?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Removed simple.h from nb_kernel_sse2_XX
[gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_sse2_double / nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_sse2_double.c
blobcde6ad370fe937af6ad6a45429c0f7e4f1c36905
1 /*
2 * This file is part of the GROMACS molecular simulation package.
3 *
4 * Copyright (c) 2012,2013,2014,2015, 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 sse2_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/math/vec.h"
46 #include "gromacs/legacyheaders/nrnb.h"
47
48 #include "gromacs/simd/math_x86_sse2_double.h"
49 #include "kernelutil_x86_sse2_double.h"
50
51 /*
52 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_VF_sse2_double
53 * Electrostatics interaction: CubicSplineTable
54 * VdW interaction: LennardJones
55 * Geometry: Water3-Particle
56 * Calculate force/pot: PotentialAndForce
57 */
58 void
59 nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_VF_sse2_double
60 (t_nblist * gmx_restrict nlist,
61 rvec * gmx_restrict xx,
62 rvec * gmx_restrict ff,
63 t_forcerec * gmx_restrict fr,
64 t_mdatoms * gmx_restrict mdatoms,
65 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
66 t_nrnb * gmx_restrict nrnb)
67 {
68 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
69 * just 0 for non-waters.
70 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
71 * jnr indices corresponding to data put in the four positions in the SIMD register.
72 */
73 int i_shift_offset,i_coord_offset,outeriter,inneriter;
74 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
75 int jnrA,jnrB;
76 int j_coord_offsetA,j_coord_offsetB;
77 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
78 real rcutoff_scalar;
79 real *shiftvec,*fshift,*x,*f;
80 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
81 int vdwioffset0;
82 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
83 int vdwioffset1;
84 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
85 int vdwioffset2;
86 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
87 int vdwjidx0A,vdwjidx0B;
88 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
89 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
90 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
91 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
92 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
93 real *charge;
94 int nvdwtype;
95 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
96 int *vdwtype;
97 real *vdwparam;
98 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
99 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
100 __m128i vfitab;
101 __m128i ifour = _mm_set1_epi32(4);
102 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
103 real *vftab;
104 __m128d dummy_mask,cutoff_mask;
105 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
106 __m128d one = _mm_set1_pd(1.0);
107 __m128d two = _mm_set1_pd(2.0);
108 x = xx[0];
109 f = ff[0];
110
111 nri = nlist->nri;
112 iinr = nlist->iinr;
113 jindex = nlist->jindex;
114 jjnr = nlist->jjnr;
115 shiftidx = nlist->shift;
116 gid = nlist->gid;
117 shiftvec = fr->shift_vec[0];
118 fshift = fr->fshift[0];
119 facel = _mm_set1_pd(fr->epsfac);
120 charge = mdatoms->chargeA;
121 nvdwtype = fr->ntype;
122 vdwparam = fr->nbfp;
123 vdwtype = mdatoms->typeA;
124
125 vftab = kernel_data->table_elec->data;
126 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
127
128 /* Setup water-specific parameters */
129 inr = nlist->iinr[0];
130 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
131 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
132 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
133 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
134
135 /* Avoid stupid compiler warnings */
136 jnrA = jnrB = 0;
137 j_coord_offsetA = 0;
138 j_coord_offsetB = 0;
139
140 outeriter = 0;
141 inneriter = 0;
142
143 /* Start outer loop over neighborlists */
144 for(iidx=0; iidx<nri; iidx++)
145 {
146 /* Load shift vector for this list */
147 i_shift_offset = DIM*shiftidx[iidx];
148
149 /* Load limits for loop over neighbors */
150 j_index_start = jindex[iidx];
151 j_index_end = jindex[iidx+1];
152
153 /* Get outer coordinate index */
154 inr = iinr[iidx];
155 i_coord_offset = DIM*inr;
156
157 /* Load i particle coords and add shift vector */
158 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
159 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
160
161 fix0 = _mm_setzero_pd();
162 fiy0 = _mm_setzero_pd();
163 fiz0 = _mm_setzero_pd();
164 fix1 = _mm_setzero_pd();
165 fiy1 = _mm_setzero_pd();
166 fiz1 = _mm_setzero_pd();
