?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Remove nb-parameters from t_forcerec
[gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_avx_256_double / nb_kernel_ElecCoul_VdwLJ_GeomW4P1_avx_256_double.c
blobdd0bfaccd2cb6c5c5346508de669e9dac69a7ff8
1 /*
2 * This file is part of the GROMACS molecular simulation package.
3 *
4 * Copyright (c) 2012,2013,2014,2015,2017, 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.
8 *
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.
13 *
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.
18 *
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.
23 *
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.
31 *
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.
34 */
35 /*
36 * Note: this file was generated by the GROMACS avx_256_double kernel generator.
37 */
38 #include "gmxpre.h"
39
40 #include "config.h"
41
42 #include <math.h>
43
44 #include "../nb_kernel.h"
45 #include "gromacs/gmxlib/nrnb.h"
46
47 #include "kernelutil_x86_avx_256_double.h"
48
49 /*
50 * Gromacs nonbonded kernel: nb_kernel_ElecCoul_VdwLJ_GeomW4P1_VF_avx_256_double
51 * Electrostatics interaction: Coulomb
52 * VdW interaction: LennardJones
53 * Geometry: Water4-Particle
54 * Calculate force/pot: PotentialAndForce
55 */
56 void
57 nb_kernel_ElecCoul_VdwLJ_GeomW4P1_VF_avx_256_double
58 (t_nblist * gmx_restrict nlist,
59 rvec * gmx_restrict xx,
60 rvec * gmx_restrict ff,
61 struct 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)
65 {
66 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
67 * just 0 for non-waters.
68 * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
69 * jnr indices corresponding to data put in the four positions in the SIMD register.
70 */
71 int i_shift_offset,i_coord_offset,outeriter,inneriter;
72 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
73 int jnrA,jnrB,jnrC,jnrD;
74 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
75 int jnrlistE,jnrlistF,jnrlistG,jnrlistH;
76 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
77 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
78 real rcutoff_scalar;
79 real *shiftvec,*fshift,*x,*f;
80 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
81 real scratch[4*DIM];
82 __m256d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
83 real * vdwioffsetptr0;
84 __m256d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
85 real * vdwioffsetptr1;
86 __m256d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
87 real * vdwioffsetptr2;
88 __m256d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
89 real * vdwioffsetptr3;
90 __m256d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
91 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
92 __m256d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
93 __m256d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
94 __m256d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
95 __m256d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
96 __m256d dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
97 __m256d velec,felec,velecsum,facel,crf,krf,krf2;
98 real *charge;
99 int nvdwtype;
100 __m256d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
101 int *vdwtype;
102 real *vdwparam;
103 __m256d one_sixth = _mm256_set1_pd(1.0/6.0);
104 __m256d one_twelfth = _mm256_set1_pd(1.0/12.0);
105 __m256d dummy_mask,cutoff_mask;
106 __m128 tmpmask0,tmpmask1;
107 __m256d signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
108 __m256d one = _mm256_set1_pd(1.0);
109 __m256d two = _mm256_set1_pd(2.0);
110 x = xx[0];
111 f = ff[0];
112
113 nri = nlist->nri;
114 iinr = nlist->iinr;
115 jindex = nlist->jindex;
116 jjnr = nlist->jjnr;
117 shiftidx = nlist->shift;
118 gid = nlist->gid;
119 shiftvec = fr->shift_vec[0];
120 fshift = fr->fshift[0];
121 facel = _mm256_set1_pd(fr->ic->epsfac);
122 charge = mdatoms->chargeA;
123 nvdwtype = fr->ntype;
124 vdwparam = fr->nbfp;
125 vdwtype = mdatoms->typeA;
126
127 /* Setup water-specific parameters */
128 inr = nlist->iinr[0];
129 iq1 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
130 iq2 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
131 iq3 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
132 vdwioffsetptr0 = vdwparam+2*nvdwtype*vdwtype[inr+0];
133
134 /* Avoid stupid compiler warnings */
135 jnrA = jnrB = jnrC = jnrD = 0;
136 j_coord_offsetA = 0;
137 j_coord_offsetB = 0;
138 j_coord_offsetC = 0;
139 j_coord_offsetD = 0;
140
141 outeriter = 0;
142 inneriter = 0;
143
144 for(iidx=0;iidx<4*DIM;iidx++)
145 {
146 scratch[iidx] = 0.0;
147 }
148
149 /* Start outer loop over neighborlists */
150 for(iidx=0; iidx<nri; iidx++)
151 {
