2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
36 * Note: this file was generated by the GROMACS avx_256_double kernel generator.
42 #include "../nb_kernel.h"
43 #include "types/simple.h"
44 #include "gromacs/math/vec.h"
47 #include "gromacs/simd/math_x86_avx_256_double.h"
48 #include "kernelutil_x86_avx_256_double.h"
51 * Gromacs nonbonded kernel: nb_kernel_ElecRF_VdwCSTab_GeomW4W4_VF_avx_256_double
52 * Electrostatics interaction: ReactionField
53 * VdW interaction: CubicSplineTable
54 * Geometry: Water4-Water4
55 * Calculate force/pot: PotentialAndForce
58 nb_kernel_ElecRF_VdwCSTab_GeomW4W4_VF_avx_256_double
59 (t_nblist * gmx_restrict nlist,
60 rvec * gmx_restrict xx,
61 rvec * gmx_restrict ff,
62 t_forcerec * gmx_restrict fr,
63 t_mdatoms * gmx_restrict mdatoms,
64 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
65 t_nrnb * gmx_restrict nrnb)
67 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
68 * just 0 for non-waters.
69 * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
70 * jnr indices corresponding to data put in the four positions in the SIMD register.
72 int i_shift_offset,i_coord_offset,outeriter,inneriter;
73 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
74 int jnrA,jnrB,jnrC,jnrD;
75 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
76 int jnrlistE,jnrlistF,jnrlistG,jnrlistH;
77 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
78 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
80 real *shiftvec,*fshift,*x,*f;
81 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
83 __m256d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
84 real * vdwioffsetptr0;
85 __m256d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
86 real * vdwioffsetptr1;
87 __m256d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
88 real * vdwioffsetptr2;
89 __m256d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
90 real * vdwioffsetptr3;
91 __m256d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
92 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
93 __m256d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
94 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
95 __m256d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
96 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
97 __m256d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
98 int vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
99 __m256d jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
100 __m256d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
101 __m256d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
102 __m256d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
103 __m256d dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
104 __m256d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
105 __m256d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
106 __m256d dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
107 __m256d dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
108 __m256d dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
109 __m256d dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
110 __m256d velec,felec,velecsum,facel,crf,krf,krf2;
113 __m256d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
116 __m256d one_sixth = _mm256_set1_pd(1.0/6.0);
117 __m256d one_twelfth = _mm256_set1_pd(1.0/12.0);
119 __m128i ifour = _mm_set1_epi32(4);
120 __m256d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
122 __m256d dummy_mask,cutoff_mask;
123 __m128 tmpmask0,tmpmask1;
124 __m256d signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
125 __m256d one = _mm256_set1_pd(1.0);
126 __m256d two = _mm256_set1_pd(2.0);
132 jindex = nlist->jindex;
134 shiftidx = nlist->shift;
136 shiftvec = fr->shift_vec[0];
137 fshift = fr->fshift[0];
138 facel = _mm256_set1_pd(fr->epsfac);
139 charge = mdatoms->chargeA;
140 krf = _mm256_set1_pd(fr->ic->k_rf);
141 krf2 = _mm256_set1_pd(fr->ic->k_rf*2.0);
142 crf = _mm256_set1_pd(fr->ic->c_rf);
143 nvdwtype = fr->ntype;
145 vdwtype = mdatoms->typeA;
147 vftab = kernel_data->table_vdw->data;
148 vftabscale = _mm256_set1_pd(kernel_data->table_vdw->scale);
150 /* Setup water-specific parameters */
151 inr = nlist->iinr[0];
152 iq1 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
153 iq2 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
154 iq3 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
155 vdwioffsetptr0 = vdwparam+2*nvdwtype*vdwtype[inr+0];
157 jq1 = _mm256_set1_pd(charge[inr+1]);
158 jq2 = _mm256_set1_pd(charge[inr+2]);
159 jq3 = _mm256_set1_pd(charge[inr+3]);
160 vdwjidx0A = 2*vdwtype[inr+0];
161 c6_00 = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
162 c12_00 = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
163 qq11 = _mm256_mul_pd(iq1,jq1);
164 qq12 = _mm256_mul_pd(iq1,jq2);
165 qq13 = _mm256_mul_pd(iq1,jq3);
166 qq21 = _mm256_mul_pd(iq2,jq1);
167 qq22 = _mm256_mul_pd(iq2,jq2);
168 qq23 = _mm256_mul_pd(iq2,jq3);
169 qq31 = _mm256_mul_pd(iq3,jq1);
170 qq32 = _mm256_mul_pd(iq3,jq2);
171 qq33 = _mm256_mul_pd(iq3,jq3);
173 /* Avoid stupid compiler warnings */
174 jnrA = jnrB = jnrC = jnrD = 0;
183 for(iidx=0;iidx<4*DIM;iidx++)
188 /* Start outer loop over neighborlists */
189 for(iidx=0; iidx<nri; iidx++)
191 /* Load shift vector for this list */
192 i_shift_offset = DIM*shiftidx[iidx];
194 /* Load limits for loop over neighbors */
195 j_index_start = jindex[iidx];
196 j_index_end = jindex[iidx+1];
198 /* Get outer coordinate index */
200 i_coord_offset = DIM*inr;
202 /* Load i particle coords and add shift vector */
203 gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
204 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
206 fix0 = _mm256_setzero_pd();
207 fiy0 = _mm256_setzero_pd();
208 fiz0 = _mm256_setzero_pd();
209 fix1 = _mm256_setzero_pd();
210 fiy1 = _mm256_setzero_pd();
211 fiz1 = _mm256_setzero_pd();
212 fix2 = _mm256_setzero_pd();
213 fiy2 = _mm256_setzero_pd();
214 fiz2 = _mm256_setzero_pd();
215 fix3 = _mm256_setzero_pd();
216 fiy3 = _mm256_setzero_pd();
217 fiz3 = _mm256_setzero_pd();
219 /* Reset potential sums */
220 velecsum = _mm256_setzero_pd();
221 vvdwsum = _mm256_setzero_pd();
223 /* Start inner kernel loop */
224 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
227 /* Get j neighbor index, and coordinate index */
232 j_coord_offsetA = DIM*jnrA;
233 j_coord_offsetB = DIM*jnrB;
234 j_coord_offsetC = DIM*jnrC;
235 j_coord_offsetD = DIM*jnrD;
237 /* load j atom coordinates */
238 gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
239 x+j_coord_offsetC,x+j_coord_offsetD,
240 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
241 &jy2,&jz2,&jx3,&jy3,&jz3);
243 /* Calculate displacement vector */
244 dx00 = _mm256_sub_pd(ix0,jx0);
245 dy00 = _mm256_sub_pd(iy0,jy0);
246 dz00 = _mm256_sub_pd(iz0,jz0);
247 dx11 = _mm256_sub_pd(ix1,jx1);
248 dy11 = _mm256_sub_pd(iy1,jy1);
249 dz11 = _mm256_sub_pd(iz1,jz1);
250 dx12 = _mm256_sub_pd(ix1,jx2);
251 dy12 = _mm256_sub_pd(iy1,jy2);
252 dz12 = _mm256_sub_pd(iz1,jz2);
253 dx13 = _mm256_sub_pd(ix1,jx3);
254 dy13 = _mm256_sub_pd(iy1,jy3);
255 dz13 = _mm256_sub_pd(iz1,jz3);
256 dx21 = _mm256_sub_pd(ix2,jx1);
257 dy21 = _mm256_sub_pd(iy2,jy1);
258 dz21 = _mm256_sub_pd(iz2,jz1);
259 dx22 = _mm256_sub_pd(ix2,jx2);
260 dy22 = _mm256_sub_pd(iy2,jy2);
261 dz22 = _mm256_sub_pd(iz2,jz2);
262 dx23 = _mm256_sub_pd(ix2,jx3);
263 dy23 = _mm256_sub_pd(iy2,jy3);
264 dz23 = _mm256_sub_pd(iz2,jz3);
265 dx31 = _mm256_sub_pd(ix3,jx1);
266 dy31 = _mm256_sub_pd(iy3,jy1);
267 dz31 = _mm256_sub_pd(iz3,jz1);
268 dx32 = _mm256_sub_pd(ix3,jx2);
269 dy32 = _mm256_sub_pd(iy3,jy2);
270 dz32 = _mm256_sub_pd(iz3,jz2);
271 dx33 = _mm256_sub_pd(ix3,jx3);
272 dy33 = _mm256_sub_pd(iy3,jy3);
273 dz33 = _mm256_sub_pd(iz3,jz3);
275 /* Calculate squared distance and things based on it */
276 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
277 rsq11 = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
278 rsq12 = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
279 rsq13 = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
280 rsq21 = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
281 rsq22 = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
282 rsq23 = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
283 rsq31 = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
284 rsq32 = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
285 rsq33 = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
287 rinv00 = gmx_mm256_invsqrt_pd(rsq00);
288 rinv11 = gmx_mm256_invsqrt_pd(rsq11);
289 rinv12 = gmx_mm256_invsqrt_pd(rsq12);
290 rinv13 = gmx_mm256_invsqrt_pd(rsq13);
291 rinv21 = gmx_mm256_invsqrt_pd(rsq21);
292 rinv22 = gmx_mm256_invsqrt_pd(rsq22);
293 rinv23 = gmx_mm256_invsqrt_pd(rsq23);
294 rinv31 = gmx_mm256_invsqrt_pd(rsq31);
295 rinv32 = gmx_mm256_invsqrt_pd(rsq32);
296 rinv33 = gmx_mm256_invsqrt_pd(rsq33);
298 rinvsq11 = _mm256_mul_pd(rinv11,rinv11);
299 rinvsq12 = _mm256_mul_pd(rinv12,rinv12);
300 rinvsq13 = _mm256_mul_pd(rinv13,rinv13);
301 rinvsq21 = _mm256_mul_pd(rinv21,rinv21);
