?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Double precision SSE2 kernels
[gromacs.git] / src / gmxlib / nonbonded / nb_kernel_sse2_double / nb_kernel_ElecCSTab_VdwNone_GeomW4P1_sse2_double.c
blobb93d71426870b4fe1bb693fed8da0ec455de71a8
1 /*
2 * Note: this file was generated by the Gromacs sse2_double kernel generator.
3 *
4 * This source code is part of
5 *
6 * G R O M A C S
7 *
8 * Copyright (c) 2001-2012, The GROMACS Development Team
9 *
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
13 *
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
17 * later version.
18 *
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.
21 */
22 #ifdef HAVE_CONFIG_H
23 #include <config.h>
24 #endif
25
26 #include <math.h>
27
28 #include "../nb_kernel.h"
29 #include "types/simple.h"
30 #include "vec.h"
31 #include "nrnb.h"
32
33 #include "gmx_math_x86_sse2_double.h"
34 #include "kernelutil_x86_sse2_double.h"
35
36 /*
37 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW4P1_VF_sse2_double
38 * Electrostatics interaction: CubicSplineTable
39 * VdW interaction: None
40 * Geometry: Water4-Particle
41 * Calculate force/pot: PotentialAndForce
42 */
43 void
44 nb_kernel_ElecCSTab_VdwNone_GeomW4P1_VF_sse2_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)
52 {
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.
57 */
58 int i_shift_offset,i_coord_offset,outeriter,inneriter;
59 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
60 int jnrA,jnrB;
61 int j_coord_offsetA,j_coord_offsetB;
62 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
63 real rcutoff_scalar;
64 real *shiftvec,*fshift,*x,*f;
65 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
66 int vdwioffset1;
67 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
68 int vdwioffset2;
69 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
70 int vdwioffset3;
71 __m128d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
72 int vdwjidx0A,vdwjidx0B;
73 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
74 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
75 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
76 __m128d dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
77 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
78 real *charge;
79 __m128i vfitab;
80 __m128i ifour = _mm_set1_epi32(4);
81 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
82 real *vftab;
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);
87 x = xx[0];
88 f = ff[0];
89
90 nri = nlist->nri;
91 iinr = nlist->iinr;
92 jindex = nlist->jindex;
93 jjnr = nlist->jjnr;
94 shiftidx = nlist->shift;
95 gid = nlist->gid;
96 shiftvec = fr->shift_vec[0];
97 fshift = fr->fshift[0];
98 facel = _mm_set1_pd(fr->epsfac);
99 charge = mdatoms->chargeA;
100
101 vftab = kernel_data->table_elec->data;
102 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
103
104 /* Setup water-specific parameters */
105 inr = nlist->iinr[0];
106 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
107 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
108 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
109
110 /* Avoid stupid compiler warnings */
111 jnrA = jnrB = 0;
112 j_coord_offsetA = 0;
113 j_coord_offsetB = 0;
114
115 outeriter = 0;
116 inneriter = 0;
117
118 /* Start outer loop over neighborlists */
119 for(iidx=0; iidx<nri; iidx++)
120 {
121 /* Load shift vector for this list */
122 i_shift_offset = DIM*shiftidx[iidx];
123
124 /* Load limits for loop over neighbors */
125 j_index_start = jindex[iidx];
126 j_index_end = jindex[iidx+1];
127
128 /* Get outer coordinate index */
129 inr = iinr[iidx];
130 i_coord_offset = DIM*inr;
131
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+DIM,
134 &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