152 /* Load shift vector for this list */
153 i_shift_offset = DIM*shiftidx[iidx];
154
155 /* Load limits for loop over neighbors */
156 j_index_start = jindex[iidx];
157 j_index_end = jindex[iidx+1];
158
159 /* Get outer coordinate index */
160 inr = iinr[iidx];
161 i_coord_offset = DIM*inr;
162
163 /* Load i particle coords and add shift vector */
164 gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
165 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
166
167 fix0 = _mm256_setzero_pd();
168 fiy0 = _mm256_setzero_pd();
169 fiz0 = _mm256_setzero_pd();
170 fix1 = _mm256_setzero_pd();
171 fiy1 = _mm256_setzero_pd();
172 fiz1 = _mm256_setzero_pd();
173 fix2 = _mm256_setzero_pd();
174 fiy2 = _mm256_setzero_pd();
175 fiz2 = _mm256_setzero_pd();
176 fix3 = _mm256_setzero_pd();
177 fiy3 = _mm256_setzero_pd();
178 fiz3 = _mm256_setzero_pd();
179
180 /* Reset potential sums */
181 velecsum = _mm256_setzero_pd();
182 vvdwsum = _mm256_setzero_pd();
183
184 /* Start inner kernel loop */
185 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
186 {
187
188 /* Get j neighbor index, and coordinate index */
189 jnrA = jjnr[jidx];
190 jnrB = jjnr[jidx+1];
191 jnrC = jjnr[jidx+2];
192 jnrD = jjnr[jidx+3];
193 j_coord_offsetA = DIM*jnrA;
194 j_coord_offsetB = DIM*jnrB;
195 j_coord_offsetC = DIM*jnrC;
196 j_coord_offsetD = DIM*jnrD;
197
198 /* load j atom coordinates */
199 gmx_mm256_load_1rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
200 x+j_coord_offsetC,x+j_coord_offsetD,
201 &jx0,&jy0,&jz0);
202
203 /* Calculate displacement vector */
204 dx00 = _mm256_sub_pd(ix0,jx0);
205 dy00 = _mm256_sub_pd(iy0,jy0);
206 dz00 = _mm256_sub_pd(iz0,jz0);
207 dx10 = _mm256_sub_pd(ix1,jx0);
208 dy10 = _mm256_sub_pd(iy1,jy0);
209 dz10 = _mm256_sub_pd(iz1,jz0);
210 dx20 = _mm256_sub_pd(ix2,jx0);
211 dy20 = _mm256_sub_pd(iy2,jy0);
212 dz20 = _mm256_sub_pd(iz2,jz0);
213 dx30 = _mm256_sub_pd(ix3,jx0);
214 dy30 = _mm256_sub_pd(iy3,jy0);
215 dz30 = _mm256_sub_pd(iz3,jz0);
216
217 /* Calculate squared distance and things based on it */
218 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
219 rsq10 = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
220 rsq20 = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
221 rsq30 = gmx_mm256_calc_rsq_pd(dx30,dy30,dz30);
222
223 rinv10 = avx256_invsqrt_d(rsq10);
224 rinv20 = avx256_invsqrt_d(rsq20);
225 rinv30 = avx256_invsqrt_d(rsq30);
226
227 rinvsq00 = avx256_inv_d(rsq00);
228 rinvsq10 = _mm256_mul_pd(rinv10,rinv10);
229 rinvsq20 = _mm256_mul_pd(rinv20,rinv20);
230 rinvsq30 = _mm256_mul_pd(rinv30,rinv30);
231
232 /* Load parameters for j particles */
233 jq0 = gmx_mm256_load_4real_swizzle_pd(charge+jnrA+0,charge+jnrB+0,
234 charge+jnrC+0,charge+jnrD+0);
235 vdwjidx0A = 2*vdwtype[jnrA+0];
236 vdwjidx0B = 2*vdwtype[jnrB+0];
237 vdwjidx0C = 2*vdwtype[jnrC+0];
238 vdwjidx0D = 2*vdwtype[jnrD+0];
239
240 fjx0 = _mm256_setzero_pd();
241 fjy0 = _mm256_setzero_pd();
242 fjz0 = _mm256_setzero_pd();
243
244 /**************************
245 * CALCULATE INTERACTIONS *
246 **************************/
247
248 /* Compute parameters for interactions between i and j atoms */
249 gmx_mm256_load_4pair_swizzle_pd(vdwioffsetptr0+vdwjidx0A,
250 vdwioffsetptr0+vdwjidx0B,
251 vdwioffsetptr0+vdwjidx0C,
252 vdwioffsetptr0+vdwjidx0D,
253 &c6_00,&c12_00);
254
255 /* LENNARD-JONES DISPERSION/REPULSION */
256
257 rinvsix = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
258 vvdw6 = _mm256_mul_pd(c6_00,rinvsix);
259 vvdw12 = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
260 vvdw = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
261 fvdw = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
262
263 /* Update potential sum for this i atom from the interaction with this j atom. */
264 vvdwsum = _mm256_add_pd(vvdwsum,vvdw);
265
266 fscal = fvdw;
267
268 /* Calculate temporary vectorial force */
269 tx = _mm256_mul_pd(fscal,dx00);
270 ty = _mm256_mul_pd(fscal,dy00);
271 tz = _mm256_mul_pd(fscal,dz00);
272
273 /* Update vectorial force */
274 fix0 = _mm256_add_pd(fix0,tx);
275 fiy0 = _mm256_add_pd(fiy0,ty);
276 fiz0 = _mm256_add_pd(fiz0,tz);
277
278 fjx0 = _mm256_add_pd(fjx0,tx);
279 fjy0 = _mm256_add_pd(fjy0,ty);
280 fjz0 = _mm256_add_pd(fjz0,tz);
281
282 /**************************
283 * CALCULATE INTERACTIONS *
284 **************************/
285
286 /* Compute parameters for interactions between i and j atoms */
287 qq10 = _mm256_mul_pd(iq1,jq0);
288
289 /* COULOMB ELECTROSTATICS */
290 velec = _mm256_mul_pd(qq10,rinv10);
291 felec = _mm256_mul_pd(velec,rinvsq10);
292
293 /* Update potential sum for this i atom from the interaction with this j atom. */
