2 * Note: this file was generated by the Gromacs sse4_1_double kernel generator.
4 * This source code is part of
8 * Copyright (c) 2001-2012, The GROMACS Development Team
10 * Gromacs is a library for molecular simulation and trajectory analysis,
11 * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
12 * a full list of developers and information, check out http://www.gromacs.org
14 * This program is free software; you can redistribute it and/or modify it under
15 * the terms of the GNU Lesser General Public License as published by the Free
16 * Software Foundation; either version 2 of the License, or (at your option) any
19 * To help fund GROMACS development, we humbly ask that you cite
20 * the papers people have written on it - you can find them on the website.
28 #include "../nb_kernel.h"
29 #include "types/simple.h"
33 #include "gmx_math_x86_sse4_1_double.h"
34 #include "kernelutil_x86_sse4_1_double.h"
37 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW3P1_VF_sse4_1_double
38 * Electrostatics interaction: CubicSplineTable
39 * VdW interaction: None
40 * Geometry: Water3-Particle
41 * Calculate force/pot: PotentialAndForce
44 nb_kernel_ElecCSTab_VdwNone_GeomW3P1_VF_sse4_1_double
45 (t_nblist * gmx_restrict nlist,
46 rvec * gmx_restrict xx,
47 rvec * gmx_restrict ff,
48 t_forcerec * gmx_restrict fr,
49 t_mdatoms * gmx_restrict mdatoms,
50 nb_kernel_data_t * gmx_restrict kernel_data,
51 t_nrnb * gmx_restrict nrnb)
53 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
54 * just 0 for non-waters.
55 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
56 * jnr indices corresponding to data put in the four positions in the SIMD register.
58 int i_shift_offset,i_coord_offset,outeriter,inneriter;
59 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
61 int j_coord_offsetA,j_coord_offsetB;
62 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
64 real *shiftvec,*fshift,*x,*f;
65 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
67 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
69 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
71 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
72 int vdwjidx0A,vdwjidx0B;
73 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
74 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
75 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
76 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
77 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
80 __m128i ifour = _mm_set1_epi32(4);
81 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
83 __m128d dummy_mask,cutoff_mask;
84 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
85 __m128d one = _mm_set1_pd(1.0);
86 __m128d two = _mm_set1_pd(2.0);
92 jindex = nlist->jindex;
94 shiftidx = nlist->shift;
96 shiftvec = fr->shift_vec[0];
97 fshift = fr->fshift[0];
98 facel = _mm_set1_pd(fr->epsfac);
99 charge = mdatoms->chargeA;
101 vftab = kernel_data->table_elec->data;
102 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
104 /* Setup water-specific parameters */
105 inr = nlist->iinr[0];
106 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
107 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
108 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
110 /* Avoid stupid compiler warnings */
118 /* Start outer loop over neighborlists */
119 for(iidx=0; iidx<nri; iidx++)
121 /* Load shift vector for this list */
122 i_shift_offset = DIM*shiftidx[iidx];
124 /* Load limits for loop over neighbors */
125 j_index_start = jindex[iidx];
126 j_index_end = jindex[iidx+1];
128 /* Get outer coordinate index */
130 i_coord_offset = DIM*inr;
132 /* Load i particle coords and add shift vector */
133 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
134 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
136 fix0 = _mm_setzero_pd();
137 fiy0 = _mm_setzero_pd();
138 fiz0 = _mm_setzero_pd();
139 fix1 = _mm_setzero_pd();
