2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013, 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.
44 #include "../nb_kernel.h"
45 #include "types/simple.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_VF_c
51 * Electrostatics interaction: CubicSplineTable
52 * VdW interaction: CubicSplineTable
53 * Geometry: Water3-Particle
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecCSTab_VdwCSTab_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_unused * 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_vdw->data;
111 vftabscale = kernel_data->table_elec_vdw->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 /* 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 /* 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 */
259 /* Calculate temporary vectorial force */
264 /* Update vectorial force */
268 f[j_coord_offset+DIM*0+XX] -= tx;
269 f[j_coord_offset+DIM*0+YY] -= ty;
270 f[j_coord_offset+DIM*0+ZZ] -= tz;
272 /**************************
273 * CALCULATE INTERACTIONS *
274 **************************/
280 /* Calculate table index by multiplying r with table scale and truncate to integer */
286 /* CUBIC SPLINE TABLE ELECTROSTATICS */
289 Geps = vfeps*vftab[vfitab+2];
290 Heps2 = vfeps*vfeps*vftab[vfitab+3];
294 FF = Fp+Geps+2.0*Heps2;
295 felec = -qq10*FF*vftabscale*rinv10;
297 /* Update potential sums from outer loop */
302 /* Calculate temporary vectorial force */
307 /* Update vectorial force */
311 f[j_coord_offset+DIM*0+XX] -= tx;
312 f[j_coord_offset+DIM*0+YY] -= ty;
313 f[j_coord_offset+DIM*0+ZZ] -= tz;
315 /**************************
316 * CALCULATE INTERACTIONS *
317 **************************/
323 /* Calculate table index by multiplying r with table scale and truncate to integer */
329 /* CUBIC SPLINE TABLE ELECTROSTATICS */
332 Geps = vfeps*vftab[vfitab+2];
333 Heps2 = vfeps*vfeps*vftab[vfitab+3];
337 FF = Fp+Geps+2.0*Heps2;
338 felec = -qq20*FF*vftabscale*rinv20;
340 /* Update potential sums from outer loop */
345 /* Calculate temporary vectorial force */
350 /* Update vectorial force */
354 f[j_coord_offset+DIM*0+XX] -= tx;
355 f[j_coord_offset+DIM*0+YY] -= ty;
356 f[j_coord_offset+DIM*0+ZZ] -= tz;
358 /* Inner loop uses 157 flops */
360 /* End of innermost loop */
363 f[i_coord_offset+DIM*0+XX] += fix0;
364 f[i_coord_offset+DIM*0+YY] += fiy0;
365 f[i_coord_offset+DIM*0+ZZ] += fiz0;
369 f[i_coord_offset+DIM*1+XX] += fix1;
370 f[i_coord_offset+DIM*1+YY] += fiy1;
371 f[i_coord_offset+DIM*1+ZZ] += fiz1;
375 f[i_coord_offset+DIM*2+XX] += fix2;
376 f[i_coord_offset+DIM*2+YY] += fiy2;
377 f[i_coord_offset+DIM*2+ZZ] += fiz2;
381 fshift[i_shift_offset+XX] += tx;
382 fshift[i_shift_offset+YY] += ty;
383 fshift[i_shift_offset+ZZ] += tz;
386 /* Update potential energies */
387 kernel_data->energygrp_elec[ggid] += velecsum;
388 kernel_data->energygrp_vdw[ggid] += vvdwsum;
390 /* Increment number of inner iterations */
391 inneriter += j_index_end - j_index_start;
393 /* Outer loop uses 32 flops */
396 /* Increment number of outer iterations */
399 /* Update outer/inner flops */
401 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*32 + inneriter*157);
404 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_F_c
405 * Electrostatics interaction: CubicSplineTable
406 * VdW interaction: CubicSplineTable
407 * Geometry: Water3-Particle
408 * Calculate force/pot: Force
411 nb_kernel_ElecCSTab_VdwCSTab_GeomW3P1_F_c
412 (t_nblist * gmx_restrict nlist,
413 rvec * gmx_restrict xx,
414 rvec * gmx_restrict ff,
415 t_forcerec * gmx_restrict fr,
416 t_mdatoms * gmx_restrict mdatoms,
417 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
418 t_nrnb * gmx_restrict nrnb)
420 int i_shift_offset,i_coord_offset,j_coord_offset;
421 int j_index_start,j_index_end;
422 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
423 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
424 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
425 real *shiftvec,*fshift,*x,*f;
427 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
429 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
431 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
433 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
434 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
435 real dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
436 real dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
437 real velec,felec,velecsum,facel,crf,krf,krf2;
440 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
444 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
452 jindex = nlist->jindex;
454 shiftidx = nlist->shift;
456 shiftvec = fr->shift_vec[0];
457 fshift = fr->fshift[0];
459 charge = mdatoms->chargeA;
460 nvdwtype = fr->ntype;
462 vdwtype = mdatoms->typeA;
464 vftab = kernel_data->table_elec_vdw->data;
465 vftabscale = kernel_data->table_elec_vdw->scale;
