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_ElecRF_VdwCSTab_GeomW4W4_VF_c
49 * Electrostatics interaction: ReactionField
50 * VdW interaction: CubicSplineTable
51 * Geometry: Water4-Water4
52 * Calculate force/pot: PotentialAndForce
55 nb_kernel_ElecRF_VdwCSTab_GeomW4W4_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;
81 real jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
83 real jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
85 real jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
86 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
87 real dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11,cexp1_11,cexp2_11;
88 real dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12,cexp1_12,cexp2_12;
89 real dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13,cexp1_13,cexp2_13;
90 real dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21,cexp1_21,cexp2_21;
91 real dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22,cexp1_22,cexp2_22;
92 real dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23,cexp1_23,cexp2_23;
93 real dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31,cexp1_31,cexp2_31;
94 real dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32,cexp1_32,cexp2_32;
95 real dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33,cexp1_33,cexp2_33;
96 real velec,felec,velecsum,facel,crf,krf,krf2;
99 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
103 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
111 jindex = nlist->jindex;
113 shiftidx = nlist->shift;
115 shiftvec = fr->shift_vec[0];
116 fshift = fr->fshift[0];
118 charge = mdatoms->chargeA;
122 nvdwtype = fr->ntype;
124 vdwtype = mdatoms->typeA;
126 vftab = kernel_data->table_vdw->data;
127 vftabscale = kernel_data->table_vdw->scale;
129 /* Setup water-specific parameters */
130 inr = nlist->iinr[0];
131 iq1 = facel*charge[inr+1];
132 iq2 = facel*charge[inr+2];
133 iq3 = facel*charge[inr+3];
134 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
139 vdwjidx0 = 2*vdwtype[inr+0];
140 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
141 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
155 /* Start outer loop over neighborlists */
156 for(iidx=0; iidx<nri; iidx++)
158 /* Load shift vector for this list */
159 i_shift_offset = DIM*shiftidx[iidx];
160 shX = shiftvec[i_shift_offset+XX];
161 shY = shiftvec[i_shift_offset+YY];
162 shZ = shiftvec[i_shift_offset+ZZ];
164 /* Load limits for loop over neighbors */
165 j_index_start = jindex[iidx];
166 j_index_end = jindex[iidx+1];
168 /* Get outer coordinate index */
170 i_coord_offset = DIM*inr;
172 /* Load i particle coords and add shift vector */
173 ix0 = shX + x[i_coord_offset+DIM*0+XX];
174 iy0 = shY + x[i_coord_offset+DIM*0+YY];
175 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
176 ix1 = shX + x[i_coord_offset+DIM*1+XX];
177 iy1 = shY + x[i_coord_offset+DIM*1+YY];
178 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
179 ix2 = shX + x[i_coord_offset+DIM*2+XX];
180 iy2 = shY + x[i_coord_offset+DIM*2+YY];
181 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