140 fiy1 = _mm_setzero_pd();
141 fiz1 = _mm_setzero_pd();
142 fix2 = _mm_setzero_pd();
143 fiy2 = _mm_setzero_pd();
144 fiz2 = _mm_setzero_pd();
146 /* Reset potential sums */
147 velecsum = _mm_setzero_pd();
149 /* Start inner kernel loop */
150 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
153 /* Get j neighbor index, and coordinate index */
156 j_coord_offsetA = DIM*jnrA;
157 j_coord_offsetB = DIM*jnrB;
159 /* load j atom coordinates */
160 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
163 /* Calculate displacement vector */
164 dx00 = _mm_sub_pd(ix0,jx0);
165 dy00 = _mm_sub_pd(iy0,jy0);
166 dz00 = _mm_sub_pd(iz0,jz0);
167 dx10 = _mm_sub_pd(ix1,jx0);
168 dy10 = _mm_sub_pd(iy1,jy0);
169 dz10 = _mm_sub_pd(iz1,jz0);
170 dx20 = _mm_sub_pd(ix2,jx0);
171 dy20 = _mm_sub_pd(iy2,jy0);
172 dz20 = _mm_sub_pd(iz2,jz0);
174 /* Calculate squared distance and things based on it */
175 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
176 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
177 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
179 rinv00 = gmx_mm_invsqrt_pd(rsq00);
180 rinv10 = gmx_mm_invsqrt_pd(rsq10);
181 rinv20 = gmx_mm_invsqrt_pd(rsq20);
183 /* Load parameters for j particles */
184 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
186 fjx0 = _mm_setzero_pd();
187 fjy0 = _mm_setzero_pd();
188 fjz0 = _mm_setzero_pd();
190 /**************************
191 * CALCULATE INTERACTIONS *
192 **************************/
194 r00 = _mm_mul_pd(rsq00,rinv00);
196 /* Compute parameters for interactions between i and j atoms */
197 qq00 = _mm_mul_pd(iq0,jq0);
199 /* Calculate table index by multiplying r with table scale and truncate to integer */
200 rt = _mm_mul_pd(r00,vftabscale);
201 vfitab = _mm_cvttpd_epi32(rt);
202 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
203 vfitab = _mm_slli_epi32(vfitab,2);
205 /* CUBIC SPLINE TABLE ELECTROSTATICS */
206 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
207 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
208 GMX_MM_TRANSPOSE2_PD(Y,F);
209 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
210 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
211 GMX_MM_TRANSPOSE2_PD(G,H);
212 Heps = _mm_mul_pd(vfeps,H);
213 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
214 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
215 velec = _mm_mul_pd(qq00,VV);
216 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
217 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
219 /* Update potential sum for this i atom from the interaction with this j atom. */
220 velecsum = _mm_add_pd(velecsum,velec);
224 /* Calculate temporary vectorial force */
225 tx = _mm_mul_pd(fscal,dx00);
226 ty = _mm_mul_pd(fscal,dy00);
227 tz = _mm_mul_pd(fscal,dz00);
229 /* Update vectorial force */
230 fix0 = _mm_add_pd(fix0,tx);
231 fiy0 = _mm_add_pd(fiy0,ty);
232 fiz0 = _mm_add_pd(fiz0,tz);
234 fjx0 = _mm_add_pd(fjx0,tx);
235 fjy0 = _mm_add_pd(fjy0,ty);
236 fjz0 = _mm_add_pd(fjz0,tz);
238 /**************************
239 * CALCULATE INTERACTIONS *
240 **************************/
242 r10 = _mm_mul_pd(rsq10,rinv10);
244 /* Compute parameters for interactions between i and j atoms */
245 qq10 = _mm_mul_pd(iq1,jq0);
247 /* Calculate table index by multiplying r with table scale and truncate to integer */
248 rt = _mm_mul_pd(r10,vftabscale);
249 vfitab = _mm_cvttpd_epi32(rt);
250 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
251 vfitab = _mm_slli_epi32(vfitab,2);
253 /* CUBIC SPLINE TABLE ELECTROSTATICS */
254 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
255 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
256 GMX_MM_TRANSPOSE2_PD(Y,F);
257 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
258 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
259 GMX_MM_TRANSPOSE2_PD(G,H);
260 Heps = _mm_mul_pd(vfeps,H);
261 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
