2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012, by the GROMACS development team, led by
5 * David van der Spoel, Berk Hess, Erik Lindahl, and including many
6 * others, as listed in the AUTHORS file in the top-level source
7 * 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 "types/simple.h"
50 * GROMACS nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_VF_c
51 * Electrostatics interaction: CubicSplineTable
52 * VdW interaction: LennardJones
53 * Geometry: Water3-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_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_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 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
77 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
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 velec,felec,velecsum,facel,crf,krf,krf2;
86 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
90 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
98 jindex = nlist->jindex;
100 shiftidx = nlist->shift;
102 shiftvec = fr->shift_vec[0];
103 fshift = fr->fshift[0];
105 charge = mdatoms->chargeA;
106 nvdwtype = fr->ntype;
108 vdwtype = mdatoms->typeA;
110 vftab = kernel_data->table_elec->data;
111 vftabscale = kernel_data->table_elec->scale;
113 /* Setup water-specific parameters */
114 inr = nlist->iinr[0];
115 iq0 = facel*charge[inr+0];
116 iq1 = facel*charge[inr+1];
117 iq2 = facel*charge[inr+2];
118 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
123 /* Start outer loop over neighborlists */
124 for(iidx=0; iidx<nri; iidx++)
126 /* Load shift vector for this list */
127 i_shift_offset = DIM*shiftidx[iidx];
128 shX = shiftvec[i_shift_offset+XX];
129 shY = shiftvec[i_shift_offset+YY];
130 shZ = shiftvec[i_shift_offset+ZZ];
132 /* Load limits for loop over neighbors */
133 j_index_start = jindex[iidx];
134 j_index_end = jindex[iidx+1];
136 /* Get outer coordinate index */
138 i_coord_offset = DIM*inr;
140 /* Load i particle coords and add shift vector */
141 ix0 = shX + x[i_coord_offset+DIM*0+XX];
142 iy0 = shY + x[i_coord_offset+DIM*0+YY];
143 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
144 ix1 = shX + x[i_coord_offset+DIM*1+XX];
145 iy1 = shY + x[i_coord_offset+DIM*1+YY];
146 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
147 ix2 = shX + x[i_coord_offset+DIM*2+XX];
148 iy2 = shY + x[i_coord_offset+DIM*2+YY];
149 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
161 /* Reset potential sums */
165 /* Start inner kernel loop */
166 for(jidx=j_index_start; jidx<j_index_end; jidx++)
168 /* Get j neighbor index, and coordinate index */
170 j_coord_offset = DIM*jnr;
172 /* load j atom coordinates */
173 jx0 = x[j_coord_offset+DIM*0+XX];
174 jy0 = x[j_coord_offset+DIM*0+YY];
175 jz0 = x[j_coord_offset+DIM*0+ZZ];
177 /* Calculate displacement vector */
188 /* Calculate squared distance and things based on it */
189 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
190 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
191 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
193 rinv00 = gmx_invsqrt(rsq00);
194 rinv10 = gmx_invsqrt(rsq10);
195 rinv20 = gmx_invsqrt(rsq20);
197 rinvsq00 = rinv00*rinv00;
199 /* Load parameters for j particles */
201 vdwjidx0 = 2*vdwtype[jnr+0];
203 /**************************
204 * CALCULATE INTERACTIONS *
205 **************************/
210 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
211 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
213 /* Calculate table index by multiplying r with table scale and truncate to integer */
219 /* CUBIC SPLINE TABLE ELECTROSTATICS */
222 Geps = vfeps*vftab[vfitab+2];
223 Heps2 = vfeps*vfeps*vftab[vfitab+3];
227 FF = Fp+Geps+2.0*Heps2;
228 felec = -qq00*FF*vftabscale*rinv00;
230 /* LENNARD-JONES DISPERSION/REPULSION */
232 rinvsix = rinvsq00*rinvsq00*rinvsq00;
233 vvdw6 = c6_00*rinvsix;
234 vvdw12 = c12_00*rinvsix*rinvsix;
235 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
236 fvdw = (vvdw12-vvdw6)*rinvsq00;
238 /* Update potential sums from outer loop */
244 /* Calculate temporary vectorial force */
249 /* Update vectorial force */
253 f[j_coord_offset+DIM*0+XX] -= tx;
254 f[j_coord_offset+DIM*0+YY] -= ty;
255 f[j_coord_offset+DIM*0+ZZ] -= tz;
257 /**************************
258 * CALCULATE INTERACTIONS *