182 ix3 = shX + x[i_coord_offset+DIM*3+XX];
183 iy3 = shY + x[i_coord_offset+DIM*3+YY];
184 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
199 /* Reset potential sums */
203 /* Start inner kernel loop */
204 for(jidx=j_index_start; jidx<j_index_end; jidx++)
206 /* Get j neighbor index, and coordinate index */
208 j_coord_offset = DIM*jnr;
210 /* load j atom coordinates */
211 jx0 = x[j_coord_offset+DIM*0+XX];
212 jy0 = x[j_coord_offset+DIM*0+YY];
213 jz0 = x[j_coord_offset+DIM*0+ZZ];
214 jx1 = x[j_coord_offset+DIM*1+XX];
215 jy1 = x[j_coord_offset+DIM*1+YY];
216 jz1 = x[j_coord_offset+DIM*1+ZZ];
217 jx2 = x[j_coord_offset+DIM*2+XX];
218 jy2 = x[j_coord_offset+DIM*2+YY];
219 jz2 = x[j_coord_offset+DIM*2+ZZ];
220 jx3 = x[j_coord_offset+DIM*3+XX];
221 jy3 = x[j_coord_offset+DIM*3+YY];
222 jz3 = x[j_coord_offset+DIM*3+ZZ];
224 /* Calculate displacement vector */
256 /* Calculate squared distance and things based on it */
257 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
258 rsq11 = dx11*dx11+dy11*dy11+dz11*dz11;
259 rsq12 = dx12*dx12+dy12*dy12+dz12*dz12;
260 rsq13 = dx13*dx13+dy13*dy13+dz13*dz13;
261 rsq21 = dx21*dx21+dy21*dy21+dz21*dz21;
262 rsq22 = dx22*dx22+dy22*dy22+dz22*dz22;
263 rsq23 = dx23*dx23+dy23*dy23+dz23*dz23;
264 rsq31 = dx31*dx31+dy31*dy31+dz31*dz31;
265 rsq32 = dx32*dx32+dy32*dy32+dz32*dz32;
266 rsq33 = dx33*dx33+dy33*dy33+dz33*dz33;
268 rinv00 = gmx_invsqrt(rsq00);
269 rinv11 = gmx_invsqrt(rsq11);
270 rinv12 = gmx_invsqrt(rsq12);
271 rinv13 = gmx_invsqrt(rsq13);
272 rinv21 = gmx_invsqrt(rsq21);
273 rinv22 = gmx_invsqrt(rsq22);
274 rinv23 = gmx_invsqrt(rsq23);
275 rinv31 = gmx_invsqrt(rsq31);
276 rinv32 = gmx_invsqrt(rsq32);
277 rinv33 = gmx_invsqrt(rsq33);
279 rinvsq11 = rinv11*rinv11;
280 rinvsq12 = rinv12*rinv12;
281 rinvsq13 = rinv13*rinv13;
282 rinvsq21 = rinv21*rinv21;
283 rinvsq22 = rinv22*rinv22;
284 rinvsq23 = rinv23*rinv23;
285 rinvsq31 = rinv31*rinv31;
286 rinvsq32 = rinv32*rinv32;
287 rinvsq33 = rinv33*rinv33;
289 /**************************
290 * CALCULATE INTERACTIONS *
291 **************************/
295 /* Calculate table index by multiplying r with table scale and truncate to integer */
301 /* CUBIC SPLINE TABLE DISPERSION */
305 Geps = vfeps*vftab[vfitab+2];
306 Heps2 = vfeps*vfeps*vftab[vfitab+3];
310 FF = Fp+Geps+2.0*Heps2;
313 /* CUBIC SPLINE TABLE REPULSION */
316 Geps = vfeps*vftab[vfitab+6];
317 Heps2 = vfeps*vfeps*vftab[vfitab+7];
321 FF = Fp+Geps+2.0*Heps2;
324 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
326 /* Update potential sums from outer loop */
331 /* Calculate temporary vectorial force */
336 /* Update vectorial force */
340 f[j_coord_offset+DIM*0+XX] -= tx;
341 f[j_coord_offset+DIM*0+YY] -= ty;
342 f[j_coord_offset+DIM*0+ZZ] -= tz;
344 /**************************
345 * CALCULATE INTERACTIONS *
346 **************************/
348 /* REACTION-FIELD ELECTROSTATICS */
349 velec = qq11*(rinv11+krf*rsq11-crf);
350 felec = qq11*(rinv11*rinvsq11-krf2);
352 /* Update potential sums from outer loop */
357 /* Calculate temporary vectorial force */
362 /* Update vectorial force */
366 f[j_coord_offset+DIM*1+XX] -= tx;
367 f[j_coord_offset+DIM*1+YY] -= ty;