167 fix2 = _mm_setzero_pd();
168 fiy2 = _mm_setzero_pd();
169 fiz2 = _mm_setzero_pd();
170
171 /* Reset potential sums */
172 velecsum = _mm_setzero_pd();
173 vvdwsum = _mm_setzero_pd();
174
175 /* Start inner kernel loop */
176 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
177 {
178
179 /* Get j neighbor index, and coordinate index */
180 jnrA = jjnr[jidx];
181 jnrB = jjnr[jidx+1];
182 j_coord_offsetA = DIM*jnrA;
183 j_coord_offsetB = DIM*jnrB;
184
185 /* load j atom coordinates */
186 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
187 &jx0,&jy0,&jz0);
188
189 /* Calculate displacement vector */
190 dx00 = _mm_sub_pd(ix0,jx0);
191 dy00 = _mm_sub_pd(iy0,jy0);
192 dz00 = _mm_sub_pd(iz0,jz0);
193 dx10 = _mm_sub_pd(ix1,jx0);
194 dy10 = _mm_sub_pd(iy1,jy0);
195 dz10 = _mm_sub_pd(iz1,jz0);
196 dx20 = _mm_sub_pd(ix2,jx0);
197 dy20 = _mm_sub_pd(iy2,jy0);
198 dz20 = _mm_sub_pd(iz2,jz0);
199
200 /* Calculate squared distance and things based on it */
201 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
202 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
203 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
204
205 rinv00 = gmx_mm_invsqrt_pd(rsq00);
206 rinv10 = gmx_mm_invsqrt_pd(rsq10);
207 rinv20 = gmx_mm_invsqrt_pd(rsq20);
208
209 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
210
211 /* Load parameters for j particles */
212 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
213 vdwjidx0A = 2*vdwtype[jnrA+0];
214 vdwjidx0B = 2*vdwtype[jnrB+0];
215
216 fjx0 = _mm_setzero_pd();
217 fjy0 = _mm_setzero_pd();
218 fjz0 = _mm_setzero_pd();
219
220 /**************************
221 * CALCULATE INTERACTIONS *
222 **************************/
223
224 r00 = _mm_mul_pd(rsq00,rinv00);
225
226 /* Compute parameters for interactions between i and j atoms */
227 qq00 = _mm_mul_pd(iq0,jq0);
228 gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
229 vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
230
231 /* Calculate table index by multiplying r with table scale and truncate to integer */
232 rt = _mm_mul_pd(r00,vftabscale);
233 vfitab = _mm_cvttpd_epi32(rt);
234 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
235 vfitab = _mm_slli_epi32(vfitab,2);
236
237 /* CUBIC SPLINE TABLE ELECTROSTATICS */
238 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
239 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
240 GMX_MM_TRANSPOSE2_PD(Y,F);
241 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
242 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
243 GMX_MM_TRANSPOSE2_PD(G,H);
244 Heps = _mm_mul_pd(vfeps,H);
245 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
246 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
247 velec = _mm_mul_pd(qq00,VV);
248 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
249 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
250
251 /* LENNARD-JONES DISPERSION/REPULSION */
252
253 rinvsix = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
254 vvdw6 = _mm_mul_pd(c6_00,rinvsix);
255 vvdw12 = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
256 vvdw = _mm_sub_pd( _mm_mul_pd(vvdw12,one_twelfth) , _mm_mul_pd(vvdw6,one_sixth) );
257 fvdw = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
258
259 /* Update potential sum for this i atom from the interaction with this j atom. */
260 velecsum = _mm_add_pd(velecsum,velec);
261 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
262
263 fscal = _mm_add_pd(felec,fvdw);
264
265 /* Calculate temporary vectorial force */
266 tx = _mm_mul_pd(fscal,dx00);
267 ty = _mm_mul_pd(fscal,dy00);
268 tz = _mm_mul_pd(fscal,dz00);
269
270 /* Update vectorial force */
271 fix0 = _mm_add_pd(fix0,tx);
272 fiy0 = _mm_add_pd(fiy0,ty);
273 fiz0 = _mm_add_pd(fiz0,tz);
274
275 fjx0 = _mm_add_pd(fjx0,tx);
276 fjy0 = _mm_add_pd(fjy0,ty);
277 fjz0 = _mm_add_pd(fjz0,tz);
278
279 /**************************
280 * CALCULATE INTERACTIONS *
281 **************************/
282
283 r10 = _mm_mul_pd(rsq10,rinv10);
284
285 /* Compute parameters for interactions between i and j atoms */
286 qq10 = _mm_mul_pd(iq1,jq0);
287
288 /* Calculate table index by multiplying r with table scale and truncate to integer */
289 rt = _mm_mul_pd(r10,vftabscale);
290 vfitab = _mm_cvttpd_epi32(rt);
291 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
292 vfitab = _mm_slli_epi32(vfitab,2);
293
294 /* CUBIC SPLINE TABLE ELECTROSTATICS */
295 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
296 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
297 GMX_MM_TRANSPOSE2_PD(Y,F);
298 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