302 rinvsq22 = _mm256_mul_pd(rinv22,rinv22);
303 rinvsq23 = _mm256_mul_pd(rinv23,rinv23);
304 rinvsq31 = _mm256_mul_pd(rinv31,rinv31);
305 rinvsq32 = _mm256_mul_pd(rinv32,rinv32);
306 rinvsq33 = _mm256_mul_pd(rinv33,rinv33);
308 fjx0 = _mm256_setzero_pd();
309 fjy0 = _mm256_setzero_pd();
310 fjz0 = _mm256_setzero_pd();
311 fjx1 = _mm256_setzero_pd();
312 fjy1 = _mm256_setzero_pd();
313 fjz1 = _mm256_setzero_pd();
314 fjx2 = _mm256_setzero_pd();
315 fjy2 = _mm256_setzero_pd();
316 fjz2 = _mm256_setzero_pd();
317 fjx3 = _mm256_setzero_pd();
318 fjy3 = _mm256_setzero_pd();
319 fjz3 = _mm256_setzero_pd();
321 /**************************
322 * CALCULATE INTERACTIONS *
323 **************************/
325 r00 = _mm256_mul_pd(rsq00,rinv00);
327 /* Calculate table index by multiplying r with table scale and truncate to integer */
328 rt = _mm256_mul_pd(r00,vftabscale);
329 vfitab = _mm256_cvttpd_epi32(rt);
330 vfeps = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
331 vfitab = _mm_slli_epi32(vfitab,3);
333 /* CUBIC SPLINE TABLE DISPERSION */
334 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
335 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
336 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
337 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
338 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
339 Heps = _mm256_mul_pd(vfeps,H);
340 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
341 VV = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
342 vvdw6 = _mm256_mul_pd(c6_00,VV);
343 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
344 fvdw6 = _mm256_mul_pd(c6_00,FF);
346 /* CUBIC SPLINE TABLE REPULSION */
347 vfitab = _mm_add_epi32(vfitab,ifour);
348 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
349 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
350 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
351 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
352 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
353 Heps = _mm256_mul_pd(vfeps,H);
354 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
355 VV = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
356 vvdw12 = _mm256_mul_pd(c12_00,VV);
357 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
358 fvdw12 = _mm256_mul_pd(c12_00,FF);
359 vvdw = _mm256_add_pd(vvdw12,vvdw6);
360 fvdw = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
362 /* Update potential sum for this i atom from the interaction with this j atom. */
363 vvdwsum = _mm256_add_pd(vvdwsum,vvdw);
367 /* Calculate temporary vectorial force */
368 tx = _mm256_mul_pd(fscal,dx00);
369 ty = _mm256_mul_pd(fscal,dy00);
370 tz = _mm256_mul_pd(fscal,dz00);
372 /* Update vectorial force */
373 fix0 = _mm256_add_pd(fix0,tx);
374 fiy0 = _mm256_add_pd(fiy0,ty);
375 fiz0 = _mm256_add_pd(fiz0,tz);
377 fjx0 = _mm256_add_pd(fjx0,tx);
378 fjy0 = _mm256_add_pd(fjy0,ty);
379 fjz0 = _mm256_add_pd(fjz0,tz);
381 /**************************
382 * CALCULATE INTERACTIONS *
383 **************************/
385 /* REACTION-FIELD ELECTROSTATICS */
386 velec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
387 felec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
389 /* Update potential sum for this i atom from the interaction with this j atom. */
390 velecsum = _mm256_add_pd(velecsum,velec);
394 /* Calculate temporary vectorial force */
395 tx = _mm256_mul_pd(fscal,dx11);
396 ty = _mm256_mul_pd(fscal,dy11);
397 tz = _mm256_mul_pd(fscal,dz11);
399 /* Update vectorial force */
400 fix1 = _mm256_add_pd(fix1,tx);
401 fiy1 = _mm256_add_pd(fiy1,ty);
402 fiz1 = _mm256_add_pd(fiz1,tz);
404 fjx1 = _mm256_add_pd(fjx1,tx);
405 fjy1 = _mm256_add_pd(fjy1,ty);
406 fjz1 = _mm256_add_pd(fjz1,tz);
408 /**************************
409 * CALCULATE INTERACTIONS *
410 **************************/
412 /* REACTION-FIELD ELECTROSTATICS */
413 velec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
414 felec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
416 /* Update potential sum for this i atom from the interaction with this j atom. */
417 velecsum = _mm256_add_pd(velecsum,velec);
421 /* Calculate temporary vectorial force */
422 tx = _mm256_mul_pd(fscal,dx12);
423 ty = _mm256_mul_pd(fscal,dy12);
424 tz = _mm256_mul_pd(fscal,dz12);
426 /* Update vectorial force */
427 fix1 = _mm256_add_pd(fix1,tx);
428 fiy1 = _mm256_add_pd(fiy1,ty);
429 fiz1 = _mm256_add_pd(fiz1,tz);
431 fjx2 = _mm256_add_pd(fjx2,tx);
432 fjy2 = _mm256_add_pd(fjy2,ty);
433 fjz2 = _mm256_add_pd(fjz2,tz);
435 /**************************
436 * CALCULATE INTERACTIONS *
437 **************************/
439 /* REACTION-FIELD ELECTROSTATICS */
440 velec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_add_pd(rinv13,_mm256_mul_pd(krf,rsq13)),crf));
441 felec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
443 /* Update potential sum for this i atom from the interaction with this j atom. */
444 velecsum = _mm256_add_pd(velecsum,velec);
448 /* Calculate temporary vectorial force */
449 tx = _mm256_mul_pd(fscal,dx13);
450 ty = _mm256_mul_pd(fscal,dy13);
451 tz = _mm256_mul_pd(fscal,dz13);
453 /* Update vectorial force */
454 fix1 = _mm256_add_pd(fix1,tx);
455 fiy1 = _mm256_add_pd(fiy1,ty);
456 fiz1 = _mm256_add_pd(fiz1,tz);
458 fjx3 = _mm256_add_pd(fjx3,tx);
459 fjy3 = _mm256_add_pd(fjy3,ty);
460 fjz3 = _mm256_add_pd(fjz3,tz);
462 /**************************
463 * CALCULATE INTERACTIONS *
464 **************************/
466 /* REACTION-FIELD ELECTROSTATICS */
467 velec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
468 felec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
470 /* Update potential sum for this i atom from the interaction with this j atom. */
471 velecsum = _mm256_add_pd(velecsum,velec);
475 /* Calculate temporary vectorial force */
476 tx = _mm256_mul_pd(fscal,dx21);
477 ty = _mm256_mul_pd(fscal,dy21);
478 tz = _mm256_mul_pd(fscal,dz21);
480 /* Update vectorial force */
481 fix2 = _mm256_add_pd(fix2,tx);
482 fiy2 = _mm256_add_pd(fiy2,ty);
483 fiz2 = _mm256_add_pd(fiz2,tz);
485 fjx1 = _mm256_add_pd(fjx1,tx);
486 fjy1 = _mm256_add_pd(fjy1,ty);
487 fjz1 = _mm256_add_pd(fjz1,tz);
489 /**************************
490 * CALCULATE INTERACTIONS *
491 **************************/
493 /* REACTION-FIELD ELECTROSTATICS */
494 velec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
495 felec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
497 /* Update potential sum for this i atom from the interaction with this j atom. */
498 velecsum = _mm256_add_pd(velecsum,velec);
502 /* Calculate temporary vectorial force */
503 tx = _mm256_mul_pd(fscal,dx22);
504 ty = _mm256_mul_pd(fscal,dy22);
505 tz = _mm256_mul_pd(fscal,dz22);
507 /* Update vectorial force */
508 fix2 = _mm256_add_pd(fix2,tx);
509 fiy2 = _mm256_add_pd(fiy2,ty);
510 fiz2 = _mm256_add_pd(fiz2,tz);
512 fjx2 = _mm256_add_pd(fjx2,tx);
513 fjy2 = _mm256_add_pd(fjy2,ty);
514 fjz2 = _mm256_add_pd(fjz2,tz);
516 /**************************
517 * CALCULATE INTERACTIONS *
518 **************************/
520 /* REACTION-FIELD ELECTROSTATICS */
521 velec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_add_pd(rinv23,_mm256_mul_pd(krf,rsq23)),crf));
522 felec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
524 /* Update potential sum for this i atom from the interaction with this j atom. */
525 velecsum = _mm256_add_pd(velecsum,velec);
529 /* Calculate temporary vectorial force */
530 tx = _mm256_mul_pd(fscal,dx23);
531 ty = _mm256_mul_pd(fscal,dy23);
532 tz = _mm256_mul_pd(fscal,dz23);
534 /* Update vectorial force */
535 fix2 = _mm256_add_pd(fix2,tx);
536 fiy2 = _mm256_add_pd(fiy2,ty);
537 fiz2 = _mm256_add_pd(fiz2,tz);
539 fjx3 = _mm256_add_pd(fjx3,tx);
540 fjy3 = _mm256_add_pd(fjy3,ty);
541 fjz3 = _mm256_add_pd(fjz3,tz);
543 /**************************
544 * CALCULATE INTERACTIONS *
545 **************************/
547 /* REACTION-FIELD ELECTROSTATICS */
548 velec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_add_pd(rinv31,_mm256_mul_pd(krf,rsq31)),crf));
549 felec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
551 /* Update potential sum for this i atom from the interaction with this j atom. */
552 velecsum = _mm256_add_pd(velecsum,velec);
556 /* Calculate temporary vectorial force */
557 tx = _mm256_mul_pd(fscal,dx31);
558 ty = _mm256_mul_pd(fscal,dy31);
559 tz = _mm256_mul_pd(fscal,dz31);
561 /* Update vectorial force */
562 fix3 = _mm256_add_pd(fix3,tx);
563 fiy3 = _mm256_add_pd(fiy3,ty);
564 fiz3 = _mm256_add_pd(fiz3,tz);
566 fjx1 = _mm256_add_pd(fjx1,tx);
567 fjy1 = _mm256_add_pd(fjy1,ty);
568 fjz1 = _mm256_add_pd(fjz1,tz);
570 /**************************
571 * CALCULATE INTERACTIONS *
572 **************************/
574 /* REACTION-FIELD ELECTROSTATICS */
575 velec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_add_pd(rinv32,_mm256_mul_pd(krf,rsq32)),crf));
576 felec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
578 /* Update potential sum for this i atom from the interaction with this j atom. */
579 velecsum = _mm256_add_pd(velecsum,velec);
583 /* Calculate temporary vectorial force */