294 velecsum = _mm256_add_pd(velecsum,velec);
295
296 fscal = felec;
297
298 /* Calculate temporary vectorial force */
299 tx = _mm256_mul_pd(fscal,dx10);
300 ty = _mm256_mul_pd(fscal,dy10);
301 tz = _mm256_mul_pd(fscal,dz10);
302
303 /* Update vectorial force */
304 fix1 = _mm256_add_pd(fix1,tx);
305 fiy1 = _mm256_add_pd(fiy1,ty);
306 fiz1 = _mm256_add_pd(fiz1,tz);
307
308 fjx0 = _mm256_add_pd(fjx0,tx);
309 fjy0 = _mm256_add_pd(fjy0,ty);
310 fjz0 = _mm256_add_pd(fjz0,tz);
311
312 /**************************
313 * CALCULATE INTERACTIONS *
314 **************************/
315
316 /* Compute parameters for interactions between i and j atoms */
317 qq20 = _mm256_mul_pd(iq2,jq0);
318
319 /* COULOMB ELECTROSTATICS */
320 velec = _mm256_mul_pd(qq20,rinv20);
321 felec = _mm256_mul_pd(velec,rinvsq20);
322
323 /* Update potential sum for this i atom from the interaction with this j atom. */
324 velecsum = _mm256_add_pd(velecsum,velec);
325
326 fscal = felec;
327
328 /* Calculate temporary vectorial force */
329 tx = _mm256_mul_pd(fscal,dx20);
330 ty = _mm256_mul_pd(fscal,dy20);
331 tz = _mm256_mul_pd(fscal,dz20);
332
333 /* Update vectorial force */
334 fix2 = _mm256_add_pd(fix2,tx);
335 fiy2 = _mm256_add_pd(fiy2,ty);
336 fiz2 = _mm256_add_pd(fiz2,tz);
337
338 fjx0 = _mm256_add_pd(fjx0,tx);
339 fjy0 = _mm256_add_pd(fjy0,ty);
340 fjz0 = _mm256_add_pd(fjz0,tz);
341
342 /**************************
343 * CALCULATE INTERACTIONS *
344 **************************/
345
346 /* Compute parameters for interactions between i and j atoms */
347 qq30 = _mm256_mul_pd(iq3,jq0);
348
349 /* COULOMB ELECTROSTATICS */
350 velec = _mm256_mul_pd(qq30,rinv30);
351 felec = _mm256_mul_pd(velec,rinvsq30);
352
353 /* Update potential sum for this i atom from the interaction with this j atom. */
354 velecsum = _mm256_add_pd(velecsum,velec);
355
356 fscal = felec;
357
358 /* Calculate temporary vectorial force */
359 tx = _mm256_mul_pd(fscal,dx30);
360 ty = _mm256_mul_pd(fscal,dy30);
361 tz = _mm256_mul_pd(fscal,dz30);
362
363 /* Update vectorial force */
364 fix3 = _mm256_add_pd(fix3,tx);
365 fiy3 = _mm256_add_pd(fiy3,ty);
366 fiz3 = _mm256_add_pd(fiz3,tz);
367
368 fjx0 = _mm256_add_pd(fjx0,tx);
369 fjy0 = _mm256_add_pd(fjy0,ty);
370 fjz0 = _mm256_add_pd(fjz0,tz);
371
372 fjptrA = f+j_coord_offsetA;
373 fjptrB = f+j_coord_offsetB;
374 fjptrC = f+j_coord_offsetC;
375 fjptrD = f+j_coord_offsetD;
376
377 gmx_mm256_decrement_1rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
378
379 /* Inner loop uses 116 flops */
380 }
381
382 if(jidx<j_index_end)
383 {
384
385 /* Get j neighbor index, and coordinate index */
386 jnrlistA = jjnr[jidx];
387 jnrlistB = jjnr[jidx+1];
388 jnrlistC = jjnr[jidx+2];
389 jnrlistD = jjnr[jidx+3];
390 /* Sign of each element will be negative for non-real atoms.
391 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
392 * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
393 */
394 tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
395
396 tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
397 tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
398 dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
399
400 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
401 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
402 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
403 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
404 j_coord_offsetA = DIM*jnrA;
405 j_coord_offsetB = DIM*jnrB;
406 j_coord_offsetC = DIM*jnrC;
407 j_coord_offsetD = DIM*jnrD;
408
409 /* load j atom coordinates */
410 gmx_mm256_load_1rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
411 x+j_coord_offsetC,x+j_coord_offsetD,
412 &jx0,&jy0,&jz0);
413
414 /* Calculate displacement vector */
415 dx00 = _mm256_sub_pd(ix0,jx0);
416 dy00 = _mm256_sub_pd(iy0,jy0);
417 dz00 = _mm256_sub_pd(iz0,jz0);
418 dx10 = _mm256_sub_pd(ix1,jx0);
419 dy10 = _mm256_sub_pd(iy1,jy0);
420 dz10 = _mm256_sub_pd(iz1,jz0);
421 dx20 = _mm256_sub_pd(ix2,jx0);
422 dy20 = _mm256_sub_pd(iy2,jy0);
423 dz20 = _mm256_sub_pd(iz2,jz0);
424 dx30 = _mm256_sub_pd(ix3,jx0);
425 dy30 = _mm256_sub_pd(iy3,jy0);
426 dz30 = _mm256_sub_pd(iz3,jz0);
427
428 /* Calculate squared distance and things based on it */
429 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
430 rsq10 = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
431 rsq20 = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
432 rsq30 = gmx_mm256_calc_rsq_pd(dx30,dy30,dz30);
433
434 rinv10 = avx256_invsqrt_d(rsq10);
435 rinv20 = avx256_invsqrt_d(rsq20);
436 rinv30 = avx256_invsqrt_d(rsq30);
437
438 rinvsq00 = avx256_inv_d(rsq00);
439 rinvsq10 = _mm256_mul_pd(rinv10,rinv10);