299 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
300 GMX_MM_TRANSPOSE2_PD(G,H);
301 Heps = _mm_mul_pd(vfeps,H);
302 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
303 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
304 velec = _mm_mul_pd(qq10,VV);
305 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
306 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
307
308 /* Update potential sum for this i atom from the interaction with this j atom. */
309 velecsum = _mm_add_pd(velecsum,velec);
310
311 fscal = felec;
312
313 /* Calculate temporary vectorial force */
314 tx = _mm_mul_pd(fscal,dx10);
315 ty = _mm_mul_pd(fscal,dy10);
316 tz = _mm_mul_pd(fscal,dz10);
317
318 /* Update vectorial force */
319 fix1 = _mm_add_pd(fix1,tx);
320 fiy1 = _mm_add_pd(fiy1,ty);
321 fiz1 = _mm_add_pd(fiz1,tz);
322
323 fjx0 = _mm_add_pd(fjx0,tx);
324 fjy0 = _mm_add_pd(fjy0,ty);
325 fjz0 = _mm_add_pd(fjz0,tz);
326
327 /**************************
328 * CALCULATE INTERACTIONS *
329 **************************/
330
331 r20 = _mm_mul_pd(rsq20,rinv20);
332
333 /* Compute parameters for interactions between i and j atoms */
334 qq20 = _mm_mul_pd(iq2,jq0);
335
336 /* Calculate table index by multiplying r with table scale and truncate to integer */
337 rt = _mm_mul_pd(r20,vftabscale);
338 vfitab = _mm_cvttpd_epi32(rt);
339 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
340 vfitab = _mm_slli_epi32(vfitab,2);
341
342 /* CUBIC SPLINE TABLE ELECTROSTATICS */
343 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
344 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
345 GMX_MM_TRANSPOSE2_PD(Y,F);
346 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
347 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
348 GMX_MM_TRANSPOSE2_PD(G,H);
349 Heps = _mm_mul_pd(vfeps,H);
350 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
351 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
352 velec = _mm_mul_pd(qq20,VV);
353 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
354 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
355
356 /* Update potential sum for this i atom from the interaction with this j atom. */
357 velecsum = _mm_add_pd(velecsum,velec);
358
359 fscal = felec;
360
361 /* Calculate temporary vectorial force */
362 tx = _mm_mul_pd(fscal,dx20);
363 ty = _mm_mul_pd(fscal,dy20);
364 tz = _mm_mul_pd(fscal,dz20);
365
366 /* Update vectorial force */
367 fix2 = _mm_add_pd(fix2,tx);
368 fiy2 = _mm_add_pd(fiy2,ty);
369 fiz2 = _mm_add_pd(fiz2,tz);
370
371 fjx0 = _mm_add_pd(fjx0,tx);
372 fjy0 = _mm_add_pd(fjy0,ty);
373 fjz0 = _mm_add_pd(fjz0,tz);
374
375 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
376
377 /* Inner loop uses 145 flops */
378 }
379
380 if(jidx<j_index_end)
381 {
382
383 jnrA = jjnr[jidx];
384 j_coord_offsetA = DIM*jnrA;
385
386 /* load j atom coordinates */
387 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
388 &jx0,&jy0,&jz0);
389
390 /* Calculate displacement vector */
391 dx00 = _mm_sub_pd(ix0,jx0);
392 dy00 = _mm_sub_pd(iy0,jy0);
393 dz00 = _mm_sub_pd(iz0,jz0);
394 dx10 = _mm_sub_pd(ix1,jx0);
395 dy10 = _mm_sub_pd(iy1,jy0);
396 dz10 = _mm_sub_pd(iz1,jz0);
397 dx20 = _mm_sub_pd(ix2,jx0);
398 dy20 = _mm_sub_pd(iy2,jy0);
399 dz20 = _mm_sub_pd(iz2,jz0);
400
401 /* Calculate squared distance and things based on it */
402 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
403 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
404 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
405
406 rinv00 = gmx_mm_invsqrt_pd(rsq00);
407 rinv10 = gmx_mm_invsqrt_pd(rsq10);
408 rinv20 = gmx_mm_invsqrt_pd(rsq20);
409
410 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
411
412 /* Load parameters for j particles */
413 jq0 = _mm_load_sd(charge+jnrA+0);
414 vdwjidx0A = 2*vdwtype[jnrA+0];
415
416 fjx0 = _mm_setzero_pd();
417 fjy0 = _mm_setzero_pd();
418 fjz0 = _mm_setzero_pd();
419
420 /**************************
421 * CALCULATE INTERACTIONS *
422 **************************/
423
424 r00 = _mm_mul_pd(rsq00,rinv00);
425
426 /* Compute parameters for interactions between i and j atoms */
427 qq00 = _mm_mul_pd(iq0,jq0);
428 gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
429
430 /* Calculate table index by multiplying r with table scale and truncate to integer */
431 rt = _mm_mul_pd(r00,vftabscale);
432 vfitab = _mm_cvttpd_epi32(rt);
433 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
434 vfitab = _mm_slli_epi32(vfitab,2);
435
436 /* CUBIC SPLINE TABLE ELECTROSTATICS */
437 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