584 tx = _mm256_mul_pd(fscal,dx32);
585 ty = _mm256_mul_pd(fscal,dy32);
586 tz = _mm256_mul_pd(fscal,dz32);
588 /* Update vectorial force */
589 fix3 = _mm256_add_pd(fix3,tx);
590 fiy3 = _mm256_add_pd(fiy3,ty);
591 fiz3 = _mm256_add_pd(fiz3,tz);
593 fjx2 = _mm256_add_pd(fjx2,tx);
594 fjy2 = _mm256_add_pd(fjy2,ty);
595 fjz2 = _mm256_add_pd(fjz2,tz);
597 /**************************
598 * CALCULATE INTERACTIONS *
599 **************************/
601 /* REACTION-FIELD ELECTROSTATICS */
602 velec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_add_pd(rinv33,_mm256_mul_pd(krf,rsq33)),crf));
603 felec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
605 /* Update potential sum for this i atom from the interaction with this j atom. */
606 velecsum = _mm256_add_pd(velecsum,velec);
610 /* Calculate temporary vectorial force */
611 tx = _mm256_mul_pd(fscal,dx33);
612 ty = _mm256_mul_pd(fscal,dy33);
613 tz = _mm256_mul_pd(fscal,dz33);
615 /* Update vectorial force */
616 fix3 = _mm256_add_pd(fix3,tx);
617 fiy3 = _mm256_add_pd(fiy3,ty);
618 fiz3 = _mm256_add_pd(fiz3,tz);
620 fjx3 = _mm256_add_pd(fjx3,tx);
621 fjy3 = _mm256_add_pd(fjy3,ty);
622 fjz3 = _mm256_add_pd(fjz3,tz);
624 fjptrA = f+j_coord_offsetA;
625 fjptrB = f+j_coord_offsetB;
626 fjptrC = f+j_coord_offsetC;
627 fjptrD = f+j_coord_offsetD;
629 gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
630 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
631 fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
633 /* Inner loop uses 347 flops */
639 /* Get j neighbor index, and coordinate index */
640 jnrlistA = jjnr[jidx];
641 jnrlistB = jjnr[jidx+1];
642 jnrlistC = jjnr[jidx+2];
643 jnrlistD = jjnr[jidx+3];
644 /* Sign of each element will be negative for non-real atoms.
645 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
646 * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
648 tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
650 tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
651 tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
652 dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
654 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
655 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
656 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
657 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
658 j_coord_offsetA = DIM*jnrA;
659 j_coord_offsetB = DIM*jnrB;
660 j_coord_offsetC = DIM*jnrC;
661 j_coord_offsetD = DIM*jnrD;
663 /* load j atom coordinates */
664 gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
665 x+j_coord_offsetC,x+j_coord_offsetD,
666 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
667 &jy2,&jz2,&jx3,&jy3,&jz3);
669 /* Calculate displacement vector */
670 dx00 = _mm256_sub_pd(ix0,jx0);
671 dy00 = _mm256_sub_pd(iy0,jy0);
672 dz00 = _mm256_sub_pd(iz0,jz0);
673 dx11 = _mm256_sub_pd(ix1,jx1);
674 dy11 = _mm256_sub_pd(iy1,jy1);
675 dz11 = _mm256_sub_pd(iz1,jz1);
676 dx12 = _mm256_sub_pd(ix1,jx2);
677 dy12 = _mm256_sub_pd(iy1,jy2);
678 dz12 = _mm256_sub_pd(iz1,jz2);
679 dx13 = _mm256_sub_pd(ix1,jx3);
680 dy13 = _mm256_sub_pd(iy1,jy3);
681 dz13 = _mm256_sub_pd(iz1,jz3);
682 dx21 = _mm256_sub_pd(ix2,jx1);
683 dy21 = _mm256_sub_pd(iy2,jy1);
684 dz21 = _mm256_sub_pd(iz2,jz1);
685 dx22 = _mm256_sub_pd(ix2,jx2);
686 dy22 = _mm256_sub_pd(iy2,jy2);
687 dz22 = _mm256_sub_pd(iz2,jz2);
688 dx23 = _mm256_sub_pd(ix2,jx3);
689 dy23 = _mm256_sub_pd(iy2,jy3);
690 dz23 = _mm256_sub_pd(iz2,jz3);
691 dx31 = _mm256_sub_pd(ix3,jx1);
692 dy31 = _mm256_sub_pd(iy3,jy1);
693 dz31 = _mm256_sub_pd(iz3,jz1);
694 dx32 = _mm256_sub_pd(ix3,jx2);
695 dy32 = _mm256_sub_pd(iy3,jy2);
696 dz32 = _mm256_sub_pd(iz3,jz2);
697 dx33 = _mm256_sub_pd(ix3,jx3);
698 dy33 = _mm256_sub_pd(iy3,jy3);
699 dz33 = _mm256_sub_pd(iz3,jz3);
701 /* Calculate squared distance and things based on it */
702 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
703 rsq11 = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
704 rsq12 = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
705 rsq13 = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
706 rsq21 = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
707 rsq22 = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
708 rsq23 = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
709 rsq31 = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
710 rsq32 = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
711 rsq33 = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
713 rinv00 = gmx_mm256_invsqrt_pd(rsq00);
714 rinv11 = gmx_mm256_invsqrt_pd(rsq11);
715 rinv12 = gmx_mm256_invsqrt_pd(rsq12);
716 rinv13 = gmx_mm256_invsqrt_pd(rsq13);
717 rinv21 = gmx_mm256_invsqrt_pd(rsq21);
718 rinv22 = gmx_mm256_invsqrt_pd(rsq22);
719 rinv23 = gmx_mm256_invsqrt_pd(rsq23);
720 rinv31 = gmx_mm256_invsqrt_pd(rsq31);
721 rinv32 = gmx_mm256_invsqrt_pd(rsq32);
722 rinv33 = gmx_mm256_invsqrt_pd(rsq33);
724 rinvsq11 = _mm256_mul_pd(rinv11,rinv11);
725 rinvsq12 = _mm256_mul_pd(rinv12,rinv12);
726 rinvsq13 = _mm256_mul_pd(rinv13,rinv13);
727 rinvsq21 = _mm256_mul_pd(rinv21,rinv21);
728 rinvsq22 = _mm256_mul_pd(rinv22,rinv22);
729 rinvsq23 = _mm256_mul_pd(rinv23,rinv23);
730 rinvsq31 = _mm256_mul_pd(rinv31,rinv31);
731 rinvsq32 = _mm256_mul_pd(rinv32,rinv32);
732 rinvsq33 = _mm256_mul_pd(rinv33,rinv33);
734 fjx0 = _mm256_setzero_pd();
735 fjy0 = _mm256_setzero_pd();
736 fjz0 = _mm256_setzero_pd();
737 fjx1 = _mm256_setzero_pd();
738 fjy1 = _mm256_setzero_pd();
739 fjz1 = _mm256_setzero_pd();
740 fjx2 = _mm256_setzero_pd();
741 fjy2 = _mm256_setzero_pd();
742 fjz2 = _mm256_setzero_pd();
743 fjx3 = _mm256_setzero_pd();
744 fjy3 = _mm256_setzero_pd();
745 fjz3 = _mm256_setzero_pd();
747 /**************************
748 * CALCULATE INTERACTIONS *
749 **************************/
751 r00 = _mm256_mul_pd(rsq00,rinv00);
752 r00 = _mm256_andnot_pd(dummy_mask,r00);
754 /* Calculate table index by multiplying r with table scale and truncate to integer */
755 rt = _mm256_mul_pd(r00,vftabscale);
756 vfitab = _mm256_cvttpd_epi32(rt);
757 vfeps = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
758 vfitab = _mm_slli_epi32(vfitab,3);
760 /* CUBIC SPLINE TABLE DISPERSION */
761 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
762 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
763 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
764 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
765 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
766 Heps = _mm256_mul_pd(vfeps,H);
767 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
768 VV = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
769 vvdw6 = _mm256_mul_pd(c6_00,VV);
770 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
771 fvdw6 = _mm256_mul_pd(c6_00,FF);
773 /* CUBIC SPLINE TABLE REPULSION */
774 vfitab = _mm_add_epi32(vfitab,ifour);
775 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
776 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
777 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
778 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
779 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
780 Heps = _mm256_mul_pd(vfeps,H);
781 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
782 VV = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
783 vvdw12 = _mm256_mul_pd(c12_00,VV);
784 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
785 fvdw12 = _mm256_mul_pd(c12_00,FF);
786 vvdw = _mm256_add_pd(vvdw12,vvdw6);
787 fvdw = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
789 /* Update potential sum for this i atom from the interaction with this j atom. */
790 vvdw = _mm256_andnot_pd(dummy_mask,vvdw);
791 vvdwsum = _mm256_add_pd(vvdwsum,vvdw);
795 fscal = _mm256_andnot_pd(dummy_mask,fscal);
797 /* Calculate temporary vectorial force */
798 tx = _mm256_mul_pd(fscal,dx00);
799 ty = _mm256_mul_pd(fscal,dy00);
800 tz = _mm256_mul_pd(fscal,dz00);
802 /* Update vectorial force */
803 fix0 = _mm256_add_pd(fix0,tx);
804 fiy0 = _mm256_add_pd(fiy0,ty);
805 fiz0 = _mm256_add_pd(fiz0,tz);
807 fjx0 = _mm256_add_pd(fjx0,tx);
808 fjy0 = _mm256_add_pd(fjy0,ty);
809 fjz0 = _mm256_add_pd(fjz0,tz);
811 /**************************
812 * CALCULATE INTERACTIONS *
813 **************************/
815 /* REACTION-FIELD ELECTROSTATICS */
816 velec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
817 felec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
819 /* Update potential sum for this i atom from the interaction with this j atom. */
820 velec = _mm256_andnot_pd(dummy_mask,velec);
821 velecsum = _mm256_add_pd(velecsum,velec);
825 fscal = _mm256_andnot_pd(dummy_mask,fscal);
827 /* Calculate temporary vectorial force */
828 tx = _mm256_mul_pd(fscal,dx11);
829 ty = _mm256_mul_pd(fscal,dy11);
830 tz = _mm256_mul_pd(fscal,dz11);
832 /* Update vectorial force */
833 fix1 = _mm256_add_pd(fix1,tx);
834 fiy1 = _mm256_add_pd(fiy1,ty);
835 fiz1 = _mm256_add_pd(fiz1,tz);
837 fjx1 = _mm256_add_pd(fjx1,tx);
838 fjy1 = _mm256_add_pd(fjy1,ty);
839 fjz1 = _mm256_add_pd(fjz1,tz);