440 rinvsq20 = _mm256_mul_pd(rinv20,rinv20);
441 rinvsq30 = _mm256_mul_pd(rinv30,rinv30);
442
443 /* Load parameters for j particles */
444 jq0 = gmx_mm256_load_4real_swizzle_pd(charge+jnrA+0,charge+jnrB+0,
445 charge+jnrC+0,charge+jnrD+0);
446 vdwjidx0A = 2*vdwtype[jnrA+0];
447 vdwjidx0B = 2*vdwtype[jnrB+0];
448 vdwjidx0C = 2*vdwtype[jnrC+0];
449 vdwjidx0D = 2*vdwtype[jnrD+0];
450
451 fjx0 = _mm256_setzero_pd();
452 fjy0 = _mm256_setzero_pd();
453 fjz0 = _mm256_setzero_pd();
454
455 /**************************
456 * CALCULATE INTERACTIONS *
457 **************************/
458
459 /* Compute parameters for interactions between i and j atoms */
460 gmx_mm256_load_4pair_swizzle_pd(vdwioffsetptr0+vdwjidx0A,
461 vdwioffsetptr0+vdwjidx0B,
462 vdwioffsetptr0+vdwjidx0C,
463 vdwioffsetptr0+vdwjidx0D,
464 &c6_00,&c12_00);
465
466 /* LENNARD-JONES DISPERSION/REPULSION */
467
468 rinvsix = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
469 vvdw6 = _mm256_mul_pd(c6_00,rinvsix);
470 vvdw12 = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
471 vvdw = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
472 fvdw = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
473
474 /* Update potential sum for this i atom from the interaction with this j atom. */
475 vvdw = _mm256_andnot_pd(dummy_mask,vvdw);
476 vvdwsum = _mm256_add_pd(vvdwsum,vvdw);
477
478 fscal = fvdw;
479
480 fscal = _mm256_andnot_pd(dummy_mask,fscal);
481
482 /* Calculate temporary vectorial force */
483 tx = _mm256_mul_pd(fscal,dx00);
484 ty = _mm256_mul_pd(fscal,dy00);
485 tz = _mm256_mul_pd(fscal,dz00);
486
487 /* Update vectorial force */
488 fix0 = _mm256_add_pd(fix0,tx);
489 fiy0 = _mm256_add_pd(fiy0,ty);
490 fiz0 = _mm256_add_pd(fiz0,tz);
491
492 fjx0 = _mm256_add_pd(fjx0,tx);
493 fjy0 = _mm256_add_pd(fjy0,ty);
494 fjz0 = _mm256_add_pd(fjz0,tz);
495
496 /**************************
497 * CALCULATE INTERACTIONS *
498 **************************/
499
500 /* Compute parameters for interactions between i and j atoms */
501 qq10 = _mm256_mul_pd(iq1,jq0);
502
503 /* COULOMB ELECTROSTATICS */
504 velec = _mm256_mul_pd(qq10,rinv10);
505 felec = _mm256_mul_pd(velec,rinvsq10);
506
507 /* Update potential sum for this i atom from the interaction with this j atom. */
508 velec = _mm256_andnot_pd(dummy_mask,velec);
509 velecsum = _mm256_add_pd(velecsum,velec);
510
511 fscal = felec;
512
513 fscal = _mm256_andnot_pd(dummy_mask,fscal);
514
515 /* Calculate temporary vectorial force */
516 tx = _mm256_mul_pd(fscal,dx10);
517 ty = _mm256_mul_pd(fscal,dy10);
518 tz = _mm256_mul_pd(fscal,dz10);
519
520 /* Update vectorial force */
521 fix1 = _mm256_add_pd(fix1,tx);
522 fiy1 = _mm256_add_pd(fiy1,ty);
523 fiz1 = _mm256_add_pd(fiz1,tz);
524
525 fjx0 = _mm256_add_pd(fjx0,tx);
526 fjy0 = _mm256_add_pd(fjy0,ty);
527 fjz0 = _mm256_add_pd(fjz0,tz);
528
529 /**************************
530 * CALCULATE INTERACTIONS *
531 **************************/
532
533 /* Compute parameters for interactions between i and j atoms */
534 qq20 = _mm256_mul_pd(iq2,jq0);
535
536 /* COULOMB ELECTROSTATICS */
537 velec = _mm256_mul_pd(qq20,rinv20);
538 felec = _mm256_mul_pd(velec,rinvsq20);
539
540 /* Update potential sum for this i atom from the interaction with this j atom. */
541 velec = _mm256_andnot_pd(dummy_mask,velec);
542 velecsum = _mm256_add_pd(velecsum,velec);
543
544 fscal = felec;
545
546 fscal = _mm256_andnot_pd(dummy_mask,fscal);
547
548 /* Calculate temporary vectorial force */
549 tx = _mm256_mul_pd(fscal,dx20);
550 ty = _mm256_mul_pd(fscal,dy20);
551 tz = _mm256_mul_pd(fscal,dz20);
552
553 /* Update vectorial force */
554 fix2 = _mm256_add_pd(fix2,tx);
555 fiy2 = _mm256_add_pd(fiy2,ty);
556 fiz2 = _mm256_add_pd(fiz2,tz);
557
558 fjx0 = _mm256_add_pd(fjx0,tx);
559 fjy0 = _mm256_add_pd(fjy0,ty);
560 fjz0 = _mm256_add_pd(fjz0,tz);
561
562 /**************************
563 * CALCULATE INTERACTIONS *
564 **************************/
565
566 /* Compute parameters for interactions between i and j atoms */
567 qq30 = _mm256_mul_pd(iq3,jq0);
568
569 /* COULOMB ELECTROSTATICS */
570 velec = _mm256_mul_pd(qq30,rinv30);
571 felec = _mm256_mul_pd(velec,rinvsq30);
572
573 /* Update potential sum for this i atom from the interaction with this j atom. */
574 velec = _mm256_andnot_pd(dummy_mask,velec);
575 velecsum = _mm256_add_pd(velecsum,velec);
576
577 fscal = felec;
578
579 fscal = _mm256_andnot_pd(dummy_mask,fscal);
580
581 /* Calculate temporary vectorial force */
582 tx = _mm256_mul_pd(fscal,dx30);
583 ty = _mm256_mul_pd(fscal,dy30);
584 tz = _mm256_mul_pd(fscal,dz30);
585
586 /* Update vectorial force */
587 fix3 = _mm256_add_pd(fix3,tx);