135
136 fix1 = _mm_setzero_pd();
137 fiy1 = _mm_setzero_pd();
138 fiz1 = _mm_setzero_pd();
139 fix2 = _mm_setzero_pd();
140 fiy2 = _mm_setzero_pd();
141 fiz2 = _mm_setzero_pd();
142 fix3 = _mm_setzero_pd();
143 fiy3 = _mm_setzero_pd();
144 fiz3 = _mm_setzero_pd();
145
146 /* Reset potential sums */
147 velecsum = _mm_setzero_pd();
148
149 /* Start inner kernel loop */
150 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
151 {
152
153 /* Get j neighbor index, and coordinate index */
154 jnrA = jjnr[jidx];
155 jnrB = jjnr[jidx+1];
156 j_coord_offsetA = DIM*jnrA;
157 j_coord_offsetB = DIM*jnrB;
158
159 /* load j atom coordinates */
160 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
161 &jx0,&jy0,&jz0);
162
163 /* Calculate displacement vector */
164 dx10 = _mm_sub_pd(ix1,jx0);
165 dy10 = _mm_sub_pd(iy1,jy0);
166 dz10 = _mm_sub_pd(iz1,jz0);
167 dx20 = _mm_sub_pd(ix2,jx0);
168 dy20 = _mm_sub_pd(iy2,jy0);
169 dz20 = _mm_sub_pd(iz2,jz0);
170 dx30 = _mm_sub_pd(ix3,jx0);
171 dy30 = _mm_sub_pd(iy3,jy0);
172 dz30 = _mm_sub_pd(iz3,jz0);
173
174 /* Calculate squared distance and things based on it */
175 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
176 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
177 rsq30 = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
178
179 rinv10 = gmx_mm_invsqrt_pd(rsq10);
180 rinv20 = gmx_mm_invsqrt_pd(rsq20);
181 rinv30 = gmx_mm_invsqrt_pd(rsq30);
182
183 /* Load parameters for j particles */
184 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
185
186 fjx0 = _mm_setzero_pd();
187 fjy0 = _mm_setzero_pd();
188 fjz0 = _mm_setzero_pd();
189
190 /**************************
191 * CALCULATE INTERACTIONS *
192 **************************/
193
194 r10 = _mm_mul_pd(rsq10,rinv10);
195
196 /* Compute parameters for interactions between i and j atoms */
197 qq10 = _mm_mul_pd(iq1,jq0);
198
199 /* Calculate table index by multiplying r with table scale and truncate to integer */
200 rt = _mm_mul_pd(r10,vftabscale);
201 vfitab = _mm_cvttpd_epi32(rt);
202 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
203 vfitab = _mm_slli_epi32(vfitab,2);
204
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(qq10,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(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
218
219 /* Update potential sum for this i atom from the interaction with this j atom. */
220 velecsum = _mm_add_pd(velecsum,velec);
221
222 fscal = felec;
223
224 /* Calculate temporary vectorial force */
225 tx = _mm_mul_pd(fscal,dx10);
226 ty = _mm_mul_pd(fscal,dy10);
227 tz = _mm_mul_pd(fscal,dz10);
228
229 /* Update vectorial force */
230 fix1 = _mm_add_pd(fix1,tx);
231 fiy1 = _mm_add_pd(fiy1,ty);
232 fiz1 = _mm_add_pd(fiz1,tz);
233
234 fjx0 = _mm_add_pd(fjx0,tx);
235 fjy0 = _mm_add_pd(fjy0,ty);
236 fjz0 = _mm_add_pd(fjz0,tz);
237
238 /**************************
239 * CALCULATE INTERACTIONS *
240 **************************/
241
242 r20 = _mm_mul_pd(rsq20,rinv20);
243
244 /* Compute parameters for interactions between i and j atoms */
245 qq20 = _mm_mul_pd(iq2,jq0);
246
247 /* Calculate table index by multiplying r with table scale and truncate to integer */
248 rt = _mm_mul_pd(r20,vftabscale);
249 vfitab = _mm_cvttpd_epi32(rt);
250 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
251 vfitab = _mm_slli_epi32(vfitab,2);
252
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(qq20,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(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
266
267 /* Update potential sum for this i atom from the interaction with this j atom. */
268 velecsum = _mm_add_pd(velecsum,velec);
269
270 fscal = felec;
271
272 /* Calculate temporary vectorial force */