368 f[j_coord_offset+DIM*1+ZZ] -= tz;
370 /**************************
371 * CALCULATE INTERACTIONS *
372 **************************/
374 /* REACTION-FIELD ELECTROSTATICS */
375 velec = qq12*(rinv12+krf*rsq12-crf);
376 felec = qq12*(rinv12*rinvsq12-krf2);
378 /* Update potential sums from outer loop */
383 /* Calculate temporary vectorial force */
388 /* Update vectorial force */
392 f[j_coord_offset+DIM*2+XX] -= tx;
393 f[j_coord_offset+DIM*2+YY] -= ty;
394 f[j_coord_offset+DIM*2+ZZ] -= tz;
396 /**************************
397 * CALCULATE INTERACTIONS *
398 **************************/
400 /* REACTION-FIELD ELECTROSTATICS */
401 velec = qq13*(rinv13+krf*rsq13-crf);
402 felec = qq13*(rinv13*rinvsq13-krf2);
404 /* Update potential sums from outer loop */
409 /* Calculate temporary vectorial force */
414 /* Update vectorial force */
418 f[j_coord_offset+DIM*3+XX] -= tx;
419 f[j_coord_offset+DIM*3+YY] -= ty;
420 f[j_coord_offset+DIM*3+ZZ] -= tz;
422 /**************************
423 * CALCULATE INTERACTIONS *
424 **************************/
426 /* REACTION-FIELD ELECTROSTATICS */
427 velec = qq21*(rinv21+krf*rsq21-crf);
428 felec = qq21*(rinv21*rinvsq21-krf2);
430 /* Update potential sums from outer loop */
435 /* Calculate temporary vectorial force */
440 /* Update vectorial force */
444 f[j_coord_offset+DIM*1+XX] -= tx;
445 f[j_coord_offset+DIM*1+YY] -= ty;
446 f[j_coord_offset+DIM*1+ZZ] -= tz;
448 /**************************
449 * CALCULATE INTERACTIONS *
450 **************************/
452 /* REACTION-FIELD ELECTROSTATICS */
453 velec = qq22*(rinv22+krf*rsq22-crf);
454 felec = qq22*(rinv22*rinvsq22-krf2);
456 /* Update potential sums from outer loop */
461 /* Calculate temporary vectorial force */
466 /* Update vectorial force */
470 f[j_coord_offset+DIM*2+XX] -= tx;
471 f[j_coord_offset+DIM*2+YY] -= ty;
472 f[j_coord_offset+DIM*2+ZZ] -= tz;
474 /**************************
475 * CALCULATE INTERACTIONS *
476 **************************/
478 /* REACTION-FIELD ELECTROSTATICS */
479 velec = qq23*(rinv23+krf*rsq23-crf);
480 felec = qq23*(rinv23*rinvsq23-krf2);
482 /* Update potential sums from outer loop */
487 /* Calculate temporary vectorial force */
492 /* Update vectorial force */
496 f[j_coord_offset+DIM*3+XX] -= tx;
497 f[j_coord_offset+DIM*3+YY] -= ty;
498 f[j_coord_offset+DIM*3+ZZ] -= tz;
500 /**************************
501 * CALCULATE INTERACTIONS *
502 **************************/
504 /* REACTION-FIELD ELECTROSTATICS */
505 velec = qq31*(rinv31+krf*rsq31-crf);
506 felec = qq31*(rinv31*rinvsq31-krf2);
508 /* Update potential sums from outer loop */
513 /* Calculate temporary vectorial force */
518 /* Update vectorial force */
522 f[j_coord_offset+DIM*1+XX] -= tx;
523 f[j_coord_offset+DIM*1+YY] -= ty;
524 f[j_coord_offset+DIM*1+ZZ] -= tz;
526 /**************************
527 * CALCULATE INTERACTIONS *
528 **************************/
530 /* REACTION-FIELD ELECTROSTATICS */
531 velec = qq32*(rinv32+krf*rsq32-crf);
532 felec = qq32*(rinv32*rinvsq32-krf2);
534 /* Update potential sums from outer loop */
539 /* Calculate temporary vectorial force */
544 /* Update vectorial force */