262 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
263 velec = _mm_mul_pd(qq10,VV);
264 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
265 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
267 /* Update potential sum for this i atom from the interaction with this j atom. */
268 velecsum = _mm_add_pd(velecsum,velec);
272 /* Calculate temporary vectorial force */
273 tx = _mm_mul_pd(fscal,dx10);
274 ty = _mm_mul_pd(fscal,dy10);
275 tz = _mm_mul_pd(fscal,dz10);
277 /* Update vectorial force */
278 fix1 = _mm_add_pd(fix1,tx);
279 fiy1 = _mm_add_pd(fiy1,ty);
280 fiz1 = _mm_add_pd(fiz1,tz);
282 fjx0 = _mm_add_pd(fjx0,tx);
283 fjy0 = _mm_add_pd(fjy0,ty);
284 fjz0 = _mm_add_pd(fjz0,tz);
286 /**************************
287 * CALCULATE INTERACTIONS *
288 **************************/
290 r20 = _mm_mul_pd(rsq20,rinv20);
292 /* Compute parameters for interactions between i and j atoms */
293 qq20 = _mm_mul_pd(iq2,jq0);
295 /* Calculate table index by multiplying r with table scale and truncate to integer */
296 rt = _mm_mul_pd(r20,vftabscale);
297 vfitab = _mm_cvttpd_epi32(rt);
298 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
299 vfitab = _mm_slli_epi32(vfitab,2);
301 /* CUBIC SPLINE TABLE ELECTROSTATICS */
302 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
303 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
304 GMX_MM_TRANSPOSE2_PD(Y,F);
305 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
306 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
307 GMX_MM_TRANSPOSE2_PD(G,H);
308 Heps = _mm_mul_pd(vfeps,H);
309 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
310 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
311 velec = _mm_mul_pd(qq20,VV);
312 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
313 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
315 /* Update potential sum for this i atom from the interaction with this j atom. */
316 velecsum = _mm_add_pd(velecsum,velec);
320 /* Calculate temporary vectorial force */
321 tx = _mm_mul_pd(fscal,dx20);
322 ty = _mm_mul_pd(fscal,dy20);
323 tz = _mm_mul_pd(fscal,dz20);
325 /* Update vectorial force */
326 fix2 = _mm_add_pd(fix2,tx);
327 fiy2 = _mm_add_pd(fiy2,ty);
328 fiz2 = _mm_add_pd(fiz2,tz);
330 fjx0 = _mm_add_pd(fjx0,tx);
331 fjy0 = _mm_add_pd(fjy0,ty);
332 fjz0 = _mm_add_pd(fjz0,tz);
334 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
336 /* Inner loop uses 132 flops */
343 j_coord_offsetA = DIM*jnrA;
345 /* load j atom coordinates */
346 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
349 /* Calculate displacement vector */
350 dx00 = _mm_sub_pd(ix0,jx0);
351 dy00 = _mm_sub_pd(iy0,jy0);
352 dz00 = _mm_sub_pd(iz0,jz0);
353 dx10 = _mm_sub_pd(ix1,jx0);
354 dy10 = _mm_sub_pd(iy1,jy0);
355 dz10 = _mm_sub_pd(iz1,jz0);
356 dx20 = _mm_sub_pd(ix2,jx0);
357 dy20 = _mm_sub_pd(iy2,jy0);
358 dz20 = _mm_sub_pd(iz2,jz0);
360 /* Calculate squared distance and things based on it */
361 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
362 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
363 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
365 rinv00 = gmx_mm_invsqrt_pd(rsq00);
366 rinv10 = gmx_mm_invsqrt_pd(rsq10);
367 rinv20 = gmx_mm_invsqrt_pd(rsq20);
369 /* Load parameters for j particles */
370 jq0 = _mm_load_sd(charge+jnrA+0);
372 fjx0 = _mm_setzero_pd();
373 fjy0 = _mm_setzero_pd();
374 fjz0 = _mm_setzero_pd();
376 /**************************
377 * CALCULATE INTERACTIONS *
378 **************************/
380 r00 = _mm_mul_pd(rsq00,rinv00);
382 /* Compute parameters for interactions between i and j atoms */
383 qq00 = _mm_mul_pd(iq0,jq0);
385 /* Calculate table index by multiplying r with table scale and truncate to integer */
386 rt = _mm_mul_pd(r00,vftabscale);
387 vfitab = _mm_cvttpd_epi32(rt);
388 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
389 vfitab = _mm_slli_epi32(vfitab,2);