273 tx = _mm_mul_pd(fscal,dx20);
274 ty = _mm_mul_pd(fscal,dy20);
275 tz = _mm_mul_pd(fscal,dz20);
276
277 /* Update vectorial force */
278 fix2 = _mm_add_pd(fix2,tx);
279 fiy2 = _mm_add_pd(fiy2,ty);
280 fiz2 = _mm_add_pd(fiz2,tz);
281
282 fjx0 = _mm_add_pd(fjx0,tx);
283 fjy0 = _mm_add_pd(fjy0,ty);
284 fjz0 = _mm_add_pd(fjz0,tz);
285
286 /**************************
287 * CALCULATE INTERACTIONS *
288 **************************/
289
290 r30 = _mm_mul_pd(rsq30,rinv30);
291
292 /* Compute parameters for interactions between i and j atoms */
293 qq30 = _mm_mul_pd(iq3,jq0);
294
295 /* Calculate table index by multiplying r with table scale and truncate to integer */
296 rt = _mm_mul_pd(r30,vftabscale);
297 vfitab = _mm_cvttpd_epi32(rt);
298 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
299 vfitab = _mm_slli_epi32(vfitab,2);
300
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(qq30,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(qq30,FF),_mm_mul_pd(vftabscale,rinv30)));
314
315 /* Update potential sum for this i atom from the interaction with this j atom. */
316 velecsum = _mm_add_pd(velecsum,velec);
317
318 fscal = felec;
319
320 /* Calculate temporary vectorial force */
321 tx = _mm_mul_pd(fscal,dx30);
322 ty = _mm_mul_pd(fscal,dy30);
323 tz = _mm_mul_pd(fscal,dz30);
324
325 /* Update vectorial force */
326 fix3 = _mm_add_pd(fix3,tx);
327 fiy3 = _mm_add_pd(fiy3,ty);
328 fiz3 = _mm_add_pd(fiz3,tz);
329
330 fjx0 = _mm_add_pd(fjx0,tx);
331 fjy0 = _mm_add_pd(fjy0,ty);
332 fjz0 = _mm_add_pd(fjz0,tz);
333
334 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
335
336 /* Inner loop uses 132 flops */
337 }
338
339 if(jidx<j_index_end)
340 {
341
342 jnrA = jjnr[jidx];
343 j_coord_offsetA = DIM*jnrA;
344
345 /* load j atom coordinates */
346 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
347 &jx0,&jy0,&jz0);
348
349 /* Calculate displacement vector */
350 dx10 = _mm_sub_pd(ix1,jx0);
351 dy10 = _mm_sub_pd(iy1,jy0);
352 dz10 = _mm_sub_pd(iz1,jz0);
353 dx20 = _mm_sub_pd(ix2,jx0);
354 dy20 = _mm_sub_pd(iy2,jy0);
355 dz20 = _mm_sub_pd(iz2,jz0);
356 dx30 = _mm_sub_pd(ix3,jx0);
357 dy30 = _mm_sub_pd(iy3,jy0);
358 dz30 = _mm_sub_pd(iz3,jz0);
359
360 /* Calculate squared distance and things based on it */
361 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
362 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
363 rsq30 = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
364
365 rinv10 = gmx_mm_invsqrt_pd(rsq10);
366 rinv20 = gmx_mm_invsqrt_pd(rsq20);
367 rinv30 = gmx_mm_invsqrt_pd(rsq30);
368
369 /* Load parameters for j particles */
370 jq0 = _mm_load_sd(charge+jnrA+0);
371
372 fjx0 = _mm_setzero_pd();
373 fjy0 = _mm_setzero_pd();
374 fjz0 = _mm_setzero_pd();
375
376 /**************************
377 * CALCULATE INTERACTIONS *
378 **************************/
379
380 r10 = _mm_mul_pd(rsq10,rinv10);
381
382 /* Compute parameters for interactions between i and j atoms */
383 qq10 = _mm_mul_pd(iq1,jq0);
384
385 /* Calculate table index by multiplying r with table scale and truncate to integer */
386 rt = _mm_mul_pd(r10,vftabscale);
387 vfitab = _mm_cvttpd_epi32(rt);
388 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
389 vfitab = _mm_slli_epi32(vfitab,2);
390
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(qq10,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(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
404
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);
408
409 fscal = felec;
410
411 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
412
413 /* Calculate temporary vectorial force */
414 tx = _mm_mul_pd(fscal,dx10);
415 ty = _mm_mul_pd(fscal,dy10);
416 tz = _mm_mul_pd(fscal,dz10);
417
418 /* Update vectorial force */