467 /* Setup water-specific parameters */
468 inr = nlist->iinr[0];
469 iq0 = facel*charge[inr+0];
470 iq1 = facel*charge[inr+1];
471 iq2 = facel*charge[inr+2];
472 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
477 /* Start outer loop over neighborlists */
478 for(iidx=0; iidx<nri; iidx++)
480 /* Load shift vector for this list */
481 i_shift_offset = DIM*shiftidx[iidx];
482 shX = shiftvec[i_shift_offset+XX];
483 shY = shiftvec[i_shift_offset+YY];
484 shZ = shiftvec[i_shift_offset+ZZ];
486 /* Load limits for loop over neighbors */
487 j_index_start = jindex[iidx];
488 j_index_end = jindex[iidx+1];
490 /* Get outer coordinate index */
492 i_coord_offset = DIM*inr;
494 /* Load i particle coords and add shift vector */
495 ix0 = shX + x[i_coord_offset+DIM*0+XX];
496 iy0 = shY + x[i_coord_offset+DIM*0+YY];
497 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
498 ix1 = shX + x[i_coord_offset+DIM*1+XX];
499 iy1 = shY + x[i_coord_offset+DIM*1+YY];
500 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
501 ix2 = shX + x[i_coord_offset+DIM*2+XX];
502 iy2 = shY + x[i_coord_offset+DIM*2+YY];
503 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
515 /* Start inner kernel loop */
516 for(jidx=j_index_start; jidx<j_index_end; jidx++)
518 /* Get j neighbor index, and coordinate index */
520 j_coord_offset = DIM*jnr;
522 /* load j atom coordinates */
523 jx0 = x[j_coord_offset+DIM*0+XX];
524 jy0 = x[j_coord_offset+DIM*0+YY];
525 jz0 = x[j_coord_offset+DIM*0+ZZ];
527 /* Calculate displacement vector */
538 /* Calculate squared distance and things based on it */
539 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
540 rsq10 = dx10*dx10+dy10*dy10+dz10*dz10;
541 rsq20 = dx20*dx20+dy20*dy20+dz20*dz20;
543 rinv00 = gmx_invsqrt(rsq00);
544 rinv10 = gmx_invsqrt(rsq10);
545 rinv20 = gmx_invsqrt(rsq20);
547 /* Load parameters for j particles */
549 vdwjidx0 = 2*vdwtype[jnr+0];
551 /**************************
552 * CALCULATE INTERACTIONS *
553 **************************/
558 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
559 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
561 /* Calculate table index by multiplying r with table scale and truncate to integer */
567 /* CUBIC SPLINE TABLE ELECTROSTATICS */
569 Geps = vfeps*vftab[vfitab+2];
570 Heps2 = vfeps*vfeps*vftab[vfitab+3];
572 FF = Fp+Geps+2.0*Heps2;
573 felec = -qq00*FF*vftabscale*rinv00;
575 /* CUBIC SPLINE TABLE DISPERSION */
578 Geps = vfeps*vftab[vfitab+2];
579 Heps2 = vfeps*vfeps*vftab[vfitab+3];
581 FF = Fp+Geps+2.0*Heps2;
584 /* CUBIC SPLINE TABLE REPULSION */
586 Geps = vfeps*vftab[vfitab+6];
587 Heps2 = vfeps*vfeps*vftab[vfitab+7];
589 FF = Fp+Geps+2.0*Heps2;
591 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
595 /* Calculate temporary vectorial force */
600 /* Update vectorial force */
604 f[j_coord_offset+DIM*0+XX] -= tx;
605 f[j_coord_offset+DIM*0+YY] -= ty;
606 f[j_coord_offset+DIM*0+ZZ] -= tz;
608 /**************************
609 * CALCULATE INTERACTIONS *
610 **************************/
616 /* Calculate table index by multiplying r with table scale and truncate to integer */
622 /* CUBIC SPLINE TABLE ELECTROSTATICS */
624 Geps = vfeps*vftab[vfitab+2];
625 Heps2 = vfeps*vfeps*vftab[vfitab+3];
627 FF = Fp+Geps+2.0*Heps2;
628 felec = -qq10*FF*vftabscale*rinv10;
632 /* Calculate temporary vectorial force */
637 /* Update vectorial force */
641 f[j_coord_offset+DIM*0+XX] -= tx;
642 f[j_coord_offset+DIM*0+YY] -= ty;
643 f[j_coord_offset+DIM*0+ZZ] -= tz;
645 /**************************
646 * CALCULATE INTERACTIONS *
647 **************************/
653 /* Calculate table index by multiplying r with table scale and truncate to integer */
659 /* CUBIC SPLINE TABLE ELECTROSTATICS */
661 Geps = vfeps*vftab[vfitab+2];
662 Heps2 = vfeps*vfeps*vftab[vfitab+3];
664 FF = Fp+Geps+2.0*Heps2;
665 felec = -qq20*FF*vftabscale*rinv20;
669 /* Calculate temporary vectorial force */
674 /* Update vectorial force */
678 f[j_coord_offset+DIM*0+XX] -= tx;
679 f[j_coord_offset+DIM*0+YY] -= ty;
680 f[j_coord_offset+DIM*0+ZZ] -= tz;
682 /* Inner loop uses 137 flops */
684 /* End of innermost loop */
687 f[i_coord_offset+DIM*0+XX] += fix0;
688 f[i_coord_offset+DIM*0+YY] += fiy0;
689 f[i_coord_offset+DIM*0+ZZ] += fiz0;
693 f[i_coord_offset+DIM*1+XX] += fix1;
694 f[i_coord_offset+DIM*1+YY] += fiy1;
695 f[i_coord_offset+DIM*1+ZZ] += fiz1;
699 f[i_coord_offset+DIM*2+XX] += fix2;
700 f[i_coord_offset+DIM*2+YY] += fiy2;
701 f[i_coord_offset+DIM*2+ZZ] += fiz2;
705 fshift[i_shift_offset+XX] += tx;
706 fshift[i_shift_offset+YY] += ty;
707 fshift[i_shift_offset+ZZ] += tz;
709 /* Increment number of inner iterations */
710 inneriter += j_index_end - j_index_start;
712 /* Outer loop uses 30 flops */
715 /* Increment number of outer iterations */
718 /* Update outer/inner flops */
720 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*30 + inneriter*137);