391 /* CUBIC SPLINE TABLE ELECTROSTATICS */
392 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
393 F = _mm_setzero_pd();
394 GMX_MM_TRANSPOSE2_PD(Y,F);
395 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
396 H = _mm_setzero_pd();
397 GMX_MM_TRANSPOSE2_PD(G,H);
398 Heps = _mm_mul_pd(vfeps,H);
399 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
400 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
401 velec = _mm_mul_pd(qq00,VV);
402 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
403 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
405 /* Update potential sum for this i atom from the interaction with this j atom. */
406 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
407 velecsum = _mm_add_pd(velecsum,velec);
411 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
413 /* Calculate temporary vectorial force */
414 tx = _mm_mul_pd(fscal,dx00);
415 ty = _mm_mul_pd(fscal,dy00);
416 tz = _mm_mul_pd(fscal,dz00);
418 /* Update vectorial force */
419 fix0 = _mm_add_pd(fix0,tx);
420 fiy0 = _mm_add_pd(fiy0,ty);
421 fiz0 = _mm_add_pd(fiz0,tz);
423 fjx0 = _mm_add_pd(fjx0,tx);
424 fjy0 = _mm_add_pd(fjy0,ty);
425 fjz0 = _mm_add_pd(fjz0,tz);
427 /**************************
428 * CALCULATE INTERACTIONS *
429 **************************/
431 r10 = _mm_mul_pd(rsq10,rinv10);
433 /* Compute parameters for interactions between i and j atoms */
434 qq10 = _mm_mul_pd(iq1,jq0);
436 /* Calculate table index by multiplying r with table scale and truncate to integer */
437 rt = _mm_mul_pd(r10,vftabscale);
438 vfitab = _mm_cvttpd_epi32(rt);
439 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
440 vfitab = _mm_slli_epi32(vfitab,2);
442 /* CUBIC SPLINE TABLE ELECTROSTATICS */
443 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
444 F = _mm_setzero_pd();
445 GMX_MM_TRANSPOSE2_PD(Y,F);
446 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
447 H = _mm_setzero_pd();
448 GMX_MM_TRANSPOSE2_PD(G,H);
449 Heps = _mm_mul_pd(vfeps,H);
450 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
451 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
452 velec = _mm_mul_pd(qq10,VV);
453 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
454 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
456 /* Update potential sum for this i atom from the interaction with this j atom. */
457 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
458 velecsum = _mm_add_pd(velecsum,velec);
462 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
464 /* Calculate temporary vectorial force */
465 tx = _mm_mul_pd(fscal,dx10);
466 ty = _mm_mul_pd(fscal,dy10);
467 tz = _mm_mul_pd(fscal,dz10);
469 /* Update vectorial force */
470 fix1 = _mm_add_pd(fix1,tx);
471 fiy1 = _mm_add_pd(fiy1,ty);
472 fiz1 = _mm_add_pd(fiz1,tz);
474 fjx0 = _mm_add_pd(fjx0,tx);
475 fjy0 = _mm_add_pd(fjy0,ty);
476 fjz0 = _mm_add_pd(fjz0,tz);
478 /**************************
479 * CALCULATE INTERACTIONS *
480 **************************/
482 r20 = _mm_mul_pd(rsq20,rinv20);
484 /* Compute parameters for interactions between i and j atoms */
485 qq20 = _mm_mul_pd(iq2,jq0);
487 /* Calculate table index by multiplying r with table scale and truncate to integer */
488 rt = _mm_mul_pd(r20,vftabscale);
489 vfitab = _mm_cvttpd_epi32(rt);
490 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
491 vfitab = _mm_slli_epi32(vfitab,2);
493 /* CUBIC SPLINE TABLE ELECTROSTATICS */
494 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
495 F = _mm_setzero_pd();
496 GMX_MM_TRANSPOSE2_PD(Y,F);
497 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
498 H = _mm_setzero_pd();
499 GMX_MM_TRANSPOSE2_PD(G,H);
500 Heps = _mm_mul_pd(vfeps,H);
501 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
502 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
503 velec = _mm_mul_pd(qq20,VV);
504 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
505 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