419 fix1 = _mm_add_pd(fix1,tx);
420 fiy1 = _mm_add_pd(fiy1,ty);
421 fiz1 = _mm_add_pd(fiz1,tz);
422
423 fjx0 = _mm_add_pd(fjx0,tx);
424 fjy0 = _mm_add_pd(fjy0,ty);
425 fjz0 = _mm_add_pd(fjz0,tz);
426
427 /**************************
428 * CALCULATE INTERACTIONS *
429 **************************/
430
431 r20 = _mm_mul_pd(rsq20,rinv20);
432
433 /* Compute parameters for interactions between i and j atoms */
434 qq20 = _mm_mul_pd(iq2,jq0);
435
436 /* Calculate table index by multiplying r with table scale and truncate to integer */
437 rt = _mm_mul_pd(r20,vftabscale);
438 vfitab = _mm_cvttpd_epi32(rt);
439 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
440 vfitab = _mm_slli_epi32(vfitab,2);
441
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(qq20,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(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
455
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);
459
460 fscal = felec;
461
462 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
463
464 /* Calculate temporary vectorial force */
465 tx = _mm_mul_pd(fscal,dx20);
466 ty = _mm_mul_pd(fscal,dy20);
467 tz = _mm_mul_pd(fscal,dz20);
468
469 /* Update vectorial force */
470 fix2 = _mm_add_pd(fix2,tx);
471 fiy2 = _mm_add_pd(fiy2,ty);
472 fiz2 = _mm_add_pd(fiz2,tz);
473
474 fjx0 = _mm_add_pd(fjx0,tx);
475 fjy0 = _mm_add_pd(fjy0,ty);
476 fjz0 = _mm_add_pd(fjz0,tz);
477
478 /**************************
479 * CALCULATE INTERACTIONS *
480 **************************/
481
482 r30 = _mm_mul_pd(rsq30,rinv30);
483
484 /* Compute parameters for interactions between i and j atoms */
485 qq30 = _mm_mul_pd(iq3,jq0);
486
487 /* Calculate table index by multiplying r with table scale and truncate to integer */
488 rt = _mm_mul_pd(r30,vftabscale);
489 vfitab = _mm_cvttpd_epi32(rt);
490 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
491 vfitab = _mm_slli_epi32(vfitab,2);
492
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(qq30,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(qq30,FF),_mm_mul_pd(vftabscale,rinv30)));
506
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);
510
511 fscal = felec;
512
513 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
514
515 /* Calculate temporary vectorial force */
516 tx = _mm_mul_pd(fscal,dx30);
517 ty = _mm_mul_pd(fscal,dy30);
518 tz = _mm_mul_pd(fscal,dz30);
519
520 /* Update vectorial force */
521 fix3 = _mm_add_pd(fix3,tx);
522 fiy3 = _mm_add_pd(fiy3,ty);
523 fiz3 = _mm_add_pd(fiz3,tz);
524
525 fjx0 = _mm_add_pd(fjx0,tx);
526 fjy0 = _mm_add_pd(fjy0,ty);
527 fjz0 = _mm_add_pd(fjz0,tz);
528
529 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
530
531 /* Inner loop uses 132 flops */
532 }
533
534 /* End of innermost loop */
535
536 gmx_mm_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
537 f+i_coord_offset+DIM,fshift+i_shift_offset);
538
539 ggid = gid[iidx];
540 /* Update potential energies */
541 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
542
543 /* Increment number of inner iterations */
544 inneriter += j_index_end - j_index_start;
545
546 /* Outer loop uses 19 flops */
547 }
548
549 /* Increment number of outer iterations */
550 outeriter += nri;
551
552 /* Update outer/inner flops */
553
554 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_VF,outeriter*19 + inneriter*132);
555 }
556 /*
557 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW4P1_F_sse2_double
558 * Electrostatics interaction: CubicSplineTable
559 * VdW interaction: None
560 * Geometry: Water4-Particle
561 * Calculate force/pot: Force
562 */
563 void
564 nb_kernel_ElecCSTab_VdwNone_GeomW4P1_F_sse2_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)
572 {
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.