259 **************************/
265 /* Calculate table index by multiplying r with table scale and truncate to integer */
271 /* CUBIC SPLINE TABLE ELECTROSTATICS */
274 Geps = vfeps*vftab[vfitab+2];
275 Heps2 = vfeps*vfeps*vftab[vfitab+3];
279 FF = Fp+Geps+2.0*Heps2;
280 felec = -qq10*FF*vftabscale*rinv10;
282 /* Update potential sums from outer loop */
287 /* Calculate temporary vectorial force */
292 /* Update vectorial force */
296 f[j_coord_offset+DIM*0+XX] -= tx;
297 f[j_coord_offset+DIM*0+YY] -= ty;
298 f[j_coord_offset+DIM*0+ZZ] -= tz;
300 /**************************
301 * CALCULATE INTERACTIONS *
302 **************************/
308 /* Calculate table index by multiplying r with table scale and truncate to integer */
314 /* CUBIC SPLINE TABLE ELECTROSTATICS */
317 Geps = vfeps*vftab[vfitab+2];
318 Heps2 = vfeps*vfeps*vftab[vfitab+3];
322 FF = Fp+Geps+2.0*Heps2;
323 felec = -qq20*FF*vftabscale*rinv20;
325 /* Update potential sums from outer loop */
330 /* Calculate temporary vectorial force */
335 /* Update vectorial force */
339 f[j_coord_offset+DIM*0+XX] -= tx;
340 f[j_coord_offset+DIM*0+YY] -= ty;
341 f[j_coord_offset+DIM*0+ZZ] -= tz;
343 /* Inner loop uses 139 flops */
345 /* End of innermost loop */
348 f[i_coord_offset+DIM*0+XX] += fix0;
349 f[i_coord_offset+DIM*0+YY] += fiy0;
350 f[i_coord_offset+DIM*0+ZZ] += fiz0;
354 f[i_coord_offset+DIM*1+XX] += fix1;
355 f[i_coord_offset+DIM*1+YY] += fiy1;
356 f[i_coord_offset+DIM*1+ZZ] += fiz1;
360 f[i_coord_offset+DIM*2+XX] += fix2;
361 f[i_coord_offset+DIM*2+YY] += fiy2;
362 f[i_coord_offset+DIM*2+ZZ] += fiz2;
366 fshift[i_shift_offset+XX] += tx;
367 fshift[i_shift_offset+YY] += ty;
368 fshift[i_shift_offset+ZZ] += tz;
371 /* Update potential energies */
372 kernel_data->energygrp_elec[ggid] += velecsum;
373 kernel_data->energygrp_vdw[ggid] += vvdwsum;
375 /* Increment number of inner iterations */
376 inneriter += j_index_end - j_index_start;
378 /* Outer loop uses 32 flops */
381 /* Increment number of outer iterations */
384 /* Update outer/inner flops */
386 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*32 + inneriter*139);
389 * GROMACS nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_c
390 * Electrostatics interaction: CubicSplineTable
391 * VdW interaction: LennardJones
392 * Geometry: Water3-Particle
393 * Calculate force/pot: Force
396 nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_c
397 (t_nblist * gmx_restrict nlist,
398 rvec * gmx_restrict xx,
399 rvec * gmx_restrict ff,
400 t_forcerec * gmx_restrict fr,
401 t_mdatoms * gmx_restrict mdatoms,
402 nb_kernel_data_t * gmx_restrict kernel_data,
403 t_nrnb * gmx_restrict nrnb)
405 int i_shift_offset,i_coord_offset,j_coord_offset;
406 int j_index_start,j_index_end;
407 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
408 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
409 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
410 real *shiftvec,*fshift,*x,*f;
412 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
414 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
416 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
418 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
419 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
420 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
421 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
422 real velec,felec,velecsum,facel,crf,krf,krf2;
425 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
429 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
437 jindex = nlist->jindex;
439 shiftidx = nlist->shift;
441 shiftvec = fr->shift_vec[0];
442 fshift = fr->fshift[0];
444 charge = mdatoms->chargeA;
445 nvdwtype = fr->ntype;
447 vdwtype = mdatoms->typeA;
449 vftab = kernel_data->table_elec->data;
450 vftabscale = kernel_data->table_elec->scale;
452 /* Setup water-specific parameters */
453 inr = nlist->iinr[0];
454 iq0 = facel*charge[inr+0];
455 iq1 = facel*charge[inr+1];
456 iq2 = facel*charge[inr+2];
457 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
462 /* Start outer loop over neighborlists */
463 for(iidx=0; iidx<nri; iidx++)
465 /* Load shift vector for this list */
466 i_shift_offset = DIM*shiftidx[iidx];
467 shX = shiftvec[i_shift_offset+XX];
468 shY = shiftvec[i_shift_offset+YY];