548 f[j_coord_offset+DIM*2+XX] -= tx;
549 f[j_coord_offset+DIM*2+YY] -= ty;
550 f[j_coord_offset+DIM*2+ZZ] -= tz;
552 /**************************
553 * CALCULATE INTERACTIONS *
554 **************************/
556 /* REACTION-FIELD ELECTROSTATICS */
557 velec = qq33*(rinv33+krf*rsq33-crf);
558 felec = qq33*(rinv33*rinvsq33-krf2);
560 /* Update potential sums from outer loop */
565 /* Calculate temporary vectorial force */
570 /* Update vectorial force */
574 f[j_coord_offset+DIM*3+XX] -= tx;
575 f[j_coord_offset+DIM*3+YY] -= ty;
576 f[j_coord_offset+DIM*3+ZZ] -= tz;
578 /* Inner loop uses 334 flops */
580 /* End of innermost loop */
583 f[i_coord_offset+DIM*0+XX] += fix0;
584 f[i_coord_offset+DIM*0+YY] += fiy0;
585 f[i_coord_offset+DIM*0+ZZ] += fiz0;
589 f[i_coord_offset+DIM*1+XX] += fix1;
590 f[i_coord_offset+DIM*1+YY] += fiy1;
591 f[i_coord_offset+DIM*1+ZZ] += fiz1;
595 f[i_coord_offset+DIM*2+XX] += fix2;
596 f[i_coord_offset+DIM*2+YY] += fiy2;
597 f[i_coord_offset+DIM*2+ZZ] += fiz2;
601 f[i_coord_offset+DIM*3+XX] += fix3;
602 f[i_coord_offset+DIM*3+YY] += fiy3;
603 f[i_coord_offset+DIM*3+ZZ] += fiz3;
607 fshift[i_shift_offset+XX] += tx;
608 fshift[i_shift_offset+YY] += ty;
609 fshift[i_shift_offset+ZZ] += tz;
612 /* Update potential energies */
613 kernel_data->energygrp_elec[ggid] += velecsum;
614 kernel_data->energygrp_vdw[ggid] += vvdwsum;
616 /* Increment number of inner iterations */
617 inneriter += j_index_end - j_index_start;
619 /* Outer loop uses 41 flops */
622 /* Increment number of outer iterations */
625 /* Update outer/inner flops */
627 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*41 + inneriter*334);
630 * Gromacs nonbonded kernel: nb_kernel_ElecRF_VdwCSTab_GeomW4W4_F_c
631 * Electrostatics interaction: ReactionField
632 * VdW interaction: CubicSplineTable
633 * Geometry: Water4-Water4
634 * Calculate force/pot: Force
637 nb_kernel_ElecRF_VdwCSTab_GeomW4W4_F_c
638 (t_nblist * gmx_restrict nlist,
639 rvec * gmx_restrict xx,
640 rvec * gmx_restrict ff,
641 t_forcerec * gmx_restrict fr,
642 t_mdatoms * gmx_restrict mdatoms,
643 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
644 t_nrnb * gmx_restrict nrnb)
646 int i_shift_offset,i_coord_offset,j_coord_offset;
647 int j_index_start,j_index_end;
648 int nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
649 real shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
650 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
651 real *shiftvec,*fshift,*x,*f;
653 real ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
655 real ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
657 real ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
659 real ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
661 real jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
663 real jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
665 real jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
667 real jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
668 real dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00,cexp1_00,cexp2_00;
669 real dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11,cexp1_11,cexp2_11;