577 */
578 int i_shift_offset,i_coord_offset,outeriter,inneriter;
579 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
580 int jnrA,jnrB;
581 int j_coord_offsetA,j_coord_offsetB;
582 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
583 real rcutoff_scalar;
584 real *shiftvec,*fshift,*x,*f;
585 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
586 int vdwioffset1;
587 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
588 int vdwioffset2;
589 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
590 int vdwioffset3;
591 __m128d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
592 int vdwjidx0A,vdwjidx0B;
593 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
594 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
595 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
596 __m128d dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
597 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
598 real *charge;
599 __m128i vfitab;
600 __m128i ifour = _mm_set1_epi32(4);
601 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
602 real *vftab;
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);
607 x = xx[0];
608 f = ff[0];
609
610 nri = nlist->nri;
611 iinr = nlist->iinr;
612 jindex = nlist->jindex;
613 jjnr = nlist->jjnr;
614 shiftidx = nlist->shift;
615 gid = nlist->gid;
616 shiftvec = fr->shift_vec[0];
617 fshift = fr->fshift[0];
618 facel = _mm_set1_pd(fr->epsfac);
619 charge = mdatoms->chargeA;
620
621 vftab = kernel_data->table_elec->data;
622 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
623
624 /* Setup water-specific parameters */
625 inr = nlist->iinr[0];
626 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
627 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
628 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
629
630 /* Avoid stupid compiler warnings */
631 jnrA = jnrB = 0;
632 j_coord_offsetA = 0;
633 j_coord_offsetB = 0;
634
635 outeriter = 0;
636 inneriter = 0;
637
638 /* Start outer loop over neighborlists */
639 for(iidx=0; iidx<nri; iidx++)
640 {
641 /* Load shift vector for this list */
642 i_shift_offset = DIM*shiftidx[iidx];
643
644 /* Load limits for loop over neighbors */
645 j_index_start = jindex[iidx];
646 j_index_end = jindex[iidx+1];
647
648 /* Get outer coordinate index */
649 inr = iinr[iidx];
650 i_coord_offset = DIM*inr;
651
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+DIM,
654 &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
655
656 fix1 = _mm_setzero_pd();
657 fiy1 = _mm_setzero_pd();
658 fiz1 = _mm_setzero_pd();
659 fix2 = _mm_setzero_pd();
660 fiy2 = _mm_setzero_pd();
661 fiz2 = _mm_setzero_pd();
662 fix3 = _mm_setzero_pd();
663 fiy3 = _mm_setzero_pd();
664 fiz3 = _mm_setzero_pd();
665
666 /* Start inner kernel loop */
667 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
668 {
669
670 /* Get j neighbor index, and coordinate index */
671 jnrA = jjnr[jidx];
672 jnrB = jjnr[jidx+1];
673 j_coord_offsetA = DIM*jnrA;
674 j_coord_offsetB = DIM*jnrB;
675
676 /* load j atom coordinates */
677 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
678 &jx0,&jy0,&jz0);
679
680 /* Calculate displacement vector */
681 dx10 = _mm_sub_pd(ix1,jx0);
682 dy10 = _mm_sub_pd(iy1,jy0);
683 dz10 = _mm_sub_pd(iz1,jz0);
684 dx20 = _mm_sub_pd(ix2,jx0);
685 dy20 = _mm_sub_pd(iy2,jy0);
686 dz20 = _mm_sub_pd(iz2,jz0);
687 dx30 = _mm_sub_pd(ix3,jx0);
688 dy30 = _mm_sub_pd(iy3,jy0);
689 dz30 = _mm_sub_pd(iz3,jz0);
690
691 /* Calculate squared distance and things based on it */
692 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
693 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
694 rsq30 = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
695
696 rinv10 = gmx_mm_invsqrt_pd(rsq10);
697 rinv20 = gmx_mm_invsqrt_pd(rsq20);
698 rinv30 = gmx_mm_invsqrt_pd(rsq30);
699
700 /* Load parameters for j particles */
701 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
702
703 fjx0 = _mm_setzero_pd();
704 fjy0 = _mm_setzero_pd();
705 fjz0 = _mm_setzero_pd();
706
707 /**************************
708 * CALCULATE INTERACTIONS *