841 /**************************
842 * CALCULATE INTERACTIONS *
843 **************************/
845 /* REACTION-FIELD ELECTROSTATICS */
846 velec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
847 felec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
849 /* Update potential sum for this i atom from the interaction with this j atom. */
850 velec = _mm256_andnot_pd(dummy_mask,velec);
851 velecsum = _mm256_add_pd(velecsum,velec);
855 fscal = _mm256_andnot_pd(dummy_mask,fscal);
857 /* Calculate temporary vectorial force */
858 tx = _mm256_mul_pd(fscal,dx12);
859 ty = _mm256_mul_pd(fscal,dy12);
860 tz = _mm256_mul_pd(fscal,dz12);
862 /* Update vectorial force */
863 fix1 = _mm256_add_pd(fix1,tx);
864 fiy1 = _mm256_add_pd(fiy1,ty);
865 fiz1 = _mm256_add_pd(fiz1,tz);
867 fjx2 = _mm256_add_pd(fjx2,tx);
868 fjy2 = _mm256_add_pd(fjy2,ty);
869 fjz2 = _mm256_add_pd(fjz2,tz);
871 /**************************
872 * CALCULATE INTERACTIONS *
873 **************************/
875 /* REACTION-FIELD ELECTROSTATICS */
876 velec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_add_pd(rinv13,_mm256_mul_pd(krf,rsq13)),crf));
877 felec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
879 /* Update potential sum for this i atom from the interaction with this j atom. */
880 velec = _mm256_andnot_pd(dummy_mask,velec);
881 velecsum = _mm256_add_pd(velecsum,velec);
885 fscal = _mm256_andnot_pd(dummy_mask,fscal);
887 /* Calculate temporary vectorial force */
888 tx = _mm256_mul_pd(fscal,dx13);
889 ty = _mm256_mul_pd(fscal,dy13);
890 tz = _mm256_mul_pd(fscal,dz13);
892 /* Update vectorial force */
893 fix1 = _mm256_add_pd(fix1,tx);
894 fiy1 = _mm256_add_pd(fiy1,ty);
895 fiz1 = _mm256_add_pd(fiz1,tz);
897 fjx3 = _mm256_add_pd(fjx3,tx);
898 fjy3 = _mm256_add_pd(fjy3,ty);
899 fjz3 = _mm256_add_pd(fjz3,tz);
901 /**************************
902 * CALCULATE INTERACTIONS *
903 **************************/
905 /* REACTION-FIELD ELECTROSTATICS */
906 velec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
907 felec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
909 /* Update potential sum for this i atom from the interaction with this j atom. */
910 velec = _mm256_andnot_pd(dummy_mask,velec);
911 velecsum = _mm256_add_pd(velecsum,velec);
915 fscal = _mm256_andnot_pd(dummy_mask,fscal);
917 /* Calculate temporary vectorial force */
918 tx = _mm256_mul_pd(fscal,dx21);
919 ty = _mm256_mul_pd(fscal,dy21);
920 tz = _mm256_mul_pd(fscal,dz21);
922 /* Update vectorial force */
923 fix2 = _mm256_add_pd(fix2,tx);
924 fiy2 = _mm256_add_pd(fiy2,ty);
925 fiz2 = _mm256_add_pd(fiz2,tz);
927 fjx1 = _mm256_add_pd(fjx1,tx);
928 fjy1 = _mm256_add_pd(fjy1,ty);
929 fjz1 = _mm256_add_pd(fjz1,tz);
931 /**************************
932 * CALCULATE INTERACTIONS *
933 **************************/
935 /* REACTION-FIELD ELECTROSTATICS */
936 velec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
937 felec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
939 /* Update potential sum for this i atom from the interaction with this j atom. */
940 velec = _mm256_andnot_pd(dummy_mask,velec);
941 velecsum = _mm256_add_pd(velecsum,velec);
945 fscal = _mm256_andnot_pd(dummy_mask,fscal);
947 /* Calculate temporary vectorial force */
948 tx = _mm256_mul_pd(fscal,dx22);
949 ty = _mm256_mul_pd(fscal,dy22);
950 tz = _mm256_mul_pd(fscal,dz22);
952 /* Update vectorial force */
953 fix2 = _mm256_add_pd(fix2,tx);
954 fiy2 = _mm256_add_pd(fiy2,ty);
955 fiz2 = _mm256_add_pd(fiz2,tz);
957 fjx2 = _mm256_add_pd(fjx2,tx);
958 fjy2 = _mm256_add_pd(fjy2,ty);
959 fjz2 = _mm256_add_pd(fjz2,tz);
961 /**************************
962 * CALCULATE INTERACTIONS *
963 **************************/
965 /* REACTION-FIELD ELECTROSTATICS */
966 velec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_add_pd(rinv23,_mm256_mul_pd(krf,rsq23)),crf));
967 felec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
969 /* Update potential sum for this i atom from the interaction with this j atom. */
970 velec = _mm256_andnot_pd(dummy_mask,velec);
971 velecsum = _mm256_add_pd(velecsum,velec);
975 fscal = _mm256_andnot_pd(dummy_mask,fscal);
977 /* Calculate temporary vectorial force */
978 tx = _mm256_mul_pd(fscal,dx23);
979 ty = _mm256_mul_pd(fscal,dy23);
980 tz = _mm256_mul_pd(fscal,dz23);
982 /* Update vectorial force */
983 fix2 = _mm256_add_pd(fix2,tx);
984 fiy2 = _mm256_add_pd(fiy2,ty);
985 fiz2 = _mm256_add_pd(fiz2,tz);
987 fjx3 = _mm256_add_pd(fjx3,tx);
988 fjy3 = _mm256_add_pd(fjy3,ty);
989 fjz3 = _mm256_add_pd(fjz3,tz);
991 /**************************
992 * CALCULATE INTERACTIONS *
993 **************************/
995 /* REACTION-FIELD ELECTROSTATICS */
996 velec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_add_pd(rinv31,_mm256_mul_pd(krf,rsq31)),crf));
997 felec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
999 /* Update potential sum for this i atom from the interaction with this j atom. */
1000 velec = _mm256_andnot_pd(dummy_mask,velec);
1001 velecsum = _mm256_add_pd(velecsum,velec);
1005 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1007 /* Calculate temporary vectorial force */
1008 tx = _mm256_mul_pd(fscal,dx31);
1009 ty = _mm256_mul_pd(fscal,dy31);
1010 tz = _mm256_mul_pd(fscal,dz31);
1012 /* Update vectorial force */
1013 fix3 = _mm256_add_pd(fix3,tx);
1014 fiy3 = _mm256_add_pd(fiy3,ty);
1015 fiz3 = _mm256_add_pd(fiz3,tz);
1017 fjx1 = _mm256_add_pd(fjx1,tx);
1018 fjy1 = _mm256_add_pd(fjy1,ty);
1019 fjz1 = _mm256_add_pd(fjz1,tz);
1021 /**************************
1022 * CALCULATE INTERACTIONS *
1023 **************************/
1025 /* REACTION-FIELD ELECTROSTATICS */
1026 velec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_add_pd(rinv32,_mm256_mul_pd(krf,rsq32)),crf));
1027 felec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
1029 /* Update potential sum for this i atom from the interaction with this j atom. */
1030 velec = _mm256_andnot_pd(dummy_mask,velec);
1031 velecsum = _mm256_add_pd(velecsum,velec);
1035 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1037 /* Calculate temporary vectorial force */
1038 tx = _mm256_mul_pd(fscal,dx32);
1039 ty = _mm256_mul_pd(fscal,dy32);
1040 tz = _mm256_mul_pd(fscal,dz32);
1042 /* Update vectorial force */
1043 fix3 = _mm256_add_pd(fix3,tx);
1044 fiy3 = _mm256_add_pd(fiy3,ty);
1045 fiz3 = _mm256_add_pd(fiz3,tz);
1047 fjx2 = _mm256_add_pd(fjx2,tx);
1048 fjy2 = _mm256_add_pd(fjy2,ty);
1049 fjz2 = _mm256_add_pd(fjz2,tz);
1051 /**************************
1052 * CALCULATE INTERACTIONS *
1053 **************************/
1055 /* REACTION-FIELD ELECTROSTATICS */
1056 velec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_add_pd(rinv33,_mm256_mul_pd(krf,rsq33)),crf));
1057 felec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
1059 /* Update potential sum for this i atom from the interaction with this j atom. */
1060 velec = _mm256_andnot_pd(dummy_mask,velec);
1061 velecsum = _mm256_add_pd(velecsum,velec);
1065 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1067 /* Calculate temporary vectorial force */
1068 tx = _mm256_mul_pd(fscal,dx33);
1069 ty = _mm256_mul_pd(fscal,dy33);
1070 tz = _mm256_mul_pd(fscal,dz33);
1072 /* Update vectorial force */
1073 fix3 = _mm256_add_pd(fix3,tx);
1074 fiy3 = _mm256_add_pd(fiy3,ty);
1075 fiz3 = _mm256_add_pd(fiz3,tz);
1077 fjx3 = _mm256_add_pd(fjx3,tx);
1078 fjy3 = _mm256_add_pd(fjy3,ty);
1079 fjz3 = _mm256_add_pd(fjz3,tz);
1081 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1082 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1083 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1084 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1086 gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1087 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1088 fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1090 /* Inner loop uses 348 flops */
1093 /* End of innermost loop */
1095 gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1096 f+i_coord_offset,fshift+i_shift_offset);
1099 /* Update potential energies */
1100 gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1101 gmx_mm256_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1103 /* Increment number of inner iterations */
1104 inneriter += j_index_end - j_index_start;
1106 /* Outer loop uses 26 flops */
1109 /* Increment number of outer iterations */
1112 /* Update outer/inner flops */
1114 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*348);
1117 * Gromacs nonbonded kernel: nb_kernel_ElecRF_VdwCSTab_GeomW4W4_F_avx_256_double
1118 * Electrostatics interaction: ReactionField
1119 * VdW interaction: CubicSplineTable
1120 * Geometry: Water4-Water4
1121 * Calculate force/pot: Force
1124 nb_kernel_ElecRF_VdwCSTab_GeomW4W4_F_avx_256_double
1125 (t_nblist * gmx_restrict nlist,
1126 rvec * gmx_restrict xx,
1127 rvec * gmx_restrict ff,
1128 t_forcerec * gmx_restrict fr,
1129 t_mdatoms * gmx_restrict mdatoms,
1130 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1131 t_nrnb * gmx_restrict nrnb)
1133 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1134 * just 0 for non-waters.