670 real dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12,cexp1_12,cexp2_12;
671 real dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13,cexp1_13,cexp2_13;
672 real dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21,cexp1_21,cexp2_21;
673 real dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22,cexp1_22,cexp2_22;
674 real dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23,cexp1_23,cexp2_23;
675 real dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31,cexp1_31,cexp2_31;
676 real dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32,cexp1_32,cexp2_32;
677 real dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33,cexp1_33,cexp2_33;
678 real velec,felec,velecsum,facel,crf,krf,krf2;
681 real rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,br,vvdwexp,sh_vdw_invrcut6;
685 real rt,vfeps,vftabscale,Y,F,Geps,Heps2,Fp,VV,FF;
693 jindex = nlist->jindex;
695 shiftidx = nlist->shift;
697 shiftvec = fr->shift_vec[0];
698 fshift = fr->fshift[0];
700 charge = mdatoms->chargeA;
704 nvdwtype = fr->ntype;
706 vdwtype = mdatoms->typeA;
708 vftab = kernel_data->table_vdw->data;
709 vftabscale = kernel_data->table_vdw->scale;
711 /* Setup water-specific parameters */
712 inr = nlist->iinr[0];
713 iq1 = facel*charge[inr+1];
714 iq2 = facel*charge[inr+2];
715 iq3 = facel*charge[inr+3];
716 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
721 vdwjidx0 = 2*vdwtype[inr+0];
722 c6_00 = vdwparam[vdwioffset0+vdwjidx0];
723 c12_00 = vdwparam[vdwioffset0+vdwjidx0+1];
737 /* Start outer loop over neighborlists */
738 for(iidx=0; iidx<nri; iidx++)
740 /* Load shift vector for this list */
741 i_shift_offset = DIM*shiftidx[iidx];
742 shX = shiftvec[i_shift_offset+XX];
743 shY = shiftvec[i_shift_offset+YY];
744 shZ = shiftvec[i_shift_offset+ZZ];
746 /* Load limits for loop over neighbors */
747 j_index_start = jindex[iidx];
748 j_index_end = jindex[iidx+1];
750 /* Get outer coordinate index */
752 i_coord_offset = DIM*inr;
754 /* Load i particle coords and add shift vector */
755 ix0 = shX + x[i_coord_offset+DIM*0+XX];
756 iy0 = shY + x[i_coord_offset+DIM*0+YY];
757 iz0 = shZ + x[i_coord_offset+DIM*0+ZZ];
758 ix1 = shX + x[i_coord_offset+DIM*1+XX];
759 iy1 = shY + x[i_coord_offset+DIM*1+YY];
760 iz1 = shZ + x[i_coord_offset+DIM*1+ZZ];
761 ix2 = shX + x[i_coord_offset+DIM*2+XX];
762 iy2 = shY + x[i_coord_offset+DIM*2+YY];
763 iz2 = shZ + x[i_coord_offset+DIM*2+ZZ];
764 ix3 = shX + x[i_coord_offset+DIM*3+XX];
765 iy3 = shY + x[i_coord_offset+DIM*3+YY];
766 iz3 = shZ + x[i_coord_offset+DIM*3+ZZ];
781 /* Start inner kernel loop */
782 for(jidx=j_index_start; jidx<j_index_end; jidx++)
784 /* Get j neighbor index, and coordinate index */
786 j_coord_offset = DIM*jnr;
788 /* load j atom coordinates */
789 jx0 = x[j_coord_offset+DIM*0+XX];
790 jy0 = x[j_coord_offset+DIM*0+YY];
791 jz0 = x[j_coord_offset+DIM*0+ZZ];
792 jx1 = x[j_coord_offset+DIM*1+XX];
793 jy1 = x[j_coord_offset+DIM*1+YY];
794 jz1 = x[j_coord_offset+DIM*1+ZZ];
795 jx2 = x[j_coord_offset+DIM*2+XX];
796 jy2 = x[j_coord_offset+DIM*2+YY];
797 jz2 = x[j_coord_offset+DIM*2+ZZ];