507 /* Update potential sum for this i atom from the interaction with this j atom. */
508 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
509 velecsum = _mm_add_pd(velecsum,velec);
513 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
515 /* Calculate temporary vectorial force */
516 tx = _mm_mul_pd(fscal,dx20);
517 ty = _mm_mul_pd(fscal,dy20);
518 tz = _mm_mul_pd(fscal,dz20);
520 /* Update vectorial force */
521 fix2 = _mm_add_pd(fix2,tx);
522 fiy2 = _mm_add_pd(fiy2,ty);
523 fiz2 = _mm_add_pd(fiz2,tz);
525 fjx0 = _mm_add_pd(fjx0,tx);
526 fjy0 = _mm_add_pd(fjy0,ty);
527 fjz0 = _mm_add_pd(fjz0,tz);
529 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
531 /* Inner loop uses 132 flops */
534 /* End of innermost loop */
536 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
537 f+i_coord_offset,fshift+i_shift_offset);
540 /* Update potential energies */
541 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
543 /* Increment number of inner iterations */
544 inneriter += j_index_end - j_index_start;
546 /* Outer loop uses 19 flops */
549 /* Increment number of outer iterations */
552 /* Update outer/inner flops */
554 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3_VF,outeriter*19 + inneriter*132);
557 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW3P1_F_sse4_1_double
558 * Electrostatics interaction: CubicSplineTable
559 * VdW interaction: None
560 * Geometry: Water3-Particle
561 * Calculate force/pot: Force
564 nb_kernel_ElecCSTab_VdwNone_GeomW3P1_F_sse4_1_double
565 (t_nblist * gmx_restrict nlist,
566 rvec * gmx_restrict xx,
567 rvec * gmx_restrict ff,
568 t_forcerec * gmx_restrict fr,
569 t_mdatoms * gmx_restrict mdatoms,
570 nb_kernel_data_t * gmx_restrict kernel_data,
571 t_nrnb * gmx_restrict nrnb)
573 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
574 * just 0 for non-waters.
575 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
576 * jnr indices corresponding to data put in the four positions in the SIMD register.
578 int i_shift_offset,i_coord_offset,outeriter,inneriter;
579 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
581 int j_coord_offsetA,j_coord_offsetB;
582 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
584 real *shiftvec,*fshift,*x,*f;
585 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
587 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
589 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
591 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
592 int vdwjidx0A,vdwjidx0B;
593 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
594 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
595 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
596 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
597 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
600 __m128i ifour = _mm_set1_epi32(4);
601 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
603 __m128d dummy_mask,cutoff_mask;
604 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
605 __m128d one = _mm_set1_pd(1.0);
606 __m128d two = _mm_set1_pd(2.0);
612 jindex = nlist->jindex;
614 shiftidx = nlist->shift;
616 shiftvec = fr->shift_vec[0];
617 fshift = fr->fshift[0];
618 facel = _mm_set1_pd(fr->epsfac);
619 charge = mdatoms->chargeA;
621 vftab = kernel_data->table_elec->data;
622 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
624 /* Setup water-specific parameters */
625 inr = nlist->iinr[0];
626 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
627 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
628 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
630 /* Avoid stupid compiler warnings */
638 /* Start outer loop over neighborlists */
639 for(iidx=0; iidx<nri; iidx++)
641 /* Load shift vector for this list */
642 i_shift_offset = DIM*shiftidx[iidx];
644 /* Load limits for loop over neighbors */
645 j_index_start = jindex[iidx];