438 F = _mm_setzero_pd();
439 GMX_MM_TRANSPOSE2_PD(Y,F);
440 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
441 H = _mm_setzero_pd();
442 GMX_MM_TRANSPOSE2_PD(G,H);
443 Heps = _mm_mul_pd(vfeps,H);
444 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
445 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
446 velec = _mm_mul_pd(qq00,VV);
447 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
448 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
449
450 /* LENNARD-JONES DISPERSION/REPULSION */
451
452 rinvsix = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
453 vvdw6 = _mm_mul_pd(c6_00,rinvsix);
454 vvdw12 = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
455 vvdw = _mm_sub_pd( _mm_mul_pd(vvdw12,one_twelfth) , _mm_mul_pd(vvdw6,one_sixth) );
456 fvdw = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
457
458 /* Update potential sum for this i atom from the interaction with this j atom. */
459 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
460 velecsum = _mm_add_pd(velecsum,velec);
461 vvdw = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
462 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
463
464 fscal = _mm_add_pd(felec,fvdw);
465
466 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
467
468 /* Calculate temporary vectorial force */
469 tx = _mm_mul_pd(fscal,dx00);
470 ty = _mm_mul_pd(fscal,dy00);
471 tz = _mm_mul_pd(fscal,dz00);
472
473 /* Update vectorial force */
474 fix0 = _mm_add_pd(fix0,tx);
475 fiy0 = _mm_add_pd(fiy0,ty);
476 fiz0 = _mm_add_pd(fiz0,tz);
477
478 fjx0 = _mm_add_pd(fjx0,tx);
479 fjy0 = _mm_add_pd(fjy0,ty);
480 fjz0 = _mm_add_pd(fjz0,tz);
481
482 /**************************
483 * CALCULATE INTERACTIONS *
484 **************************/
485
486 r10 = _mm_mul_pd(rsq10,rinv10);
487
488 /* Compute parameters for interactions between i and j atoms */
489 qq10 = _mm_mul_pd(iq1,jq0);
490
491 /* Calculate table index by multiplying r with table scale and truncate to integer */
492 rt = _mm_mul_pd(r10,vftabscale);
493 vfitab = _mm_cvttpd_epi32(rt);
494 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
495 vfitab = _mm_slli_epi32(vfitab,2);
496
497 /* CUBIC SPLINE TABLE ELECTROSTATICS */
498 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
499 F = _mm_setzero_pd();
500 GMX_MM_TRANSPOSE2_PD(Y,F);
501 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
502 H = _mm_setzero_pd();
503 GMX_MM_TRANSPOSE2_PD(G,H);
504 Heps = _mm_mul_pd(vfeps,H);
505 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
506 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
507 velec = _mm_mul_pd(qq10,VV);
508 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
509 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
510
511 /* Update potential sum for this i atom from the interaction with this j atom. */
512 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
513 velecsum = _mm_add_pd(velecsum,velec);
514
515 fscal = felec;
516
517 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
518
519 /* Calculate temporary vectorial force */
520 tx = _mm_mul_pd(fscal,dx10);
521 ty = _mm_mul_pd(fscal,dy10);
522 tz = _mm_mul_pd(fscal,dz10);
523
524 /* Update vectorial force */
525 fix1 = _mm_add_pd(fix1,tx);
526 fiy1 = _mm_add_pd(fiy1,ty);
527 fiz1 = _mm_add_pd(fiz1,tz);
528
529 fjx0 = _mm_add_pd(fjx0,tx);
530 fjy0 = _mm_add_pd(fjy0,ty);
531 fjz0 = _mm_add_pd(fjz0,tz);
532
533 /**************************
534 * CALCULATE INTERACTIONS *
535 **************************/
536
537 r20 = _mm_mul_pd(rsq20,rinv20);
538
539 /* Compute parameters for interactions between i and j atoms */
540 qq20 = _mm_mul_pd(iq2,jq0);
541
542 /* Calculate table index by multiplying r with table scale and truncate to integer */
543 rt = _mm_mul_pd(r20,vftabscale);
544 vfitab = _mm_cvttpd_epi32(rt);
545 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
546 vfitab = _mm_slli_epi32(vfitab,2);
547
548 /* CUBIC SPLINE TABLE ELECTROSTATICS */
549 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
550 F = _mm_setzero_pd();
551 GMX_MM_TRANSPOSE2_PD(Y,F);
552 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
553 H = _mm_setzero_pd();
554 GMX_MM_TRANSPOSE2_PD(G,H);
555 Heps = _mm_mul_pd(vfeps,H);
556 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
557 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
558 velec = _mm_mul_pd(qq20,VV);
559 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
560 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
561
562 /* Update potential sum for this i atom from the interaction with this j atom. */
563 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
