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 "gromacs/legacyheaders/types/simple.h"
44 #include "gromacs/math/vec.h"
45 #include "gromacs/legacyheaders/nrnb.h"
48 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_VF_c
49 * Electrostatics interaction: CubicSplineTable
50 * VdW interaction: CubicSplineTable
51 * Geometry: Water3-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_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 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
78 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
79 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
80 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
81 real velec,felec,velecsum,facel,crf,krf,krf2;
84 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
88 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
96 jindex = nlist->jindex;
98 shiftidx = nlist->shift;
100 shiftvec = fr->shift_vec[0];
101 fshift = fr->fshift[0];
103 charge = mdatoms->chargeA;
104 nvdwtype = fr->ntype;
106 vdwtype = mdatoms->typeA;
108 vftab = kernel_data->table_elec_vdw->data;
109 vftabscale = kernel_data->table_elec_vdw->scale;
111 /* Setup water-specific parameters */
112 inr = nlist->iinr[0];
113 iq0 = facel*charge[inr+0];
114 iq1 = facel*charge[inr+1];
115 iq2 = facel*charge[inr+2];
116 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
121 /* Start outer loop over neighborlists */
122 for(iidx=0; iidx<nri; iidx++)
124 /* Load shift vector for this list */
125 i_shift_offset = DIM*shiftidx[iidx];
126 shX = shiftvec[i_shift_offset+XX];
127 shY = shiftvec[i_shift_offset+YY];
128 shZ = shiftvec[i_shift_offset+ZZ];
130 /* Load limits for loop over neighbors */
131 j_index_start = jindex[iidx];
132 j_index_end = jindex[iidx+1];
134 /* Get outer coordinate index */
136 i_coord_offset = DIM*inr;
138 /* Load i particle coords and add shift vector */
139 ix0 = shX + x[i_coord_offset+DIM*0+XX];
140 iy0 = shY + x[i_coord_offset+DIM*0+YY];
141 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
142 ix1 = shX + x[i_coord_offset+DIM*1+XX];
143 iy1 = shY + x[i_coord_offset+DIM*1+YY];
144 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
145 ix2 = shX + x[i_coord_offset+DIM*2+XX];
146 iy2 = shY + x[i_coord_offset+DIM*2+YY];
147 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
159 /* Reset potential sums */
163 /* Start inner kernel loop */
164 for(jidx=j_index_start; jidx<j_index_end; jidx++)
166 /* Get j neighbor index, and coordinate index */
168 j_coord_offset = DIM*jnr;
170 /* load j atom coordinates */
171 jx0 = x[j_coord_offset+DIM*0+XX];
172 jy0 = x[j_coord_offset+DIM*0+YY];
173 jz0 = x[j_coord_offset+DIM*0+ZZ];
175 /* Calculate displacement vector */
186 /* Calculate squared distance and things based on it */
187 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
188 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
189 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
191 rinv00 = gmx_invsqrt(rsq00);
192 rinv10 = gmx_invsqrt(rsq10);
193 rinv20 = gmx_invsqrt(rsq20);
195 /* Load parameters for j particles */
197 vdwjidx0 = 2*vdwtype[jnr+0];
199 /**************************
200 * CALCULATE INTERACTIONS *
201 **************************/
206 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
207 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
209 /* Calculate table index by multiplying r with table scale and truncate to integer */
215 /* CUBIC SPLINE TABLE ELECTROSTATICS */
218 Geps = vfeps*vftab[vfitab+2];
219 Heps2 = vfeps*vfeps*vftab[vfitab+3];
223 FF = Fp+Geps+2.0*Heps2;
224 felec = -qq00*FF*vftabscale*rinv00;
226 /* CUBIC SPLINE TABLE DISPERSION */
230 Geps = vfeps*vftab[vfitab+2];
231 Heps2 = vfeps*vfeps*vftab[vfitab+3];
235 FF = Fp+Geps+2.0*Heps2;
238 /* CUBIC SPLINE TABLE REPULSION */
241 Geps = vfeps*vftab[vfitab+6];
242 Heps2 = vfeps*vfeps*vftab[vfitab+7];
246 FF = Fp+Geps+2.0*Heps2;
249 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
251 /* Update potential sums from outer loop */
257 /* Calculate temporary vectorial force */
262 /* Update vectorial force */
266 f[j_coord_offset+DIM*0+XX] -= tx;
267 f[j_coord_offset+DIM*0+YY] -= ty;
268 f[j_coord_offset+DIM*0+ZZ] -= tz;
270 /**************************
271 * CALCULATE INTERACTIONS *
272 **************************/
278 /* Calculate table index by multiplying r with table scale and truncate to integer */