646 j_index_end = jindex[iidx+1];
648 /* Get outer coordinate index */
650 i_coord_offset = DIM*inr;
652 /* Load i particle coords and add shift vector */
653 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
654 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
656 fix0 = _mm_setzero_pd();
657 fiy0 = _mm_setzero_pd();
658 fiz0 = _mm_setzero_pd();
659 fix1 = _mm_setzero_pd();
660 fiy1 = _mm_setzero_pd();
661 fiz1 = _mm_setzero_pd();
662 fix2 = _mm_setzero_pd();
663 fiy2 = _mm_setzero_pd();
664 fiz2 = _mm_setzero_pd();
666 /* Start inner kernel loop */
667 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
670 /* Get j neighbor index, and coordinate index */
673 j_coord_offsetA = DIM*jnrA;
674 j_coord_offsetB = DIM*jnrB;
676 /* load j atom coordinates */
677 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
680 /* Calculate displacement vector */
681 dx00 = _mm_sub_pd(ix0,jx0);
682 dy00 = _mm_sub_pd(iy0,jy0);
683 dz00 = _mm_sub_pd(iz0,jz0);
684 dx10 = _mm_sub_pd(ix1,jx0);
685 dy10 = _mm_sub_pd(iy1,jy0);
686 dz10 = _mm_sub_pd(iz1,jz0);
687 dx20 = _mm_sub_pd(ix2,jx0);
688 dy20 = _mm_sub_pd(iy2,jy0);
689 dz20 = _mm_sub_pd(iz2,jz0);
691 /* Calculate squared distance and things based on it */
692 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
693 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
694 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
696 rinv00 = gmx_mm_invsqrt_pd(rsq00);
697 rinv10 = gmx_mm_invsqrt_pd(rsq10);
698 rinv20 = gmx_mm_invsqrt_pd(rsq20);
700 /* Load parameters for j particles */
701 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
703 fjx0 = _mm_setzero_pd();
704 fjy0 = _mm_setzero_pd();
705 fjz0 = _mm_setzero_pd();
707 /**************************
708 * CALCULATE INTERACTIONS *
709 **************************/
711 r00 = _mm_mul_pd(rsq00,rinv00);
713 /* Compute parameters for interactions between i and j atoms */
714 qq00 = _mm_mul_pd(iq0,jq0);
716 /* Calculate table index by multiplying r with table scale and truncate to integer */
717 rt = _mm_mul_pd(r00,vftabscale);
718 vfitab = _mm_cvttpd_epi32(rt);
719 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
720 vfitab = _mm_slli_epi32(vfitab,2);
722 /* CUBIC SPLINE TABLE ELECTROSTATICS */
723 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
724 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
725 GMX_MM_TRANSPOSE2_PD(Y,F);
726 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
727 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
728 GMX_MM_TRANSPOSE2_PD(G,H);
729 Heps = _mm_mul_pd(vfeps,H);
730 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
731 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
732 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
736 /* Calculate temporary vectorial force */
737 tx = _mm_mul_pd(fscal,dx00);
738 ty = _mm_mul_pd(fscal,dy00);
739 tz = _mm_mul_pd(fscal,dz00);
741 /* Update vectorial force */
742 fix0 = _mm_add_pd(fix0,tx);
743 fiy0 = _mm_add_pd(fiy0,ty);
744 fiz0 = _mm_add_pd(fiz0,tz);
746 fjx0 = _mm_add_pd(fjx0,tx);
747 fjy0 = _mm_add_pd(fjy0,ty);
748 fjz0 = _mm_add_pd(fjz0,tz);
750 /**************************
751 * CALCULATE INTERACTIONS *
752 **************************/
754 r10 = _mm_mul_pd(rsq10,rinv10);
756 /* Compute parameters for interactions between i and j atoms */
757 qq10 = _mm_mul_pd(iq1,jq0);
759 /* Calculate table index by multiplying r with table scale and truncate to integer */
760 rt = _mm_mul_pd(r10,vftabscale);
761 vfitab = _mm_cvttpd_epi32(rt);
762 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
763 vfitab = _mm_slli_epi32(vfitab,2);
765 /* CUBIC SPLINE TABLE ELECTROSTATICS */
766 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
767 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
768 GMX_MM_TRANSPOSE2_PD(Y,F);
769 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
770 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