588 fiy3 = _mm256_add_pd(fiy3,ty);
589 fiz3 = _mm256_add_pd(fiz3,tz);
590
591 fjx0 = _mm256_add_pd(fjx0,tx);
592 fjy0 = _mm256_add_pd(fjy0,ty);
593 fjz0 = _mm256_add_pd(fjz0,tz);
594
595 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
596 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
597 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
598 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
599
600 gmx_mm256_decrement_1rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
601
602 /* Inner loop uses 116 flops */
603 }
604
605 /* End of innermost loop */
606
607 gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
608 f+i_coord_offset,fshift+i_shift_offset);
609
610 ggid = gid[iidx];
611 /* Update potential energies */
612 gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
613 gmx_mm256_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
614
615 /* Increment number of inner iterations */
616 inneriter += j_index_end - j_index_start;
617
618 /* Outer loop uses 26 flops */
619 }
620
621 /* Increment number of outer iterations */
622 outeriter += nri;
623
624 /* Update outer/inner flops */
625
626 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*26 + inneriter*116);
627 }
628 /*
629 * Gromacs nonbonded kernel: nb_kernel_ElecCoul_VdwLJ_GeomW4P1_F_avx_256_double
630 * Electrostatics interaction: Coulomb
631 * VdW interaction: LennardJones
632 * Geometry: Water4-Particle
633 * Calculate force/pot: Force
634 */
635 void
636 nb_kernel_ElecCoul_VdwLJ_GeomW4P1_F_avx_256_double
637 (t_nblist * gmx_restrict nlist,
638 rvec * gmx_restrict xx,
639 rvec * gmx_restrict ff,
640 struct t_forcerec * gmx_restrict fr,
641 t_mdatoms * gmx_restrict mdatoms,
642 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
643 t_nrnb * gmx_restrict nrnb)
644 {
645 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
646 * just 0 for non-waters.
647 * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
648 * jnr indices corresponding to data put in the four positions in the SIMD register.
649 */
650 int i_shift_offset,i_coord_offset,outeriter,inneriter;
651 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
652 int jnrA,jnrB,jnrC,jnrD;
653 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
654 int jnrlistE,jnrlistF,jnrlistG,jnrlistH;
655 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
656 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
657 real rcutoff_scalar;
658 real *shiftvec,*fshift,*x,*f;
659 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
660 real scratch[4*DIM];
661 __m256d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
662 real * vdwioffsetptr0;
663 __m256d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
664 real * vdwioffsetptr1;
665 __m256d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
666 real * vdwioffsetptr2;
667 __m256d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
668 real * vdwioffsetptr3;
669 __m256d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
670 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
671 __m256d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
672 __m256d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
673 __m256d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
674 __m256d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
675 __m256d dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
676 __m256d velec,felec,velecsum,facel,crf,krf,krf2;
677 real *charge;
678 int nvdwtype;
679 __m256d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
680 int *vdwtype;
681 real *vdwparam;
682 __m256d one_sixth = _mm256_set1_pd(1.0/6.0);
683 __m256d one_twelfth = _mm256_set1_pd(1.0/12.0);
684 __m256d dummy_mask,cutoff_mask;
685 __m128 tmpmask0,tmpmask1;
686 __m256d signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
687 __m256d one = _mm256_set1_pd(1.0);
688 __m256d two = _mm256_set1_pd(2.0);
689 x = xx[0];
690 f = ff[0];
691
692 nri = nlist->nri;
693 iinr = nlist->iinr;
694 jindex = nlist->jindex;
695 jjnr = nlist->jjnr;
696 shiftidx = nlist->shift;
697 gid = nlist->gid;
698 shiftvec = fr->shift_vec[0];
699 fshift = fr->fshift[0];
700 facel = _mm256_set1_pd(fr->ic->epsfac);
701 charge = mdatoms->chargeA;
702 nvdwtype = fr->ntype;
703 vdwparam = fr->nbfp;
704 vdwtype = mdatoms->typeA;
705
706 /* Setup water-specific parameters */
707 inr = nlist->iinr[0];
708 iq1 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
709 iq2 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
710 iq3 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
711 vdwioffsetptr0 = vdwparam+2*nvdwtype*vdwtype[inr+0];
712
713 /* Avoid stupid compiler warnings */