284 /* CUBIC SPLINE TABLE ELECTROSTATICS */
287 Geps = vfeps*vftab[vfitab+2];
288 Heps2 = vfeps*vfeps*vftab[vfitab+3];
292 FF = Fp+Geps+2.0*Heps2;
293 felec = -qq10*FF*vftabscale*rinv10;
295 /* Update potential sums from outer loop */
300 /* Calculate temporary vectorial force */
305 /* Update vectorial force */
309 f[j_coord_offset+DIM*0+XX] -= tx;
310 f[j_coord_offset+DIM*0+YY] -= ty;
311 f[j_coord_offset+DIM*0+ZZ] -= tz;
313 /**************************
314 * CALCULATE INTERACTIONS *
315 **************************/
321 /* Calculate table index by multiplying r with table scale and truncate to integer */
327 /* CUBIC SPLINE TABLE ELECTROSTATICS */
330 Geps = vfeps*vftab[vfitab+2];
331 Heps2 = vfeps*vfeps*vftab[vfitab+3];
335 FF = Fp+Geps+2.0*Heps2;
336 felec = -qq20*FF*vftabscale*rinv20;
338 /* Update potential sums from outer loop */
343 /* Calculate temporary vectorial force */
348 /* Update vectorial force */
352 f[j_coord_offset+DIM*0+XX] -= tx;
353 f[j_coord_offset+DIM*0+YY] -= ty;
354 f[j_coord_offset+DIM*0+ZZ] -= tz;
356 /* Inner loop uses 157 flops */
358 /* End of innermost loop */
361 f[i_coord_offset+DIM*0+XX] += fix0;
362 f[i_coord_offset+DIM*0+YY] += fiy0;
363 f[i_coord_offset+DIM*0+ZZ] += fiz0;
367 f[i_coord_offset+DIM*1+XX] += fix1;
368 f[i_coord_offset+DIM*1+YY] += fiy1;
369 f[i_coord_offset+DIM*1+ZZ] += fiz1;
373 f[i_coord_offset+DIM*2+XX] += fix2;
374 f[i_coord_offset+DIM*2+YY] += fiy2;
375 f[i_coord_offset+DIM*2+ZZ] += fiz2;
379 fshift[i_shift_offset+XX] += tx;
380 fshift[i_shift_offset+YY] += ty;
381 fshift[i_shift_offset+ZZ] += tz;
384 /* Update potential energies */
385 kernel_data->energygrp_elec[ggid] += velecsum;
386 kernel_data->energygrp_vdw[ggid] += vvdwsum;
388 /* Increment number of inner iterations */
389 inneriter += j_index_end - j_index_start;
391 /* Outer loop uses 32 flops */
394 /* Increment number of outer iterations */
397 /* Update outer/inner flops */
399 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*32 + inneriter*157);
402 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_F_c
403 * Electrostatics interaction: CubicSplineTable
404 * VdW interaction: CubicSplineTable
405 * Geometry: Water3-Particle
406 * Calculate force/pot: Force
409 nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_F_c
410 (t_nblist * gmx_restrict nlist,
411 rvec * gmx_restrict xx,
412 rvec * gmx_restrict ff,
413 t_forcerec * gmx_restrict fr,
414 t_mdatoms * gmx_restrict mdatoms,
415 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
416 t_nrnb * gmx_restrict nrnb)
418 int i_shift_offset,i_coord_offset,j_coord_offset;
419 int j_index_start,j_index_end;
420 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
421 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
422 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
423 real *shiftvec,*fshift,*x,*f;
425 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
427 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
429 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
431 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
432 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
433 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
434 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
435 real velec,felec,velecsum,facel,crf,krf,krf2;
438 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
442 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
450 jindex = nlist->jindex;
452 shiftidx = nlist->shift;
454 shiftvec = fr->shift_vec[0];
455 fshift = fr->fshift[0];
457 charge = mdatoms->chargeA;
458 nvdwtype = fr->ntype;
460 vdwtype = mdatoms->typeA;
462 vftab = kernel_data->table_elec_vdw->data;
463 vftabscale = kernel_data->table_elec_vdw->scale;
465 /* Setup water-specific parameters */
466 inr = nlist->iinr[0];
467 iq0 = facel*charge[inr+0];
468 iq1 = facel*charge[inr+1];
469 iq2 = facel*charge[inr+2];
470 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
475 /* Start outer loop over neighborlists */
476 for(iidx=0; iidx<nri; iidx++)
478 /* Load shift vector for this list */
479 i_shift_offset = DIM*shiftidx[iidx];
480 shX = shiftvec[i_shift_offset+XX];
481 shY = shiftvec[i_shift_offset+YY];
482 shZ = shiftvec[i_shift_offset+ZZ];
484 /* Load limits for loop over neighbors */
485 j_index_start = jindex[iidx];