771 GMX_MM_TRANSPOSE2_PD(G,H);
772 Heps = _mm_mul_pd(vfeps,H);
773 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
774 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
775 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
779 /* Calculate temporary vectorial force */
780 tx = _mm_mul_pd(fscal,dx10);
781 ty = _mm_mul_pd(fscal,dy10);
782 tz = _mm_mul_pd(fscal,dz10);
784 /* Update vectorial force */
785 fix1 = _mm_add_pd(fix1,tx);
786 fiy1 = _mm_add_pd(fiy1,ty);
787 fiz1 = _mm_add_pd(fiz1,tz);
789 fjx0 = _mm_add_pd(fjx0,tx);
790 fjy0 = _mm_add_pd(fjy0,ty);
791 fjz0 = _mm_add_pd(fjz0,tz);
793 /**************************
794 * CALCULATE INTERACTIONS *
795 **************************/
797 r20 = _mm_mul_pd(rsq20,rinv20);
799 /* Compute parameters for interactions between i and j atoms */
800 qq20 = _mm_mul_pd(iq2,jq0);
802 /* Calculate table index by multiplying r with table scale and truncate to integer */
803 rt = _mm_mul_pd(r20,vftabscale);
804 vfitab = _mm_cvttpd_epi32(rt);
805 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
806 vfitab = _mm_slli_epi32(vfitab,2);
808 /* CUBIC SPLINE TABLE ELECTROSTATICS */
809 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
810 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
811 GMX_MM_TRANSPOSE2_PD(Y,F);
812 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
813 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
814 GMX_MM_TRANSPOSE2_PD(G,H);
815 Heps = _mm_mul_pd(vfeps,H);
816 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
817 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
818 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
822 /* Calculate temporary vectorial force */
823 tx = _mm_mul_pd(fscal,dx20);
824 ty = _mm_mul_pd(fscal,dy20);
825 tz = _mm_mul_pd(fscal,dz20);
827 /* Update vectorial force */
828 fix2 = _mm_add_pd(fix2,tx);
829 fiy2 = _mm_add_pd(fiy2,ty);
830 fiz2 = _mm_add_pd(fiz2,tz);
832 fjx0 = _mm_add_pd(fjx0,tx);
833 fjy0 = _mm_add_pd(fjy0,ty);
834 fjz0 = _mm_add_pd(fjz0,tz);
836 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
838 /* Inner loop uses 120 flops */
845 j_coord_offsetA = DIM*jnrA;
847 /* load j atom coordinates */
848 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
851 /* Calculate displacement vector */
852 dx00 = _mm_sub_pd(ix0,jx0);
853 dy00 = _mm_sub_pd(iy0,jy0);
854 dz00 = _mm_sub_pd(iz0,jz0);
855 dx10 = _mm_sub_pd(ix1,jx0);
856 dy10 = _mm_sub_pd(iy1,jy0);
857 dz10 = _mm_sub_pd(iz1,jz0);
858 dx20 = _mm_sub_pd(ix2,jx0);
859 dy20 = _mm_sub_pd(iy2,jy0);
860 dz20 = _mm_sub_pd(iz2,jz0);
862 /* Calculate squared distance and things based on it */
863 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
864 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
865 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
867 rinv00 = gmx_mm_invsqrt_pd(rsq00);
868 rinv10 = gmx_mm_invsqrt_pd(rsq10);
869 rinv20 = gmx_mm_invsqrt_pd(rsq20);
871 /* Load parameters for j particles */
872 jq0 = _mm_load_sd(charge+jnrA+0);
874 fjx0 = _mm_setzero_pd();
875 fjy0 = _mm_setzero_pd();
876 fjz0 = _mm_setzero_pd();
878 /**************************
879 * CALCULATE INTERACTIONS *
880 **************************/
882 r00 = _mm_mul_pd(rsq00,rinv00);
884 /* Compute parameters for interactions between i and j atoms */
885 qq00 = _mm_mul_pd(iq0,jq0);
887 /* Calculate table index by multiplying r with table scale and truncate to integer */
888 rt = _mm_mul_pd(r00,vftabscale);
889 vfitab = _mm_cvttpd_epi32(rt);
890 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
891 vfitab = _mm_slli_epi32(vfitab,2);
893 /* CUBIC SPLINE TABLE ELECTROSTATICS */
894 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
895 F = _mm_setzero_pd();
896 GMX_MM_TRANSPOSE2_PD(Y,F);
897 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
898 H = _mm_setzero_pd();
899 GMX_MM_TRANSPOSE2_PD(G,H);
900 Heps = _mm_mul_pd(vfeps,H);
901 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