1135 * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
1136 * jnr indices corresponding to data put in the four positions in the SIMD register.
1138 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1139 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1140 int jnrA,jnrB,jnrC,jnrD;
1141 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1142 int jnrlistE,jnrlistF,jnrlistG,jnrlistH;
1143 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1144 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1145 real rcutoff_scalar;
1146 real *shiftvec,*fshift,*x,*f;
1147 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1148 real scratch[4*DIM];
1149 __m256d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1150 real * vdwioffsetptr0;
1151 __m256d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1152 real * vdwioffsetptr1;
1153 __m256d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1154 real * vdwioffsetptr2;
1155 __m256d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1156 real * vdwioffsetptr3;
1157 __m256d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1158 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1159 __m256d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1160 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1161 __m256d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1162 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1163 __m256d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1164 int vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1165 __m256d jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1166 __m256d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1167 __m256d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1168 __m256d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1169 __m256d dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1170 __m256d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1171 __m256d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1172 __m256d dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1173 __m256d dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1174 __m256d dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1175 __m256d dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1176 __m256d velec,felec,velecsum,facel,crf,krf,krf2;
1179 __m256d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1182 __m256d one_sixth = _mm256_set1_pd(1.0/6.0);
1183 __m256d one_twelfth = _mm256_set1_pd(1.0/12.0);
1185 __m128i ifour = _mm_set1_epi32(4);
1186 __m256d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1188 __m256d dummy_mask,cutoff_mask;
1189 __m128 tmpmask0,tmpmask1;
1190 __m256d signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
1191 __m256d one = _mm256_set1_pd(1.0);
1192 __m256d two = _mm256_set1_pd(2.0);
1198 jindex = nlist->jindex;
1200 shiftidx = nlist->shift;
1202 shiftvec = fr->shift_vec[0];
1203 fshift = fr->fshift[0];
1204 facel = _mm256_set1_pd(fr->epsfac);
1205 charge = mdatoms->chargeA;
1206 krf = _mm256_set1_pd(fr->ic->k_rf);
1207 krf2 = _mm256_set1_pd(fr->ic->k_rf*2.0);
1208 crf = _mm256_set1_pd(fr->ic->c_rf);
1209 nvdwtype = fr->ntype;
1210 vdwparam = fr->nbfp;
1211 vdwtype = mdatoms->typeA;
1213 vftab = kernel_data->table_vdw->data;
1214 vftabscale = _mm256_set1_pd(kernel_data->table_vdw->scale);
1216 /* Setup water-specific parameters */
1217 inr = nlist->iinr[0];
1218 iq1 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1219 iq2 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1220 iq3 = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
1221 vdwioffsetptr0 = vdwparam+2*nvdwtype*vdwtype[inr+0];
1223 jq1 = _mm256_set1_pd(charge[inr+1]);
1224 jq2 = _mm256_set1_pd(charge[inr+2]);
1225 jq3 = _mm256_set1_pd(charge[inr+3]);
1226 vdwjidx0A = 2*vdwtype[inr+0];
1227 c6_00 = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
1228 c12_00 = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
1229 qq11 = _mm256_mul_pd(iq1,jq1);
1230 qq12 = _mm256_mul_pd(iq1,jq2);
1231 qq13 = _mm256_mul_pd(iq1,jq3);
1232 qq21 = _mm256_mul_pd(iq2,jq1);
1233 qq22 = _mm256_mul_pd(iq2,jq2);
1234 qq23 = _mm256_mul_pd(iq2,jq3);
1235 qq31 = _mm256_mul_pd(iq3,jq1);
1236 qq32 = _mm256_mul_pd(iq3,jq2);
1237 qq33 = _mm256_mul_pd(iq3,jq3);
1239 /* Avoid stupid compiler warnings */
1240 jnrA = jnrB = jnrC = jnrD = 0;
1241 j_coord_offsetA = 0;
1242 j_coord_offsetB = 0;
1243 j_coord_offsetC = 0;
1244 j_coord_offsetD = 0;
1249 for(iidx=0;iidx<4*DIM;iidx++)
1251 scratch[iidx] = 0.0;
1254 /* Start outer loop over neighborlists */
1255 for(iidx=0; iidx<nri; iidx++)
1257 /* Load shift vector for this list */
1258 i_shift_offset = DIM*shiftidx[iidx];
1260 /* Load limits for loop over neighbors */
1261 j_index_start = jindex[iidx];
1262 j_index_end = jindex[iidx+1];
1264 /* Get outer coordinate index */
1266 i_coord_offset = DIM*inr;
1268 /* Load i particle coords and add shift vector */
1269 gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1270 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1272 fix0 = _mm256_setzero_pd();
1273 fiy0 = _mm256_setzero_pd();
1274 fiz0 = _mm256_setzero_pd();
1275 fix1 = _mm256_setzero_pd();
1276 fiy1 = _mm256_setzero_pd();
1277 fiz1 = _mm256_setzero_pd();
1278 fix2 = _mm256_setzero_pd();
1279 fiy2 = _mm256_setzero_pd();
1280 fiz2 = _mm256_setzero_pd();
1281 fix3 = _mm256_setzero_pd();
1282 fiy3 = _mm256_setzero_pd();
1283 fiz3 = _mm256_setzero_pd();
1285 /* Start inner kernel loop */
1286 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1289 /* Get j neighbor index, and coordinate index */
1291 jnrB = jjnr[jidx+1];
1292 jnrC = jjnr[jidx+2];
1293 jnrD = jjnr[jidx+3];
1294 j_coord_offsetA = DIM*jnrA;
1295 j_coord_offsetB = DIM*jnrB;
1296 j_coord_offsetC = DIM*jnrC;
1297 j_coord_offsetD = DIM*jnrD;
1299 /* load j atom coordinates */
1300 gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1301 x+j_coord_offsetC,x+j_coord_offsetD,
1302 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1303 &jy2,&jz2,&jx3,&jy3,&jz3);
1305 /* Calculate displacement vector */
1306 dx00 = _mm256_sub_pd(ix0,jx0);
1307 dy00 = _mm256_sub_pd(iy0,jy0);
1308 dz00 = _mm256_sub_pd(iz0,jz0);
1309 dx11 = _mm256_sub_pd(ix1,jx1);
1310 dy11 = _mm256_sub_pd(iy1,jy1);
1311 dz11 = _mm256_sub_pd(iz1,jz1);
1312 dx12 = _mm256_sub_pd(ix1,jx2);
1313 dy12 = _mm256_sub_pd(iy1,jy2);
1314 dz12 = _mm256_sub_pd(iz1,jz2);
1315 dx13 = _mm256_sub_pd(ix1,jx3);
1316 dy13 = _mm256_sub_pd(iy1,jy3);
1317 dz13 = _mm256_sub_pd(iz1,jz3);
1318 dx21 = _mm256_sub_pd(ix2,jx1);
1319 dy21 = _mm256_sub_pd(iy2,jy1);
1320 dz21 = _mm256_sub_pd(iz2,jz1);
1321 dx22 = _mm256_sub_pd(ix2,jx2);
1322 dy22 = _mm256_sub_pd(iy2,jy2);
1323 dz22 = _mm256_sub_pd(iz2,jz2);
1324 dx23 = _mm256_sub_pd(ix2,jx3);
1325 dy23 = _mm256_sub_pd(iy2,jy3);
1326 dz23 = _mm256_sub_pd(iz2,jz3);
1327 dx31 = _mm256_sub_pd(ix3,jx1);
1328 dy31 = _mm256_sub_pd(iy3,jy1);
1329 dz31 = _mm256_sub_pd(iz3,jz1);
1330 dx32 = _mm256_sub_pd(ix3,jx2);
1331 dy32 = _mm256_sub_pd(iy3,jy2);
1332 dz32 = _mm256_sub_pd(iz3,jz2);
1333 dx33 = _mm256_sub_pd(ix3,jx3);
1334 dy33 = _mm256_sub_pd(iy3,jy3);
1335 dz33 = _mm256_sub_pd(iz3,jz3);
1337 /* Calculate squared distance and things based on it */
1338 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1339 rsq11 = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1340 rsq12 = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1341 rsq13 = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
1342 rsq21 = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1343 rsq22 = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1344 rsq23 = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