486 j_index_end = jindex[iidx+1];
488 /* Get outer coordinate index */
490 i_coord_offset = DIM*inr;
492 /* Load i particle coords and add shift vector */
493 ix0 = shX + x[i_coord_offset+DIM*0+XX];
494 iy0 = shY + x[i_coord_offset+DIM*0+YY];
495 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
496 ix1 = shX + x[i_coord_offset+DIM*1+XX];
497 iy1 = shY + x[i_coord_offset+DIM*1+YY];
498 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
499 ix2 = shX + x[i_coord_offset+DIM*2+XX];
500 iy2 = shY + x[i_coord_offset+DIM*2+YY];
501 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
513 /* Start inner kernel loop */
514 for(jidx=j_index_start; jidx<j_index_end; jidx++)
516 /* Get j neighbor index, and coordinate index */
518 j_coord_offset = DIM*jnr;
520 /* load j atom coordinates */
521 jx0 = x[j_coord_offset+DIM*0+XX];
522 jy0 = x[j_coord_offset+DIM*0+YY];
523 jz0 = x[j_coord_offset+DIM*0+ZZ];
525 /* Calculate displacement vector */
536 /* Calculate squared distance and things based on it */
537 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
538 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
539 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
541 rinv00 = gmx_invsqrt(rsq00);
542 rinv10 = gmx_invsqrt(rsq10);
543 rinv20 = gmx_invsqrt(rsq20);
545 /* Load parameters for j particles */
547 vdwjidx0 = 2*vdwtype[jnr+0];
549 /**************************
550 * CALCULATE INTERACTIONS *
551 **************************/
556 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
557 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
559 /* Calculate table index by multiplying r with table scale and truncate to integer */
565 /* CUBIC SPLINE TABLE ELECTROSTATICS */
567 Geps = vfeps*vftab[vfitab+2];
568 Heps2 = vfeps*vfeps*vftab[vfitab+3];
570 FF = Fp+Geps+2.0*Heps2;
571 felec = -qq00*FF*vftabscale*rinv00;
573 /* CUBIC SPLINE TABLE DISPERSION */
576 Geps = vfeps*vftab[vfitab+2];
577 Heps2 = vfeps*vfeps*vftab[vfitab+3];
579 FF = Fp+Geps+2.0*Heps2;
582 /* CUBIC SPLINE TABLE REPULSION */
584 Geps = vfeps*vftab[vfitab+6];
585 Heps2 = vfeps*vfeps*vftab[vfitab+7];
587 FF = Fp+Geps+2.0*Heps2;
589 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
593 /* Calculate temporary vectorial force */
598 /* Update vectorial force */
602 f[j_coord_offset+DIM*0+XX] -= tx;
603 f[j_coord_offset+DIM*0+YY] -= ty;
604 f[j_coord_offset+DIM*0+ZZ] -= tz;
606 /**************************
607 * CALCULATE INTERACTIONS *
608 **************************/
614 /* Calculate table index by multiplying r with table scale and truncate to integer */
620 /* CUBIC SPLINE TABLE ELECTROSTATICS */
622 Geps = vfeps*vftab[vfitab+2];
623 Heps2 = vfeps*vfeps*vftab[vfitab+3];
625 FF = Fp+Geps+2.0*Heps2;
626 felec = -qq10*FF*vftabscale*rinv10;
630 /* Calculate temporary vectorial force */
635 /* Update vectorial force */
639 f[j_coord_offset+DIM*0+XX] -= tx;
640 f[j_coord_offset+DIM*0+YY] -= ty;
641 f[j_coord_offset+DIM*0+ZZ] -= tz;
643 /**************************
644 * CALCULATE INTERACTIONS *
645 **************************/
651 /* Calculate table index by multiplying r with table scale and truncate to integer */
657 /* CUBIC SPLINE TABLE ELECTROSTATICS */
659 Geps = vfeps*vftab[vfitab+2];
660 Heps2 = vfeps*vfeps*vftab[vfitab+3];
662 FF = Fp+Geps+2.0*Heps2;
663 felec = -qq20*FF*vftabscale*rinv20;
667 /* Calculate temporary vectorial force */
672 /* Update vectorial force */
676 f[j_coord_offset+DIM*0+XX] -= tx;
677 f[j_coord_offset+DIM*0+YY] -= ty;
678 f[j_coord_offset+DIM*0+ZZ] -= tz;
680 /* Inner loop uses 137 flops */
682 /* End of innermost loop */
685 f[i_coord_offset+DIM*0+XX] += fix0;
686 f[i_coord_offset+DIM*0+YY] += fiy0;
687 f[i_coord_offset+DIM*0+ZZ] += fiz0;
691 f[i_coord_offset+DIM*1+XX] += fix1;
692 f[i_coord_offset+DIM*1+YY] += fiy1;
693 f[i_coord_offset+DIM*1+ZZ] += fiz1;
697 f[i_coord_offset+DIM*2+XX] += fix2;
698 f[i_coord_offset+DIM*2+YY] += fiy2;
699 f[i_coord_offset+DIM*2+ZZ] += fiz2;
703 fshift[i_shift_offset+XX] += tx;
704 fshift[i_shift_offset+YY] += ty;
705 fshift[i_shift_offset+ZZ] += tz;
707 /* Increment number of inner iterations */
708 inneriter += j_index_end - j_index_start;
710 /* Outer loop uses 30 flops */
713 /* Increment number of outer iterations */
716 /* Update outer/inner flops */
718 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*30 + inneriter*137);