902 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
903 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
907 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
909 /* Calculate temporary vectorial force */
910 tx = _mm_mul_pd(fscal,dx00);
911 ty = _mm_mul_pd(fscal,dy00);
912 tz = _mm_mul_pd(fscal,dz00);
914 /* Update vectorial force */
915 fix0 = _mm_add_pd(fix0,tx);
916 fiy0 = _mm_add_pd(fiy0,ty);
917 fiz0 = _mm_add_pd(fiz0,tz);
919 fjx0 = _mm_add_pd(fjx0,tx);
920 fjy0 = _mm_add_pd(fjy0,ty);
921 fjz0 = _mm_add_pd(fjz0,tz);
923 /**************************
924 * CALCULATE INTERACTIONS *
925 **************************/
927 r10 = _mm_mul_pd(rsq10,rinv10);
929 /* Compute parameters for interactions between i and j atoms */
930 qq10 = _mm_mul_pd(iq1,jq0);
932 /* Calculate table index by multiplying r with table scale and truncate to integer */
933 rt = _mm_mul_pd(r10,vftabscale);
934 vfitab = _mm_cvttpd_epi32(rt);
935 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
936 vfitab = _mm_slli_epi32(vfitab,2);
938 /* CUBIC SPLINE TABLE ELECTROSTATICS */
939 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
940 F = _mm_setzero_pd();
941 GMX_MM_TRANSPOSE2_PD(Y,F);
942 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
943 H = _mm_setzero_pd();
944 GMX_MM_TRANSPOSE2_PD(G,H);
945 Heps = _mm_mul_pd(vfeps,H);
946 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
947 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
948 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
952 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
954 /* Calculate temporary vectorial force */
955 tx = _mm_mul_pd(fscal,dx10);
956 ty = _mm_mul_pd(fscal,dy10);
957 tz = _mm_mul_pd(fscal,dz10);
959 /* Update vectorial force */
960 fix1 = _mm_add_pd(fix1,tx);
961 fiy1 = _mm_add_pd(fiy1,ty);
962 fiz1 = _mm_add_pd(fiz1,tz);
964 fjx0 = _mm_add_pd(fjx0,tx);
965 fjy0 = _mm_add_pd(fjy0,ty);
966 fjz0 = _mm_add_pd(fjz0,tz);
968 /**************************
969 * CALCULATE INTERACTIONS *
970 **************************/
972 r20 = _mm_mul_pd(rsq20,rinv20);
974 /* Compute parameters for interactions between i and j atoms */
975 qq20 = _mm_mul_pd(iq2,jq0);
977 /* Calculate table index by multiplying r with table scale and truncate to integer */
978 rt = _mm_mul_pd(r20,vftabscale);
979 vfitab = _mm_cvttpd_epi32(rt);
980 vfeps = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
981 vfitab = _mm_slli_epi32(vfitab,2);
983 /* CUBIC SPLINE TABLE ELECTROSTATICS */
984 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
985 F = _mm_setzero_pd();
986 GMX_MM_TRANSPOSE2_PD(Y,F);
987 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
988 H = _mm_setzero_pd();
989 GMX_MM_TRANSPOSE2_PD(G,H);
990 Heps = _mm_mul_pd(vfeps,H);
991 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
992 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
993 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
997 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
999 /* Calculate temporary vectorial force */
1000 tx = _mm_mul_pd(fscal,dx20);
1001 ty = _mm_mul_pd(fscal,dy20);
1002 tz = _mm_mul_pd(fscal,dz20);
1004 /* Update vectorial force */
1005 fix2 = _mm_add_pd(fix2,tx);
1006 fiy2 = _mm_add_pd(fiy2,ty);
1007 fiz2 = _mm_add_pd(fiz2,tz);
1009 fjx0 = _mm_add_pd(fjx0,tx);
1010 fjy0 = _mm_add_pd(fjy0,ty);
1011 fjz0 = _mm_add_pd(fjz0,tz);
1013 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1015 /* Inner loop uses 120 flops */
1018 /* End of innermost loop */
1020 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1021 f+i_coord_offset,fshift+i_shift_offset);
1023 /* Increment number of inner iterations */
1024 inneriter += j_index_end - j_index_start;
1026 /* Outer loop uses 18 flops */
1029 /* Increment number of outer iterations */
1032 /* Update outer/inner flops */
1034 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3_F,outeriter*18 + inneriter*120);