798 jx3 = x[j_coord_offset+DIM*3+XX];
799 jy3 = x[j_coord_offset+DIM*3+YY];
800 jz3 = x[j_coord_offset+DIM*3+ZZ];
802 /* Calculate displacement vector */
834 /* Calculate squared distance and things based on it */
835 rsq00 = dx00*dx00+dy00*dy00+dz00*dz00;
836 rsq11 = dx11*dx11+dy11*dy11+dz11*dz11;
837 rsq12 = dx12*dx12+dy12*dy12+dz12*dz12;
838 rsq13 = dx13*dx13+dy13*dy13+dz13*dz13;
839 rsq21 = dx21*dx21+dy21*dy21+dz21*dz21;
840 rsq22 = dx22*dx22+dy22*dy22+dz22*dz22;
841 rsq23 = dx23*dx23+dy23*dy23+dz23*dz23;
842 rsq31 = dx31*dx31+dy31*dy31+dz31*dz31;
843 rsq32 = dx32*dx32+dy32*dy32+dz32*dz32;
844 rsq33 = dx33*dx33+dy33*dy33+dz33*dz33;
846 rinv00 = gmx_invsqrt(rsq00);
847 rinv11 = gmx_invsqrt(rsq11);
848 rinv12 = gmx_invsqrt(rsq12);
849 rinv13 = gmx_invsqrt(rsq13);
850 rinv21 = gmx_invsqrt(rsq21);
851 rinv22 = gmx_invsqrt(rsq22);
852 rinv23 = gmx_invsqrt(rsq23);
853 rinv31 = gmx_invsqrt(rsq31);
854 rinv32 = gmx_invsqrt(rsq32);
855 rinv33 = gmx_invsqrt(rsq33);
857 rinvsq11 = rinv11*rinv11;
858 rinvsq12 = rinv12*rinv12;
859 rinvsq13 = rinv13*rinv13;
860 rinvsq21 = rinv21*rinv21;
861 rinvsq22 = rinv22*rinv22;
862 rinvsq23 = rinv23*rinv23;
863 rinvsq31 = rinv31*rinv31;
864 rinvsq32 = rinv32*rinv32;
865 rinvsq33 = rinv33*rinv33;
867 /**************************
868 * CALCULATE INTERACTIONS *
869 **************************/
873 /* Calculate table index by multiplying r with table scale and truncate to integer */
879 /* CUBIC SPLINE TABLE DISPERSION */
882 Geps = vfeps*vftab[vfitab+2];
883 Heps2 = vfeps*vfeps*vftab[vfitab+3];
885 FF = Fp+Geps+2.0*Heps2;
888 /* CUBIC SPLINE TABLE REPULSION */
890 Geps = vfeps*vftab[vfitab+6];
891 Heps2 = vfeps*vfeps*vftab[vfitab+7];
893 FF = Fp+Geps+2.0*Heps2;
895 fvdw = -(fvdw6+fvdw12)*vftabscale*rinv00;
899 /* Calculate temporary vectorial force */
904 /* Update vectorial force */
908 f[j_coord_offset+DIM*0+XX] -= tx;
909 f[j_coord_offset+DIM*0+YY] -= ty;
910 f[j_coord_offset+DIM*0+ZZ] -= tz;
912 /**************************
913 * CALCULATE INTERACTIONS *
914 **************************/
916 /* REACTION-FIELD ELECTROSTATICS */
917 felec = qq11*(rinv11*rinvsq11-krf2);
921 /* Calculate temporary vectorial force */
926 /* Update vectorial force */
930 f[j_coord_offset+DIM*1+XX] -= tx;
931 f[j_coord_offset+DIM*1+YY] -= ty;
932 f[j_coord_offset+DIM*1+ZZ] -= tz;
934 /**************************
935 * CALCULATE INTERACTIONS *
936 **************************/
938 /* REACTION-FIELD ELECTROSTATICS */
939 felec = qq12*(rinv12*rinvsq12-krf2);
943 /* Calculate temporary vectorial force */
948 /* Update vectorial force */
952 f[j_coord_offset+DIM*2+XX] -= tx;
953 f[j_coord_offset+DIM*2+YY] -= ty;
954 f[j_coord_offset+DIM*2+ZZ] -= tz;
956 /**************************
957 * CALCULATE INTERACTIONS *
958 **************************/
960 /* REACTION-FIELD ELECTROSTATICS */
961 felec = qq13*(rinv13*rinvsq13-krf2);
965 /* Calculate temporary vectorial force */
970 /* Update vectorial force */
974 f[j_coord_offset+DIM*3+XX] -= tx;
975 f[j_coord_offset+DIM*3+YY] -= ty;
976 f[j_coord_offset+DIM*3+ZZ] -= tz;
978 /**************************
979 * CALCULATE INTERACTIONS *
980 **************************/