564 velecsum = _mm_add_pd(velecsum,velec);
565
566 fscal = felec;
567
568 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
569
570 /* Calculate temporary vectorial force */
571 tx = _mm_mul_pd(fscal,dx20);
572 ty = _mm_mul_pd(fscal,dy20);
573 tz = _mm_mul_pd(fscal,dz20);
574
575 /* Update vectorial force */
576 fix2 = _mm_add_pd(fix2,tx);
577 fiy2 = _mm_add_pd(fiy2,ty);
578 fiz2 = _mm_add_pd(fiz2,tz);
579
580 fjx0 = _mm_add_pd(fjx0,tx);
581 fjy0 = _mm_add_pd(fjy0,ty);
582 fjz0 = _mm_add_pd(fjz0,tz);
583
584 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
585
586 /* Inner loop uses 145 flops */
587 }
588
589 /* End of innermost loop */
590
591 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
592 f+i_coord_offset,fshift+i_shift_offset);
593
594 ggid = gid[iidx];
595 /* Update potential energies */
596 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
597 gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
598
599 /* Increment number of inner iterations */
600 inneriter += j_index_end - j_index_start;
601
602 /* Outer loop uses 20 flops */
603 }
604
605 /* Increment number of outer iterations */
606 outeriter += nri;
607
608 /* Update outer/inner flops */
609
610 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*20 + inneriter*145);
611 }
612 /*
613 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_sse2_double
614 * Electrostatics interaction: CubicSplineTable
615 * VdW interaction: LennardJones
616 * Geometry: Water3-Particle
617 * Calculate force/pot: Force
618 */
619 void
620 nb_kernel_ElecCSTab_VdwLJ_GeomW3P1_F_sse2_double
621 (t_nblist * gmx_restrict nlist,
622 rvec * gmx_restrict xx,
623 rvec * gmx_restrict ff,
624 t_forcerec * gmx_restrict fr,
625 t_mdatoms * gmx_restrict mdatoms,
626 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
627 t_nrnb * gmx_restrict nrnb)
628 {
629 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
630 * just 0 for non-waters.
631 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
632 * jnr indices corresponding to data put in the four positions in the SIMD register.
633 */
634 int i_shift_offset,i_coord_offset,outeriter,inneriter;
635 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
636 int jnrA,jnrB;
637 int j_coord_offsetA,j_coord_offsetB;
638 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
639 real rcutoff_scalar;
640 real *shiftvec,*fshift,*x,*f;
641 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
642 int vdwioffset0;
643 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
644 int vdwioffset1;
645 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
646 int vdwioffset2;
647 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
648 int vdwjidx0A,vdwjidx0B;
649 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
650 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
651 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
652 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
653 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
654 real *charge;
655 int nvdwtype;
656 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
657 int *vdwtype;
658 real *vdwparam;
659 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
660 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
661 __m128i vfitab;
662 __m128i ifour = _mm_set1_epi32(4);
663 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
664 real *vftab;
665 __m128d dummy_mask,cutoff_mask;
666 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
667 __m128d one = _mm_set1_pd(1.0);
668 __m128d two = _mm_set1_pd(2.0);
669 x = xx[0];
670 f = ff[0];
671
672 nri = nlist->nri;
673 iinr = nlist->iinr;
674 jindex = nlist->jindex;
675 jjnr = nlist->jjnr;
676 shiftidx = nlist->shift;
677 gid = nlist->gid;
678 shiftvec = fr->shift_vec[0];
679 fshift = fr->fshift[0];
680 facel = _mm_set1_pd(fr->epsfac);
681 charge = mdatoms->chargeA;
682 nvdwtype = fr->ntype;
683 vdwparam = fr->nbfp;
684 vdwtype = mdatoms->typeA;
685
686 vftab = kernel_data->table_elec->data;
687 vftabscale = _mm_set1_pd(kernel_data->table_elec->scale);
688
689 /* Setup water-specific parameters */
690 inr = nlist->iinr[0];
691 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
692 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
693 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
694 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
695
696 /* Avoid stupid compiler warnings */
697 jnrA = jnrB = 0;
698 j_coord_offsetA = 0;
699 j_coord_offsetB = 0;
700
701 outeriter = 0;
702 inneriter = 0;
703
704 /* Start outer loop over neighborlists */
705 for(iidx=0; iidx<nri; iidx++)
