */
714 Geps = vfeps*vftab[vfitab+2];
715 Heps2 = vfeps*vfeps*vftab[vfitab+3];
717 FF = Fp+Geps+2.0*Heps2;
718 felec = -qq20*FF*vftabscale*rinv20;
722 /* Calculate temporary vectorial force */
727 /* Update vectorial force */
731 f[j_coord_offset+DIM*0+XX] -= tx;
732 f[j_coord_offset+DIM*0+YY] -= ty;
733 f[j_coord_offset+DIM*0+ZZ] -= tz;
735 /**************************
736 * CALCULATE INTERACTIONS *
737 **************************/
743 /* Calculate table index by multiplying r with table scale and truncate to integer */
749 /* CUBIC SPLINE TABLE ELECTROSTATICS */
751 Geps = vfeps*vftab[vfitab+2];
752 Heps2 = vfeps*vfeps*vftab[vfitab+3];
754 FF = Fp+Geps+2.0*Heps2;
755 felec = -qq30*FF*vftabscale*rinv30;
759 /* Calculate temporary vectorial force */
764 /* Update vectorial force */
768 f[j_coord_offset+DIM*0+XX] -= tx;
769 f[j_coord_offset+DIM*0+YY] -= ty;
770 f[j_coord_offset+DIM*0+ZZ] -= tz;
772 /* Inner loop uses 161 flops */
774 /* End of innermost loop */
777 f[i_coord_offset+DIM*0+XX] += fix0;
778 f[i_coord_offset+DIM*0+YY] += fiy0;
779 f[i_coord_offset+DIM*0+ZZ] += fiz0;
783 f[i_coord_offset+DIM*1+XX] += fix1;
784 f[i_coord_offset+DIM*1+YY] += fiy1;
785 f[i_coord_offset+DIM*1+ZZ] += fiz1;
789 f[i_coord_offset+DIM*2+XX] += fix2;
790 f[i_coord_offset+DIM*2+YY] += fiy2;
791 f[i_coord_offset+DIM*2+ZZ] += fiz2;
795 f[i_coord_offset+DIM*3+XX] += fix3;
796 f[i_coord_offset+DIM*3+YY] += fiy3;
797 f[i_coord_offset+DIM*3+ZZ] += fiz3;
801 fshift[i_shift_offset+XX] += tx;
802 fshift[i_shift_offset+YY] += ty;
803 fshift[i_shift_offset+ZZ] += tz;
805 /* Increment number of inner iterations */
806 inneriter += j_index_end - j_index_start;
808 /* Outer loop uses 39 flops */
811 /* Increment number of outer iterations */
814 /* Update outer/inner flops */
816 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*39 + inneriter*161);