982 /* REACTION-FIELD ELECTROSTATICS */
983 felec = qq21*(rinv21*rinvsq21-krf2);
987 /* Calculate temporary vectorial force */
992 /* Update vectorial force */
996 f[j_coord_offset+DIM*1+XX] -= tx;
997 f[j_coord_offset+DIM*1+YY] -= ty;
998 f[j_coord_offset+DIM*1+ZZ] -= tz;
1000 /**************************
1001 * CALCULATE INTERACTIONS *
1002 **************************/
1004 /* REACTION-FIELD ELECTROSTATICS */
1005 felec = qq22*(rinv22*rinvsq22-krf2);
1009 /* Calculate temporary vectorial force */
1014 /* Update vectorial force */
1018 f[j_coord_offset+DIM*2+XX] -= tx;
1019 f[j_coord_offset+DIM*2+YY] -= ty;
1020 f[j_coord_offset+DIM*2+ZZ] -= tz;
1022 /**************************
1023 * CALCULATE INTERACTIONS *
1024 **************************/
1026 /* REACTION-FIELD ELECTROSTATICS */
1027 felec = qq23*(rinv23*rinvsq23-krf2);
1031 /* Calculate temporary vectorial force */
1036 /* Update vectorial force */
1040 f[j_coord_offset+DIM*3+XX] -= tx;
1041 f[j_coord_offset+DIM*3+YY] -= ty;
1042 f[j_coord_offset+DIM*3+ZZ] -= tz;
1044 /**************************
1045 * CALCULATE INTERACTIONS *
1046 **************************/
1048 /* REACTION-FIELD ELECTROSTATICS */
1049 felec = qq31*(rinv31*rinvsq31-krf2);
1053 /* Calculate temporary vectorial force */
1058 /* Update vectorial force */
1062 f[j_coord_offset+DIM*1+XX] -= tx;
1063 f[j_coord_offset+DIM*1+YY] -= ty;
1064 f[j_coord_offset+DIM*1+ZZ] -= tz;
1066 /**************************
1067 * CALCULATE INTERACTIONS *
1068 **************************/
1070 /* REACTION-FIELD ELECTROSTATICS */
1071 felec = qq32*(rinv32*rinvsq32-krf2);
1075 /* Calculate temporary vectorial force */
1080 /* Update vectorial force */
1084 f[j_coord_offset+DIM*2+XX] -= tx;
1085 f[j_coord_offset+DIM*2+YY] -= ty;
1086 f[j_coord_offset+DIM*2+ZZ] -= tz;
1088 /**************************
1089 * CALCULATE INTERACTIONS *
1090 **************************/
1092 /* REACTION-FIELD ELECTROSTATICS */
1093 felec = qq33*(rinv33*rinvsq33-krf2);
1097 /* Calculate temporary vectorial force */
1102 /* Update vectorial force */
1106 f[j_coord_offset+DIM*3+XX] -= tx;
1107 f[j_coord_offset+DIM*3+YY] -= ty;
1108 f[j_coord_offset+DIM*3+ZZ] -= tz;
1110 /* Inner loop uses 281 flops */
1112 /* End of innermost loop */
1115 f[i_coord_offset+DIM*0+XX] += fix0;
1116 f[i_coord_offset+DIM*0+YY] += fiy0;
1117 f[i_coord_offset+DIM*0+ZZ] += fiz0;
1121 f[i_coord_offset+DIM*1+XX] += fix1;
1122 f[i_coord_offset+DIM*1+YY] += fiy1;
1123 f[i_coord_offset+DIM*1+ZZ] += fiz1;
1127 f[i_coord_offset+DIM*2+XX] += fix2;
1128 f[i_coord_offset+DIM*2+YY] += fiy2;
1129 f[i_coord_offset+DIM*2+ZZ] += fiz2;
1133 f[i_coord_offset+DIM*3+XX] += fix3;
1134 f[i_coord_offset+DIM*3+YY] += fiy3;
1135 f[i_coord_offset+DIM*3+ZZ] += fiz3;
1139 fshift[i_shift_offset+XX] += tx;
1140 fshift[i_shift_offset+YY] += ty;
1141 fshift[i_shift_offset+ZZ] += tz;
1143 /* Increment number of inner iterations */
1144 inneriter += j_index_end - j_index_start;
1146 /* Outer loop uses 39 flops */
1149 /* Increment number of outer iterations */
1152 /* Update outer/inner flops */
1154 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*39 + inneriter*281);