706 {
707 /* Load shift vector for this list */
708 i_shift_offset = DIM*shiftidx[iidx];
709
710 /* Load limits for loop over neighbors */
711 j_index_start = jindex[iidx];
712 j_index_end = jindex[iidx+1];
713
714 /* Get outer coordinate index */
715 inr = iinr[iidx];
716 i_coord_offset = DIM*inr;
717
718 /* Load i particle coords and add shift vector */
719 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
720 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
721
722 fix0 = _mm_setzero_pd();
723 fiy0 = _mm_setzero_pd();
724 fiz0 = _mm_setzero_pd();
725 fix1 = _mm_setzero_pd();
726 fiy1 = _mm_setzero_pd();
727 fiz1 = _mm_setzero_pd();
728 fix2 = _mm_setzero_pd();
729 fiy2 = _mm_setzero_pd();
730 fiz2 = _mm_setzero_pd();
731
732 /* Start inner kernel loop */
733 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
734 {
735
736 /* Get j neighbor index, and coordinate index */
737 jnrA = jjnr[jidx];
738 jnrB = jjnr[jidx+1];
739 j_coord_offsetA = DIM*jnrA;
740 j_coord_offsetB = DIM*jnrB;
741
742 /* load j atom coordinates */
743 gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
744 &jx0,&jy0,&jz0);
745
746 /* Calculate displacement vector */
747 dx00 = _mm_sub_pd(ix0,jx0);
748 dy00 = _mm_sub_pd(iy0,jy0);
749 dz00 = _mm_sub_pd(iz0,jz0);
750 dx10 = _mm_sub_pd(ix1,jx0);
751 dy10 = _mm_sub_pd(iy1,jy0);
752 dz10 = _mm_sub_pd(iz1,jz0);
753 dx20 = _mm_sub_pd(ix2,jx0);
754 dy20 = _mm_sub_pd(iy2,jy0);
755 dz20 = _mm_sub_pd(iz2,jz0);
756
757 /* Calculate squared distance and things based on it */
758 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
759 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
760 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
761
762 rinv00 = gmx_mm_invsqrt_pd(rsq00);
763 rinv10 = gmx_mm_invsqrt_pd(rsq10);
764 rinv20 = gmx_mm_invsqrt_pd(rsq20);
765
766 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
767
768 /* Load parameters for j particles */
769 jq0 = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
770 vdwjidx0A = 2*vdwtype[jnrA+0];
771 vdwjidx0B = 2*vdwtype[jnrB+0];
772
773 fjx0 = _mm_setzero_pd();
774 fjy0 = _mm_setzero_pd();
775 fjz0 = _mm_setzero_pd();
776
777 /**************************
778 * CALCULATE INTERACTIONS *
779 **************************/
780
781 r00 = _mm_mul_pd(rsq00,rinv00);
782
783 /* Compute parameters for interactions between i and j atoms */
784 qq00 = _mm_mul_pd(iq0,jq0);
785 gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
786 vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
787
788 /* Calculate table index by multiplying r with table scale and truncate to integer */
789 rt = _mm_mul_pd(r00,vftabscale);
790 vfitab = _mm_cvttpd_epi32(rt);
791 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
792 vfitab = _mm_slli_epi32(vfitab,2);
793
794 /* CUBIC SPLINE TABLE ELECTROSTATICS */
795 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
796 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
797 GMX_MM_TRANSPOSE2_PD(Y,F);
798 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
799 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
800 GMX_MM_TRANSPOSE2_PD(G,H);
801 Heps = _mm_mul_pd(vfeps,H);
802 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
803 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
804 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
805
806 /* LENNARD-JONES DISPERSION/REPULSION */
807
808 rinvsix = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
809 fvdw = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(c12_00,rinvsix),c6_00),_mm_mul_pd(rinvsix,rinvsq00));
810
811 fscal = _mm_add_pd(felec,fvdw);
812
813 /* Calculate temporary vectorial force */
814 tx = _mm_mul_pd(fscal,dx00);
815 ty = _mm_mul_pd(fscal,dy00);
816 tz = _mm_mul_pd(fscal,dz00);
817
818 /* Update vectorial force */
819 fix0 = _mm_add_pd(fix0,tx);
820 fiy0 = _mm_add_pd(fiy0,ty);
821 fiz0 = _mm_add_pd(fiz0,tz);
822
823 fjx0 = _mm_add_pd(fjx0,tx);
824 fjy0 = _mm_add_pd(fjy0,ty);
825 fjz0 = _mm_add_pd(fjz0,tz);
826
827 /**************************
828 * CALCULATE INTERACTIONS *
829 **************************/
830
831 r10 = _mm_mul_pd(rsq10,rinv10);
832
833 /* Compute parameters for interactions between i and j atoms */
834 qq10 = _mm_mul_pd(iq1,jq0);
835
836 /* Calculate table index by multiplying r with table scale and truncate to integer */
837 rt = _mm_mul_pd(r10,vftabscale);
838 vfitab = _mm_cvttpd_epi32(rt);
839 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
840 vfitab = _mm_slli_epi32(vfitab,2);
841
842 /* CUBIC SPLINE TABLE ELECTROSTATICS */
843 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