709 **************************/
710
711 r10 = _mm_mul_pd(rsq10,rinv10);
712
713 /* Compute parameters for interactions between i and j atoms */
714 qq10 = _mm_mul_pd(iq1,jq0);
715
716 /* Calculate table index by multiplying r with table scale and truncate to integer */
717 rt = _mm_mul_pd(r10,vftabscale);
718 vfitab = _mm_cvttpd_epi32(rt);
719 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
720 vfitab = _mm_slli_epi32(vfitab,2);
721
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(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
733
734 fscal = felec;
735
736 /* Calculate temporary vectorial force */
737 tx = _mm_mul_pd(fscal,dx10);
738 ty = _mm_mul_pd(fscal,dy10);
739 tz = _mm_mul_pd(fscal,dz10);
740
741 /* Update vectorial force */
742 fix1 = _mm_add_pd(fix1,tx);
743 fiy1 = _mm_add_pd(fiy1,ty);
744 fiz1 = _mm_add_pd(fiz1,tz);
745
746 fjx0 = _mm_add_pd(fjx0,tx);
747 fjy0 = _mm_add_pd(fjy0,ty);
748 fjz0 = _mm_add_pd(fjz0,tz);
749
750 /**************************
751 * CALCULATE INTERACTIONS *
752 **************************/
753
754 r20 = _mm_mul_pd(rsq20,rinv20);
755
756 /* Compute parameters for interactions between i and j atoms */
757 qq20 = _mm_mul_pd(iq2,jq0);
758
759 /* Calculate table index by multiplying r with table scale and truncate to integer */
760 rt = _mm_mul_pd(r20,vftabscale);
761 vfitab = _mm_cvttpd_epi32(rt);
762 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
763 vfitab = _mm_slli_epi32(vfitab,2);
764
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(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
776
777 fscal = felec;
778
779 /* Calculate temporary vectorial force */
780 tx = _mm_mul_pd(fscal,dx20);
781 ty = _mm_mul_pd(fscal,dy20);
782 tz = _mm_mul_pd(fscal,dz20);
783
784 /* Update vectorial force */
785 fix2 = _mm_add_pd(fix2,tx);
786 fiy2 = _mm_add_pd(fiy2,ty);
787 fiz2 = _mm_add_pd(fiz2,tz);
788
789 fjx0 = _mm_add_pd(fjx0,tx);
790 fjy0 = _mm_add_pd(fjy0,ty);
791 fjz0 = _mm_add_pd(fjz0,tz);
792
793 /**************************
794 * CALCULATE INTERACTIONS *
795 **************************/
796
797 r30 = _mm_mul_pd(rsq30,rinv30);
798
799 /* Compute parameters for interactions between i and j atoms */
800 qq30 = _mm_mul_pd(iq3,jq0);
801
802 /* Calculate table index by multiplying r with table scale and truncate to integer */
803 rt = _mm_mul_pd(r30,vftabscale);
804 vfitab = _mm_cvttpd_epi32(rt);
805 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
806 vfitab = _mm_slli_epi32(vfitab,2);
807
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(qq30,FF),_mm_mul_pd(vftabscale,rinv30)));
819
820 fscal = felec;
821
822 /* Calculate temporary vectorial force */
823 tx = _mm_mul_pd(fscal,dx30);
824 ty = _mm_mul_pd(fscal,dy30);
825 tz = _mm_mul_pd(fscal,dz30);
826
827 /* Update vectorial force */
828 fix3 = _mm_add_pd(fix3,tx);
829 fiy3 = _mm_add_pd(fiy3,ty);
830 fiz3 = _mm_add_pd(fiz3,tz);
831
832 fjx0 = _mm_add_pd(fjx0,tx);
833 fjy0 = _mm_add_pd(fjy0,ty);
834 fjz0 = _mm_add_pd(fjz0,tz);
835
836 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
837
838 /* Inner loop uses 120 flops */
839 }
840
841 if(jidx<j_index_end)
842 {
843
844 jnrA = jjnr[jidx];
845 j_coord_offsetA = DIM*jnrA;
846
847 /* load j atom coordinates */
848 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
849 &jx0,&jy0,&jz0);
850
851 /* Calculate displacement vector */
852 dx10 = _mm_sub_pd(ix1,jx0);
853 dy10 = _mm_sub_pd(iy1,jy0);
854 dz10 = _mm_sub_pd(iz1,jz0);
855 dx20 = _mm_sub_pd(ix2,jx0);
856 dy20 = _mm_sub_pd(iy2,jy0);
857 dz20 = _mm_sub_pd(iz2,jz0);
858 dx30 = _mm_sub_pd(ix3,jx0);
859 dy30 = _mm_sub_pd(iy3,jy0);
860 dz30 = _mm_sub_pd(iz3,jz0);
861
862 /* Calculate squared distance and things based on it */
863 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
864 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
865 rsq30 = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
866
867 rinv10 = gmx_mm_invsqrt_pd(rsq10);
868 rinv20 = gmx_mm_invsqrt_pd(rsq20);
869 rinv30 = gmx_mm_invsqrt_pd(rsq30);
870
871 /* Load parameters for j particles */