714 jnrA = jnrB = jnrC = jnrD = 0;
715 j_coord_offsetA = 0;
716 j_coord_offsetB = 0;
717 j_coord_offsetC = 0;
718 j_coord_offsetD = 0;
719
720 outeriter = 0;
721 inneriter = 0;
722
723 for(iidx=0;iidx<4*DIM;iidx++)
724 {
725 scratch[iidx] = 0.0;
726 }
727
728 /* Start outer loop over neighborlists */
729 for(iidx=0; iidx<nri; iidx++)
730 {
731 /* Load shift vector for this list */
732 i_shift_offset = DIM*shiftidx[iidx];
733
734 /* Load limits for loop over neighbors */
735 j_index_start = jindex[iidx];
736 j_index_end = jindex[iidx+1];
737
738 /* Get outer coordinate index */
739 inr = iinr[iidx];
740 i_coord_offset = DIM*inr;
741
742 /* Load i particle coords and add shift vector */
743 gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
744 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
745
746 fix0 = _mm256_setzero_pd();
747 fiy0 = _mm256_setzero_pd();
748 fiz0 = _mm256_setzero_pd();
749 fix1 = _mm256_setzero_pd();
750 fiy1 = _mm256_setzero_pd();
751 fiz1 = _mm256_setzero_pd();
752 fix2 = _mm256_setzero_pd();
753 fiy2 = _mm256_setzero_pd();
754 fiz2 = _mm256_setzero_pd();
755 fix3 = _mm256_setzero_pd();
756 fiy3 = _mm256_setzero_pd();
757 fiz3 = _mm256_setzero_pd();
758
759 /* Start inner kernel loop */
760 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
761 {
762
763 /* Get j neighbor index, and coordinate index */
764 jnrA = jjnr[jidx];
765 jnrB = jjnr[jidx+1];
766 jnrC = jjnr[jidx+2];
767 jnrD = jjnr[jidx+3];
768 j_coord_offsetA = DIM*jnrA;
769 j_coord_offsetB = DIM*jnrB;
770 j_coord_offsetC = DIM*jnrC;
771 j_coord_offsetD = DIM*jnrD;
772
773 /* load j atom coordinates */
774 gmx_mm256_load_1rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
775 x+j_coord_offsetC,x+j_coord_offsetD,
776 &jx0,&jy0,&jz0);
777
778 /* Calculate displacement vector */
779 dx00 = _mm256_sub_pd(ix0,jx0);
780 dy00 = _mm256_sub_pd(iy0,jy0);
781 dz00 = _mm256_sub_pd(iz0,jz0);
782 dx10 = _mm256_sub_pd(ix1,jx0);
783 dy10 = _mm256_sub_pd(iy1,jy0);
784 dz10 = _mm256_sub_pd(iz1,jz0);
785 dx20 = _mm256_sub_pd(ix2,jx0);
786 dy20 = _mm256_sub_pd(iy2,jy0);
787 dz20 = _mm256_sub_pd(iz2,jz0);
788 dx30 = _mm256_sub_pd(ix3,jx0);
789 dy30 = _mm256_sub_pd(iy3,jy0);
790 dz30 = _mm256_sub_pd(iz3,jz0);
791
792 /* Calculate squared distance and things based on it */
793 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
794 rsq10 = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
795 rsq20 = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
796 rsq30 = gmx_mm256_calc_rsq_pd(dx30,dy30,dz30);
797
798 rinv10 = avx256_invsqrt_d(rsq10);
799 rinv20 = avx256_invsqrt_d(rsq20);
800 rinv30 = avx256_invsqrt_d(rsq30);
801
802 rinvsq00 = avx256_inv_d(rsq00);
803 rinvsq10 = _mm256_mul_pd(rinv10,rinv10);
804 rinvsq20 = _mm256_mul_pd(rinv20,rinv20);
805 rinvsq30 = _mm256_mul_pd(rinv30,rinv30);
806
807 /* Load parameters for j particles */
808 jq0 = gmx_mm256_load_4real_swizzle_pd(charge+jnrA+0,charge+jnrB+0,
809 charge+jnrC+0,charge+jnrD+0);
810 vdwjidx0A = 2*vdwtype[jnrA+0];
811 vdwjidx0B = 2*vdwtype[jnrB+0];
812 vdwjidx0C = 2*vdwtype[jnrC+0];
813 vdwjidx0D = 2*vdwtype[jnrD+0];
814
815 fjx0 = _mm256_setzero_pd();
816 fjy0 = _mm256_setzero_pd();
817 fjz0 = _mm256_setzero_pd();
818
819 /**************************
820 * CALCULATE INTERACTIONS *
821 **************************/
822
823 /* Compute parameters for interactions between i and j atoms */
824 gmx_mm256_load_4pair_swizzle_pd(vdwioffsetptr0+vdwjidx0A,
825 vdwioffsetptr0+vdwjidx0B,
826 vdwioffsetptr0+vdwjidx0C,
827 vdwioffsetptr0+vdwjidx0D,
828 &c6_00,&c12_00);
829
830 /* LENNARD-JONES DISPERSION/REPULSION */
831
832 rinvsix = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
833 fvdw = _mm256_mul_pd(_mm256_sub_pd(_mm256_mul_pd(c12_00,rinvsix),c6_00),_mm256_mul_pd(rinvsix,rinvsq00));
834
835 fscal = fvdw;
836
837 /* Calculate temporary vectorial force */
838 tx = _mm256_mul_pd(fscal,dx00);
839 ty = _mm256_mul_pd(fscal,dy00);
840 tz = _mm256_mul_pd(fscal,dz00);
841
842 /* Update vectorial force */
843 fix0 = _mm256_add_pd(fix0,tx);
844 fiy0 = _mm256_add_pd(fiy0,ty);
845 fiz0 = _mm256_add_pd(fiz0,tz);
846
847 fjx0 = _mm256_add_pd(fjx0,tx);
848 fjy0 = _mm256_add_pd(fjy0,ty);
849 fjz0 = _mm256_add_pd(fjz0,tz);
850
851 /**************************
852 * CALCULATE INTERACTIONS *
853 **************************/
854
855 /* Compute parameters for interactions between i and j atoms */
856 qq10 = _mm256_mul_pd(iq1,jq0);
857
858 /* COULOMB ELECTROSTATICS */
859 velec = _mm256_mul_pd(qq10,rinv10);
860 felec = _mm256_mul_pd(velec,rinvsq10);
861
862 fscal = felec;
863
864 /* Calculate temporary vectorial force */
865 tx = _mm256_mul_pd(fscal,dx10);