844 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
845 GMX_MM_TRANSPOSE2_PD(Y,F);
846 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
847 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
848 GMX_MM_TRANSPOSE2_PD(G,H);
849 Heps = _mm_mul_pd(vfeps,H);
850 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
851 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
852 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
853
854 fscal = felec;
855
856 /* Calculate temporary vectorial force */
857 tx = _mm_mul_pd(fscal,dx10);
858 ty = _mm_mul_pd(fscal,dy10);
859 tz = _mm_mul_pd(fscal,dz10);
860
861 /* Update vectorial force */
862 fix1 = _mm_add_pd(fix1,tx);
863 fiy1 = _mm_add_pd(fiy1,ty);
864 fiz1 = _mm_add_pd(fiz1,tz);
865
866 fjx0 = _mm_add_pd(fjx0,tx);
867 fjy0 = _mm_add_pd(fjy0,ty);
868 fjz0 = _mm_add_pd(fjz0,tz);
869
870 /**************************
871 * CALCULATE INTERACTIONS *
872 **************************/
873
874 r20 = _mm_mul_pd(rsq20,rinv20);
875
876 /* Compute parameters for interactions between i and j atoms */
877 qq20 = _mm_mul_pd(iq2,jq0);
878
879 /* Calculate table index by multiplying r with table scale and truncate to integer */
880 rt = _mm_mul_pd(r20,vftabscale);
881 vfitab = _mm_cvttpd_epi32(rt);
882 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
883 vfitab = _mm_slli_epi32(vfitab,2);
884
885 /* CUBIC SPLINE TABLE ELECTROSTATICS */
886 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
887 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
888 GMX_MM_TRANSPOSE2_PD(Y,F);
889 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
890 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
891 GMX_MM_TRANSPOSE2_PD(G,H);
892 Heps = _mm_mul_pd(vfeps,H);
893 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
894 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
895 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
896
897 fscal = felec;
898
899 /* Calculate temporary vectorial force */
900 tx = _mm_mul_pd(fscal,dx20);
901 ty = _mm_mul_pd(fscal,dy20);
902 tz = _mm_mul_pd(fscal,dz20);
903
904 /* Update vectorial force */
905 fix2 = _mm_add_pd(fix2,tx);
906 fiy2 = _mm_add_pd(fiy2,ty);
907 fiz2 = _mm_add_pd(fiz2,tz);
908
909 fjx0 = _mm_add_pd(fjx0,tx);
910 fjy0 = _mm_add_pd(fjy0,ty);
911 fjz0 = _mm_add_pd(fjz0,tz);
912
913 gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
914
915 /* Inner loop uses 128 flops */
916 }
917
918 if(jidx<j_index_end)
919 {
920
921 jnrA = jjnr[jidx];
922 j_coord_offsetA = DIM*jnrA;
923
924 /* load j atom coordinates */
925 gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
926 &jx0,&jy0,&jz0);
927
928 /* Calculate displacement vector */
929 dx00 = _mm_sub_pd(ix0,jx0);
930 dy00 = _mm_sub_pd(iy0,jy0);
931 dz00 = _mm_sub_pd(iz0,jz0);
932 dx10 = _mm_sub_pd(ix1,jx0);
933 dy10 = _mm_sub_pd(iy1,jy0);
934 dz10 = _mm_sub_pd(iz1,jz0);
935 dx20 = _mm_sub_pd(ix2,jx0);
936 dy20 = _mm_sub_pd(iy2,jy0);
937 dz20 = _mm_sub_pd(iz2,jz0);
938
939 /* Calculate squared distance and things based on it */
940 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
941 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
942 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
943
944 rinv00 = gmx_mm_invsqrt_pd(rsq00);
945 rinv10 = gmx_mm_invsqrt_pd(rsq10);
946 rinv20 = gmx_mm_invsqrt_pd(rsq20);
947
948 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
949
950 /* Load parameters for j particles */
951 jq0 = _mm_load_sd(charge+jnrA+0);
952 vdwjidx0A = 2*vdwtype[jnrA+0];
953
954 fjx0 = _mm_setzero_pd();
955 fjy0 = _mm_setzero_pd();
956 fjz0 = _mm_setzero_pd();
957
958 /**************************
959 * CALCULATE INTERACTIONS *
960 **************************/
961
962 r00 = _mm_mul_pd(rsq00,rinv00);
963
964 /* Compute parameters for interactions between i and j atoms */
965 qq00 = _mm_mul_pd(iq0,jq0);
966 gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
967
968 /* Calculate table index by multiplying r with table scale and truncate to integer */
969 rt = _mm_mul_pd(r00,vftabscale);
970 vfitab = _mm_cvttpd_epi32(rt);
971 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
972 vfitab = _mm_slli_epi32(vfitab,2);
973
974 /* CUBIC SPLINE TABLE ELECTROSTATICS */
975 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
976 F = _mm_setzero_pd();
977 GMX_MM_TRANSPOSE2_PD(Y,F);
978 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
979 H = _mm_setzero_pd();
980 GMX_MM_TRANSPOSE2_PD(G,H);
981 Heps = _mm_mul_pd(vfeps,H);
982 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
983 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