469 shZ = shiftvec[i_shift_offset+ZZ];
471 /* Load limits for loop over neighbors */
472 j_index_start = jindex[iidx];
473 j_index_end = jindex[iidx+1];
475 /* Get outer coordinate index */
477 i_coord_offset = DIM*inr;
479 /* Load i particle coords and add shift vector */
480 ix0 = shX + x[i_coord_offset+DIM*0+XX];
481 iy0 = shY + x[i_coord_offset+DIM*0+YY];
482 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
483 ix1 = shX + x[i_coord_offset+DIM*1+XX];
484 iy1 = shY + x[i_coord_offset+DIM*1+YY];
485 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
486 ix2 = shX + x[i_coord_offset+DIM*2+XX];
487 iy2 = shY + x[i_coord_offset+DIM*2+YY];
488 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
500 /* Start inner kernel loop */
501 for(jidx=j_index_start; jidx<j_index_end; jidx++)
503 /* Get j neighbor index, and coordinate index */
505 j_coord_offset = DIM*jnr;
507 /* load j atom coordinates */
508 jx0 = x[j_coord_offset+DIM*0+XX];
509 jy0 = x[j_coord_offset+DIM*0+YY];
510 jz0 = x[j_coord_offset+DIM*0+ZZ];
512 /* Calculate displacement vector */
523 /* Calculate squared distance and things based on it */
524 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
525 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
526 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
528 rinv00 = gmx_invsqrt(rsq00);
529 rinv10 = gmx_invsqrt(rsq10);
530 rinv20 = gmx_invsqrt(rsq20);
532 rinvsq00 = rinv00*rinv00;
534 /* Load parameters for j particles */
536 vdwjidx0 = 2*vdwtype[jnr+0];
538 /**************************
539 * CALCULATE INTERACTIONS *
540 **************************/
545 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
546 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
548 /* Calculate table index by multiplying r with table scale and truncate to integer */
554 /* CUBIC SPLINE TABLE ELECTROSTATICS */
556 Geps = vfeps*vftab[vfitab+2];
557 Heps2 = vfeps*vfeps*vftab[vfitab+3];
559 FF = Fp+Geps+2.0*Heps2;
560 felec = -qq00*FF*vftabscale*rinv00;
562 /* LENNARD-JONES DISPERSION/REPULSION */
564 rinvsix = rinvsq00*rinvsq00*rinvsq00;
565 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
569 /* Calculate temporary vectorial force */
574 /* Update vectorial force */
578 f[j_coord_offset+DIM*0+XX] -= tx;
579 f[j_coord_offset+DIM*0+YY] -= ty;
580 f[j_coord_offset+DIM*0+ZZ] -= tz;
582 /**************************
583 * CALCULATE INTERACTIONS *
584 **************************/
590 /* Calculate table index by multiplying r with table scale and truncate to integer */
596 /* CUBIC SPLINE TABLE ELECTROSTATICS */
598 Geps = vfeps*vftab[vfitab+2];
599 Heps2 = vfeps*vfeps*vftab[vfitab+3];
601 FF = Fp+Geps+2.0*Heps2;
602 felec = -qq10*FF*vftabscale*rinv10;
606 /* Calculate temporary vectorial force */
611 /* Update vectorial force */
615 f[j_coord_offset+DIM*0+XX] -= tx;
616 f[j_coord_offset+DIM*0+YY] -= ty;
617 f[j_coord_offset+DIM*0+ZZ] -= tz;
619 /**************************
620 * CALCULATE INTERACTIONS *
621 **************************/
627 /* Calculate table index by multiplying r with table scale and truncate to integer */
633 /* CUBIC SPLINE TABLE ELECTROSTATICS */
635 Geps = vfeps*vftab[vfitab+2];
636 Heps2 = vfeps*vfeps*vftab[vfitab+3];
638 FF = Fp+Geps+2.0*Heps2;
639 felec = -qq20*FF*vftabscale*rinv20;
643 /* Calculate temporary vectorial force */
648 /* Update vectorial force */
652 f[j_coord_offset+DIM*0+XX] -= tx;
653 f[j_coord_offset+DIM*0+YY] -= ty;
654 f[j_coord_offset+DIM*0+ZZ] -= tz;
656 /* Inner loop uses 122 flops */
658 /* End of innermost loop */
661 f[i_coord_offset+DIM*0+XX] += fix0;
662 f[i_coord_offset+DIM*0+YY] += fiy0;
663 f[i_coord_offset+DIM*0+ZZ] += fiz0;
667 f[i_coord_offset+DIM*1+XX] += fix1;
668 f[i_coord_offset+DIM*1+YY] += fiy1;
669 f[i_coord_offset+DIM*1+ZZ] += fiz1;
673 f[i_coord_offset+DIM*2+XX] += fix2;
674 f[i_coord_offset+DIM*2+YY] += fiy2;
675 f[i_coord_offset+DIM*2+ZZ] += fiz2;
679 fshift[i_shift_offset+XX] += tx;
680 fshift[i_shift_offset+YY] += ty;
681 fshift[i_shift_offset+ZZ] += tz;
683 /* Increment number of inner iterations */
684 inneriter += j_index_end - j_index_start;
686 /* Outer loop uses 30 flops */
689 /* Increment number of outer iterations */
692 /* Update outer/inner flops */
694 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*30 + inneriter*122);