866 ty = _mm256_mul_pd(fscal,dy10);
867 tz = _mm256_mul_pd(fscal,dz10);
868
869 /* Update vectorial force */
870 fix1 = _mm256_add_pd(fix1,tx);
871 fiy1 = _mm256_add_pd(fiy1,ty);
872 fiz1 = _mm256_add_pd(fiz1,tz);
873
874 fjx0 = _mm256_add_pd(fjx0,tx);
875 fjy0 = _mm256_add_pd(fjy0,ty);
876 fjz0 = _mm256_add_pd(fjz0,tz);
877
878 /**************************
879 * CALCULATE INTERACTIONS *
880 **************************/
881
882 /* Compute parameters for interactions between i and j atoms */
883 qq20 = _mm256_mul_pd(iq2,jq0);
884
885 /* COULOMB ELECTROSTATICS */
886 velec = _mm256_mul_pd(qq20,rinv20);
887 felec = _mm256_mul_pd(velec,rinvsq20);
888
889 fscal = felec;
890
891 /* Calculate temporary vectorial force */
892 tx = _mm256_mul_pd(fscal,dx20);
893 ty = _mm256_mul_pd(fscal,dy20);
894 tz = _mm256_mul_pd(fscal,dz20);
895
896 /* Update vectorial force */
897 fix2 = _mm256_add_pd(fix2,tx);
898 fiy2 = _mm256_add_pd(fiy2,ty);
899 fiz2 = _mm256_add_pd(fiz2,tz);
900
901 fjx0 = _mm256_add_pd(fjx0,tx);
902 fjy0 = _mm256_add_pd(fjy0,ty);
903 fjz0 = _mm256_add_pd(fjz0,tz);
904
905 /**************************
906 * CALCULATE INTERACTIONS *
907 **************************/
908
909 /* Compute parameters for interactions between i and j atoms */
910 qq30 = _mm256_mul_pd(iq3,jq0);
911
912 /* COULOMB ELECTROSTATICS */
913 velec = _mm256_mul_pd(qq30,rinv30);
914 felec = _mm256_mul_pd(velec,rinvsq30);
915
916 fscal = felec;
917
918 /* Calculate temporary vectorial force */
919 tx = _mm256_mul_pd(fscal,dx30);
920 ty = _mm256_mul_pd(fscal,dy30);
921 tz = _mm256_mul_pd(fscal,dz30);
922
923 /* Update vectorial force */
924 fix3 = _mm256_add_pd(fix3,tx);
925 fiy3 = _mm256_add_pd(fiy3,ty);
926 fiz3 = _mm256_add_pd(fiz3,tz);
927
928 fjx0 = _mm256_add_pd(fjx0,tx);
929 fjy0 = _mm256_add_pd(fjy0,ty);
930 fjz0 = _mm256_add_pd(fjz0,tz);
931
932 fjptrA = f+j_coord_offsetA;
933 fjptrB = f+j_coord_offsetB;
934 fjptrC = f+j_coord_offsetC;
935 fjptrD = f+j_coord_offsetD;
936
937 gmx_mm256_decrement_1rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
938
939 /* Inner loop uses 108 flops */
940 }
941
942 if(jidx<j_index_end)
943 {
944
945 /* Get j neighbor index, and coordinate index */
946 jnrlistA = jjnr[jidx];
947 jnrlistB = jjnr[jidx+1];
948 jnrlistC = jjnr[jidx+2];
949 jnrlistD = jjnr[jidx+3];
950 /* Sign of each element will be negative for non-real atoms.
951 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
952 * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
953 */
954 tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
955
956 tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
957 tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
958 dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
959
960 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
961 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
962 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
963 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
964 j_coord_offsetA = DIM*jnrA;
965 j_coord_offsetB = DIM*jnrB;
966 j_coord_offsetC = DIM*jnrC;
967 j_coord_offsetD = DIM*jnrD;
968
969 /* load j atom coordinates */
970 gmx_mm256_load_1rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
971 x+j_coord_offsetC,x+j_coord_offsetD,
972 &jx0,&jy0,&jz0);
973
974 /* Calculate displacement vector */
975 dx00 = _mm256_sub_pd(ix0,jx0);
976 dy00 = _mm256_sub_pd(iy0,jy0);
977 dz00 = _mm256_sub_pd(iz0,jz0);
978 dx10 = _mm256_sub_pd(ix1,jx0);
979 dy10 = _mm256_sub_pd(iy1,jy0);
980 dz10 = _mm256_sub_pd(iz1,jz0);
981 dx20 = _mm256_sub_pd(ix2,jx0);
982 dy20 = _mm256_sub_pd(iy2,jy0);
983 dz20 = _mm256_sub_pd(iz2,jz0);
984 dx30 = _mm256_sub_pd(ix3,jx0);
985 dy30 = _mm256_sub_pd(iy3,jy0);
986 dz30 = _mm256_sub_pd(iz3,jz0);
987
988 /* Calculate squared distance and things based on it */
989 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
990 rsq10 = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
991 rsq20 = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
992 rsq30 = gmx_mm256_calc_rsq_pd(dx30,dy30,dz30);
993
994 rinv10 = avx256_invsqrt_d(rsq10);
995 rinv20 = avx256_invsqrt_d(rsq20);
996 rinv30 = avx256_invsqrt_d(rsq30);
997
998 rinvsq00 = avx256_inv_d(rsq00);
999 rinvsq10 = _mm256_mul_pd(rinv10,rinv10);
1000 rinvsq20 = _mm256_mul_pd(rinv20,rinv20);
1001 rinvsq30 = _mm256_mul_pd(rinv30,rinv30);
1002
1003 /* Load parameters for j particles */
1004 jq0 = gmx_mm256_load_4real_swizzle_pd(charge+jnrA+0,charge+jnrB+0,
1005 charge+jnrC+0,charge+jnrD+0);
1006 vdwjidx0A = 2*vdwtype[jnrA+0];
1007 vdwjidx0B = 2*vdwtype[jnrB+0];
1008 vdwjidx0C = 2*vdwtype[jnrC+0];
1009 vdwjidx0D = 2*vdwtype[jnrD+0];
1010
1011 fjx0 = _mm256_setzero_pd();