872 jq0 = _mm_load_sd(charge+jnrA+0);
873
874 fjx0 = _mm_setzero_pd();
875 fjy0 = _mm_setzero_pd();
876 fjz0 = _mm_setzero_pd();
877
878 /**************************
879 * CALCULATE INTERACTIONS *
880 **************************/
881
882 r10 = _mm_mul_pd(rsq10,rinv10);
883
884 /* Compute parameters for interactions between i and j atoms */
885 qq10 = _mm_mul_pd(iq1,jq0);
886
887 /* Calculate table index by multiplying r with table scale and truncate to integer */
888 rt = _mm_mul_pd(r10,vftabscale);
889 vfitab = _mm_cvttpd_epi32(rt);
890 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
891 vfitab = _mm_slli_epi32(vfitab,2);
892
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(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
904
905 fscal = felec;
906
907 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
908
909 /* Calculate temporary vectorial force */
910 tx = _mm_mul_pd(fscal,dx10);
911 ty = _mm_mul_pd(fscal,dy10);
912 tz = _mm_mul_pd(fscal,dz10);
913
914 /* Update vectorial force */
915 fix1 = _mm_add_pd(fix1,tx);
916 fiy1 = _mm_add_pd(fiy1,ty);
917 fiz1 = _mm_add_pd(fiz1,tz);
918
919 fjx0 = _mm_add_pd(fjx0,tx);
920 fjy0 = _mm_add_pd(fjy0,ty);
921 fjz0 = _mm_add_pd(fjz0,tz);
922
923 /**************************
924 * CALCULATE INTERACTIONS *
925 **************************/
926
927 r20 = _mm_mul_pd(rsq20,rinv20);
928
929 /* Compute parameters for interactions between i and j atoms */
930 qq20 = _mm_mul_pd(iq2,jq0);
931
932 /* Calculate table index by multiplying r with table scale and truncate to integer */
933 rt = _mm_mul_pd(r20,vftabscale);
934 vfitab = _mm_cvttpd_epi32(rt);
935 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
936 vfitab = _mm_slli_epi32(vfitab,2);
937
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(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
949
950 fscal = felec;
951
952 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
953
954 /* Calculate temporary vectorial force */
955 tx = _mm_mul_pd(fscal,dx20);
956 ty = _mm_mul_pd(fscal,dy20);
957 tz = _mm_mul_pd(fscal,dz20);
958
959 /* Update vectorial force */
960 fix2 = _mm_add_pd(fix2,tx);
961 fiy2 = _mm_add_pd(fiy2,ty);
962 fiz2 = _mm_add_pd(fiz2,tz);
963
964 fjx0 = _mm_add_pd(fjx0,tx);
965 fjy0 = _mm_add_pd(fjy0,ty);
966 fjz0 = _mm_add_pd(fjz0,tz);
967
968 /**************************
969 * CALCULATE INTERACTIONS *
970 **************************/
971
972 r30 = _mm_mul_pd(rsq30,rinv30);
973
974 /* Compute parameters for interactions between i and j atoms */
975 qq30 = _mm_mul_pd(iq3,jq0);
976
977 /* Calculate table index by multiplying r with table scale and truncate to integer */
978 rt = _mm_mul_pd(r30,vftabscale);
979 vfitab = _mm_cvttpd_epi32(rt);
980 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
981 vfitab = _mm_slli_epi32(vfitab,2);
982
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(qq30,FF),_mm_mul_pd(vftabscale,rinv30)));
994
995 fscal = felec;
996
997 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
998
999 /* Calculate temporary vectorial force */
1000 tx = _mm_mul_pd(fscal,dx30);
1001 ty = _mm_mul_pd(fscal,dy30);
1002 tz = _mm_mul_pd(fscal,dz30);
1003
1004 /* Update vectorial force */
1005 fix3 = _mm_add_pd(fix3,tx);
1006 fiy3 = _mm_add_pd(fiy3,ty);
1007 fiz3 = _mm_add_pd(fiz3,tz);
1008
1009 fjx0 = _mm_add_pd(fjx0,tx);
1010 fjy0 = _mm_add_pd(fjy0,ty);
1011 fjz0 = _mm_add_pd(fjz0,tz);
1012
1013 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1014
1015 /* Inner loop uses 120 flops */
1016 }
1017
1018 /* End of innermost loop */
1019
1020 gmx_mm_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1021 f+i_coord_offset+DIM,fshift+i_shift_offset);
1022
1023 /* Increment number of inner iterations */
1024 inneriter += j_index_end - j_index_start;
1025
1026 /* Outer loop uses 18 flops */
1027 }
1028
1029 /* Increment number of outer iterations */
1030 outeriter += nri;
1031
1032 /* Update outer/inner flops */
1033
1034 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_F,outeriter*18 + inneriter*120);
1035 }
The ultimate molecular dynamics simulation package