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_GeomW3P1_VF_c
49 * Electrostatics interaction: CubicSplineTable
50 * VdW interaction: LennardJones
51 * Geometry: Water3-Particle
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecCSTab_VdwLJ_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->data;
109 vftabscale = kernel_data->table_elec->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 rinvsq00 = rinv00*rinv00;
197 /* Load parameters for j particles */
199 vdwjidx0 = 2*vdwtype[jnr+0];
201 /**************************
202 * CALCULATE INTERACTIONS *
203 **************************/
208 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
209 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
211 /* Calculate table index by multiplying r with table scale and truncate to integer */
217 /* CUBIC SPLINE TABLE ELECTROSTATICS */
220 Geps = vfeps*vftab[vfitab+2];
221 Heps2 = vfeps*vfeps*vftab[vfitab+3];
225 FF = Fp+Geps+2.0*Heps2;
226 felec = -qq00*FF*vftabscale*rinv00;
228 /* LENNARD-JONES DISPERSION/REPULSION */
230 rinvsix = rinvsq00*rinvsq00*rinvsq00;
231 vvdw6 = c6_00*rinvsix;
232 vvdw12 = c12_00*rinvsix*rinvsix;
233 vvdw = vvdw12*(1.0/12.0) - vvdw6*(1.0/6.0);
234 fvdw = (vvdw12-vvdw6)*rinvsq00;
236 /* Update potential sums from outer loop */
242 /* Calculate temporary vectorial force */
247 /* Update vectorial force */
251 f[j_coord_offset+DIM*0+XX] -= tx;
252 f[j_coord_offset+DIM*0+YY] -= ty;
253 f[j_coord_offset+DIM*0+ZZ] -= tz;
255 /**************************
256 * CALCULATE INTERACTIONS *
257 **************************/
263 /* Calculate table index by multiplying r with table scale and truncate to integer */
269 /* CUBIC SPLINE TABLE ELECTROSTATICS */
272 Geps = vfeps*vftab[vfitab+2];
273 Heps2 = vfeps*vfeps*vftab[vfitab+3];
277 FF = Fp+Geps+2.0*Heps2;
278 felec = -qq10*FF*vftabscale*rinv10;
280 /* Update potential sums from outer loop */
285 /* Calculate temporary vectorial force */
290 /* Update vectorial force */
294 f[j_coord_offset+DIM*0+XX] -= tx;
295 f[j_coord_offset+DIM*0+YY] -= ty;
296 f[j_coord_offset+DIM*0+ZZ] -= tz;
298 /**************************
299 * CALCULATE INTERACTIONS *
300 **************************/
306 /* Calculate table index by multiplying r with table scale and truncate to integer */
312 /* CUBIC SPLINE TABLE ELECTROSTATICS */
315 Geps = vfeps*vftab[vfitab+2];
316 Heps2 = vfeps*vfeps*vftab[vfitab+3];
320 FF = Fp+Geps+2.0*Heps2;
321 felec = -qq20*FF*vftabscale*rinv20;
323 /* Update potential sums from outer loop */
328 /* Calculate temporary vectorial force */
333 /* Update vectorial force */
337 f[j_coord_offset+DIM*0+XX] -= tx;
338 f[j_coord_offset+DIM*0+YY] -= ty;
339 f[j_coord_offset+DIM*0+ZZ] -= tz;
341 /* Inner loop uses 139 flops */
343 /* End of innermost loop */
346 f[i_coord_offset+DIM*0+XX] += fix0;
347 f[i_coord_offset+DIM*0+YY] += fiy0;
348 f[i_coord_offset+DIM*0+ZZ] += fiz0;
352 f[i_coord_offset+DIM*1+XX] += fix1;
353 f[i_coord_offset+DIM*1+YY] += fiy1;
354 f[i_coord_offset+DIM*1+ZZ] += fiz1;
358 f[i_coord_offset+DIM*2+XX] += fix2;
359 f[i_coord_offset+DIM*2+YY] += fiy2;
360 f[i_coord_offset+DIM*2+ZZ] += fiz2;
364 fshift[i_shift_offset+XX] += tx;
365 fshift[i_shift_offset+YY] += ty;
366 fshift[i_shift_offset+ZZ] += tz;
369 /* Update potential energies */
370 kernel_data->energygrp_elec[ggid] += velecsum;
371 kernel_data->energygrp_vdw[ggid] += vvdwsum;
373 /* Increment number of inner iterations */
374 inneriter += j_index_end - j_index_start;
376 /* Outer loop uses 32 flops */
379 /* Increment number of outer iterations */
382 /* Update outer/inner flops */
384 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*32 + inneriter*139);
387 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_c
388 * Electrostatics interaction: CubicSplineTable
389 * VdW interaction: LennardJones
390 * Geometry: Water3-Particle
391 * Calculate force/pot: Force
394 nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_c
395 (t_nblist * gmx_restrict nlist,
396 rvec * gmx_restrict xx,
397 rvec * gmx_restrict ff,
398 t_forcerec * gmx_restrict fr,
399 t_mdatoms * gmx_restrict mdatoms,
400 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
401 t_nrnb * gmx_restrict nrnb)
403 int i_shift_offset,i_coord_offset,j_coord_offset;
404 int j_index_start,j_index_end;
405 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
406 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
407 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
408 real *shiftvec,*fshift,*x,*f;
410 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
412 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
414 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
416 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
417 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
418 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
419 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
420 real velec,felec,velecsum,facel,crf,krf,krf2;
423 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
427 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
435 jindex = nlist->jindex;
437 shiftidx = nlist->shift;
439 shiftvec = fr->shift_vec[0];
440 fshift = fr->fshift[0];
442 charge = mdatoms->chargeA;
443 nvdwtype = fr->ntype;
445 vdwtype = mdatoms->typeA;
447 vftab = kernel_data->table_elec->data;
448 vftabscale = kernel_data->table_elec->scale;
450 /* Setup water-specific parameters */
451 inr = nlist->iinr[0];
452 iq0 = facel*charge[inr+0];
453 iq1 = facel*charge[inr+1];
454 iq2 = facel*charge[inr+2];
455 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
460 /* Start outer loop over neighborlists */
461 for(iidx=0; iidx<nri; iidx++)
463 /* Load shift vector for this list */
464 i_shift_offset = DIM*shiftidx[iidx];
465 shX = shiftvec[i_shift_offset+XX];
466 shY = shiftvec[i_shift_offset+YY];
467 shZ = shiftvec[i_shift_offset+ZZ];
469 /* Load limits for loop over neighbors */
470 j_index_start = jindex[iidx];
471 j_index_end = jindex[iidx+1];
473 /* Get outer coordinate index */
475 i_coord_offset = DIM*inr;
477 /* Load i particle coords and add shift vector */
478 ix0 = shX + x[i_coord_offset+DIM*0+XX];
479 iy0 = shY + x[i_coord_offset+DIM*0+YY];
480 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
481 ix1 = shX + x[i_coord_offset+DIM*1+XX];
482 iy1 = shY + x[i_coord_offset+DIM*1+YY];
483 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
484 ix2 = shX + x[i_coord_offset+DIM*2+XX];
485 iy2 = shY + x[i_coord_offset+DIM*2+YY];
486 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
498 /* Start inner kernel loop */
499 for(jidx=j_index_start; jidx<j_index_end; jidx++)
501 /* Get j neighbor index, and coordinate index */
503 j_coord_offset = DIM*jnr;
505 /* load j atom coordinates */
506 jx0 = x[j_coord_offset+DIM*0+XX];
507 jy0 = x[j_coord_offset+DIM*0+YY];
508 jz0 = x[j_coord_offset+DIM*0+ZZ];
510 /* Calculate displacement vector */
521 /* Calculate squared distance and things based on it */
522 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
523 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
524 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
526 rinv00 = gmx_invsqrt(rsq00);
527 rinv10 = gmx_invsqrt(rsq10);
528 rinv20 = gmx_invsqrt(rsq20);
530 rinvsq00 = rinv00*rinv00;
532 /* Load parameters for j particles */
534 vdwjidx0 = 2*vdwtype[jnr+0];
536 /**************************
537 * CALCULATE INTERACTIONS *
538 **************************/
543 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
544 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
546 /* Calculate table index by multiplying r with table scale and truncate to integer */
552 /* CUBIC SPLINE TABLE ELECTROSTATICS */
554 Geps = vfeps*vftab[vfitab+2];
555 Heps2 = vfeps*vfeps*vftab[vfitab+3];
557 FF = Fp+Geps+2.0*Heps2;
558 felec = -qq00*FF*vftabscale*rinv00;
560 /* LENNARD-JONES DISPERSION/REPULSION */
562 rinvsix = rinvsq00*rinvsq00*rinvsq00;
563 fvdw = (c12_00*rinvsix-c6_00)*rinvsix*rinvsq00;
567 /* Calculate temporary vectorial force */
572 /* Update vectorial force */
576 f[j_coord_offset+DIM*0+XX] -= tx;
577 f[j_coord_offset+DIM*0+YY] -= ty;
578 f[j_coord_offset+DIM*0+ZZ] -= tz;
580 /**************************
581 * CALCULATE INTERACTIONS *
582 **************************/
588 /* Calculate table index by multiplying r with table scale and truncate to integer */
594 /* CUBIC SPLINE TABLE ELECTROSTATICS */
596 Geps = vfeps*vftab[vfitab+2];
597 Heps2 = vfeps*vfeps*vftab[vfitab+3];
599 FF = Fp+Geps+2.0*Heps2;
600 felec = -qq10*FF*vftabscale*rinv10;
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 = -qq20*FF*vftabscale*rinv20;
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 /* Inner loop uses 122 flops */
656 /* End of innermost loop */
659 f[i_coord_offset+DIM*0+XX] += fix0;
660 f[i_coord_offset+DIM*0+YY] += fiy0;
661 f[i_coord_offset+DIM*0+ZZ] += fiz0;
665 f[i_coord_offset+DIM*1+XX] += fix1;
666 f[i_coord_offset+DIM*1+YY] += fiy1;
667 f[i_coord_offset+DIM*1+ZZ] += fiz1;
671 f[i_coord_offset+DIM*2+XX] += fix2;
672 f[i_coord_offset+DIM*2+YY] += fiy2;
673 f[i_coord_offset+DIM*2+ZZ] += fiz2;
677 fshift[i_shift_offset+XX] += tx;
678 fshift[i_shift_offset+YY] += ty;
679 fshift[i_shift_offset+ZZ] += tz;
681 /* Increment number of inner iterations */
682 inneriter += j_index_end - j_index_start;
684 /* Outer loop uses 30 flops */
687 /* Increment number of outer iterations */
690 /* Update outer/inner flops */
692 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*30 + inneriter*122);