1012 fjy0 = _mm256_setzero_pd();
1013 fjz0 = _mm256_setzero_pd();
1014
1015 /**************************
1016 * CALCULATE INTERACTIONS *
1017 **************************/
1018
1019 /* Compute parameters for interactions between i and j atoms */
1020 gmx_mm256_load_4pair_swizzle_pd(vdwioffsetptr0+vdwjidx0A,
1021 vdwioffsetptr0+vdwjidx0B,
1022 vdwioffsetptr0+vdwjidx0C,
1023 vdwioffsetptr0+vdwjidx0D,
1024 &c6_00,&c12_00);
1025
1026 /* LENNARD-JONES DISPERSION/REPULSION */
1027
1028 rinvsix = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1029 fvdw = _mm256_mul_pd(_mm256_sub_pd(_mm256_mul_pd(c12_00,rinvsix),c6_00),_mm256_mul_pd(rinvsix,rinvsq00));
1030
1031 fscal = fvdw;
1032
1033 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1034
1035 /* Calculate temporary vectorial force */
1036 tx = _mm256_mul_pd(fscal,dx00);
1037 ty = _mm256_mul_pd(fscal,dy00);
1038 tz = _mm256_mul_pd(fscal,dz00);
1039
1040 /* Update vectorial force */
1041 fix0 = _mm256_add_pd(fix0,tx);
1042 fiy0 = _mm256_add_pd(fiy0,ty);
1043 fiz0 = _mm256_add_pd(fiz0,tz);
1044
1045 fjx0 = _mm256_add_pd(fjx0,tx);
1046 fjy0 = _mm256_add_pd(fjy0,ty);
1047 fjz0 = _mm256_add_pd(fjz0,tz);
1048
1049 /**************************
1050 * CALCULATE INTERACTIONS *
1051 **************************/
1052
1053 /* Compute parameters for interactions between i and j atoms */
1054 qq10 = _mm256_mul_pd(iq1,jq0);
1055
1056 /* COULOMB ELECTROSTATICS */
1057 velec = _mm256_mul_pd(qq10,rinv10);
1058 felec = _mm256_mul_pd(velec,rinvsq10);
1059
1060 fscal = felec;
1061
1062 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1063
1064 /* Calculate temporary vectorial force */
1065 tx = _mm256_mul_pd(fscal,dx10);
1066 ty = _mm256_mul_pd(fscal,dy10);
1067 tz = _mm256_mul_pd(fscal,dz10);
1068
1069 /* Update vectorial force */
1070 fix1 = _mm256_add_pd(fix1,tx);
1071 fiy1 = _mm256_add_pd(fiy1,ty);
1072 fiz1 = _mm256_add_pd(fiz1,tz);
1073
1074 fjx0 = _mm256_add_pd(fjx0,tx);
1075 fjy0 = _mm256_add_pd(fjy0,ty);
1076 fjz0 = _mm256_add_pd(fjz0,tz);
1077
1078 /**************************
1079 * CALCULATE INTERACTIONS *
1080 **************************/
1081
1082 /* Compute parameters for interactions between i and j atoms */
1083 qq20 = _mm256_mul_pd(iq2,jq0);
1084
1085 /* COULOMB ELECTROSTATICS */
1086 velec = _mm256_mul_pd(qq20,rinv20);
1087 felec = _mm256_mul_pd(velec,rinvsq20);
1088
1089 fscal = felec;
1090
1091 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1092
1093 /* Calculate temporary vectorial force */
1094 tx = _mm256_mul_pd(fscal,dx20);
1095 ty = _mm256_mul_pd(fscal,dy20);
1096 tz = _mm256_mul_pd(fscal,dz20);
1097
1098 /* Update vectorial force */
1099 fix2 = _mm256_add_pd(fix2,tx);
1100 fiy2 = _mm256_add_pd(fiy2,ty);
1101 fiz2 = _mm256_add_pd(fiz2,tz);
1102
1103 fjx0 = _mm256_add_pd(fjx0,tx);
1104 fjy0 = _mm256_add_pd(fjy0,ty);
1105 fjz0 = _mm256_add_pd(fjz0,tz);
1106
1107 /**************************
1108 * CALCULATE INTERACTIONS *
1109 **************************/
1110
1111 /* Compute parameters for interactions between i and j atoms */
1112 qq30 = _mm256_mul_pd(iq3,jq0);
1113
1114 /* COULOMB ELECTROSTATICS */
1115 velec = _mm256_mul_pd(qq30,rinv30);
1116 felec = _mm256_mul_pd(velec,rinvsq30);
1117
1118 fscal = felec;
1119
1120 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1121
1122 /* Calculate temporary vectorial force */
1123 tx = _mm256_mul_pd(fscal,dx30);
1124 ty = _mm256_mul_pd(fscal,dy30);
1125 tz = _mm256_mul_pd(fscal,dz30);
1126
1127 /* Update vectorial force */
1128 fix3 = _mm256_add_pd(fix3,tx);
1129 fiy3 = _mm256_add_pd(fiy3,ty);
1130 fiz3 = _mm256_add_pd(fiz3,tz);
1131
1132 fjx0 = _mm256_add_pd(fjx0,tx);
1133 fjy0 = _mm256_add_pd(fjy0,ty);
1134 fjz0 = _mm256_add_pd(fjz0,tz);
1135
1136 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1137 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1138 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1139 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1140
1141 gmx_mm256_decrement_1rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
1142
1143 /* Inner loop uses 108 flops */
1144 }
1145
1146 /* End of innermost loop */
1147
1148 gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1149 f+i_coord_offset,fshift+i_shift_offset);
1150
1151 /* Increment number of inner iterations */
1152 inneriter += j_index_end - j_index_start;
1153
1154 /* Outer loop uses 24 flops */
1155 }
1156
1157 /* Increment number of outer iterations */
1158 outeriter += nri;
1159
1160 /* Update outer/inner flops */
1161
1162 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*24 + inneriter*108);
1163 }
The ultimate molecular dynamics simulation package