1345 rsq31 = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
1346 rsq32 = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
1347 rsq33 = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
1349 rinv00 = gmx_mm256_invsqrt_pd(rsq00);
1350 rinv11 = gmx_mm256_invsqrt_pd(rsq11);
1351 rinv12 = gmx_mm256_invsqrt_pd(rsq12);
1352 rinv13 = gmx_mm256_invsqrt_pd(rsq13);
1353 rinv21 = gmx_mm256_invsqrt_pd(rsq21);
1354 rinv22 = gmx_mm256_invsqrt_pd(rsq22);
1355 rinv23 = gmx_mm256_invsqrt_pd(rsq23);
1356 rinv31 = gmx_mm256_invsqrt_pd(rsq31);
1357 rinv32 = gmx_mm256_invsqrt_pd(rsq32);
1358 rinv33 = gmx_mm256_invsqrt_pd(rsq33);
1360 rinvsq11 = _mm256_mul_pd(rinv11,rinv11);
1361 rinvsq12 = _mm256_mul_pd(rinv12,rinv12);
1362 rinvsq13 = _mm256_mul_pd(rinv13,rinv13);
1363 rinvsq21 = _mm256_mul_pd(rinv21,rinv21);
1364 rinvsq22 = _mm256_mul_pd(rinv22,rinv22);
1365 rinvsq23 = _mm256_mul_pd(rinv23,rinv23);
1366 rinvsq31 = _mm256_mul_pd(rinv31,rinv31);
1367 rinvsq32 = _mm256_mul_pd(rinv32,rinv32);
1368 rinvsq33 = _mm256_mul_pd(rinv33,rinv33);
1370 fjx0 = _mm256_setzero_pd();
1371 fjy0 = _mm256_setzero_pd();
1372 fjz0 = _mm256_setzero_pd();
1373 fjx1 = _mm256_setzero_pd();
1374 fjy1 = _mm256_setzero_pd();
1375 fjz1 = _mm256_setzero_pd();
1376 fjx2 = _mm256_setzero_pd();
1377 fjy2 = _mm256_setzero_pd();
1378 fjz2 = _mm256_setzero_pd();
1379 fjx3 = _mm256_setzero_pd();
1380 fjy3 = _mm256_setzero_pd();
1381 fjz3 = _mm256_setzero_pd();
1383 /**************************
1384 * CALCULATE INTERACTIONS *
1385 **************************/
1387 r00 = _mm256_mul_pd(rsq00,rinv00);
1389 /* Calculate table index by multiplying r with table scale and truncate to integer */
1390 rt = _mm256_mul_pd(r00,vftabscale);
1391 vfitab = _mm256_cvttpd_epi32(rt);
1392 vfeps = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
1393 vfitab = _mm_slli_epi32(vfitab,3);
1395 /* CUBIC SPLINE TABLE DISPERSION */
1396 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1397 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1398 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1399 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1400 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1401 Heps = _mm256_mul_pd(vfeps,H);
1402 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1403 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1404 fvdw6 = _mm256_mul_pd(c6_00,FF);
1406 /* CUBIC SPLINE TABLE REPULSION */
1407 vfitab = _mm_add_epi32(vfitab,ifour);
1408 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1409 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1410 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1411 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1412 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1413 Heps = _mm256_mul_pd(vfeps,H);
1414 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1415 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1416 fvdw12 = _mm256_mul_pd(c12_00,FF);
1417 fvdw = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
1421 /* Calculate temporary vectorial force */
1422 tx = _mm256_mul_pd(fscal,dx00);
1423 ty = _mm256_mul_pd(fscal,dy00);
1424 tz = _mm256_mul_pd(fscal,dz00);
1426 /* Update vectorial force */
1427 fix0 = _mm256_add_pd(fix0,tx);
1428 fiy0 = _mm256_add_pd(fiy0,ty);
1429 fiz0 = _mm256_add_pd(fiz0,tz);
1431 fjx0 = _mm256_add_pd(fjx0,tx);
1432 fjy0 = _mm256_add_pd(fjy0,ty);
1433 fjz0 = _mm256_add_pd(fjz0,tz);
1435 /**************************
1436 * CALCULATE INTERACTIONS *
1437 **************************/
1439 /* REACTION-FIELD ELECTROSTATICS */
1440 felec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1444 /* Calculate temporary vectorial force */
1445 tx = _mm256_mul_pd(fscal,dx11);
1446 ty = _mm256_mul_pd(fscal,dy11);
1447 tz = _mm256_mul_pd(fscal,dz11);
1449 /* Update vectorial force */
1450 fix1 = _mm256_add_pd(fix1,tx);
1451 fiy1 = _mm256_add_pd(fiy1,ty);
1452 fiz1 = _mm256_add_pd(fiz1,tz);
1454 fjx1 = _mm256_add_pd(fjx1,tx);
1455 fjy1 = _mm256_add_pd(fjy1,ty);
1456 fjz1 = _mm256_add_pd(fjz1,tz);
1458 /**************************
1459 * CALCULATE INTERACTIONS *
1460 **************************/
1462 /* REACTION-FIELD ELECTROSTATICS */
1463 felec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1467 /* Calculate temporary vectorial force */
1468 tx = _mm256_mul_pd(fscal,dx12);
1469 ty = _mm256_mul_pd(fscal,dy12);
1470 tz = _mm256_mul_pd(fscal,dz12);
1472 /* Update vectorial force */
1473 fix1 = _mm256_add_pd(fix1,tx);
1474 fiy1 = _mm256_add_pd(fiy1,ty);
1475 fiz1 = _mm256_add_pd(fiz1,tz);
1477 fjx2 = _mm256_add_pd(fjx2,tx);
1478 fjy2 = _mm256_add_pd(fjy2,ty);
1479 fjz2 = _mm256_add_pd(fjz2,tz);
1481 /**************************
1482 * CALCULATE INTERACTIONS *
1483 **************************/
1485 /* REACTION-FIELD ELECTROSTATICS */
1486 felec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
1490 /* Calculate temporary vectorial force */
1491 tx = _mm256_mul_pd(fscal,dx13);
1492 ty = _mm256_mul_pd(fscal,dy13);
1493 tz = _mm256_mul_pd(fscal,dz13);
1495 /* Update vectorial force */
1496 fix1 = _mm256_add_pd(fix1,tx);
1497 fiy1 = _mm256_add_pd(fiy1,ty);
1498 fiz1 = _mm256_add_pd(fiz1,tz);
1500 fjx3 = _mm256_add_pd(fjx3,tx);
1501 fjy3 = _mm256_add_pd(fjy3,ty);
1502 fjz3 = _mm256_add_pd(fjz3,tz);
1504 /**************************
1505 * CALCULATE INTERACTIONS *
1506 **************************/
1508 /* REACTION-FIELD ELECTROSTATICS */
1509 felec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1513 /* Calculate temporary vectorial force */
1514 tx = _mm256_mul_pd(fscal,dx21);
1515 ty = _mm256_mul_pd(fscal,dy21);
1516 tz = _mm256_mul_pd(fscal,dz21);
1518 /* Update vectorial force */
1519 fix2 = _mm256_add_pd(fix2,tx);
1520 fiy2 = _mm256_add_pd(fiy2,ty);
1521 fiz2 = _mm256_add_pd(fiz2,tz);
1523 fjx1 = _mm256_add_pd(fjx1,tx);
1524 fjy1 = _mm256_add_pd(fjy1,ty);
1525 fjz1 = _mm256_add_pd(fjz1,tz);
1527 /**************************
1528 * CALCULATE INTERACTIONS *
1529 **************************/
1531 /* REACTION-FIELD ELECTROSTATICS */
1532 felec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1536 /* Calculate temporary vectorial force */
1537 tx = _mm256_mul_pd(fscal,dx22);
1538 ty = _mm256_mul_pd(fscal,dy22);
1539 tz = _mm256_mul_pd(fscal,dz22);
1541 /* Update vectorial force */
1542 fix2 = _mm256_add_pd(fix2,tx);
1543 fiy2 = _mm256_add_pd(fiy2,ty);
1544 fiz2 = _mm256_add_pd(fiz2,tz);
1546 fjx2 = _mm256_add_pd(fjx2,tx);
1547 fjy2 = _mm256_add_pd(fjy2,ty);
1548 fjz2 = _mm256_add_pd(fjz2,tz);
1550 /**************************
1551 * CALCULATE INTERACTIONS *
1552 **************************/
1554 /* REACTION-FIELD ELECTROSTATICS */
1555 felec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
1559 /* Calculate temporary vectorial force */
1560 tx = _mm256_mul_pd(fscal,dx23);
1561 ty = _mm256_mul_pd(fscal,dy23);
1562 tz = _mm256_mul_pd(fscal,dz23);
1564 /* Update vectorial force */
1565 fix2 = _mm256_add_pd(fix2,tx);
1566 fiy2 = _mm256_add_pd(fiy2,ty);
1567 fiz2 = _mm256_add_pd(fiz2,tz);
1569 fjx3 = _mm256_add_pd(fjx3,tx);
1570 fjy3 = _mm256_add_pd(fjy3,ty);
1571 fjz3 = _mm256_add_pd(fjz3,tz);
1573 /**************************
1574 * CALCULATE INTERACTIONS *
1575 **************************/
1577 /* REACTION-FIELD ELECTROSTATICS */
1578 felec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
1582 /* Calculate temporary vectorial force */
1583 tx = _mm256_mul_pd(fscal,dx31);
1584 ty = _mm256_mul_pd(fscal,dy31);
1585 tz = _mm256_mul_pd(fscal,dz31);
1587 /* Update vectorial force */
1588 fix3 = _mm256_add_pd(fix3,tx);
1589 fiy3 = _mm256_add_pd(fiy3,ty);
1590 fiz3 = _mm256_add_pd(fiz3,tz);
1592 fjx1 = _mm256_add_pd(fjx1,tx);
1593 fjy1 = _mm256_add_pd(fjy1,ty);
1594 fjz1 = _mm256_add_pd(fjz1,tz);
1596 /**************************
1597 * CALCULATE INTERACTIONS *
1598 **************************/
1600 /* REACTION-FIELD ELECTROSTATICS */
1601 felec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
1605 /* Calculate temporary vectorial force */
1606 tx = _mm256_mul_pd(fscal,dx32);
1607 ty = _mm256_mul_pd(fscal,dy32);
1608 tz = _mm256_mul_pd(fscal,dz32);
1610 /* Update vectorial force */
1611 fix3 = _mm256_add_pd(fix3,tx);
1612 fiy3 = _mm256_add_pd(fiy3,ty);
1613 fiz3 = _mm256_add_pd(fiz3,tz);
1615 fjx2 = _mm256_add_pd(fjx2,tx);
1616 fjy2 = _mm256_add_pd(fjy2,ty);
1617 fjz2 = _mm256_add_pd(fjz2,tz);
1619 /**************************
1620 * CALCULATE INTERACTIONS *
1621 **************************/
1623 /* REACTION-FIELD ELECTROSTATICS */
1624 felec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
1628 /* Calculate temporary vectorial force */
1629 tx = _mm256_mul_pd(fscal,dx33);
1630 ty = _mm256_mul_pd(fscal,dy33);
1631 tz = _mm256_mul_pd(fscal,dz33);
1633 /* Update vectorial force */
1634 fix3 = _mm256_add_pd(fix3,tx);
1635 fiy3 = _mm256_add_pd(fiy3,ty);
1636 fiz3 = _mm256_add_pd(fiz3,tz);
1638 fjx3 = _mm256_add_pd(fjx3,tx);
1639 fjy3 = _mm256_add_pd(fjy3,ty);
1640 fjz3 = _mm256_add_pd(fjz3,tz);
1642 fjptrA = f+j_coord_offsetA;
1643 fjptrB = f+j_coord_offsetB;
1644 fjptrC = f+j_coord_offsetC;
1645 fjptrD = f+j_coord_offsetD;
1647 gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1648 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1649 fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1651 /* Inner loop uses 294 flops */
1654 if(jidx<j_index_end)
1657 /* Get j neighbor index, and coordinate index */
1658 jnrlistA = jjnr[jidx];
1659 jnrlistB = jjnr[jidx+1];
1660 jnrlistC = jjnr[jidx+2];
1661 jnrlistD = jjnr[jidx+3];
1662 /* Sign of each element will be negative for non-real atoms.
1663 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1664 * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
1666 tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1668 tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
1669 tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
1670 dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
1672 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
1673 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
1674 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
1675 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
1676 j_coord_offsetA = DIM*jnrA;
1677 j_coord_offsetB = DIM*jnrB;
1678 j_coord_offsetC = DIM*jnrC;
1679 j_coord_offsetD = DIM*jnrD;
1681 /* load j atom coordinates */
1682 gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1683 x+j_coord_offsetC,x+j_coord_offsetD,
1684 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1685 &jy2,&jz2,&jx3,&jy3,&jz3);
1687 /* Calculate displacement vector */
1688 dx00 = _mm256_sub_pd(ix0,jx0);
1689 dy00 = _mm256_sub_pd(iy0,jy0);
1690 dz00 = _mm256_sub_pd(iz0,jz0);
1691 dx11 = _mm256_sub_pd(ix1,jx1);
1692 dy11 = _mm256_sub_pd(iy1,jy1);
1693 dz11 = _mm256_sub_pd(iz1,jz1);
1694 dx12 = _mm256_sub_pd(ix1,jx2);
1695 dy12 = _mm256_sub_pd(iy1,jy2);
1696 dz12 = _mm256_sub_pd(iz1,jz2);
1697 dx13 = _mm256_sub_pd(ix1,jx3);
1698 dy13 = _mm256_sub_pd(iy1,jy3);
1699 dz13 = _mm256_sub_pd(iz1,jz3);
1700 dx21 = _mm256_sub_pd(ix2,jx1);
1701 dy21 = _mm256_sub_pd(iy2,jy1);
1702 dz21 = _mm256_sub_pd(iz2,jz1);
1703 dx22 = _mm256_sub_pd(ix2,jx2);
1704 dy22 = _mm256_sub_pd(iy2,jy2);
1705 dz22 = _mm256_sub_pd(iz2,jz2);
1706 dx23 = _mm256_sub_pd(ix2,jx3);
1707 dy23 = _mm256_sub_pd(iy2,jy3);
1708 dz23 = _mm256_sub_pd(iz2,jz3);
1709 dx31 = _mm256_sub_pd(ix3,jx1);
1710 dy31 = _mm256_sub_pd(iy3,jy1);
1711 dz31 = _mm256_sub_pd(iz3,jz1);
1712 dx32 = _mm256_sub_pd(ix3,jx2);
1713 dy32 = _mm256_sub_pd(iy3,jy2);
1714 dz32 = _mm256_sub_pd(iz3,jz2);
1715 dx33 = _mm256_sub_pd(ix3,jx3);
1716 dy33 = _mm256_sub_pd(iy3,jy3);
1717 dz33 = _mm256_sub_pd(iz3,jz3);
1719 /* Calculate squared distance and things based on it */
1720 rsq00 = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1721 rsq11 = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1722 rsq12 = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1723 rsq13 = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
1724 rsq21 = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1725 rsq22 = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1726 rsq23 = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
1727 rsq31 = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
1728 rsq32 = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
1729 rsq33 = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
1731 rinv00 = gmx_mm256_invsqrt_pd(rsq00);
1732 rinv11 = gmx_mm256_invsqrt_pd(rsq11);
1733 rinv12 = gmx_mm256_invsqrt_pd(rsq12);
1734 rinv13 = gmx_mm256_invsqrt_pd(rsq13);
1735 rinv21 = gmx_mm256_invsqrt_pd(rsq21);
1736 rinv22 = gmx_mm256_invsqrt_pd(rsq22);
1737 rinv23 = gmx_mm256_invsqrt_pd(rsq23);
1738 rinv31 = gmx_mm256_invsqrt_pd(rsq31);
1739 rinv32 = gmx_mm256_invsqrt_pd(rsq32);
1740 rinv33 = gmx_mm256_invsqrt_pd(rsq33);
1742 rinvsq11 = _mm256_mul_pd(rinv11,rinv11);
1743 rinvsq12 = _mm256_mul_pd(rinv12,rinv12);
1744 rinvsq13 = _mm256_mul_pd(rinv13,rinv13);
1745 rinvsq21 = _mm256_mul_pd(rinv21,rinv21);
1746 rinvsq22 = _mm256_mul_pd(rinv22,rinv22);
1747 rinvsq23 = _mm256_mul_pd(rinv23,rinv23);
1748 rinvsq31 = _mm256_mul_pd(rinv31,rinv31);
1749 rinvsq32 = _mm256_mul_pd(rinv32,rinv32);
1750 rinvsq33 = _mm256_mul_pd(rinv33,rinv33);
1752 fjx0 = _mm256_setzero_pd();
1753 fjy0 = _mm256_setzero_pd();
1754 fjz0 = _mm256_setzero_pd();
1755 fjx1 = _mm256_setzero_pd();
1756 fjy1 = _mm256_setzero_pd();
1757 fjz1 = _mm256_setzero_pd();
1758 fjx2 = _mm256_setzero_pd();
1759 fjy2 = _mm256_setzero_pd();
1760 fjz2 = _mm256_setzero_pd();
1761 fjx3 = _mm256_setzero_pd();
1762 fjy3 = _mm256_setzero_pd();
1763 fjz3 = _mm256_setzero_pd();
1765 /**************************
1766 * CALCULATE INTERACTIONS *
1767 **************************/
1769 r00 = _mm256_mul_pd(rsq00,rinv00);
1770 r00 = _mm256_andnot_pd(dummy_mask,r00);
1772 /* Calculate table index by multiplying r with table scale and truncate to integer */
1773 rt = _mm256_mul_pd(r00,vftabscale);
1774 vfitab = _mm256_cvttpd_epi32(rt);
1775 vfeps = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
1776 vfitab = _mm_slli_epi32(vfitab,3);
1778 /* CUBIC SPLINE TABLE DISPERSION */
1779 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1780 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1781 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1782 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1783 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1784 Heps = _mm256_mul_pd(vfeps,H);
1785 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1786 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1787 fvdw6 = _mm256_mul_pd(c6_00,FF);
1789 /* CUBIC SPLINE TABLE REPULSION */
1790 vfitab = _mm_add_epi32(vfitab,ifour);
1791 Y = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1792 F = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1793 G = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1794 H = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1795 GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1796 Heps = _mm256_mul_pd(vfeps,H);
1797 Fp = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1798 FF = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1799 fvdw12 = _mm256_mul_pd(c12_00,FF);
1800 fvdw = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
1804 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1806 /* Calculate temporary vectorial force */
1807 tx = _mm256_mul_pd(fscal,dx00);
1808 ty = _mm256_mul_pd(fscal,dy00);
1809 tz = _mm256_mul_pd(fscal,dz00);
1811 /* Update vectorial force */
1812 fix0 = _mm256_add_pd(fix0,tx);
1813 fiy0 = _mm256_add_pd(fiy0,ty);
1814 fiz0 = _mm256_add_pd(fiz0,tz);
1816 fjx0 = _mm256_add_pd(fjx0,tx);
1817 fjy0 = _mm256_add_pd(fjy0,ty);
1818 fjz0 = _mm256_add_pd(fjz0,tz);
1820 /**************************
1821 * CALCULATE INTERACTIONS *
1822 **************************/
1824 /* REACTION-FIELD ELECTROSTATICS */
1825 felec = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1829 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1831 /* Calculate temporary vectorial force */
1832 tx = _mm256_mul_pd(fscal,dx11);
1833 ty = _mm256_mul_pd(fscal,dy11);
1834 tz = _mm256_mul_pd(fscal,dz11);
1836 /* Update vectorial force */
1837 fix1 = _mm256_add_pd(fix1,tx);
1838 fiy1 = _mm256_add_pd(fiy1,ty);
1839 fiz1 = _mm256_add_pd(fiz1,tz);
1841 fjx1 = _mm256_add_pd(fjx1,tx);
1842 fjy1 = _mm256_add_pd(fjy1,ty);
1843 fjz1 = _mm256_add_pd(fjz1,tz);
1845 /**************************
1846 * CALCULATE INTERACTIONS *
1847 **************************/
1849 /* REACTION-FIELD ELECTROSTATICS */
1850 felec = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1854 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1856 /* Calculate temporary vectorial force */
1857 tx = _mm256_mul_pd(fscal,dx12);
1858 ty = _mm256_mul_pd(fscal,dy12);
1859 tz = _mm256_mul_pd(fscal,dz12);
1861 /* Update vectorial force */
1862 fix1 = _mm256_add_pd(fix1,tx);
1863 fiy1 = _mm256_add_pd(fiy1,ty);
1864 fiz1 = _mm256_add_pd(fiz1,tz);
1866 fjx2 = _mm256_add_pd(fjx2,tx);
1867 fjy2 = _mm256_add_pd(fjy2,ty);
1868 fjz2 = _mm256_add_pd(fjz2,tz);
1870 /**************************
1871 * CALCULATE INTERACTIONS *
1872 **************************/
1874 /* REACTION-FIELD ELECTROSTATICS */
1875 felec = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
1879 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1881 /* Calculate temporary vectorial force */
1882 tx = _mm256_mul_pd(fscal,dx13);
1883 ty = _mm256_mul_pd(fscal,dy13);
1884 tz = _mm256_mul_pd(fscal,dz13);
1886 /* Update vectorial force */
1887 fix1 = _mm256_add_pd(fix1,tx);
1888 fiy1 = _mm256_add_pd(fiy1,ty);
1889 fiz1 = _mm256_add_pd(fiz1,tz);
1891 fjx3 = _mm256_add_pd(fjx3,tx);
1892 fjy3 = _mm256_add_pd(fjy3,ty);
1893 fjz3 = _mm256_add_pd(fjz3,tz);
1895 /**************************
1896 * CALCULATE INTERACTIONS *
1897 **************************/
1899 /* REACTION-FIELD ELECTROSTATICS */
1900 felec = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1904 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1906 /* Calculate temporary vectorial force */
1907 tx = _mm256_mul_pd(fscal,dx21);
1908 ty = _mm256_mul_pd(fscal,dy21);
1909 tz = _mm256_mul_pd(fscal,dz21);
1911 /* Update vectorial force */
1912 fix2 = _mm256_add_pd(fix2,tx);
1913 fiy2 = _mm256_add_pd(fiy2,ty);
1914 fiz2 = _mm256_add_pd(fiz2,tz);
1916 fjx1 = _mm256_add_pd(fjx1,tx);
1917 fjy1 = _mm256_add_pd(fjy1,ty);
1918 fjz1 = _mm256_add_pd(fjz1,tz);
1920 /**************************
1921 * CALCULATE INTERACTIONS *
1922 **************************/
1924 /* REACTION-FIELD ELECTROSTATICS */
1925 felec = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1929 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1931 /* Calculate temporary vectorial force */
1932 tx = _mm256_mul_pd(fscal,dx22);
1933 ty = _mm256_mul_pd(fscal,dy22);
1934 tz = _mm256_mul_pd(fscal,dz22);
1936 /* Update vectorial force */
1937 fix2 = _mm256_add_pd(fix2,tx);
1938 fiy2 = _mm256_add_pd(fiy2,ty);
1939 fiz2 = _mm256_add_pd(fiz2,tz);
1941 fjx2 = _mm256_add_pd(fjx2,tx);
1942 fjy2 = _mm256_add_pd(fjy2,ty);
1943 fjz2 = _mm256_add_pd(fjz2,tz);
1945 /**************************
1946 * CALCULATE INTERACTIONS *
1947 **************************/
1949 /* REACTION-FIELD ELECTROSTATICS */
1950 felec = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
1954 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1956 /* Calculate temporary vectorial force */
1957 tx = _mm256_mul_pd(fscal,dx23);
1958 ty = _mm256_mul_pd(fscal,dy23);
1959 tz = _mm256_mul_pd(fscal,dz23);
1961 /* Update vectorial force */
1962 fix2 = _mm256_add_pd(fix2,tx);
1963 fiy2 = _mm256_add_pd(fiy2,ty);
1964 fiz2 = _mm256_add_pd(fiz2,tz);
1966 fjx3 = _mm256_add_pd(fjx3,tx);
1967 fjy3 = _mm256_add_pd(fjy3,ty);
1968 fjz3 = _mm256_add_pd(fjz3,tz);
1970 /**************************
1971 * CALCULATE INTERACTIONS *
1972 **************************/
1974 /* REACTION-FIELD ELECTROSTATICS */
1975 felec = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
1979 fscal = _mm256_andnot_pd(dummy_mask,fscal);
1981 /* Calculate temporary vectorial force */
1982 tx = _mm256_mul_pd(fscal,dx31);
1983 ty = _mm256_mul_pd(fscal,dy31);
1984 tz = _mm256_mul_pd(fscal,dz31);
1986 /* Update vectorial force */
1987 fix3 = _mm256_add_pd(fix3,tx);
1988 fiy3 = _mm256_add_pd(fiy3,ty);
1989 fiz3 = _mm256_add_pd(fiz3,tz);
1991 fjx1 = _mm256_add_pd(fjx1,tx);
1992 fjy1 = _mm256_add_pd(fjy1,ty);
1993 fjz1 = _mm256_add_pd(fjz1,tz);
1995 /**************************
1996 * CALCULATE INTERACTIONS *
1997 **************************/
1999 /* REACTION-FIELD ELECTROSTATICS */
2000 felec = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
2004 fscal = _mm256_andnot_pd(dummy_mask,fscal);
2006 /* Calculate temporary vectorial force */
2007 tx = _mm256_mul_pd(fscal,dx32);
2008 ty = _mm256_mul_pd(fscal,dy32);
2009 tz = _mm256_mul_pd(fscal,dz32);
2011 /* Update vectorial force */
2012 fix3 = _mm256_add_pd(fix3,tx);
2013 fiy3 = _mm256_add_pd(fiy3,ty);
2014 fiz3 = _mm256_add_pd(fiz3,tz);
2016 fjx2 = _mm256_add_pd(fjx2,tx);
2017 fjy2 = _mm256_add_pd(fjy2,ty);
2018 fjz2 = _mm256_add_pd(fjz2,tz);
2020 /**************************
2021 * CALCULATE INTERACTIONS *
2022 **************************/
2024 /* REACTION-FIELD ELECTROSTATICS */
2025 felec = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
2029 fscal = _mm256_andnot_pd(dummy_mask,fscal);
2031 /* Calculate temporary vectorial force */
2032 tx = _mm256_mul_pd(fscal,dx33);
2033 ty = _mm256_mul_pd(fscal,dy33);
2034 tz = _mm256_mul_pd(fscal,dz33);
2036 /* Update vectorial force */
2037 fix3 = _mm256_add_pd(fix3,tx);
2038 fiy3 = _mm256_add_pd(fiy3,ty);
2039 fiz3 = _mm256_add_pd(fiz3,tz);
2041 fjx3 = _mm256_add_pd(fjx3,tx);
2042 fjy3 = _mm256_add_pd(fjy3,ty);
2043 fjz3 = _mm256_add_pd(fjz3,tz);
2045 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2046 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2047 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2048 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2050 gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2051 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2052 fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2054 /* Inner loop uses 295 flops */
2057 /* End of innermost loop */
2059 gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2060 f+i_coord_offset,fshift+i_shift_offset);
2062 /* Increment number of inner iterations */
2063 inneriter += j_index_end - j_index_start;
2065 /* Outer loop uses 24 flops */
2068 /* Increment number of outer iterations */
2071 /* Update outer/inner flops */
2073 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*295);