984 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq00,FF),_mm_mul_pd(vftabscale,rinv00)));
985
986 /* LENNARD-JONES DISPERSION/REPULSION */
987
988 rinvsix = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
989 fvdw = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(c12_00,rinvsix),c6_00),_mm_mul_pd(rinvsix,rinvsq00));
990
991 fscal = _mm_add_pd(felec,fvdw);
992
993 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
994
995 /* Calculate temporary vectorial force */
996 tx = _mm_mul_pd(fscal,dx00);
997 ty = _mm_mul_pd(fscal,dy00);
998 tz = _mm_mul_pd(fscal,dz00);
999
1000 /* Update vectorial force */
1001 fix0 = _mm_add_pd(fix0,tx);
1002 fiy0 = _mm_add_pd(fiy0,ty);
1003 fiz0 = _mm_add_pd(fiz0,tz);
1004
1005 fjx0 = _mm_add_pd(fjx0,tx);
1006 fjy0 = _mm_add_pd(fjy0,ty);
1007 fjz0 = _mm_add_pd(fjz0,tz);
1008
1009 /**************************
1010 * CALCULATE INTERACTIONS *
1011 **************************/
1012
1013 r10 = _mm_mul_pd(rsq10,rinv10);
1014
1015 /* Compute parameters for interactions between i and j atoms */
1016 qq10 = _mm_mul_pd(iq1,jq0);
1017
1018 /* Calculate table index by multiplying r with table scale and truncate to integer */
1019 rt = _mm_mul_pd(r10,vftabscale);
1020 vfitab = _mm_cvttpd_epi32(rt);
1021 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1022 vfitab = _mm_slli_epi32(vfitab,2);
1023
1024 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1025 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1026 F = _mm_setzero_pd();
1027 GMX_MM_TRANSPOSE2_PD(Y,F);
1028 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1029 H = _mm_setzero_pd();
1030 GMX_MM_TRANSPOSE2_PD(G,H);
1031 Heps = _mm_mul_pd(vfeps,H);
1032 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1033 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1034 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq10,FF),_mm_mul_pd(vftabscale,rinv10)));
1035
1036 fscal = felec;
1037
1038 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1039
1040 /* Calculate temporary vectorial force */
1041 tx = _mm_mul_pd(fscal,dx10);
1042 ty = _mm_mul_pd(fscal,dy10);
1043 tz = _mm_mul_pd(fscal,dz10);
1044
1045 /* Update vectorial force */
1046 fix1 = _mm_add_pd(fix1,tx);
1047 fiy1 = _mm_add_pd(fiy1,ty);
1048 fiz1 = _mm_add_pd(fiz1,tz);
1049
1050 fjx0 = _mm_add_pd(fjx0,tx);
1051 fjy0 = _mm_add_pd(fjy0,ty);
1052 fjz0 = _mm_add_pd(fjz0,tz);
1053
1054 /**************************
1055 * CALCULATE INTERACTIONS *
1056 **************************/
1057
1058 r20 = _mm_mul_pd(rsq20,rinv20);
1059
1060 /* Compute parameters for interactions between i and j atoms */
1061 qq20 = _mm_mul_pd(iq2,jq0);
1062
1063 /* Calculate table index by multiplying r with table scale and truncate to integer */
1064 rt = _mm_mul_pd(r20,vftabscale);
1065 vfitab = _mm_cvttpd_epi32(rt);
1066 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1067 vfitab = _mm_slli_epi32(vfitab,2);
1068
1069 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1070 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1071 F = _mm_setzero_pd();
1072 GMX_MM_TRANSPOSE2_PD(Y,F);
1073 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1074 H = _mm_setzero_pd();
1075 GMX_MM_TRANSPOSE2_PD(G,H);
1076 Heps = _mm_mul_pd(vfeps,H);
1077 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1078 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1079 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq20,FF),_mm_mul_pd(vftabscale,rinv20)));
1080
1081 fscal = felec;
1082
1083 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1084
1085 /* Calculate temporary vectorial force */
1086 tx = _mm_mul_pd(fscal,dx20);
1087 ty = _mm_mul_pd(fscal,dy20);
1088 tz = _mm_mul_pd(fscal,dz20);
1089
1090 /* Update vectorial force */
1091 fix2 = _mm_add_pd(fix2,tx);
1092 fiy2 = _mm_add_pd(fiy2,ty);
1093 fiz2 = _mm_add_pd(fiz2,tz);
1094
1095 fjx0 = _mm_add_pd(fjx0,tx);
1096 fjy0 = _mm_add_pd(fjy0,ty);
1097 fjz0 = _mm_add_pd(fjz0,tz);
1098
1099 gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1100
1101 /* Inner loop uses 128 flops */
1102 }
1103
1104 /* End of innermost loop */
1105
1106 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1107 f+i_coord_offset,fshift+i_shift_offset);
1108
1109 /* Increment number of inner iterations */
1110 inneriter += j_index_end - j_index_start;
1111
1112 /* Outer loop uses 18 flops */
1113 }
1114
1115 /* Increment number of outer iterations */
1116 outeriter += nri;
1117
1118 /* Update outer/inner flops */
1119
1120 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*18 + inneriter*128);
1121 }
The ultimate molecular dynamics simulation package