2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013,2014,2015,2017,2018, 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/gmxlib/nrnb.h"
47 #include "kernelutil_x86_sse2_double.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW3W3_VF_sse2_double
51 * Electrostatics interaction: Ewald
52 * VdW interaction: CubicSplineTable
53 * Geometry: Water3-Water3
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecEw_VdwCSTab_GeomW3W3_VF_sse2_double
58 (t_nblist * gmx_restrict nlist,
59 rvec * gmx_restrict xx,
60 rvec * gmx_restrict ff,
61 struct t_forcerec * gmx_restrict fr,
62 t_mdatoms * gmx_restrict mdatoms,
63 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
64 t_nrnb * gmx_restrict nrnb)
66 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
67 * just 0 for non-waters.
68 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
69 * jnr indices corresponding to data put in the four positions in the SIMD register.
71 int i_shift_offset,i_coord_offset,outeriter,inneriter;
72 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
74 int j_coord_offsetA,j_coord_offsetB;
75 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
77 real *shiftvec,*fshift,*x,*f;
78 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
80 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
82 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
84 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
85 int vdwjidx0A,vdwjidx0B;
86 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
87 int vdwjidx1A,vdwjidx1B;
88 __m128d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
89 int vdwjidx2A,vdwjidx2B;
90 __m128d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
91 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
92 __m128d dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
93 __m128d dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
94 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
95 __m128d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
96 __m128d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
97 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
98 __m128d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
99 __m128d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
100 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
103 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
106 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
107 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
109 __m128i ifour = _mm_set1_epi32(4);
110 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
113 __m128d ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
115 __m128d dummy_mask,cutoff_mask;
116 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
117 __m128d one = _mm_set1_pd(1.0);
118 __m128d two = _mm_set1_pd(2.0);
124 jindex = nlist->jindex;
126 shiftidx = nlist->shift;
128 shiftvec = fr->shift_vec[0];
129 fshift = fr->fshift[0];
130 facel = _mm_set1_pd(fr->ic->epsfac);
131 charge = mdatoms->chargeA;
132 nvdwtype = fr->ntype;
134 vdwtype = mdatoms->typeA;
136 vftab = kernel_data->table_vdw->data;
137 vftabscale = _mm_set1_pd(kernel_data->table_vdw->scale);
139 sh_ewald = _mm_set1_pd(fr->ic->sh_ewald);
140 ewtab = fr->ic->tabq_coul_FDV0;
141 ewtabscale = _mm_set1_pd(fr->ic->tabq_scale);
142 ewtabhalfspace = _mm_set1_pd(0.5/fr->ic->tabq_scale);
144 /* Setup water-specific parameters */
145 inr = nlist->iinr[0];
146 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
147 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
148 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
149 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
151 jq0 = _mm_set1_pd(charge[inr+0]);
152 jq1 = _mm_set1_pd(charge[inr+1]);
153 jq2 = _mm_set1_pd(charge[inr+2]);
154 vdwjidx0A = 2*vdwtype[inr+0];
155 qq00 = _mm_mul_pd(iq0,jq0);
156 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
157 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
158 qq01 = _mm_mul_pd(iq0,jq1);
159 qq02 = _mm_mul_pd(iq0,jq2);
160 qq10 = _mm_mul_pd(iq1,jq0);
161 qq11 = _mm_mul_pd(iq1,jq1);
162 qq12 = _mm_mul_pd(iq1,jq2);
163 qq20 = _mm_mul_pd(iq2,jq0);
164 qq21 = _mm_mul_pd(iq2,jq1);
165 qq22 = _mm_mul_pd(iq2,jq2);
167 /* Avoid stupid compiler warnings */
175 /* Start outer loop over neighborlists */
176 for(iidx=0; iidx<nri; iidx++)
178 /* Load shift vector for this list */
179 i_shift_offset = DIM*shiftidx[iidx];
181 /* Load limits for loop over neighbors */
182 j_index_start = jindex[iidx];
183 j_index_end = jindex[iidx+1];
185 /* Get outer coordinate index */
187 i_coord_offset = DIM*inr;
189 /* Load i particle coords and add shift vector */
190 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
191 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
193 fix0 = _mm_setzero_pd();
194 fiy0 = _mm_setzero_pd();
195 fiz0 = _mm_setzero_pd();
196 fix1 = _mm_setzero_pd();
197 fiy1 = _mm_setzero_pd();
198 fiz1 = _mm_setzero_pd();
199 fix2 = _mm_setzero_pd();
200 fiy2 = _mm_setzero_pd();
201 fiz2 = _mm_setzero_pd();
203 /* Reset potential sums */
204 velecsum = _mm_setzero_pd();
205 vvdwsum = _mm_setzero_pd();
207 /* Start inner kernel loop */
208 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
211 /* Get j neighbor index, and coordinate index */
214 j_coord_offsetA = DIM*jnrA;
215 j_coord_offsetB = DIM*jnrB;
217 /* load j atom coordinates */
218 gmx_mm_load_3rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
219 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
221 /* Calculate displacement vector */
222 dx00 = _mm_sub_pd(ix0,jx0);
223 dy00 = _mm_sub_pd(iy0,jy0);
224 dz00 = _mm_sub_pd(iz0,jz0);
225 dx01 = _mm_sub_pd(ix0,jx1);
226 dy01 = _mm_sub_pd(iy0,jy1);
227 dz01 = _mm_sub_pd(iz0,jz1);
228 dx02 = _mm_sub_pd(ix0,jx2);
229 dy02 = _mm_sub_pd(iy0,jy2);
230 dz02 = _mm_sub_pd(iz0,jz2);
231 dx10 = _mm_sub_pd(ix1,jx0);
232 dy10 = _mm_sub_pd(iy1,jy0);
233 dz10 = _mm_sub_pd(iz1,jz0);
234 dx11 = _mm_sub_pd(ix1,jx1);
235 dy11 = _mm_sub_pd(iy1,jy1);
236 dz11 = _mm_sub_pd(iz1,jz1);
237 dx12 = _mm_sub_pd(ix1,jx2);
238 dy12 = _mm_sub_pd(iy1,jy2);
239 dz12 = _mm_sub_pd(iz1,jz2);
240 dx20 = _mm_sub_pd(ix2,jx0);
241 dy20 = _mm_sub_pd(iy2,jy0);
242 dz20 = _mm_sub_pd(iz2,jz0);
243 dx21 = _mm_sub_pd(ix2,jx1);
244 dy21 = _mm_sub_pd(iy2,jy1);
245 dz21 = _mm_sub_pd(iz2,jz1);
246 dx22 = _mm_sub_pd(ix2,jx2);
247 dy22 = _mm_sub_pd(iy2,jy2);
248 dz22 = _mm_sub_pd(iz2,jz2);
250 /* Calculate squared distance and things based on it */
251 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
252 rsq01 = gmx_mm_calc_rsq_pd(dx01,dy01,dz01);
253 rsq02 = gmx_mm_calc_rsq_pd(dx02,dy02,dz02);
254 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
255 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
256 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
257 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
258 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
259 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
261 rinv00 = sse2_invsqrt_d(rsq00);
262 rinv01 = sse2_invsqrt_d(rsq01);
263 rinv02 = sse2_invsqrt_d(rsq02);
264 rinv10 = sse2_invsqrt_d(rsq10);
265 rinv11 = sse2_invsqrt_d(rsq11);
266 rinv12 = sse2_invsqrt_d(rsq12);
267 rinv20 = sse2_invsqrt_d(rsq20);
268 rinv21 = sse2_invsqrt_d(rsq21);
269 rinv22 = sse2_invsqrt_d(rsq22);
271 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
272 rinvsq01 = _mm_mul_pd(rinv01,rinv01);
273 rinvsq02 = _mm_mul_pd(rinv02,rinv02);
274 rinvsq10 = _mm_mul_pd(rinv10,rinv10);
275 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
276 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
277 rinvsq20 = _mm_mul_pd(rinv20,rinv20);
278 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
279 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
281 fjx0 = _mm_setzero_pd();
282 fjy0 = _mm_setzero_pd();
283 fjz0 = _mm_setzero_pd();
284 fjx1 = _mm_setzero_pd();
285 fjy1 = _mm_setzero_pd();
286 fjz1 = _mm_setzero_pd();
287 fjx2 = _mm_setzero_pd();
288 fjy2 = _mm_setzero_pd();
289 fjz2 = _mm_setzero_pd();
291 /**************************
292 * CALCULATE INTERACTIONS *
293 **************************/
295 r00 = _mm_mul_pd(rsq00,rinv00);
297 /* Calculate table index by multiplying r with table scale and truncate to integer */
298 rt = _mm_mul_pd(r00,vftabscale);
299 vfitab = _mm_cvttpd_epi32(rt);
300 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
301 vfitab = _mm_slli_epi32(vfitab,3);
303 /* EWALD ELECTROSTATICS */
305 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
306 ewrt = _mm_mul_pd(r00,ewtabscale);
307 ewitab = _mm_cvttpd_epi32(ewrt);
308 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
309 ewitab = _mm_slli_epi32(ewitab,2);
310 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
311 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
312 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
313 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
314 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
315 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
316 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
317 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
318 velec = _mm_mul_pd(qq00,_mm_sub_pd(rinv00,velec));
319 felec = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
321 /* CUBIC SPLINE TABLE DISPERSION */
322 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
323 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
324 GMX_MM_TRANSPOSE2_PD(Y,F);
325 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
326 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
327 GMX_MM_TRANSPOSE2_PD(G,H);
328 Heps = _mm_mul_pd(vfeps,H);
329 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
330 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
331 vvdw6 = _mm_mul_pd(c6_00,VV);
332 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
333 fvdw6 = _mm_mul_pd(c6_00,FF);
335 /* CUBIC SPLINE TABLE REPULSION */
336 vfitab = _mm_add_epi32(vfitab,ifour);
337 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
338 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
339 GMX_MM_TRANSPOSE2_PD(Y,F);
340 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
341 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
342 GMX_MM_TRANSPOSE2_PD(G,H);
343 Heps = _mm_mul_pd(vfeps,H);
344 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
345 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
346 vvdw12 = _mm_mul_pd(c12_00,VV);
347 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
348 fvdw12 = _mm_mul_pd(c12_00,FF);
349 vvdw = _mm_add_pd(vvdw12,vvdw6);
350 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
352 /* Update potential sum for this i atom from the interaction with this j atom. */
353 velecsum = _mm_add_pd(velecsum,velec);
354 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
356 fscal = _mm_add_pd(felec,fvdw);
358 /* Calculate temporary vectorial force */
359 tx = _mm_mul_pd(fscal,dx00);
360 ty = _mm_mul_pd(fscal,dy00);
361 tz = _mm_mul_pd(fscal,dz00);
363 /* Update vectorial force */
364 fix0 = _mm_add_pd(fix0,tx);
365 fiy0 = _mm_add_pd(fiy0,ty);
366 fiz0 = _mm_add_pd(fiz0,tz);
368 fjx0 = _mm_add_pd(fjx0,tx);
369 fjy0 = _mm_add_pd(fjy0,ty);
370 fjz0 = _mm_add_pd(fjz0,tz);
372 /**************************
373 * CALCULATE INTERACTIONS *
374 **************************/
376 r01 = _mm_mul_pd(rsq01,rinv01);
378 /* EWALD ELECTROSTATICS */
380 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
381 ewrt = _mm_mul_pd(r01,ewtabscale);
382 ewitab = _mm_cvttpd_epi32(ewrt);
383 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
384 ewitab = _mm_slli_epi32(ewitab,2);
385 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
386 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
387 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
388 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
389 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
390 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
391 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
392 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
393 velec = _mm_mul_pd(qq01,_mm_sub_pd(rinv01,velec));
394 felec = _mm_mul_pd(_mm_mul_pd(qq01,rinv01),_mm_sub_pd(rinvsq01,felec));
396 /* Update potential sum for this i atom from the interaction with this j atom. */
397 velecsum = _mm_add_pd(velecsum,velec);
401 /* Calculate temporary vectorial force */
402 tx = _mm_mul_pd(fscal,dx01);
403 ty = _mm_mul_pd(fscal,dy01);
404 tz = _mm_mul_pd(fscal,dz01);
406 /* Update vectorial force */
407 fix0 = _mm_add_pd(fix0,tx);
408 fiy0 = _mm_add_pd(fiy0,ty);
409 fiz0 = _mm_add_pd(fiz0,tz);
411 fjx1 = _mm_add_pd(fjx1,tx);
412 fjy1 = _mm_add_pd(fjy1,ty);
413 fjz1 = _mm_add_pd(fjz1,tz);
415 /**************************
416 * CALCULATE INTERACTIONS *
417 **************************/
419 r02 = _mm_mul_pd(rsq02,rinv02);
421 /* EWALD ELECTROSTATICS */
423 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
424 ewrt = _mm_mul_pd(r02,ewtabscale);
425 ewitab = _mm_cvttpd_epi32(ewrt);
426 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
427 ewitab = _mm_slli_epi32(ewitab,2);
428 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
429 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
430 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
431 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
432 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
433 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
434 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
435 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
436 velec = _mm_mul_pd(qq02,_mm_sub_pd(rinv02,velec));
437 felec = _mm_mul_pd(_mm_mul_pd(qq02,rinv02),_mm_sub_pd(rinvsq02,felec));
439 /* Update potential sum for this i atom from the interaction with this j atom. */
440 velecsum = _mm_add_pd(velecsum,velec);
444 /* Calculate temporary vectorial force */
445 tx = _mm_mul_pd(fscal,dx02);
446 ty = _mm_mul_pd(fscal,dy02);
447 tz = _mm_mul_pd(fscal,dz02);
449 /* Update vectorial force */
450 fix0 = _mm_add_pd(fix0,tx);
451 fiy0 = _mm_add_pd(fiy0,ty);
452 fiz0 = _mm_add_pd(fiz0,tz);
454 fjx2 = _mm_add_pd(fjx2,tx);
455 fjy2 = _mm_add_pd(fjy2,ty);
456 fjz2 = _mm_add_pd(fjz2,tz);
458 /**************************
459 * CALCULATE INTERACTIONS *
460 **************************/
462 r10 = _mm_mul_pd(rsq10,rinv10);
464 /* EWALD ELECTROSTATICS */
466 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
467 ewrt = _mm_mul_pd(r10,ewtabscale);
468 ewitab = _mm_cvttpd_epi32(ewrt);
469 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
470 ewitab = _mm_slli_epi32(ewitab,2);
471 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
472 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
473 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
474 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
475 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
476 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
477 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
478 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
479 velec = _mm_mul_pd(qq10,_mm_sub_pd(rinv10,velec));
480 felec = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
482 /* Update potential sum for this i atom from the interaction with this j atom. */
483 velecsum = _mm_add_pd(velecsum,velec);
487 /* Calculate temporary vectorial force */
488 tx = _mm_mul_pd(fscal,dx10);
489 ty = _mm_mul_pd(fscal,dy10);
490 tz = _mm_mul_pd(fscal,dz10);
492 /* Update vectorial force */
493 fix1 = _mm_add_pd(fix1,tx);
494 fiy1 = _mm_add_pd(fiy1,ty);
495 fiz1 = _mm_add_pd(fiz1,tz);
497 fjx0 = _mm_add_pd(fjx0,tx);
498 fjy0 = _mm_add_pd(fjy0,ty);
499 fjz0 = _mm_add_pd(fjz0,tz);
501 /**************************
502 * CALCULATE INTERACTIONS *
503 **************************/
505 r11 = _mm_mul_pd(rsq11,rinv11);
507 /* EWALD ELECTROSTATICS */
509 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
510 ewrt = _mm_mul_pd(r11,ewtabscale);
511 ewitab = _mm_cvttpd_epi32(ewrt);
512 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
513 ewitab = _mm_slli_epi32(ewitab,2);
514 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
515 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
516 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
517 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
518 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
519 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
520 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
521 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
522 velec = _mm_mul_pd(qq11,_mm_sub_pd(rinv11,velec));
523 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
525 /* Update potential sum for this i atom from the interaction with this j atom. */
526 velecsum = _mm_add_pd(velecsum,velec);
530 /* Calculate temporary vectorial force */
531 tx = _mm_mul_pd(fscal,dx11);
532 ty = _mm_mul_pd(fscal,dy11);
533 tz = _mm_mul_pd(fscal,dz11);
535 /* Update vectorial force */
536 fix1 = _mm_add_pd(fix1,tx);
537 fiy1 = _mm_add_pd(fiy1,ty);
538 fiz1 = _mm_add_pd(fiz1,tz);
540 fjx1 = _mm_add_pd(fjx1,tx);
541 fjy1 = _mm_add_pd(fjy1,ty);
542 fjz1 = _mm_add_pd(fjz1,tz);
544 /**************************
545 * CALCULATE INTERACTIONS *
546 **************************/
548 r12 = _mm_mul_pd(rsq12,rinv12);
550 /* EWALD ELECTROSTATICS */
552 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
553 ewrt = _mm_mul_pd(r12,ewtabscale);
554 ewitab = _mm_cvttpd_epi32(ewrt);
555 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
556 ewitab = _mm_slli_epi32(ewitab,2);
557 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
558 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
559 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
560 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
561 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
562 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
563 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
564 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
565 velec = _mm_mul_pd(qq12,_mm_sub_pd(rinv12,velec));
566 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
568 /* Update potential sum for this i atom from the interaction with this j atom. */
569 velecsum = _mm_add_pd(velecsum,velec);
573 /* Calculate temporary vectorial force */
574 tx = _mm_mul_pd(fscal,dx12);
575 ty = _mm_mul_pd(fscal,dy12);
576 tz = _mm_mul_pd(fscal,dz12);
578 /* Update vectorial force */
579 fix1 = _mm_add_pd(fix1,tx);
580 fiy1 = _mm_add_pd(fiy1,ty);
581 fiz1 = _mm_add_pd(fiz1,tz);
583 fjx2 = _mm_add_pd(fjx2,tx);
584 fjy2 = _mm_add_pd(fjy2,ty);
585 fjz2 = _mm_add_pd(fjz2,tz);
587 /**************************
588 * CALCULATE INTERACTIONS *
589 **************************/
591 r20 = _mm_mul_pd(rsq20,rinv20);
593 /* EWALD ELECTROSTATICS */
595 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
596 ewrt = _mm_mul_pd(r20,ewtabscale);
597 ewitab = _mm_cvttpd_epi32(ewrt);
598 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
599 ewitab = _mm_slli_epi32(ewitab,2);
600 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
601 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
602 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
603 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
604 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
605 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
606 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
607 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
608 velec = _mm_mul_pd(qq20,_mm_sub_pd(rinv20,velec));
609 felec = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
611 /* Update potential sum for this i atom from the interaction with this j atom. */
612 velecsum = _mm_add_pd(velecsum,velec);
616 /* Calculate temporary vectorial force */
617 tx = _mm_mul_pd(fscal,dx20);
618 ty = _mm_mul_pd(fscal,dy20);
619 tz = _mm_mul_pd(fscal,dz20);
621 /* Update vectorial force */
622 fix2 = _mm_add_pd(fix2,tx);
623 fiy2 = _mm_add_pd(fiy2,ty);
624 fiz2 = _mm_add_pd(fiz2,tz);
626 fjx0 = _mm_add_pd(fjx0,tx);
627 fjy0 = _mm_add_pd(fjy0,ty);
628 fjz0 = _mm_add_pd(fjz0,tz);
630 /**************************
631 * CALCULATE INTERACTIONS *
632 **************************/
634 r21 = _mm_mul_pd(rsq21,rinv21);
636 /* EWALD ELECTROSTATICS */
638 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
639 ewrt = _mm_mul_pd(r21,ewtabscale);
640 ewitab = _mm_cvttpd_epi32(ewrt);
641 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
642 ewitab = _mm_slli_epi32(ewitab,2);
643 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
644 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
645 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
646 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
647 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
648 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
649 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
650 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
651 velec = _mm_mul_pd(qq21,_mm_sub_pd(rinv21,velec));
652 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
654 /* Update potential sum for this i atom from the interaction with this j atom. */
655 velecsum = _mm_add_pd(velecsum,velec);
659 /* Calculate temporary vectorial force */
660 tx = _mm_mul_pd(fscal,dx21);
661 ty = _mm_mul_pd(fscal,dy21);
662 tz = _mm_mul_pd(fscal,dz21);
664 /* Update vectorial force */
665 fix2 = _mm_add_pd(fix2,tx);
666 fiy2 = _mm_add_pd(fiy2,ty);
667 fiz2 = _mm_add_pd(fiz2,tz);
669 fjx1 = _mm_add_pd(fjx1,tx);
670 fjy1 = _mm_add_pd(fjy1,ty);
671 fjz1 = _mm_add_pd(fjz1,tz);
673 /**************************
674 * CALCULATE INTERACTIONS *
675 **************************/
677 r22 = _mm_mul_pd(rsq22,rinv22);
679 /* EWALD ELECTROSTATICS */
681 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
682 ewrt = _mm_mul_pd(r22,ewtabscale);
683 ewitab = _mm_cvttpd_epi32(ewrt);
684 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
685 ewitab = _mm_slli_epi32(ewitab,2);
686 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
687 ewtabD = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
688 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
689 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
690 ewtabFn = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
691 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
692 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
693 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
694 velec = _mm_mul_pd(qq22,_mm_sub_pd(rinv22,velec));
695 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
697 /* Update potential sum for this i atom from the interaction with this j atom. */
698 velecsum = _mm_add_pd(velecsum,velec);
702 /* Calculate temporary vectorial force */
703 tx = _mm_mul_pd(fscal,dx22);
704 ty = _mm_mul_pd(fscal,dy22);
705 tz = _mm_mul_pd(fscal,dz22);
707 /* Update vectorial force */
708 fix2 = _mm_add_pd(fix2,tx);
709 fiy2 = _mm_add_pd(fiy2,ty);
710 fiz2 = _mm_add_pd(fiz2,tz);
712 fjx2 = _mm_add_pd(fjx2,tx);
713 fjy2 = _mm_add_pd(fjy2,ty);
714 fjz2 = _mm_add_pd(fjz2,tz);
716 gmx_mm_decrement_3rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
718 /* Inner loop uses 403 flops */
725 j_coord_offsetA = DIM*jnrA;
727 /* load j atom coordinates */
728 gmx_mm_load_3rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
729 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
731 /* Calculate displacement vector */
732 dx00 = _mm_sub_pd(ix0,jx0);
733 dy00 = _mm_sub_pd(iy0,jy0);
734 dz00 = _mm_sub_pd(iz0,jz0);
735 dx01 = _mm_sub_pd(ix0,jx1);
736 dy01 = _mm_sub_pd(iy0,jy1);
737 dz01 = _mm_sub_pd(iz0,jz1);
738 dx02 = _mm_sub_pd(ix0,jx2);
739 dy02 = _mm_sub_pd(iy0,jy2);
740 dz02 = _mm_sub_pd(iz0,jz2);
741 dx10 = _mm_sub_pd(ix1,jx0);
742 dy10 = _mm_sub_pd(iy1,jy0);
743 dz10 = _mm_sub_pd(iz1,jz0);
744 dx11 = _mm_sub_pd(ix1,jx1);
745 dy11 = _mm_sub_pd(iy1,jy1);
746 dz11 = _mm_sub_pd(iz1,jz1);
747 dx12 = _mm_sub_pd(ix1,jx2);
748 dy12 = _mm_sub_pd(iy1,jy2);
749 dz12 = _mm_sub_pd(iz1,jz2);
750 dx20 = _mm_sub_pd(ix2,jx0);
751 dy20 = _mm_sub_pd(iy2,jy0);
752 dz20 = _mm_sub_pd(iz2,jz0);
753 dx21 = _mm_sub_pd(ix2,jx1);
754 dy21 = _mm_sub_pd(iy2,jy1);
755 dz21 = _mm_sub_pd(iz2,jz1);
756 dx22 = _mm_sub_pd(ix2,jx2);
757 dy22 = _mm_sub_pd(iy2,jy2);
758 dz22 = _mm_sub_pd(iz2,jz2);
760 /* Calculate squared distance and things based on it */
761 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
762 rsq01 = gmx_mm_calc_rsq_pd(dx01,dy01,dz01);
763 rsq02 = gmx_mm_calc_rsq_pd(dx02,dy02,dz02);
764 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
765 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
766 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
767 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
768 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
769 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
771 rinv00 = sse2_invsqrt_d(rsq00);
772 rinv01 = sse2_invsqrt_d(rsq01);
773 rinv02 = sse2_invsqrt_d(rsq02);
774 rinv10 = sse2_invsqrt_d(rsq10);
775 rinv11 = sse2_invsqrt_d(rsq11);
776 rinv12 = sse2_invsqrt_d(rsq12);
777 rinv20 = sse2_invsqrt_d(rsq20);
778 rinv21 = sse2_invsqrt_d(rsq21);
779 rinv22 = sse2_invsqrt_d(rsq22);
781 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
782 rinvsq01 = _mm_mul_pd(rinv01,rinv01);
783 rinvsq02 = _mm_mul_pd(rinv02,rinv02);
784 rinvsq10 = _mm_mul_pd(rinv10,rinv10);
785 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
786 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
787 rinvsq20 = _mm_mul_pd(rinv20,rinv20);
788 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
789 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
791 fjx0 = _mm_setzero_pd();
792 fjy0 = _mm_setzero_pd();
793 fjz0 = _mm_setzero_pd();
794 fjx1 = _mm_setzero_pd();
795 fjy1 = _mm_setzero_pd();
796 fjz1 = _mm_setzero_pd();
797 fjx2 = _mm_setzero_pd();
798 fjy2 = _mm_setzero_pd();
799 fjz2 = _mm_setzero_pd();
801 /**************************
802 * CALCULATE INTERACTIONS *
803 **************************/
805 r00 = _mm_mul_pd(rsq00,rinv00);
807 /* Calculate table index by multiplying r with table scale and truncate to integer */
808 rt = _mm_mul_pd(r00,vftabscale);
809 vfitab = _mm_cvttpd_epi32(rt);
810 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
811 vfitab = _mm_slli_epi32(vfitab,3);
813 /* EWALD ELECTROSTATICS */
815 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
816 ewrt = _mm_mul_pd(r00,ewtabscale);
817 ewitab = _mm_cvttpd_epi32(ewrt);
818 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
819 ewitab = _mm_slli_epi32(ewitab,2);
820 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
821 ewtabD = _mm_setzero_pd();
822 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
823 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
824 ewtabFn = _mm_setzero_pd();
825 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
826 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
827 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
828 velec = _mm_mul_pd(qq00,_mm_sub_pd(rinv00,velec));
829 felec = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
831 /* CUBIC SPLINE TABLE DISPERSION */
832 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
833 F = _mm_setzero_pd();
834 GMX_MM_TRANSPOSE2_PD(Y,F);
835 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
836 H = _mm_setzero_pd();
837 GMX_MM_TRANSPOSE2_PD(G,H);
838 Heps = _mm_mul_pd(vfeps,H);
839 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
840 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
841 vvdw6 = _mm_mul_pd(c6_00,VV);
842 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
843 fvdw6 = _mm_mul_pd(c6_00,FF);
845 /* CUBIC SPLINE TABLE REPULSION */
846 vfitab = _mm_add_epi32(vfitab,ifour);
847 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
848 F = _mm_setzero_pd();
849 GMX_MM_TRANSPOSE2_PD(Y,F);
850 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
851 H = _mm_setzero_pd();
852 GMX_MM_TRANSPOSE2_PD(G,H);
853 Heps = _mm_mul_pd(vfeps,H);
854 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
855 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
856 vvdw12 = _mm_mul_pd(c12_00,VV);
857 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
858 fvdw12 = _mm_mul_pd(c12_00,FF);
859 vvdw = _mm_add_pd(vvdw12,vvdw6);
860 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
862 /* Update potential sum for this i atom from the interaction with this j atom. */
863 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
864 velecsum = _mm_add_pd(velecsum,velec);
865 vvdw = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
866 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
868 fscal = _mm_add_pd(felec,fvdw);
870 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
872 /* Calculate temporary vectorial force */
873 tx = _mm_mul_pd(fscal,dx00);
874 ty = _mm_mul_pd(fscal,dy00);
875 tz = _mm_mul_pd(fscal,dz00);
877 /* Update vectorial force */
878 fix0 = _mm_add_pd(fix0,tx);
879 fiy0 = _mm_add_pd(fiy0,ty);
880 fiz0 = _mm_add_pd(fiz0,tz);
882 fjx0 = _mm_add_pd(fjx0,tx);
883 fjy0 = _mm_add_pd(fjy0,ty);
884 fjz0 = _mm_add_pd(fjz0,tz);
886 /**************************
887 * CALCULATE INTERACTIONS *
888 **************************/
890 r01 = _mm_mul_pd(rsq01,rinv01);
892 /* EWALD ELECTROSTATICS */
894 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
895 ewrt = _mm_mul_pd(r01,ewtabscale);
896 ewitab = _mm_cvttpd_epi32(ewrt);
897 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
898 ewitab = _mm_slli_epi32(ewitab,2);
899 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
900 ewtabD = _mm_setzero_pd();
901 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
902 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
903 ewtabFn = _mm_setzero_pd();
904 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
905 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
906 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
907 velec = _mm_mul_pd(qq01,_mm_sub_pd(rinv01,velec));
908 felec = _mm_mul_pd(_mm_mul_pd(qq01,rinv01),_mm_sub_pd(rinvsq01,felec));
910 /* Update potential sum for this i atom from the interaction with this j atom. */
911 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
912 velecsum = _mm_add_pd(velecsum,velec);
916 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
918 /* Calculate temporary vectorial force */
919 tx = _mm_mul_pd(fscal,dx01);
920 ty = _mm_mul_pd(fscal,dy01);
921 tz = _mm_mul_pd(fscal,dz01);
923 /* Update vectorial force */
924 fix0 = _mm_add_pd(fix0,tx);
925 fiy0 = _mm_add_pd(fiy0,ty);
926 fiz0 = _mm_add_pd(fiz0,tz);
928 fjx1 = _mm_add_pd(fjx1,tx);
929 fjy1 = _mm_add_pd(fjy1,ty);
930 fjz1 = _mm_add_pd(fjz1,tz);
932 /**************************
933 * CALCULATE INTERACTIONS *
934 **************************/
936 r02 = _mm_mul_pd(rsq02,rinv02);
938 /* EWALD ELECTROSTATICS */
940 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
941 ewrt = _mm_mul_pd(r02,ewtabscale);
942 ewitab = _mm_cvttpd_epi32(ewrt);
943 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
944 ewitab = _mm_slli_epi32(ewitab,2);
945 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
946 ewtabD = _mm_setzero_pd();
947 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
948 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
949 ewtabFn = _mm_setzero_pd();
950 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
951 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
952 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
953 velec = _mm_mul_pd(qq02,_mm_sub_pd(rinv02,velec));
954 felec = _mm_mul_pd(_mm_mul_pd(qq02,rinv02),_mm_sub_pd(rinvsq02,felec));
956 /* Update potential sum for this i atom from the interaction with this j atom. */
957 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
958 velecsum = _mm_add_pd(velecsum,velec);
962 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
964 /* Calculate temporary vectorial force */
965 tx = _mm_mul_pd(fscal,dx02);
966 ty = _mm_mul_pd(fscal,dy02);
967 tz = _mm_mul_pd(fscal,dz02);
969 /* Update vectorial force */
970 fix0 = _mm_add_pd(fix0,tx);
971 fiy0 = _mm_add_pd(fiy0,ty);
972 fiz0 = _mm_add_pd(fiz0,tz);
974 fjx2 = _mm_add_pd(fjx2,tx);
975 fjy2 = _mm_add_pd(fjy2,ty);
976 fjz2 = _mm_add_pd(fjz2,tz);
978 /**************************
979 * CALCULATE INTERACTIONS *
980 **************************/
982 r10 = _mm_mul_pd(rsq10,rinv10);
984 /* EWALD ELECTROSTATICS */
986 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
987 ewrt = _mm_mul_pd(r10,ewtabscale);
988 ewitab = _mm_cvttpd_epi32(ewrt);
989 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
990 ewitab = _mm_slli_epi32(ewitab,2);
991 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
992 ewtabD = _mm_setzero_pd();
993 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
994 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
995 ewtabFn = _mm_setzero_pd();
996 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
997 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
998 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
999 velec = _mm_mul_pd(qq10,_mm_sub_pd(rinv10,velec));
1000 felec = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1002 /* Update potential sum for this i atom from the interaction with this j atom. */
1003 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1004 velecsum = _mm_add_pd(velecsum,velec);
1008 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1010 /* Calculate temporary vectorial force */
1011 tx = _mm_mul_pd(fscal,dx10);
1012 ty = _mm_mul_pd(fscal,dy10);
1013 tz = _mm_mul_pd(fscal,dz10);
1015 /* Update vectorial force */
1016 fix1 = _mm_add_pd(fix1,tx);
1017 fiy1 = _mm_add_pd(fiy1,ty);
1018 fiz1 = _mm_add_pd(fiz1,tz);
1020 fjx0 = _mm_add_pd(fjx0,tx);
1021 fjy0 = _mm_add_pd(fjy0,ty);
1022 fjz0 = _mm_add_pd(fjz0,tz);
1024 /**************************
1025 * CALCULATE INTERACTIONS *
1026 **************************/
1028 r11 = _mm_mul_pd(rsq11,rinv11);
1030 /* EWALD ELECTROSTATICS */
1032 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1033 ewrt = _mm_mul_pd(r11,ewtabscale);
1034 ewitab = _mm_cvttpd_epi32(ewrt);
1035 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1036 ewitab = _mm_slli_epi32(ewitab,2);
1037 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1038 ewtabD = _mm_setzero_pd();
1039 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1040 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1041 ewtabFn = _mm_setzero_pd();
1042 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1043 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1044 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1045 velec = _mm_mul_pd(qq11,_mm_sub_pd(rinv11,velec));
1046 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
1048 /* Update potential sum for this i atom from the interaction with this j atom. */
1049 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1050 velecsum = _mm_add_pd(velecsum,velec);
1054 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1056 /* Calculate temporary vectorial force */
1057 tx = _mm_mul_pd(fscal,dx11);
1058 ty = _mm_mul_pd(fscal,dy11);
1059 tz = _mm_mul_pd(fscal,dz11);
1061 /* Update vectorial force */
1062 fix1 = _mm_add_pd(fix1,tx);
1063 fiy1 = _mm_add_pd(fiy1,ty);
1064 fiz1 = _mm_add_pd(fiz1,tz);
1066 fjx1 = _mm_add_pd(fjx1,tx);
1067 fjy1 = _mm_add_pd(fjy1,ty);
1068 fjz1 = _mm_add_pd(fjz1,tz);
1070 /**************************
1071 * CALCULATE INTERACTIONS *
1072 **************************/
1074 r12 = _mm_mul_pd(rsq12,rinv12);
1076 /* EWALD ELECTROSTATICS */
1078 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1079 ewrt = _mm_mul_pd(r12,ewtabscale);
1080 ewitab = _mm_cvttpd_epi32(ewrt);
1081 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1082 ewitab = _mm_slli_epi32(ewitab,2);
1083 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1084 ewtabD = _mm_setzero_pd();
1085 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1086 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1087 ewtabFn = _mm_setzero_pd();
1088 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1089 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1090 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1091 velec = _mm_mul_pd(qq12,_mm_sub_pd(rinv12,velec));
1092 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
1094 /* Update potential sum for this i atom from the interaction with this j atom. */
1095 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1096 velecsum = _mm_add_pd(velecsum,velec);
1100 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1102 /* Calculate temporary vectorial force */
1103 tx = _mm_mul_pd(fscal,dx12);
1104 ty = _mm_mul_pd(fscal,dy12);
1105 tz = _mm_mul_pd(fscal,dz12);
1107 /* Update vectorial force */
1108 fix1 = _mm_add_pd(fix1,tx);
1109 fiy1 = _mm_add_pd(fiy1,ty);
1110 fiz1 = _mm_add_pd(fiz1,tz);
1112 fjx2 = _mm_add_pd(fjx2,tx);
1113 fjy2 = _mm_add_pd(fjy2,ty);
1114 fjz2 = _mm_add_pd(fjz2,tz);
1116 /**************************
1117 * CALCULATE INTERACTIONS *
1118 **************************/
1120 r20 = _mm_mul_pd(rsq20,rinv20);
1122 /* EWALD ELECTROSTATICS */
1124 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1125 ewrt = _mm_mul_pd(r20,ewtabscale);
1126 ewitab = _mm_cvttpd_epi32(ewrt);
1127 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1128 ewitab = _mm_slli_epi32(ewitab,2);
1129 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1130 ewtabD = _mm_setzero_pd();
1131 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1132 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1133 ewtabFn = _mm_setzero_pd();
1134 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1135 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1136 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1137 velec = _mm_mul_pd(qq20,_mm_sub_pd(rinv20,velec));
1138 felec = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1140 /* Update potential sum for this i atom from the interaction with this j atom. */
1141 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1142 velecsum = _mm_add_pd(velecsum,velec);
1146 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1148 /* Calculate temporary vectorial force */
1149 tx = _mm_mul_pd(fscal,dx20);
1150 ty = _mm_mul_pd(fscal,dy20);
1151 tz = _mm_mul_pd(fscal,dz20);
1153 /* Update vectorial force */
1154 fix2 = _mm_add_pd(fix2,tx);
1155 fiy2 = _mm_add_pd(fiy2,ty);
1156 fiz2 = _mm_add_pd(fiz2,tz);
1158 fjx0 = _mm_add_pd(fjx0,tx);
1159 fjy0 = _mm_add_pd(fjy0,ty);
1160 fjz0 = _mm_add_pd(fjz0,tz);
1162 /**************************
1163 * CALCULATE INTERACTIONS *
1164 **************************/
1166 r21 = _mm_mul_pd(rsq21,rinv21);
1168 /* EWALD ELECTROSTATICS */
1170 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1171 ewrt = _mm_mul_pd(r21,ewtabscale);
1172 ewitab = _mm_cvttpd_epi32(ewrt);
1173 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1174 ewitab = _mm_slli_epi32(ewitab,2);
1175 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1176 ewtabD = _mm_setzero_pd();
1177 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1178 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1179 ewtabFn = _mm_setzero_pd();
1180 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1181 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1182 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1183 velec = _mm_mul_pd(qq21,_mm_sub_pd(rinv21,velec));
1184 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1186 /* Update potential sum for this i atom from the interaction with this j atom. */
1187 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1188 velecsum = _mm_add_pd(velecsum,velec);
1192 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1194 /* Calculate temporary vectorial force */
1195 tx = _mm_mul_pd(fscal,dx21);
1196 ty = _mm_mul_pd(fscal,dy21);
1197 tz = _mm_mul_pd(fscal,dz21);
1199 /* Update vectorial force */
1200 fix2 = _mm_add_pd(fix2,tx);
1201 fiy2 = _mm_add_pd(fiy2,ty);
1202 fiz2 = _mm_add_pd(fiz2,tz);
1204 fjx1 = _mm_add_pd(fjx1,tx);
1205 fjy1 = _mm_add_pd(fjy1,ty);
1206 fjz1 = _mm_add_pd(fjz1,tz);
1208 /**************************
1209 * CALCULATE INTERACTIONS *
1210 **************************/
1212 r22 = _mm_mul_pd(rsq22,rinv22);
1214 /* EWALD ELECTROSTATICS */
1216 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1217 ewrt = _mm_mul_pd(r22,ewtabscale);
1218 ewitab = _mm_cvttpd_epi32(ewrt);
1219 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1220 ewitab = _mm_slli_epi32(ewitab,2);
1221 ewtabF = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1222 ewtabD = _mm_setzero_pd();
1223 GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1224 ewtabV = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
1225 ewtabFn = _mm_setzero_pd();
1226 GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1227 felec = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
1228 velec = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
1229 velec = _mm_mul_pd(qq22,_mm_sub_pd(rinv22,velec));
1230 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1232 /* Update potential sum for this i atom from the interaction with this j atom. */
1233 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1234 velecsum = _mm_add_pd(velecsum,velec);
1238 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1240 /* Calculate temporary vectorial force */
1241 tx = _mm_mul_pd(fscal,dx22);
1242 ty = _mm_mul_pd(fscal,dy22);
1243 tz = _mm_mul_pd(fscal,dz22);
1245 /* Update vectorial force */
1246 fix2 = _mm_add_pd(fix2,tx);
1247 fiy2 = _mm_add_pd(fiy2,ty);
1248 fiz2 = _mm_add_pd(fiz2,tz);
1250 fjx2 = _mm_add_pd(fjx2,tx);
1251 fjy2 = _mm_add_pd(fjy2,ty);
1252 fjz2 = _mm_add_pd(fjz2,tz);
1254 gmx_mm_decrement_3rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1256 /* Inner loop uses 403 flops */
1259 /* End of innermost loop */
1261 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1262 f+i_coord_offset,fshift+i_shift_offset);
1265 /* Update potential energies */
1266 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1267 gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1269 /* Increment number of inner iterations */
1270 inneriter += j_index_end - j_index_start;
1272 /* Outer loop uses 20 flops */
1275 /* Increment number of outer iterations */
1278 /* Update outer/inner flops */
1280 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_VF,outeriter*20 + inneriter*403);
1283 * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW3W3_F_sse2_double
1284 * Electrostatics interaction: Ewald
1285 * VdW interaction: CubicSplineTable
1286 * Geometry: Water3-Water3
1287 * Calculate force/pot: Force
1290 nb_kernel_ElecEw_VdwCSTab_GeomW3W3_F_sse2_double
1291 (t_nblist * gmx_restrict nlist,
1292 rvec * gmx_restrict xx,
1293 rvec * gmx_restrict ff,
1294 struct t_forcerec * gmx_restrict fr,
1295 t_mdatoms * gmx_restrict mdatoms,
1296 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1297 t_nrnb * gmx_restrict nrnb)
1299 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1300 * just 0 for non-waters.
1301 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
1302 * jnr indices corresponding to data put in the four positions in the SIMD register.
1304 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1305 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1307 int j_coord_offsetA,j_coord_offsetB;
1308 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1309 real rcutoff_scalar;
1310 real *shiftvec,*fshift,*x,*f;
1311 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1313 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1315 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1317 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1318 int vdwjidx0A,vdwjidx0B;
1319 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1320 int vdwjidx1A,vdwjidx1B;
1321 __m128d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1322 int vdwjidx2A,vdwjidx2B;
1323 __m128d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1324 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1325 __m128d dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1326 __m128d dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1327 __m128d dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1328 __m128d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1329 __m128d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1330 __m128d dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1331 __m128d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1332 __m128d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1333 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
1336 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1339 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
1340 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
1342 __m128i ifour = _mm_set1_epi32(4);
1343 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1346 __m128d ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1348 __m128d dummy_mask,cutoff_mask;
1349 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
1350 __m128d one = _mm_set1_pd(1.0);
1351 __m128d two = _mm_set1_pd(2.0);
1357 jindex = nlist->jindex;
1359 shiftidx = nlist->shift;
1361 shiftvec = fr->shift_vec[0];
1362 fshift = fr->fshift[0];
1363 facel = _mm_set1_pd(fr->ic->epsfac);
1364 charge = mdatoms->chargeA;
1365 nvdwtype = fr->ntype;
1366 vdwparam = fr->nbfp;
1367 vdwtype = mdatoms->typeA;
1369 vftab = kernel_data->table_vdw->data;
1370 vftabscale = _mm_set1_pd(kernel_data->table_vdw->scale);
1372 sh_ewald = _mm_set1_pd(fr->ic->sh_ewald);
1373 ewtab = fr->ic->tabq_coul_F;
1374 ewtabscale = _mm_set1_pd(fr->ic->tabq_scale);
1375 ewtabhalfspace = _mm_set1_pd(0.5/fr->ic->tabq_scale);
1377 /* Setup water-specific parameters */
1378 inr = nlist->iinr[0];
1379 iq0 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
1380 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
1381 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
1382 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
1384 jq0 = _mm_set1_pd(charge[inr+0]);
1385 jq1 = _mm_set1_pd(charge[inr+1]);
1386 jq2 = _mm_set1_pd(charge[inr+2]);
1387 vdwjidx0A = 2*vdwtype[inr+0];
1388 qq00 = _mm_mul_pd(iq0,jq0);
1389 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
1390 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
1391 qq01 = _mm_mul_pd(iq0,jq1);
1392 qq02 = _mm_mul_pd(iq0,jq2);
1393 qq10 = _mm_mul_pd(iq1,jq0);
1394 qq11 = _mm_mul_pd(iq1,jq1);
1395 qq12 = _mm_mul_pd(iq1,jq2);
1396 qq20 = _mm_mul_pd(iq2,jq0);
1397 qq21 = _mm_mul_pd(iq2,jq1);
1398 qq22 = _mm_mul_pd(iq2,jq2);
1400 /* Avoid stupid compiler warnings */
1402 j_coord_offsetA = 0;
1403 j_coord_offsetB = 0;
1408 /* Start outer loop over neighborlists */
1409 for(iidx=0; iidx<nri; iidx++)
1411 /* Load shift vector for this list */
1412 i_shift_offset = DIM*shiftidx[iidx];
1414 /* Load limits for loop over neighbors */
1415 j_index_start = jindex[iidx];
1416 j_index_end = jindex[iidx+1];
1418 /* Get outer coordinate index */
1420 i_coord_offset = DIM*inr;
1422 /* Load i particle coords and add shift vector */
1423 gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1424 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1426 fix0 = _mm_setzero_pd();
1427 fiy0 = _mm_setzero_pd();
1428 fiz0 = _mm_setzero_pd();
1429 fix1 = _mm_setzero_pd();
1430 fiy1 = _mm_setzero_pd();
1431 fiz1 = _mm_setzero_pd();
1432 fix2 = _mm_setzero_pd();
1433 fiy2 = _mm_setzero_pd();
1434 fiz2 = _mm_setzero_pd();
1436 /* Start inner kernel loop */
1437 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
1440 /* Get j neighbor index, and coordinate index */
1442 jnrB = jjnr[jidx+1];
1443 j_coord_offsetA = DIM*jnrA;
1444 j_coord_offsetB = DIM*jnrB;
1446 /* load j atom coordinates */
1447 gmx_mm_load_3rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1448 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1450 /* Calculate displacement vector */
1451 dx00 = _mm_sub_pd(ix0,jx0);
1452 dy00 = _mm_sub_pd(iy0,jy0);
1453 dz00 = _mm_sub_pd(iz0,jz0);
1454 dx01 = _mm_sub_pd(ix0,jx1);
1455 dy01 = _mm_sub_pd(iy0,jy1);
1456 dz01 = _mm_sub_pd(iz0,jz1);
1457 dx02 = _mm_sub_pd(ix0,jx2);
1458 dy02 = _mm_sub_pd(iy0,jy2);
1459 dz02 = _mm_sub_pd(iz0,jz2);
1460 dx10 = _mm_sub_pd(ix1,jx0);
1461 dy10 = _mm_sub_pd(iy1,jy0);
1462 dz10 = _mm_sub_pd(iz1,jz0);
1463 dx11 = _mm_sub_pd(ix1,jx1);
1464 dy11 = _mm_sub_pd(iy1,jy1);
1465 dz11 = _mm_sub_pd(iz1,jz1);
1466 dx12 = _mm_sub_pd(ix1,jx2);
1467 dy12 = _mm_sub_pd(iy1,jy2);
1468 dz12 = _mm_sub_pd(iz1,jz2);
1469 dx20 = _mm_sub_pd(ix2,jx0);
1470 dy20 = _mm_sub_pd(iy2,jy0);
1471 dz20 = _mm_sub_pd(iz2,jz0);
1472 dx21 = _mm_sub_pd(ix2,jx1);
1473 dy21 = _mm_sub_pd(iy2,jy1);
1474 dz21 = _mm_sub_pd(iz2,jz1);
1475 dx22 = _mm_sub_pd(ix2,jx2);
1476 dy22 = _mm_sub_pd(iy2,jy2);
1477 dz22 = _mm_sub_pd(iz2,jz2);
1479 /* Calculate squared distance and things based on it */
1480 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1481 rsq01 = gmx_mm_calc_rsq_pd(dx01,dy01,dz01);
1482 rsq02 = gmx_mm_calc_rsq_pd(dx02,dy02,dz02);
1483 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
1484 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
1485 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
1486 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
1487 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
1488 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
1490 rinv00 = sse2_invsqrt_d(rsq00);
1491 rinv01 = sse2_invsqrt_d(rsq01);
1492 rinv02 = sse2_invsqrt_d(rsq02);
1493 rinv10 = sse2_invsqrt_d(rsq10);
1494 rinv11 = sse2_invsqrt_d(rsq11);
1495 rinv12 = sse2_invsqrt_d(rsq12);
1496 rinv20 = sse2_invsqrt_d(rsq20);
1497 rinv21 = sse2_invsqrt_d(rsq21);
1498 rinv22 = sse2_invsqrt_d(rsq22);
1500 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
1501 rinvsq01 = _mm_mul_pd(rinv01,rinv01);
1502 rinvsq02 = _mm_mul_pd(rinv02,rinv02);
1503 rinvsq10 = _mm_mul_pd(rinv10,rinv10);
1504 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
1505 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
1506 rinvsq20 = _mm_mul_pd(rinv20,rinv20);
1507 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
1508 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
1510 fjx0 = _mm_setzero_pd();
1511 fjy0 = _mm_setzero_pd();
1512 fjz0 = _mm_setzero_pd();
1513 fjx1 = _mm_setzero_pd();
1514 fjy1 = _mm_setzero_pd();
1515 fjz1 = _mm_setzero_pd();
1516 fjx2 = _mm_setzero_pd();
1517 fjy2 = _mm_setzero_pd();
1518 fjz2 = _mm_setzero_pd();
1520 /**************************
1521 * CALCULATE INTERACTIONS *
1522 **************************/
1524 r00 = _mm_mul_pd(rsq00,rinv00);
1526 /* Calculate table index by multiplying r with table scale and truncate to integer */
1527 rt = _mm_mul_pd(r00,vftabscale);
1528 vfitab = _mm_cvttpd_epi32(rt);
1529 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1530 vfitab = _mm_slli_epi32(vfitab,3);
1532 /* EWALD ELECTROSTATICS */
1534 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1535 ewrt = _mm_mul_pd(r00,ewtabscale);
1536 ewitab = _mm_cvttpd_epi32(ewrt);
1537 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1538 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1540 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1541 felec = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
1543 /* CUBIC SPLINE TABLE DISPERSION */
1544 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1545 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1546 GMX_MM_TRANSPOSE2_PD(Y,F);
1547 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1548 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1549 GMX_MM_TRANSPOSE2_PD(G,H);
1550 Heps = _mm_mul_pd(vfeps,H);
1551 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1552 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1553 fvdw6 = _mm_mul_pd(c6_00,FF);
1555 /* CUBIC SPLINE TABLE REPULSION */
1556 vfitab = _mm_add_epi32(vfitab,ifour);
1557 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1558 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1559 GMX_MM_TRANSPOSE2_PD(Y,F);
1560 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1561 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1562 GMX_MM_TRANSPOSE2_PD(G,H);
1563 Heps = _mm_mul_pd(vfeps,H);
1564 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1565 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1566 fvdw12 = _mm_mul_pd(c12_00,FF);
1567 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
1569 fscal = _mm_add_pd(felec,fvdw);
1571 /* Calculate temporary vectorial force */
1572 tx = _mm_mul_pd(fscal,dx00);
1573 ty = _mm_mul_pd(fscal,dy00);
1574 tz = _mm_mul_pd(fscal,dz00);
1576 /* Update vectorial force */
1577 fix0 = _mm_add_pd(fix0,tx);
1578 fiy0 = _mm_add_pd(fiy0,ty);
1579 fiz0 = _mm_add_pd(fiz0,tz);
1581 fjx0 = _mm_add_pd(fjx0,tx);
1582 fjy0 = _mm_add_pd(fjy0,ty);
1583 fjz0 = _mm_add_pd(fjz0,tz);
1585 /**************************
1586 * CALCULATE INTERACTIONS *
1587 **************************/
1589 r01 = _mm_mul_pd(rsq01,rinv01);
1591 /* EWALD ELECTROSTATICS */
1593 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1594 ewrt = _mm_mul_pd(r01,ewtabscale);
1595 ewitab = _mm_cvttpd_epi32(ewrt);
1596 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1597 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1599 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1600 felec = _mm_mul_pd(_mm_mul_pd(qq01,rinv01),_mm_sub_pd(rinvsq01,felec));
1604 /* Calculate temporary vectorial force */
1605 tx = _mm_mul_pd(fscal,dx01);
1606 ty = _mm_mul_pd(fscal,dy01);
1607 tz = _mm_mul_pd(fscal,dz01);
1609 /* Update vectorial force */
1610 fix0 = _mm_add_pd(fix0,tx);
1611 fiy0 = _mm_add_pd(fiy0,ty);
1612 fiz0 = _mm_add_pd(fiz0,tz);
1614 fjx1 = _mm_add_pd(fjx1,tx);
1615 fjy1 = _mm_add_pd(fjy1,ty);
1616 fjz1 = _mm_add_pd(fjz1,tz);
1618 /**************************
1619 * CALCULATE INTERACTIONS *
1620 **************************/
1622 r02 = _mm_mul_pd(rsq02,rinv02);
1624 /* EWALD ELECTROSTATICS */
1626 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1627 ewrt = _mm_mul_pd(r02,ewtabscale);
1628 ewitab = _mm_cvttpd_epi32(ewrt);
1629 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1630 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1632 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1633 felec = _mm_mul_pd(_mm_mul_pd(qq02,rinv02),_mm_sub_pd(rinvsq02,felec));
1637 /* Calculate temporary vectorial force */
1638 tx = _mm_mul_pd(fscal,dx02);
1639 ty = _mm_mul_pd(fscal,dy02);
1640 tz = _mm_mul_pd(fscal,dz02);
1642 /* