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 sse2_double kernel generator.
44 #include "../nb_kernel.h"
45 #include "gromacs/legacyheaders/types/simple.h"
46 #include "gromacs/math/vec.h"
47 #include "gromacs/legacyheaders/nrnb.h"
49 #include "gromacs/simd/math_x86_sse2_double.h"
50 #include "kernelutil_x86_sse2_double.h"
53 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW4W4_VF_sse2_double
54 * Electrostatics interaction: Ewald
55 * VdW interaction: CubicSplineTable
56 * Geometry: Water4-Water4
57 * Calculate force/pot: PotentialAndForce
60 nb_kernel_ElecEw_VdwCSTab_GeomW4W4_VF_sse2_double
61 (t_nblist * gmx_restrict nlist,
62 rvec * gmx_restrict xx,
63 rvec * gmx_restrict ff,
64 t_forcerec * gmx_restrict fr,
65 t_mdatoms * gmx_restrict mdatoms,
66 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
67 t_nrnb * gmx_restrict nrnb)
69 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
70 * just 0 for non-waters.
71 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
72 * jnr indices corresponding to data put in the four positions in the SIMD register.
74 int i_shift_offset,i_coord_offset,outeriter,inneriter;
75 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
77 int j_coord_offsetA,j_coord_offsetB;
78 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
80 real *shiftvec,*fshift,*x,*f;
81 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
83 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
85 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
87 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
89 __m128d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
90 int vdwjidx0A,vdwjidx0B;
91 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
92 int vdwjidx1A,vdwjidx1B;
93 __m128d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
94 int vdwjidx2A,vdwjidx2B;
95 __m128d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
96 int vdwjidx3A,vdwjidx3B;
97 __m128d jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
98 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
99 __m128d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
100 __m128d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
101 __m128d dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
102 __m128d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
103 __m128d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
104 __m128d dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
105 __m128d dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
106 __m128d dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
107 __m128d dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
108 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
111 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
114 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
115 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
117 __m128i ifour = _mm_set1_epi32(4);
118 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
121 __m128d ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
123 __m128d dummy_mask,cutoff_mask;
124 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
125 __m128d one = _mm_set1_pd(1.0);
126 __m128d two = _mm_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 = _mm_set1_pd(fr->epsfac);
139 charge = mdatoms->chargeA;
140 nvdwtype = fr->ntype;
142 vdwtype = mdatoms->typeA;
144 vftab = kernel_data->table_vdw->data;
145 vftabscale = _mm_set1_pd(kernel_data->table_vdw->scale);
147 sh_ewald = _mm_set1_pd(fr->ic->sh_ewald);
148 ewtab = fr->ic->tabq_coul_FDV0;
149 ewtabscale = _mm_set1_pd(fr->ic->tabq_scale);
150 ewtabhalfspace = _mm_set1_pd(0.5/fr->ic->tabq_scale);
152 /* Setup water-specific parameters */
153 inr = nlist->iinr[0];
154 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
155 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
156 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
157 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
159 jq1 = _mm_set1_pd(charge[inr+1]);
160 jq2 = _mm_set1_pd(charge[inr+2]);
161 jq3 = _mm_set1_pd(charge[inr+3]);
162 vdwjidx0A = 2*vdwtype[inr+0];
163 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
164 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
165 qq11 = _mm_mul_pd(iq1,jq1);
166 qq12 = _mm_mul_pd(iq1,jq2);
167 qq13 = _mm_mul_pd(iq1,jq3);
168 qq21 = _mm_mul_pd(iq2,jq1);
169 qq22 = _mm_mul_pd(iq2,jq2);
170 qq23 = _mm_mul_pd(iq2,jq3);
171 qq31 = _mm_mul_pd(iq3,jq1);
172 qq32 = _mm_mul_pd(iq3,jq2);
173 qq33 = _mm_mul_pd(iq3,jq3);
175 /* Avoid stupid compiler warnings */
183 /* Start outer loop over neighborlists */
184 for(iidx=0; iidx<nri; iidx++)
186 /* Load shift vector for this list */
187 i_shift_offset = DIM*shiftidx[iidx];
189 /* Load limits for loop over neighbors */
190 j_index_start = jindex[iidx];
191 j_index_end = jindex[iidx+1];
193 /* Get outer coordinate index */
195 i_coord_offset = DIM*inr;
197 /* Load i particle coords and add shift vector */
198 gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
199 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
201 fix0 = _mm_setzero_pd();
202 fiy0 = _mm_setzero_pd();
203 fiz0 = _mm_setzero_pd();
204 fix1 = _mm_setzero_pd();
205 fiy1 = _mm_setzero_pd();
206 fiz1 = _mm_setzero_pd();
207 fix2 = _mm_setzero_pd();
208 fiy2 = _mm_setzero_pd();
209 fiz2 = _mm_setzero_pd();
210 fix3 = _mm_setzero_pd();
211 fiy3 = _mm_setzero_pd();
212 fiz3 = _mm_setzero_pd();
214 /* Reset potential sums */
215 velecsum = _mm_setzero_pd();
216 vvdwsum = _mm_setzero_pd();
218 /* Start inner kernel loop */
219 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
222 /* Get j neighbor index, and coordinate index */
225 j_coord_offsetA = DIM*jnrA;
226 j_coord_offsetB = DIM*jnrB;
228 /* load j atom coordinates */
229 gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
230 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
231 &jy2,&jz2,&jx3,&jy3,&jz3);
233 /* Calculate displacement vector */
234 dx00 = _mm_sub_pd(ix0,jx0);
235 dy00 = _mm_sub_pd(iy0,jy0);
236 dz00 = _mm_sub_pd(iz0,jz0);
237 dx11 = _mm_sub_pd(ix1,jx1);
238 dy11 = _mm_sub_pd(iy1,jy1);
239 dz11 = _mm_sub_pd(iz1,jz1);
240 dx12 = _mm_sub_pd(ix1,jx2);
241 dy12 = _mm_sub_pd(iy1,jy2);
242 dz12 = _mm_sub_pd(iz1,jz2);
243 dx13 = _mm_sub_pd(ix1,jx3);
244 dy13 = _mm_sub_pd(iy1,jy3);
245 dz13 = _mm_sub_pd(iz1,jz3);
246 dx21 = _mm_sub_pd(ix2,jx1);
247 dy21 = _mm_sub_pd(iy2,jy1);
248 dz21 = _mm_sub_pd(iz2,jz1);
249 dx22 = _mm_sub_pd(ix2,jx2);
250 dy22 = _mm_sub_pd(iy2,jy2);
251 dz22 = _mm_sub_pd(iz2,jz2);
252 dx23 = _mm_sub_pd(ix2,jx3);
253 dy23 = _mm_sub_pd(iy2,jy3);
254 dz23 = _mm_sub_pd(iz2,jz3);
255 dx31 = _mm_sub_pd(ix3,jx1);
256 dy31 = _mm_sub_pd(iy3,jy1);
257 dz31 = _mm_sub_pd(iz3,jz1);
258 dx32 = _mm_sub_pd(ix3,jx2);
259 dy32 = _mm_sub_pd(iy3,jy2);
260 dz32 = _mm_sub_pd(iz3,jz2);
261 dx33 = _mm_sub_pd(ix3,jx3);
262 dy33 = _mm_sub_pd(iy3,jy3);
263 dz33 = _mm_sub_pd(iz3,jz3);
265 /* Calculate squared distance and things based on it */
266 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
267 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
268 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
269 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
270 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
271 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
272 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
273 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
274 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
275 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
277 rinv00 = gmx_mm_invsqrt_pd(rsq00);
278 rinv11 = gmx_mm_invsqrt_pd(rsq11);
279 rinv12 = gmx_mm_invsqrt_pd(rsq12);
280 rinv13 = gmx_mm_invsqrt_pd(rsq13);
281 rinv21 = gmx_mm_invsqrt_pd(rsq21);
282 rinv22 = gmx_mm_invsqrt_pd(rsq22);
283 rinv23 = gmx_mm_invsqrt_pd(rsq23);
284 rinv31 = gmx_mm_invsqrt_pd(rsq31);
285 rinv32 = gmx_mm_invsqrt_pd(rsq32);
286 rinv33 = gmx_mm_invsqrt_pd(rsq33);
288 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
289 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
290 rinvsq13 = _mm_mul_pd(rinv13,rinv13);
291 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
292 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
293 rinvsq23 = _mm_mul_pd(rinv23,rinv23);
294 rinvsq31 = _mm_mul_pd(rinv31,rinv31);
295 rinvsq32 = _mm_mul_pd(rinv32,rinv32);
296 rinvsq33 = _mm_mul_pd(rinv33,rinv33);
298 fjx0 = _mm_setzero_pd();
299 fjy0 = _mm_setzero_pd();
300 fjz0 = _mm_setzero_pd();
301 fjx1 = _mm_setzero_pd();
302 fjy1 = _mm_setzero_pd();
303 fjz1 = _mm_setzero_pd();
304 fjx2 = _mm_setzero_pd();
305 fjy2 = _mm_setzero_pd();
306 fjz2 = _mm_setzero_pd();
307 fjx3 = _mm_setzero_pd();
308 fjy3 = _mm_setzero_pd();
309 fjz3 = _mm_setzero_pd();
311 /**************************
312 * CALCULATE INTERACTIONS *
313 **************************/
315 r00 = _mm_mul_pd(rsq00,rinv00);
317 /* Calculate table index by multiplying r with table scale and truncate to integer */
318 rt = _mm_mul_pd(r00,vftabscale);
319 vfitab = _mm_cvttpd_epi32(rt);
320 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
321 vfitab = _mm_slli_epi32(vfitab,3);
323 /* CUBIC SPLINE TABLE DISPERSION */
324 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
325 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
326 GMX_MM_TRANSPOSE2_PD(Y,F);
327 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
328 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
329 GMX_MM_TRANSPOSE2_PD(G,H);
330 Heps = _mm_mul_pd(vfeps,H);
331 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
332 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
333 vvdw6 = _mm_mul_pd(c6_00,VV);
334 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
335 fvdw6 = _mm_mul_pd(c6_00,FF);
337 /* CUBIC SPLINE TABLE REPULSION */
338 vfitab = _mm_add_epi32(vfitab,ifour);
339 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
340 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
341 GMX_MM_TRANSPOSE2_PD(Y,F);
342 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
343 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
344 GMX_MM_TRANSPOSE2_PD(G,H);
345 Heps = _mm_mul_pd(vfeps,H);
346 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
347 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
348 vvdw12 = _mm_mul_pd(c12_00,VV);
349 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
350 fvdw12 = _mm_mul_pd(c12_00,FF);
351 vvdw = _mm_add_pd(vvdw12,vvdw6);
352 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
354 /* Update potential sum for this i atom from the interaction with this j atom. */
355 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
359 /* Calculate temporary vectorial force */
360 tx = _mm_mul_pd(fscal,dx00);
361 ty = _mm_mul_pd(fscal,dy00);
362 tz = _mm_mul_pd(fscal,dz00);
364 /* Update vectorial force */
365 fix0 = _mm_add_pd(fix0,tx);
366 fiy0 = _mm_add_pd(fiy0,ty);
367 fiz0 = _mm_add_pd(fiz0,tz);
369 fjx0 = _mm_add_pd(fjx0,tx);
370 fjy0 = _mm_add_pd(fjy0,ty);
371 fjz0 = _mm_add_pd(fjz0,tz);
373 /**************************
374 * CALCULATE INTERACTIONS *
375 **************************/
377 r11 = _mm_mul_pd(rsq11,rinv11);
379 /* EWALD ELECTROSTATICS */
381 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
382 ewrt = _mm_mul_pd(r11,ewtabscale);
383 ewitab = _mm_cvttpd_epi32(ewrt);
384 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
385 ewitab = _mm_slli_epi32(ewitab,2);
386 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
387 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
388 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
389 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
390 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
391 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
392 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
393 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
394 velec = _mm_mul_pd(qq11,_mm_sub_pd(rinv11,velec));
395 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
397 /* Update potential sum for this i atom from the interaction with this j atom. */
398 velecsum = _mm_add_pd(velecsum,velec);
402 /* Calculate temporary vectorial force */
403 tx = _mm_mul_pd(fscal,dx11);
404 ty = _mm_mul_pd(fscal,dy11);
405 tz = _mm_mul_pd(fscal,dz11);
407 /* Update vectorial force */
408 fix1 = _mm_add_pd(fix1,tx);
409 fiy1 = _mm_add_pd(fiy1,ty);
410 fiz1 = _mm_add_pd(fiz1,tz);
412 fjx1 = _mm_add_pd(fjx1,tx);
413 fjy1 = _mm_add_pd(fjy1,ty);
414 fjz1 = _mm_add_pd(fjz1,tz);
416 /**************************
417 * CALCULATE INTERACTIONS *
418 **************************/
420 r12 = _mm_mul_pd(rsq12,rinv12);
422 /* EWALD ELECTROSTATICS */
424 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
425 ewrt = _mm_mul_pd(r12,ewtabscale);
426 ewitab = _mm_cvttpd_epi32(ewrt);
427 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
428 ewitab = _mm_slli_epi32(ewitab,2);
429 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
430 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
431 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
432 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
433 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
434 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
435 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
436 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
437 velec = _mm_mul_pd(qq12,_mm_sub_pd(rinv12,velec));
438 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
440 /* Update potential sum for this i atom from the interaction with this j atom. */
441 velecsum = _mm_add_pd(velecsum,velec);
445 /* Calculate temporary vectorial force */
446 tx = _mm_mul_pd(fscal,dx12);
447 ty = _mm_mul_pd(fscal,dy12);
448 tz = _mm_mul_pd(fscal,dz12);
450 /* Update vectorial force */
451 fix1 = _mm_add_pd(fix1,tx);
452 fiy1 = _mm_add_pd(fiy1,ty);
453 fiz1 = _mm_add_pd(fiz1,tz);
455 fjx2 = _mm_add_pd(fjx2,tx);
456 fjy2 = _mm_add_pd(fjy2,ty);
457 fjz2 = _mm_add_pd(fjz2,tz);
459 /**************************
460 * CALCULATE INTERACTIONS *
461 **************************/
463 r13 = _mm_mul_pd(rsq13,rinv13);
465 /* EWALD ELECTROSTATICS */
467 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
468 ewrt = _mm_mul_pd(r13,ewtabscale);
469 ewitab = _mm_cvttpd_epi32(ewrt);
470 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
471 ewitab = _mm_slli_epi32(ewitab,2);
472 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
473 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
474 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
475 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
476 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
477 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
478 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
479 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
480 velec = _mm_mul_pd(qq13,_mm_sub_pd(rinv13,velec));
481 felec = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
483 /* Update potential sum for this i atom from the interaction with this j atom. */
484 velecsum = _mm_add_pd(velecsum,velec);
488 /* Calculate temporary vectorial force */
489 tx = _mm_mul_pd(fscal,dx13);
490 ty = _mm_mul_pd(fscal,dy13);
491 tz = _mm_mul_pd(fscal,dz13);
493 /* Update vectorial force */
494 fix1 = _mm_add_pd(fix1,tx);
495 fiy1 = _mm_add_pd(fiy1,ty);
496 fiz1 = _mm_add_pd(fiz1,tz);
498 fjx3 = _mm_add_pd(fjx3,tx);
499 fjy3 = _mm_add_pd(fjy3,ty);
500 fjz3 = _mm_add_pd(fjz3,tz);
502 /**************************
503 * CALCULATE INTERACTIONS *
504 **************************/
506 r21 = _mm_mul_pd(rsq21,rinv21);
508 /* EWALD ELECTROSTATICS */
510 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
511 ewrt = _mm_mul_pd(r21,ewtabscale);
512 ewitab = _mm_cvttpd_epi32(ewrt);
513 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
514 ewitab = _mm_slli_epi32(ewitab,2);
515 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
516 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
517 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
518 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
519 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
520 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
521 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
522 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
523 velec = _mm_mul_pd(qq21,_mm_sub_pd(rinv21,velec));
524 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
526 /* Update potential sum for this i atom from the interaction with this j atom. */
527 velecsum = _mm_add_pd(velecsum,velec);
531 /* Calculate temporary vectorial force */
532 tx = _mm_mul_pd(fscal,dx21);
533 ty = _mm_mul_pd(fscal,dy21);
534 tz = _mm_mul_pd(fscal,dz21);
536 /* Update vectorial force */
537 fix2 = _mm_add_pd(fix2,tx);
538 fiy2 = _mm_add_pd(fiy2,ty);
539 fiz2 = _mm_add_pd(fiz2,tz);
541 fjx1 = _mm_add_pd(fjx1,tx);
542 fjy1 = _mm_add_pd(fjy1,ty);
543 fjz1 = _mm_add_pd(fjz1,tz);
545 /**************************
546 * CALCULATE INTERACTIONS *
547 **************************/
549 r22 = _mm_mul_pd(rsq22,rinv22);
551 /* EWALD ELECTROSTATICS */
553 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
554 ewrt = _mm_mul_pd(r22,ewtabscale);
555 ewitab = _mm_cvttpd_epi32(ewrt);
556 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
557 ewitab = _mm_slli_epi32(ewitab,2);
558 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
559 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
560 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
561 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
562 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
563 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
564 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
565 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
566 velec = _mm_mul_pd(qq22,_mm_sub_pd(rinv22,velec));
567 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
569 /* Update potential sum for this i atom from the interaction with this j atom. */
570 velecsum = _mm_add_pd(velecsum,velec);
574 /* Calculate temporary vectorial force */
575 tx = _mm_mul_pd(fscal,dx22);
576 ty = _mm_mul_pd(fscal,dy22);
577 tz = _mm_mul_pd(fscal,dz22);
579 /* Update vectorial force */
580 fix2 = _mm_add_pd(fix2,tx);
581 fiy2 = _mm_add_pd(fiy2,ty);
582 fiz2 = _mm_add_pd(fiz2,tz);
584 fjx2 = _mm_add_pd(fjx2,tx);
585 fjy2 = _mm_add_pd(fjy2,ty);
586 fjz2 = _mm_add_pd(fjz2,tz);
588 /**************************
589 * CALCULATE INTERACTIONS *
590 **************************/
592 r23 = _mm_mul_pd(rsq23,rinv23);
594 /* EWALD ELECTROSTATICS */
596 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
597 ewrt = _mm_mul_pd(r23,ewtabscale);
598 ewitab = _mm_cvttpd_epi32(ewrt);
599 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
600 ewitab = _mm_slli_epi32(ewitab,2);
601 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
602 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
603 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
604 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
605 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
606 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
607 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
608 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
609 velec = _mm_mul_pd(qq23,_mm_sub_pd(rinv23,velec));
610 felec = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
612 /* Update potential sum for this i atom from the interaction with this j atom. */
613 velecsum = _mm_add_pd(velecsum,velec);
617 /* Calculate temporary vectorial force */
618 tx = _mm_mul_pd(fscal,dx23);
619 ty = _mm_mul_pd(fscal,dy23);
620 tz = _mm_mul_pd(fscal,dz23);
622 /* Update vectorial force */
623 fix2 = _mm_add_pd(fix2,tx);
624 fiy2 = _mm_add_pd(fiy2,ty);
625 fiz2 = _mm_add_pd(fiz2,tz);
627 fjx3 = _mm_add_pd(fjx3,tx);
628 fjy3 = _mm_add_pd(fjy3,ty);
629 fjz3 = _mm_add_pd(fjz3,tz);
631 /**************************
632 * CALCULATE INTERACTIONS *
633 **************************/
635 r31 = _mm_mul_pd(rsq31,rinv31);
637 /* EWALD ELECTROSTATICS */
639 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
640 ewrt = _mm_mul_pd(r31,ewtabscale);
641 ewitab = _mm_cvttpd_epi32(ewrt);
642 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
643 ewitab = _mm_slli_epi32(ewitab,2);
644 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
645 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
646 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
647 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
648 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
649 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
650 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
651 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
652 velec = _mm_mul_pd(qq31,_mm_sub_pd(rinv31,velec));
653 felec = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
655 /* Update potential sum for this i atom from the interaction with this j atom. */
656 velecsum = _mm_add_pd(velecsum,velec);
660 /* Calculate temporary vectorial force */
661 tx = _mm_mul_pd(fscal,dx31);
662 ty = _mm_mul_pd(fscal,dy31);
663 tz = _mm_mul_pd(fscal,dz31);
665 /* Update vectorial force */
666 fix3 = _mm_add_pd(fix3,tx);
667 fiy3 = _mm_add_pd(fiy3,ty);
668 fiz3 = _mm_add_pd(fiz3,tz);
670 fjx1 = _mm_add_pd(fjx1,tx);
671 fjy1 = _mm_add_pd(fjy1,ty);
672 fjz1 = _mm_add_pd(fjz1,tz);
674 /**************************
675 * CALCULATE INTERACTIONS *
676 **************************/
678 r32 = _mm_mul_pd(rsq32,rinv32);
680 /* EWALD ELECTROSTATICS */
682 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
683 ewrt = _mm_mul_pd(r32,ewtabscale);
684 ewitab = _mm_cvttpd_epi32(ewrt);
685 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
686 ewitab = _mm_slli_epi32(ewitab,2);
687 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
688 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
689 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
690 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
691 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
692 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
693 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
694 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
695 velec = _mm_mul_pd(qq32,_mm_sub_pd(rinv32,velec));
696 felec = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
698 /* Update potential sum for this i atom from the interaction with this j atom. */
699 velecsum = _mm_add_pd(velecsum,velec);
703 /* Calculate temporary vectorial force */
704 tx = _mm_mul_pd(fscal,dx32);
705 ty = _mm_mul_pd(fscal,dy32);
706 tz = _mm_mul_pd(fscal,dz32);
708 /* Update vectorial force */
709 fix3 = _mm_add_pd(fix3,tx);
710 fiy3 = _mm_add_pd(fiy3,ty);
711 fiz3 = _mm_add_pd(fiz3,tz);
713 fjx2 = _mm_add_pd(fjx2,tx);
714 fjy2 = _mm_add_pd(fjy2,ty);
715 fjz2 = _mm_add_pd(fjz2,tz);
717 /**************************
718 * CALCULATE INTERACTIONS *
719 **************************/
721 r33 = _mm_mul_pd(rsq33,rinv33);
723 /* EWALD ELECTROSTATICS */
725 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
726 ewrt = _mm_mul_pd(r33,ewtabscale);
727 ewitab = _mm_cvttpd_epi32(ewrt);
728 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
729 ewitab = _mm_slli_epi32(ewitab,2);
730 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
731 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
732 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
733 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
734 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
735 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
736 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
737 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
738 velec = _mm_mul_pd(qq33,_mm_sub_pd(rinv33,velec));
739 felec = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
741 /* Update potential sum for this i atom from the interaction with this j atom. */
742 velecsum = _mm_add_pd(velecsum,velec);
746 /* Calculate temporary vectorial force */
747 tx = _mm_mul_pd(fscal,dx33);
748 ty = _mm_mul_pd(fscal,dy33);
749 tz = _mm_mul_pd(fscal,dz33);
751 /* Update vectorial force */
752 fix3 = _mm_add_pd(fix3,tx);
753 fiy3 = _mm_add_pd(fiy3,ty);
754 fiz3 = _mm_add_pd(fiz3,tz);
756 fjx3 = _mm_add_pd(fjx3,tx);
757 fjy3 = _mm_add_pd(fjy3,ty);
758 fjz3 = _mm_add_pd(fjz3,tz);
760 gmx_mm_decrement_4rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
762 /* Inner loop uses 428 flops */
769 j_coord_offsetA = DIM*jnrA;
771 /* load j atom coordinates */
772 gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
773 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
774 &jy2,&jz2,&jx3,&jy3,&jz3);
776 /* Calculate displacement vector */
777 dx00 = _mm_sub_pd(ix0,jx0);
778 dy00 = _mm_sub_pd(iy0,jy0);
779 dz00 = _mm_sub_pd(iz0,jz0);
780 dx11 = _mm_sub_pd(ix1,jx1);
781 dy11 = _mm_sub_pd(iy1,jy1);
782 dz11 = _mm_sub_pd(iz1,jz1);
783 dx12 = _mm_sub_pd(ix1,jx2);
784 dy12 = _mm_sub_pd(iy1,jy2);
785 dz12 = _mm_sub_pd(iz1,jz2);
786 dx13 = _mm_sub_pd(ix1,jx3);
787 dy13 = _mm_sub_pd(iy1,jy3);
788 dz13 = _mm_sub_pd(iz1,jz3);
789 dx21 = _mm_sub_pd(ix2,jx1);
790 dy21 = _mm_sub_pd(iy2,jy1);
791 dz21 = _mm_sub_pd(iz2,jz1);
792 dx22 = _mm_sub_pd(ix2,jx2);
793 dy22 = _mm_sub_pd(iy2,jy2);
794 dz22 = _mm_sub_pd(iz2,jz2);
795 dx23 = _mm_sub_pd(ix2,jx3);
796 dy23 = _mm_sub_pd(iy2,jy3);
797 dz23 = _mm_sub_pd(iz2,jz3);
798 dx31 = _mm_sub_pd(ix3,jx1);
799 dy31 = _mm_sub_pd(iy3,jy1);
800 dz31 = _mm_sub_pd(iz3,jz1);
801 dx32 = _mm_sub_pd(ix3,jx2);
802 dy32 = _mm_sub_pd(iy3,jy2);
803 dz32 = _mm_sub_pd(iz3,jz2);
804 dx33 = _mm_sub_pd(ix3,jx3);
805 dy33 = _mm_sub_pd(iy3,jy3);
806 dz33 = _mm_sub_pd(iz3,jz3);
808 /* Calculate squared distance and things based on it */
809 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
810 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
811 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
812 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
813 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
814 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
815 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
816 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
817 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
818 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
820 rinv00 = gmx_mm_invsqrt_pd(rsq00);
821 rinv11 = gmx_mm_invsqrt_pd(rsq11);
822 rinv12 = gmx_mm_invsqrt_pd(rsq12);
823 rinv13 = gmx_mm_invsqrt_pd(rsq13);
824 rinv21 = gmx_mm_invsqrt_pd(rsq21);
825 rinv22 = gmx_mm_invsqrt_pd(rsq22);
826 rinv23 = gmx_mm_invsqrt_pd(rsq23);
827 rinv31 = gmx_mm_invsqrt_pd(rsq31);
828 rinv32 = gmx_mm_invsqrt_pd(rsq32);
829 rinv33 = gmx_mm_invsqrt_pd(rsq33);
831 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
832 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
833 rinvsq13 = _mm_mul_pd(rinv13,rinv13);
834 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
835 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
836 rinvsq23 = _mm_mul_pd(rinv23,rinv23);
837 rinvsq31 = _mm_mul_pd(rinv31,rinv31);
838 rinvsq32 = _mm_mul_pd(rinv32,rinv32);
839 rinvsq33 = _mm_mul_pd(rinv33,rinv33);
841 fjx0 = _mm_setzero_pd();
842 fjy0 = _mm_setzero_pd();
843 fjz0 = _mm_setzero_pd();
844 fjx1 = _mm_setzero_pd();
845 fjy1 = _mm_setzero_pd();
846 fjz1 = _mm_setzero_pd();
847 fjx2 = _mm_setzero_pd();
848 fjy2 = _mm_setzero_pd();
849 fjz2 = _mm_setzero_pd();
850 fjx3 = _mm_setzero_pd();
851 fjy3 = _mm_setzero_pd();
852 fjz3 = _mm_setzero_pd();
854 /**************************
855 * CALCULATE INTERACTIONS *
856 **************************/
858 r00 = _mm_mul_pd(rsq00,rinv00);
860 /* Calculate table index by multiplying r with table scale and truncate to integer */
861 rt = _mm_mul_pd(r00,vftabscale);
862 vfitab = _mm_cvttpd_epi32(rt);
863 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
864 vfitab = _mm_slli_epi32(vfitab,3);
866 /* CUBIC SPLINE TABLE DISPERSION */
867 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
868 F = _mm_setzero_pd();
869 GMX_MM_TRANSPOSE2_PD(Y,F);
870 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
871 H = _mm_setzero_pd();
872 GMX_MM_TRANSPOSE2_PD(G,H);
873 Heps = _mm_mul_pd(vfeps,H);
874 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
875 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
876 vvdw6 = _mm_mul_pd(c6_00,VV);
877 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
878 fvdw6 = _mm_mul_pd(c6_00,FF);
880 /* CUBIC SPLINE TABLE REPULSION */
881 vfitab = _mm_add_epi32(vfitab,ifour);
882 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
883 F = _mm_setzero_pd();
884 GMX_MM_TRANSPOSE2_PD(Y,F);
885 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
886 H = _mm_setzero_pd();
887 GMX_MM_TRANSPOSE2_PD(G,H);
888 Heps = _mm_mul_pd(vfeps,H);
889 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
890 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
891 vvdw12 = _mm_mul_pd(c12_00,VV);
892 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
893 fvdw12 = _mm_mul_pd(c12_00,FF);
894 vvdw = _mm_add_pd(vvdw12,vvdw6);
895 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
897 /* Update potential sum for this i atom from the interaction with this j atom. */
898 vvdw = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
899 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
903 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
905 /* Calculate temporary vectorial force */
906 tx = _mm_mul_pd(fscal,dx00);
907 ty = _mm_mul_pd(fscal,dy00);
908 tz = _mm_mul_pd(fscal,dz00);
910 /* Update vectorial force */
911 fix0 = _mm_add_pd(fix0,tx);
912 fiy0 = _mm_add_pd(fiy0,ty);
913 fiz0 = _mm_add_pd(fiz0,tz);
915 fjx0 = _mm_add_pd(fjx0,tx);
916 fjy0 = _mm_add_pd(fjy0,ty);
917 fjz0 = _mm_add_pd(fjz0,tz);
919 /**************************
920 * CALCULATE INTERACTIONS *
921 **************************/
923 r11 = _mm_mul_pd(rsq11,rinv11);
925 /* EWALD ELECTROSTATICS */
927 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
928 ewrt = _mm_mul_pd(r11,ewtabscale);
929 ewitab = _mm_cvttpd_epi32(ewrt);
930 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
931 ewitab = _mm_slli_epi32(ewitab,2);
932 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
933 ewtabD = _mm_setzero_pd();
934 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
935 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
936 ewtabFn = _mm_setzero_pd();
937 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
938 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
939 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
940 velec = _mm_mul_pd(qq11,_mm_sub_pd(rinv11,velec));
941 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
943 /* Update potential sum for this i atom from the interaction with this j atom. */
944 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
945 velecsum = _mm_add_pd(velecsum,velec);
949 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
951 /* Calculate temporary vectorial force */
952 tx = _mm_mul_pd(fscal,dx11);
953 ty = _mm_mul_pd(fscal,dy11);
954 tz = _mm_mul_pd(fscal,dz11);
956 /* Update vectorial force */
957 fix1 = _mm_add_pd(fix1,tx);
958 fiy1 = _mm_add_pd(fiy1,ty);
959 fiz1 = _mm_add_pd(fiz1,tz);
961 fjx1 = _mm_add_pd(fjx1,tx);
962 fjy1 = _mm_add_pd(fjy1,ty);
963 fjz1 = _mm_add_pd(fjz1,tz);
965 /**************************
966 * CALCULATE INTERACTIONS *
967 **************************/
969 r12 = _mm_mul_pd(rsq12,rinv12);
971 /* EWALD ELECTROSTATICS */
973 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
974 ewrt = _mm_mul_pd(r12,ewtabscale);
975 ewitab = _mm_cvttpd_epi32(ewrt);
976 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
977 ewitab = _mm_slli_epi32(ewitab,2);
978 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
979 ewtabD = _mm_setzero_pd();
980 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
981 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
982 ewtabFn = _mm_setzero_pd();
983 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
984 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
985 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
986 velec = _mm_mul_pd(qq12,_mm_sub_pd(rinv12,velec));
987 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
989 /* Update potential sum for this i atom from the interaction with this j atom. */
990 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
991 velecsum = _mm_add_pd(velecsum,velec);
995 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
997 /* Calculate temporary vectorial force */
998 tx = _mm_mul_pd(fscal,dx12);
999 ty = _mm_mul_pd(fscal,dy12);
1000 tz = _mm_mul_pd(fscal,dz12);
1002 /* Update vectorial force */
1003 fix1 = _mm_add_pd(fix1,tx);
1004 fiy1 = _mm_add_pd(fiy1,ty);
1005 fiz1 = _mm_add_pd(fiz1,tz);
1007 fjx2 = _mm_add_pd(fjx2,tx);
1008 fjy2 = _mm_add_pd(fjy2,ty);
1009 fjz2 = _mm_add_pd(fjz2,tz);
1011 /**************************
1012 * CALCULATE INTERACTIONS *
1013 **************************/
1015 r13 = _mm_mul_pd(rsq13,rinv13);
1017 /* EWALD ELECTROSTATICS */
1019 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1020 ewrt = _mm_mul_pd(r13,ewtabscale);
1021 ewitab = _mm_cvttpd_epi32(ewrt);
1022 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1023 ewitab = _mm_slli_epi32(ewitab,2);
1024 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1025 ewtabD = _mm_setzero_pd();
1026 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1027 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1028 ewtabFn = _mm_setzero_pd();
1029 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1030 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1031 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1032 velec = _mm_mul_pd(qq13,_mm_sub_pd(rinv13,velec));
1033 felec = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
1035 /* Update potential sum for this i atom from the interaction with this j atom. */
1036 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1037 velecsum = _mm_add_pd(velecsum,velec);
1041 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1043 /* Calculate temporary vectorial force */
1044 tx = _mm_mul_pd(fscal,dx13);
1045 ty = _mm_mul_pd(fscal,dy13);
1046 tz = _mm_mul_pd(fscal,dz13);
1048 /* Update vectorial force */
1049 fix1 = _mm_add_pd(fix1,tx);
1050 fiy1 = _mm_add_pd(fiy1,ty);
1051 fiz1 = _mm_add_pd(fiz1,tz);
1053 fjx3 = _mm_add_pd(fjx3,tx);
1054 fjy3 = _mm_add_pd(fjy3,ty);
1055 fjz3 = _mm_add_pd(fjz3,tz);
1057 /**************************
1058 * CALCULATE INTERACTIONS *
1059 **************************/
1061 r21 = _mm_mul_pd(rsq21,rinv21);
1063 /* EWALD ELECTROSTATICS */
1065 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1066 ewrt = _mm_mul_pd(r21,ewtabscale);
1067 ewitab = _mm_cvttpd_epi32(ewrt);
1068 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1069 ewitab = _mm_slli_epi32(ewitab,2);
1070 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1071 ewtabD = _mm_setzero_pd();
1072 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1073 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1074 ewtabFn = _mm_setzero_pd();
1075 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1076 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1077 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1078 velec = _mm_mul_pd(qq21,_mm_sub_pd(rinv21,velec));
1079 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1081 /* Update potential sum for this i atom from the interaction with this j atom. */
1082 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1083 velecsum = _mm_add_pd(velecsum,velec);
1087 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1089 /* Calculate temporary vectorial force */
1090 tx = _mm_mul_pd(fscal,dx21);
1091 ty = _mm_mul_pd(fscal,dy21);
1092 tz = _mm_mul_pd(fscal,dz21);
1094 /* Update vectorial force */
1095 fix2 = _mm_add_pd(fix2,tx);
1096 fiy2 = _mm_add_pd(fiy2,ty);
1097 fiz2 = _mm_add_pd(fiz2,tz);
1099 fjx1 = _mm_add_pd(fjx1,tx);
1100 fjy1 = _mm_add_pd(fjy1,ty);
1101 fjz1 = _mm_add_pd(fjz1,tz);
1103 /**************************
1104 * CALCULATE INTERACTIONS *
1105 **************************/
1107 r22 = _mm_mul_pd(rsq22,rinv22);
1109 /* EWALD ELECTROSTATICS */
1111 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1112 ewrt = _mm_mul_pd(r22,ewtabscale);
1113 ewitab = _mm_cvttpd_epi32(ewrt);
1114 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1115 ewitab = _mm_slli_epi32(ewitab,2);
1116 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1117 ewtabD = _mm_setzero_pd();
1118 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1119 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1120 ewtabFn = _mm_setzero_pd();
1121 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1122 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1123 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1124 velec = _mm_mul_pd(qq22,_mm_sub_pd(rinv22,velec));
1125 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1127 /* Update potential sum for this i atom from the interaction with this j atom. */
1128 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1129 velecsum = _mm_add_pd(velecsum,velec);
1133 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1135 /* Calculate temporary vectorial force */
1136 tx = _mm_mul_pd(fscal,dx22);
1137 ty = _mm_mul_pd(fscal,dy22);
1138 tz = _mm_mul_pd(fscal,dz22);
1140 /* Update vectorial force */
1141 fix2 = _mm_add_pd(fix2,tx);
1142 fiy2 = _mm_add_pd(fiy2,ty);
1143 fiz2 = _mm_add_pd(fiz2,tz);
1145 fjx2 = _mm_add_pd(fjx2,tx);
1146 fjy2 = _mm_add_pd(fjy2,ty);
1147 fjz2 = _mm_add_pd(fjz2,tz);
1149 /**************************
1150 * CALCULATE INTERACTIONS *
1151 **************************/
1153 r23 = _mm_mul_pd(rsq23,rinv23);
1155 /* EWALD ELECTROSTATICS */
1157 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1158 ewrt = _mm_mul_pd(r23,ewtabscale);
1159 ewitab = _mm_cvttpd_epi32(ewrt);
1160 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1161 ewitab = _mm_slli_epi32(ewitab,2);
1162 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1163 ewtabD = _mm_setzero_pd();
1164 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1165 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1166 ewtabFn = _mm_setzero_pd();
1167 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1168 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1169 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1170 velec = _mm_mul_pd(qq23,_mm_sub_pd(rinv23,velec));
1171 felec = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
1173 /* Update potential sum for this i atom from the interaction with this j atom. */
1174 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1175 velecsum = _mm_add_pd(velecsum,velec);
1179 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1181 /* Calculate temporary vectorial force */
1182 tx = _mm_mul_pd(fscal,dx23);
1183 ty = _mm_mul_pd(fscal,dy23);
1184 tz = _mm_mul_pd(fscal,dz23);
1186 /* Update vectorial force */
1187 fix2 = _mm_add_pd(fix2,tx);
1188 fiy2 = _mm_add_pd(fiy2,ty);
1189 fiz2 = _mm_add_pd(fiz2,tz);
1191 fjx3 = _mm_add_pd(fjx3,tx);
1192 fjy3 = _mm_add_pd(fjy3,ty);
1193 fjz3 = _mm_add_pd(fjz3,tz);
1195 /**************************
1196 * CALCULATE INTERACTIONS *
1197 **************************/
1199 r31 = _mm_mul_pd(rsq31,rinv31);
1201 /* EWALD ELECTROSTATICS */
1203 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1204 ewrt = _mm_mul_pd(r31,ewtabscale);
1205 ewitab = _mm_cvttpd_epi32(ewrt);
1206 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1207 ewitab = _mm_slli_epi32(ewitab,2);
1208 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1209 ewtabD = _mm_setzero_pd();
1210 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1211 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1212 ewtabFn = _mm_setzero_pd();
1213 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1214 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1215 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1216 velec = _mm_mul_pd(qq31,_mm_sub_pd(rinv31,velec));
1217 felec = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
1219 /* Update potential sum for this i atom from the interaction with this j atom. */
1220 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1221 velecsum = _mm_add_pd(velecsum,velec);
1225 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1227 /* Calculate temporary vectorial force */
1228 tx = _mm_mul_pd(fscal,dx31);
1229 ty = _mm_mul_pd(fscal,dy31);
1230 tz = _mm_mul_pd(fscal,dz31);
1232 /* Update vectorial force */
1233 fix3 = _mm_add_pd(fix3,tx);
1234 fiy3 = _mm_add_pd(fiy3,ty);
1235 fiz3 = _mm_add_pd(fiz3,tz);
1237 fjx1 = _mm_add_pd(fjx1,tx);
1238 fjy1 = _mm_add_pd(fjy1,ty);
1239 fjz1 = _mm_add_pd(fjz1,tz);
1241 /**************************
1242 * CALCULATE INTERACTIONS *
1243 **************************/
1245 r32 = _mm_mul_pd(rsq32,rinv32);
1247 /* EWALD ELECTROSTATICS */
1249 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1250 ewrt = _mm_mul_pd(r32,ewtabscale);
1251 ewitab = _mm_cvttpd_epi32(ewrt);
1252 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1253 ewitab = _mm_slli_epi32(ewitab,2);
1254 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1255 ewtabD = _mm_setzero_pd();
1256 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1257 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1258 ewtabFn = _mm_setzero_pd();
1259 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1260 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1261 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1262 velec = _mm_mul_pd(qq32,_mm_sub_pd(rinv32,velec));
1263 felec = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
1265 /* Update potential sum for this i atom from the interaction with this j atom. */
1266 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1267 velecsum = _mm_add_pd(velecsum,velec);
1271 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1273 /* Calculate temporary vectorial force */
1274 tx = _mm_mul_pd(fscal,dx32);
1275 ty = _mm_mul_pd(fscal,dy32);
1276 tz = _mm_mul_pd(fscal,dz32);
1278 /* Update vectorial force */
1279 fix3 = _mm_add_pd(fix3,tx);
1280 fiy3 = _mm_add_pd(fiy3,ty);
1281 fiz3 = _mm_add_pd(fiz3,tz);
1283 fjx2 = _mm_add_pd(fjx2,tx);
1284 fjy2 = _mm_add_pd(fjy2,ty);
1285 fjz2 = _mm_add_pd(fjz2,tz);
1287 /**************************
1288 * CALCULATE INTERACTIONS *
1289 **************************/
1291 r33 = _mm_mul_pd(rsq33,rinv33);
1293 /* EWALD ELECTROSTATICS */
1295 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1296 ewrt = _mm_mul_pd(r33,ewtabscale);
1297 ewitab = _mm_cvttpd_epi32(ewrt);
1298 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1299 ewitab = _mm_slli_epi32(ewitab,2);
1300 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1301 ewtabD = _mm_setzero_pd();
1302 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1303 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1304 ewtabFn = _mm_setzero_pd();
1305 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1306 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1307 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1308 velec = _mm_mul_pd(qq33,_mm_sub_pd(rinv33,velec));
1309 felec = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
1311 /* Update potential sum for this i atom from the interaction with this j atom. */
1312 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1313 velecsum = _mm_add_pd(velecsum,velec);
1317 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1319 /* Calculate temporary vectorial force */
1320 tx = _mm_mul_pd(fscal,dx33);
1321 ty = _mm_mul_pd(fscal,dy33);
1322 tz = _mm_mul_pd(fscal,dz33);
1324 /* Update vectorial force */
1325 fix3 = _mm_add_pd(fix3,tx);
1326 fiy3 = _mm_add_pd(fiy3,ty);
1327 fiz3 = _mm_add_pd(fiz3,tz);
1329 fjx3 = _mm_add_pd(fjx3,tx);
1330 fjy3 = _mm_add_pd(fjy3,ty);
1331 fjz3 = _mm_add_pd(fjz3,tz);
1333 gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1335 /* Inner loop uses 428 flops */
1338 /* End of innermost loop */
1340 gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1341 f+i_coord_offset,fshift+i_shift_offset);
1344 /* Update potential energies */
1345 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1346 gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1348 /* Increment number of inner iterations */
1349 inneriter += j_index_end - j_index_start;
1351 /* Outer loop uses 26 flops */
1354 /* Increment number of outer iterations */
1357 /* Update outer/inner flops */
1359 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*428);
1362 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW4W4_F_sse2_double
1363 * Electrostatics interaction: Ewald
1364 * VdW interaction: CubicSplineTable
1365 * Geometry: Water4-Water4
1366 * Calculate force/pot: Force
1369 nb_kernel_ElecEw_VdwCSTab_GeomW4W4_F_sse2_double
1370 (t_nblist * gmx_restrict nlist,
1371 rvec * gmx_restrict xx,
1372 rvec * gmx_restrict ff,
1373 t_forcerec * gmx_restrict fr,
1374 t_mdatoms * gmx_restrict mdatoms,
1375 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1376 t_nrnb * gmx_restrict nrnb)
1378 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1379 * just 0 for non-waters.
1380 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
1381 * jnr indices corresponding to data put in the four positions in the SIMD register.
1383 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1384 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1386 int j_coord_offsetA,j_coord_offsetB;
1387 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1388 real rcutoff_scalar;
1389 real *shiftvec,*fshift,*x,*f;
1390 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1392 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1394 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1396 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1398 __m128d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1399 int vdwjidx0A,vdwjidx0B;
1400 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1401 int vdwjidx1A,vdwjidx1B;
1402 __m128d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1403 int vdwjidx2A,vdwjidx2B;
1404 __m128d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1405 int vdwjidx3A,vdwjidx3B;
1406 __m128d jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1407 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1408 __m128d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1409 __m128d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1410 __m128d dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1411 __m128d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1412 __m128d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1413 __m128d dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1414 __m128d dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1415 __m128d dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1416 __m128d dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1417 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
1420 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1423 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
1424 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
1426 __m128i ifour = _mm_set1_epi32(4);
1427 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1430 __m128d ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1432 __m128d dummy_mask,cutoff_mask;
1433 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
1434 __m128d one = _mm_set1_pd(1.0);
1435 __m128d two = _mm_set1_pd(2.0);
1441 jindex = nlist->jindex;
1443 shiftidx = nlist->shift;
1445 shiftvec = fr->shift_vec[0];
1446 fshift = fr->fshift[0];
1447 facel = _mm_set1_pd(fr->epsfac);
1448 charge = mdatoms->chargeA;
1449 nvdwtype = fr->ntype;
1450 vdwparam = fr->nbfp;
1451 vdwtype = mdatoms->typeA;
1453 vftab = kernel_data->table_vdw->data;
1454 vftabscale = _mm_set1_pd(kernel_data->table_vdw->scale);
1456 sh_ewald = _mm_set1_pd(fr->ic->sh_ewald);
1457 ewtab = fr->ic->tabq_coul_F;
1458 ewtabscale = _mm_set1_pd(fr->ic->tabq_scale);
1459 ewtabhalfspace = _mm_set1_pd(0.5/fr->ic->tabq_scale);
1461 /* Setup water-specific parameters */
1462 inr = nlist->iinr[0];
1463 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
1464 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
1465 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
1466 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
1468 jq1 = _mm_set1_pd(charge[inr+1]);
1469 jq2 = _mm_set1_pd(charge[inr+2]);
1470 jq3 = _mm_set1_pd(charge[inr+3]);
1471 vdwjidx0A = 2*vdwtype[inr+0];
1472 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
1473 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
1474 qq11 = _mm_mul_pd(iq1,jq1);
1475 qq12 = _mm_mul_pd(iq1,jq2);
1476 qq13 = _mm_mul_pd(iq1,jq3);
1477 qq21 = _mm_mul_pd(iq2,jq1);
1478 qq22 = _mm_mul_pd(iq2,jq2);
1479 qq23 = _mm_mul_pd(iq2,jq3);
1480 qq31 = _mm_mul_pd(iq3,jq1);
1481 qq32 = _mm_mul_pd(iq3,jq2);
1482 qq33 = _mm_mul_pd(iq3,jq3);
1484 /* Avoid stupid compiler warnings */
1486 j_coord_offsetA = 0;
1487 j_coord_offsetB = 0;
1492 /* Start outer loop over neighborlists */
1493 for(iidx=0; iidx<nri; iidx++)
1495 /* Load shift vector for this list */
1496 i_shift_offset = DIM*shiftidx[iidx];
1498 /* Load limits for loop over neighbors */
1499 j_index_start = jindex[iidx];
1500 j_index_end = jindex[iidx+1];
1502 /* Get outer coordinate index */
1504 i_coord_offset = DIM*inr;
1506 /* Load i particle coords and add shift vector */
1507 gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1508 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1510 fix0 = _mm_setzero_pd();
1511 fiy0 = _mm_setzero_pd();
1512 fiz0 = _mm_setzero_pd();
1513 fix1 = _mm_setzero_pd();
1514 fiy1 = _mm_setzero_pd();
1515 fiz1 = _mm_setzero_pd();
1516 fix2 = _mm_setzero_pd();
1517 fiy2 = _mm_setzero_pd();
1518 fiz2 = _mm_setzero_pd();
1519 fix3 = _mm_setzero_pd();
1520 fiy3 = _mm_setzero_pd();
1521 fiz3 = _mm_setzero_pd();
1523 /* Start inner kernel loop */
1524 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
1527 /* Get j neighbor index, and coordinate index */
1529 jnrB = jjnr[jidx+1];
1530 j_coord_offsetA = DIM*jnrA;
1531 j_coord_offsetB = DIM*jnrB;
1533 /* load j atom coordinates */
1534 gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1535 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1536 &jy2,&jz2,&jx3,&jy3,&jz3);
1538 /* Calculate displacement vector */
1539 dx00 = _mm_sub_pd(ix0,jx0);
1540 dy00 = _mm_sub_pd(iy0,jy0);
1541 dz00 = _mm_sub_pd(iz0,jz0);
1542 dx11 = _mm_sub_pd(ix1,jx1);
1543 dy11 = _mm_sub_pd(iy1,jy1);
1544 dz11 = _mm_sub_pd(iz1,jz1);
1545 dx12 = _mm_sub_pd(ix1,jx2);
1546 dy12 = _mm_sub_pd(iy1,jy2);
1547 dz12 = _mm_sub_pd(iz1,jz2);
1548 dx13 = _mm_sub_pd(ix1,jx3);
1549 dy13 = _mm_sub_pd(iy1,jy3);
1550 dz13 = _mm_sub_pd(iz1,jz3);
1551 dx21 = _mm_sub_pd(ix2,jx1);
1552 dy21 = _mm_sub_pd(iy2,jy1);
1553 dz21 = _mm_sub_pd(iz2,jz1);
1554 dx22 = _mm_sub_pd(ix2,jx2);
1555 dy22 = _mm_sub_pd(iy2,jy2);
1556 dz22 = _mm_sub_pd(iz2,jz2);
1557 dx23 = _mm_sub_pd(ix2,jx3);
1558 dy23 = _mm_sub_pd(iy2,jy3);
1559 dz23 = _mm_sub_pd(iz2,jz3);
1560 dx31 = _mm_sub_pd(ix3,jx1);
1561 dy31 = _mm_sub_pd(iy3,jy1);
1562 dz31 = _mm_sub_pd(iz3,jz1);
1563 dx32 = _mm_sub_pd(ix3,jx2);
1564 dy32 = _mm_sub_pd(iy3,jy2);
1565 dz32 = _mm_sub_pd(iz3,jz2);
1566 dx33 = _mm_sub_pd(ix3,jx3);
1567 dy33 = _mm_sub_pd(iy3,jy3);
1568 dz33 = _mm_sub_pd(iz3,jz3);
1570 /* Calculate squared distance and things based on it */
1571 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1572 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
1573 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
1574 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
1575 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
1576 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
1577 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
1578 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
1579 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
1580 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
1582 rinv00 = gmx_mm_invsqrt_pd(rsq00);
1583 rinv11 = gmx_mm_invsqrt_pd(rsq11);
1584 rinv12 = gmx_mm_invsqrt_pd(rsq12);
1585 rinv13 = gmx_mm_invsqrt_pd(rsq13);
1586 rinv21 = gmx_mm_invsqrt_pd(rsq21);
1587 rinv22 = gmx_mm_invsqrt_pd(rsq22);
1588 rinv23 = gmx_mm_invsqrt_pd(rsq23);
1589 rinv31 = gmx_mm_invsqrt_pd(rsq31);
1590 rinv32 = gmx_mm_invsqrt_pd(rsq32);
1591 rinv33 = gmx_mm_invsqrt_pd(rsq33);
1593 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
1594 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
1595 rinvsq13 = _mm_mul_pd(rinv13,rinv13);
1596 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
1597 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
1598 rinvsq23 = _mm_mul_pd(rinv23,rinv23);
1599 rinvsq31 = _mm_mul_pd(rinv31,rinv31);
1600 rinvsq32 = _mm_mul_pd(rinv32,rinv32);
1601 rinvsq33 = _mm_mul_pd(rinv33,rinv33);
1603 fjx0 = _mm_setzero_pd();
1604 fjy0 = _mm_setzero_pd();
1605 fjz0 = _mm_setzero_pd();
1606 fjx1 = _mm_setzero_pd();
1607 fjy1 = _mm_setzero_pd();
1608 fjz1 = _mm_setzero_pd();
1609 fjx2 = _mm_setzero_pd();
1610 fjy2 = _mm_setzero_pd();
1611 fjz2 = _mm_setzero_pd();
1612 fjx3 = _mm_setzero_pd();
1613 fjy3 = _mm_setzero_pd();
1614 fjz3 = _mm_setzero_pd();
1616 /**************************
1617 * CALCULATE INTERACTIONS *
1618 **************************/
1620 r00 = _mm_mul_pd(rsq00,rinv00);
1622 /* Calculate table index by multiplying r with table scale and truncate to integer */
1623 rt = _mm_mul_pd(r00,vftabscale);
1624 vfitab = _mm_cvttpd_epi32(rt);
1625 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1626 vfitab = _mm_slli_epi32(vfitab,3);
1628 /* CUBIC SPLINE TABLE DISPERSION */
1629 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1630 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1631 GMX_MM_TRANSPOSE2_PD(Y,F);
1632 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1633 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1634 GMX_MM_TRANSPOSE2_PD(G,H);
1635 Heps = _mm_mul_pd(vfeps,H);
1636 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1637 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1638 fvdw6 = _mm_mul_pd(c6_00,FF);
1640 /* CUBIC SPLINE TABLE REPULSION */
1641 vfitab = _mm_add_epi32(vfitab,ifour);
1642 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1643 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1644 GMX_MM_TRANSPOSE2_PD(Y,F);
1645 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1646 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1647 GMX_MM_TRANSPOSE2_PD(G,H);
1648 Heps = _mm_mul_pd(vfeps,H);
1649 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1650 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1651 fvdw12 = _mm_mul_pd(c12_00,FF);
1652 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
1656 /* Calculate temporary vectorial force */
1657 tx = _mm_mul_pd(fscal,dx00);
1658 ty = _mm_mul_pd(fscal,dy00);
1659 tz = _mm_mul_pd(fscal,dz00);
1661 /* Update vectorial force */
1662 fix0 = _mm_add_pd(fix0,tx);
1663 fiy0 = _mm_add_pd(fiy0,ty);
1664 fiz0 = _mm_add_pd(fiz0,tz);
1666 fjx0 = _mm_add_pd(fjx0,tx);
1667 fjy0 = _mm_add_pd(fjy0,ty);
1668 fjz0 = _mm_add_pd(fjz0,tz);
1670 /**************************
1671 * CALCULATE INTERACTIONS *
1672 **************************/
1674 r11 = _mm_mul_pd(rsq11,rinv11);
1676 /* EWALD ELECTROSTATICS */
1678 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1679 ewrt = _mm_mul_pd(r11,ewtabscale);
1680 ewitab = _mm_cvttpd_epi32(ewrt);
1681 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1682 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1684 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1685 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
1689 /* Calculate temporary vectorial force */
1690 tx = _mm_mul_pd(fscal,dx11);
1691 ty = _mm_mul_pd(fscal,dy11);
1692 tz = _mm_mul_pd(fscal,dz11);
1694 /* Update vectorial force */
1695 fix1 = _mm_add_pd(fix1,tx);
1696 fiy1 = _mm_add_pd(fiy1,ty);
1697 fiz1 = _mm_add_pd(fiz1,tz);
1699 fjx1 = _mm_add_pd(fjx1,tx);
1700 fjy1 = _mm_add_pd(fjy1,ty);
1701 fjz1 = _mm_add_pd(fjz1,tz);
1703 /**************************
1704 * CALCULATE INTERACTIONS *
1705 **************************/
1707 r12 = _mm_mul_pd(rsq12,rinv12);
1709 /* EWALD ELECTROSTATICS */
1711 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1712 ewrt = _mm_mul_pd(r12,ewtabscale);
1713 ewitab = _mm_cvttpd_epi32(ewrt);
1714 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1715 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1717 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1718 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
1722 /* Calculate temporary vectorial force */
1723 tx = _mm_mul_pd(fscal,dx12);
1724 ty = _mm_mul_pd(fscal,dy12);
1725 tz = _mm_mul_pd(fscal,dz12);
1727 /* Update vectorial force */
1728 fix1 = _mm_add_pd(fix1,tx);
1729 fiy1 = _mm_add_pd(fiy1,ty);
1730 fiz1 = _mm_add_pd(fiz1,tz);
1732 fjx2 = _mm_add_pd(fjx2,tx);
1733 fjy2 = _mm_add_pd(fjy2,ty);
1734 fjz2 = _mm_add_pd(fjz2,tz);
1736 /**************************
1737 * CALCULATE INTERACTIONS *
1738 **************************/
1740 r13 = _mm_mul_pd(rsq13,rinv13);
1742 /* EWALD ELECTROSTATICS */
1744 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1745 ewrt = _mm_mul_pd(r13,ewtabscale);
1746 ewitab = _mm_cvttpd_epi32(ewrt);
1747 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1748 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1750 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1751 felec = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
1755 /* Calculate temporary vectorial force */
1756 tx = _mm_mul_pd(fscal,dx13);
1757 ty = _mm_mul_pd(fscal,dy13);
1758 tz = _mm_mul_pd(fscal,dz13);
1760 /* Update vectorial force */
1761 fix1 = _mm_add_pd(fix1,tx);
1762 fiy1 = _mm_add_pd(fiy1,ty);
1763 fiz1 = _mm_add_pd(fiz1,tz);
1765 fjx3 = _mm_add_pd(fjx3,tx);
1766 fjy3 = _mm_add_pd(fjy3,ty);
1767 fjz3 = _mm_add_pd(fjz3,tz);
1769 /**************************
1770 * CALCULATE INTERACTIONS *
1771 **************************/
1773 r21 = _mm_mul_pd(rsq21,rinv21);
1775 /* EWALD ELECTROSTATICS */
1777 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1778 ewrt = _mm_mul_pd(r21,ewtabscale);
1779 ewitab = _mm_cvttpd_epi32(ewrt);
1780 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1781 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1783 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1784 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1788 /* Calculate temporary vectorial force */
1789 tx = _mm_mul_pd(fscal,dx21);
1790 ty = _mm_mul_pd(fscal,dy21);
1791 tz = _mm_mul_pd(fscal,dz21);
1793 /* Update vectorial force */
1794 fix2 = _mm_add_pd(fix2,tx);
1795 fiy2 = _mm_add_pd(fiy2,ty);
1796 fiz2 = _mm_add_pd(fiz2,tz);
1798 fjx1 = _mm_add_pd(fjx1,tx);
1799 fjy1 = _mm_add_pd(fjy1,ty);
1800 fjz1 = _mm_add_pd(fjz1,tz);
1802 /**************************
1803 * CALCULATE INTERACTIONS *
1804 **************************/
1806 r22 = _mm_mul_pd(rsq22,rinv22);
1808 /* EWALD ELECTROSTATICS */
1810 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1811 ewrt = _mm_mul_pd(r22,ewtabscale);
1812 ewitab = _mm_cvttpd_epi32(ewrt);
1813 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1814 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1816 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1817 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1821 /* Calculate temporary vectorial force */
1822 tx = _mm_mul_pd(fscal,dx22);
1823 ty = _mm_mul_pd(fscal,dy22);
1824 tz = _mm_mul_pd(fscal,dz22);
1826 /* Update vectorial force */
1827 fix2 = _mm_add_pd(fix2,tx);
1828 fiy2 = _mm_add_pd(fiy2,ty);
1829 fiz2 = _mm_add_pd(fiz2,tz);
1831 fjx2 = _mm_add_pd(fjx2,tx);
1832 fjy2 = _mm_add_pd(fjy2,ty);
1833 fjz2 = _mm_add_pd(fjz2,tz);
1835 /**************************
1836 * CALCULATE INTERACTIONS *
1837 **************************/
1839 r23 = _mm_mul_pd(rsq23,rinv23);
1841 /* EWALD ELECTROSTATICS */
1843 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1844 ewrt = _mm_mul_pd(r23,ewtabscale);
1845 ewitab = _mm_cvttpd_epi32(ewrt);
1846 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1847 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1849 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1850 felec = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
1854 /* Calculate temporary vectorial force */
1855 tx = _mm_mul_pd(fscal,dx23);
1856 ty = _mm_mul_pd(fscal,dy23);
1857 tz = _mm_mul_pd(fscal,dz23);
1859 /* Update vectorial force */
1860 fix2 = _mm_add_pd(fix2,tx);
1861 fiy2 = _mm_add_pd(fiy2,ty);
1862 fiz2 = _mm_add_pd(fiz2,tz);
1864 fjx3 = _mm_add_pd(fjx3,tx);
1865 fjy3 = _mm_add_pd(fjy3,ty);
1866 fjz3 = _mm_add_pd(fjz3,tz);
1868 /**************************
1869 * CALCULATE INTERACTIONS *
1870 **************************/
1872 r31 = _mm_mul_pd(rsq31,rinv31);
1874 /* EWALD ELECTROSTATICS */
1876 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1877 ewrt = _mm_mul_pd(r31,ewtabscale);
1878 ewitab = _mm_cvttpd_epi32(ewrt);
1879 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1880 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1882 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1883 felec = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
1887 /* Calculate temporary vectorial force */
1888 tx = _mm_mul_pd(fscal,dx31);
1889 ty = _mm_mul_pd(fscal,dy31);
1890 tz = _mm_mul_pd(fscal,dz31);
1892 /* Update vectorial force */
1893 fix3 = _mm_add_pd(fix3,tx);
1894 fiy3 = _mm_add_pd(fiy3,ty);
1895 fiz3 = _mm_add_pd(fiz3,tz);
1897 fjx1 = _mm_add_pd(fjx1,tx);
1898 fjy1 = _mm_add_pd(fjy1,ty);
1899 fjz1 = _mm_add_pd(fjz1,tz);
1901 /**************************
1902 * CALCULATE INTERACTIONS *
1903 **************************/
1905 r32 = _mm_mul_pd(rsq32,rinv32);
1907 /* EWALD ELECTROSTATICS */
1909 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1910 ewrt = _mm_mul_pd(r32,ewtabscale);
1911 ewitab = _mm_cvttpd_epi32(ewrt);
1912 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1913 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1915 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1916 felec = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
1920 /* Calculate temporary vectorial force */
1921 tx = _mm_mul_pd(fscal,dx32);
1922 ty = _mm_mul_pd(fscal,dy32);
1923 tz = _mm_mul_pd(fscal,dz32);
1925 /* Update vectorial force */
1926 fix3 = _mm_add_pd(fix3,tx);
1927 fiy3 = _mm_add_pd(fiy3,ty);
1928 fiz3 = _mm_add_pd(fiz3,tz);
1930 fjx2 = _mm_add_pd(fjx2,tx);
1931 fjy2 = _mm_add_pd(fjy2,ty);
1932 fjz2 = _mm_add_pd(fjz2,tz);
1934 /**************************
1935 * CALCULATE INTERACTIONS *
1936 **************************/
1938 r33 = _mm_mul_pd(rsq33,rinv33);
1940 /* EWALD ELECTROSTATICS */
1942 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1943 ewrt = _mm_mul_pd(r33,ewtabscale);
1944 ewitab = _mm_cvttpd_epi32(ewrt);
1945 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1946 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1948 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1949 felec = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
1953 /* Calculate temporary vectorial force */
1954 tx = _mm_mul_pd(fscal,dx33);
1955 ty = _mm_mul_pd(fscal,dy33);
1956 tz = _mm_mul_pd(fscal,dz33);
1958 /* Update vectorial force */
1959 fix3 = _mm_add_pd(fix3,tx);
1960 fiy3 = _mm_add_pd(fiy3,ty);
1961 fiz3 = _mm_add_pd(fiz3,tz);
1963 fjx3 = _mm_add_pd(fjx3,tx);
1964 fjy3 = _mm_add_pd(fjy3,ty);
1965 fjz3 = _mm_add_pd(fjz3,tz);
1967 gmx_mm_decrement_4rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1969 /* Inner loop uses 375 flops */
1972 if(jidx<j_index_end)
1976 j_coord_offsetA = DIM*jnrA;
1978 /* load j atom coordinates */
1979 gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
1980 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1981 &jy2,&jz2,&jx3,&jy3,&jz3);
1983 /* Calculate displacement vector */
1984 dx00 = _mm_sub_pd(ix0,jx0);
1985 dy00 = _mm_sub_pd(iy0,jy0);
1986 dz00 = _mm_sub_pd(iz0,jz0);
1987 dx11 = _mm_sub_pd(ix1,jx1);
1988 dy11 = _mm_sub_pd(iy1,jy1);
1989 dz11 = _mm_sub_pd(iz1,jz1);
1990 dx12 = _mm_sub_pd(ix1,jx2);
1991 dy12 = _mm_sub_pd(iy1,jy2);
1992 dz12 = _mm_sub_pd(iz1,jz2);
1993 dx13 = _mm_sub_pd(ix1,jx3);
1994 dy13 = _mm_sub_pd(iy1,jy3);
1995 dz13 = _mm_sub_pd(iz1,jz3);
1996 dx21 = _mm_sub_pd(ix2,jx1);
1997 dy21 = _mm_sub_pd(iy2,jy1);
1998 dz21 = _mm_sub_pd(iz2,jz1);
1999 dx22 = _mm_sub_pd(ix2,jx2);
2000 dy22 = _mm_sub_pd(iy2,jy2);
2001 dz22 = _mm_sub_pd(iz2,jz2);
2002 dx23 = _mm_sub_pd(ix2,jx3);
2003 dy23 = _mm_sub_pd(iy2,jy3);
2004 dz23 = _mm_sub_pd(iz2,jz3);
2005 dx31 = _mm_sub_pd(ix3,jx1);
2006 dy31 = _mm_sub_pd(iy3,jy1);
2007 dz31 = _mm_sub_pd(iz3,jz1);
2008 dx32 = _mm_sub_pd(ix3,jx2);
2009 dy32 = _mm_sub_pd(iy3,jy2);
2010 dz32 = _mm_sub_pd(iz3,jz2);
2011 dx33 = _mm_sub_pd(ix3,jx3);
2012 dy33 = _mm_sub_pd(iy3,jy3);
2013 dz33 = _mm_sub_pd(iz3,jz3);
2015 /* Calculate squared distance and things based on it */
2016 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
2017 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
2018 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
2019 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
2020 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
2021 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
2022 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
2023 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
2024 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
2025 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
2027 rinv00 = gmx_mm_invsqrt_pd(rsq00);
2028 rinv11 = gmx_mm_invsqrt_pd(rsq11);
2029 rinv12 = gmx_mm_invsqrt_pd(rsq12);
2030 rinv13 = gmx_mm_invsqrt_pd(rsq13);
2031 rinv21 = gmx_mm_invsqrt_pd(rsq21);
2032 rinv22 = gmx_mm_invsqrt_pd(rsq22);
2033 rinv23 = gmx_mm_invsqrt_pd(rsq23);
2034 rinv31 = gmx_mm_invsqrt_pd(rsq31);
2035 rinv32 = gmx_mm_invsqrt_pd(rsq32);
2036 rinv33 = gmx_mm_invsqrt_pd(rsq33);
2038 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
2039 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
2040 rinvsq13 = _mm_mul_pd(rinv13,rinv13);
2041 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
2042 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
2043 rinvsq23 = _mm_mul_pd(rinv23,rinv23);
2044 rinvsq31 = _mm_mul_pd(rinv31,rinv31);
2045 rinvsq32 = _mm_mul_pd(rinv32,rinv32);
2046 rinvsq33 = _mm_mul_pd(rinv33,rinv33);
2048 fjx0 = _mm_setzero_pd();
2049 fjy0 = _mm_setzero_pd();
2050 fjz0 = _mm_setzero_pd();
2051 fjx1 = _mm_setzero_pd();
2052 fjy1 = _mm_setzero_pd();
2053 fjz1 = _mm_setzero_pd();
2054 fjx2 = _mm_setzero_pd();
2055 fjy2 = _mm_setzero_pd();
2056 fjz2 = _mm_setzero_pd();
2057 fjx3 = _mm_setzero_pd();
2058 fjy3 = _mm_setzero_pd();
2059 fjz3 = _mm_setzero_pd();
2061 /**************************
2062 * CALCULATE INTERACTIONS *
2063 **************************/
2065 r00 = _mm_mul_pd(rsq00,rinv00);
2067 /* Calculate table index by multiplying r with table scale and truncate to integer */
2068 rt = _mm_mul_pd(r00,vftabscale);
2069 vfitab = _mm_cvttpd_epi32(rt);
2070 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2071 vfitab = _mm_slli_epi32(vfitab,3);
2073 /* CUBIC SPLINE TABLE DISPERSION */
2074 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2075 F = _mm_setzero_pd();
2076 GMX_MM_TRANSPOSE2_PD(Y,F);
2077 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2078 H = _mm_setzero_pd();
2079 GMX_MM_TRANSPOSE2_PD(G,H);
2080 Heps = _mm_mul_pd(vfeps,H);
2081 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2082 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2083 fvdw6 = _mm_mul_pd(c6_00,FF);
2085 /* CUBIC SPLINE TABLE REPULSION */
2086 vfitab = _mm_add_epi32(vfitab,ifour);
2087 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2088 F = _mm_setzero_pd();
2089 GMX_MM_TRANSPOSE2_PD(Y,F);
2090 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2091 H = _mm_setzero_pd();
2092 GMX_MM_TRANSPOSE2_PD(G,H);
2093 Heps = _mm_mul_pd(vfeps,H);
2094 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2095 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2096 fvdw12 = _mm_mul_pd(c12_00,FF);
2097 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
2101 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2103 /* Calculate temporary vectorial force */
2104 tx = _mm_mul_pd(fscal,dx00);
2105 ty = _mm_mul_pd(fscal,dy00);
2106 tz = _mm_mul_pd(fscal,dz00);
2108 /* Update vectorial force */
2109 fix0 = _mm_add_pd(fix0,tx);
2110 fiy0 = _mm_add_pd(fiy0,ty);
2111 fiz0 = _mm_add_pd(fiz0,tz);
2113 fjx0 = _mm_add_pd(fjx0,tx);
2114 fjy0 = _mm_add_pd(fjy0,ty);
2115 fjz0 = _mm_add_pd(fjz0,tz);
2117 /**************************
2118 * CALCULATE INTERACTIONS *
2119 **************************/
2121 r11 = _mm_mul_pd(rsq11,rinv11);
2123 /* EWALD ELECTROSTATICS */
2125 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2126 ewrt = _mm_mul_pd(r11,ewtabscale);
2127 ewitab = _mm_cvttpd_epi32(ewrt);
2128 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2129 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2130 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2131 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
2135 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2137 /* Calculate temporary vectorial force */
2138 tx = _mm_mul_pd(fscal,dx11);
2139 ty = _mm_mul_pd(fscal,dy11);
2140 tz = _mm_mul_pd(fscal,dz11);
2142 /* Update vectorial force */
2143 fix1 = _mm_add_pd(fix1,tx);
2144 fiy1 = _mm_add_pd(fiy1,ty);
2145 fiz1 = _mm_add_pd(fiz1,tz);
2147 fjx1 = _mm_add_pd(fjx1,tx);
2148 fjy1 = _mm_add_pd(fjy1,ty);
2149 fjz1 = _mm_add_pd(fjz1,tz);
2151 /**************************
2152 * CALCULATE INTERACTIONS *
2153 **************************/
2155 r12 = _mm_mul_pd(rsq12,rinv12);
2157 /* EWALD ELECTROSTATICS */
2159 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2160 ewrt = _mm_mul_pd(r12,ewtabscale);
2161 ewitab = _mm_cvttpd_epi32(ewrt);
2162 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2163 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2164 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2165 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
2169 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2171 /* Calculate temporary vectorial force */
2172 tx = _mm_mul_pd(fscal,dx12);
2173 ty = _mm_mul_pd(fscal,dy12);
2174 tz = _mm_mul_pd(fscal,dz12);
2176 /* Update vectorial force */
2177 fix1 = _mm_add_pd(fix1,tx);
2178 fiy1 = _mm_add_pd(fiy1,ty);
2179 fiz1 = _mm_add_pd(fiz1,tz);
2181 fjx2 = _mm_add_pd(fjx2,tx);
2182 fjy2 = _mm_add_pd(fjy2,ty);
2183 fjz2 = _mm_add_pd(fjz2,tz);
2185 /**************************
2186 * CALCULATE INTERACTIONS *
2187 **************************/
2189 r13 = _mm_mul_pd(rsq13,rinv13);
2191 /* EWALD ELECTROSTATICS */
2193 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2194 ewrt = _mm_mul_pd(r13,ewtabscale);
2195 ewitab = _mm_cvttpd_epi32(ewrt);
2196 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2197 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2198 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2199 felec = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
2203 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2205 /* Calculate temporary vectorial force */
2206 tx = _mm_mul_pd(fscal,dx13);
2207 ty = _mm_mul_pd(fscal,dy13);
2208 tz = _mm_mul_pd(fscal,dz13);
2210 /* Update vectorial force */
2211 fix1 = _mm_add_pd(fix1,tx);
2212 fiy1 = _mm_add_pd(fiy1,ty);
2213 fiz1 = _mm_add_pd(fiz1,tz);
2215 fjx3 = _mm_add_pd(fjx3,tx);
2216 fjy3 = _mm_add_pd(fjy3,ty);
2217 fjz3 = _mm_add_pd(fjz3,tz);
2219 /**************************
2220 * CALCULATE INTERACTIONS *
2221 **************************/
2223 r21 = _mm_mul_pd(rsq21,rinv21);
2225 /* EWALD ELECTROSTATICS */
2227 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2228 ewrt = _mm_mul_pd(r21,ewtabscale);
2229 ewitab = _mm_cvttpd_epi32(ewrt);
2230 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2231 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2232 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2233 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
2237 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2239 /* Calculate temporary vectorial force */
2240 tx = _mm_mul_pd(fscal,dx21);
2241 ty = _mm_mul_pd(fscal,dy21);
2242 tz = _mm_mul_pd(fscal,dz21);
2244 /* Update vectorial force */
2245 fix2 = _mm_add_pd(fix2,tx);
2246 fiy2 = _mm_add_pd(fiy2,ty);
2247 fiz2 = _mm_add_pd(fiz2,tz);
2249 fjx1 = _mm_add_pd(fjx1,tx);
2250 fjy1 = _mm_add_pd(fjy1,ty);
2251 fjz1 = _mm_add_pd(fjz1,tz);
2253 /**************************
2254 * CALCULATE INTERACTIONS *
2255 **************************/
2257 r22 = _mm_mul_pd(rsq22,rinv22);
2259 /* EWALD ELECTROSTATICS */
2261 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2262 ewrt = _mm_mul_pd(r22,ewtabscale);
2263 ewitab = _mm_cvttpd_epi32(ewrt);
2264 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2265 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2266 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2267 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
2271 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2273 /* Calculate temporary vectorial force */
2274 tx = _mm_mul_pd(fscal,dx22);
2275 ty = _mm_mul_pd(fscal,dy22);
2276 tz = _mm_mul_pd(fscal,dz22);
2278 /* Update vectorial force */
2279 fix2 = _mm_add_pd(fix2,tx);
2280 fiy2 = _mm_add_pd(fiy2,ty);
2281 fiz2 = _mm_add_pd(fiz2,tz);
2283 fjx2 = _mm_add_pd(fjx2,tx);
2284 fjy2 = _mm_add_pd(fjy2,ty);
2285 fjz2 = _mm_add_pd(fjz2,tz);
2287 /**************************
2288 * CALCULATE INTERACTIONS *
2289 **************************/
2291 r23 = _mm_mul_pd(rsq23,rinv23);
2293 /* EWALD ELECTROSTATICS */
2295 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2296 ewrt = _mm_mul_pd(r23,ewtabscale);
2297 ewitab = _mm_cvttpd_epi32(ewrt);
2298 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2299 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2300 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2301 felec = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
2305 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2307 /* Calculate temporary vectorial force */
2308 tx = _mm_mul_pd(fscal,dx23);
2309 ty = _mm_mul_pd(fscal,dy23);
2310 tz = _mm_mul_pd(fscal,dz23);
2312 /* Update vectorial force */
2313 fix2 = _mm_add_pd(fix2,tx);
2314 fiy2 = _mm_add_pd(fiy2,ty);
2315 fiz2 = _mm_add_pd(fiz2,tz);
2317 fjx3 = _mm_add_pd(fjx3,tx);
2318 fjy3 = _mm_add_pd(fjy3,ty);
2319 fjz3 = _mm_add_pd(fjz3,tz);
2321 /**************************
2322 * CALCULATE INTERACTIONS *
2323 **************************/
2325 r31 = _mm_mul_pd(rsq31,rinv31);
2327 /* EWALD ELECTROSTATICS */
2329 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2330 ewrt = _mm_mul_pd(r31,ewtabscale);
2331 ewitab = _mm_cvttpd_epi32(ewrt);
2332 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2333 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2334 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2335 felec = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
2339 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2341 /* Calculate temporary vectorial force */
2342 tx = _mm_mul_pd(fscal,dx31);
2343 ty = _mm_mul_pd(fscal,dy31);
2344 tz = _mm_mul_pd(fscal,dz31);
2346 /* Update vectorial force */
2347 fix3 = _mm_add_pd(fix3,tx);
2348 fiy3 = _mm_add_pd(fiy3,ty);
2349 fiz3 = _mm_add_pd(fiz3,tz);
2351 fjx1 = _mm_add_pd(fjx1,tx);
2352 fjy1 = _mm_add_pd(fjy1,ty);
2353 fjz1 = _mm_add_pd(fjz1,tz);
2355 /**************************
2356 * CALCULATE INTERACTIONS *
2357 **************************/
2359 r32 = _mm_mul_pd(rsq32,rinv32);
2361 /* EWALD ELECTROSTATICS */
2363 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2364 ewrt = _mm_mul_pd(r32,ewtabscale);
2365 ewitab = _mm_cvttpd_epi32(ewrt);
2366 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2367 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2368 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2369 felec = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
2373 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2375 /* Calculate temporary vectorial force */
2376 tx = _mm_mul_pd(fscal,dx32);
2377 ty = _mm_mul_pd(fscal,dy32);
2378 tz = _mm_mul_pd(fscal,dz32);
2380 /* Update vectorial force */
2381 fix3 = _mm_add_pd(fix3,tx);
2382 fiy3 = _mm_add_pd(fiy3,ty);
2383 fiz3 = _mm_add_pd(fiz3,tz);
2385 fjx2 = _mm_add_pd(fjx2,tx);
2386 fjy2 = _mm_add_pd(fjy2,ty);
2387 fjz2 = _mm_add_pd(fjz2,tz);
2389 /**************************
2390 * CALCULATE INTERACTIONS *
2391 **************************/
2393 r33 = _mm_mul_pd(rsq33,rinv33);
2395 /* EWALD ELECTROSTATICS */
2397 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2398 ewrt = _mm_mul_pd(r33,ewtabscale);
2399 ewitab = _mm_cvttpd_epi32(ewrt);
2400 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2401 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2402 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2403 felec = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
2407 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2409 /* Calculate temporary vectorial force */
2410 tx = _mm_mul_pd(fscal,dx33);
2411 ty = _mm_mul_pd(fscal,dy33);
2412 tz = _mm_mul_pd(fscal,dz33);
2414 /* Update vectorial force */
2415 fix3 = _mm_add_pd(fix3,tx);
2416 fiy3 = _mm_add_pd(fiy3,ty);
2417 fiz3 = _mm_add_pd(fiz3,tz);
2419 fjx3 = _mm_add_pd(fjx3,tx);
2420 fjy3 = _mm_add_pd(fjy3,ty);
2421 fjz3 = _mm_add_pd(fjz3,tz);
2423 gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2425 /* Inner loop uses 375 flops */
2428 /* End of innermost loop */
2430 gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2431 f+i_coord_offset,fshift+i_shift_offset);
2433 /* Increment number of inner iterations */
2434 inneriter += j_index_end - j_index_start;
2436 /* Outer loop uses 24 flops */
2439 /* Increment number of outer iterations */
2442 /* Update outer/inner flops */
2444 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*375);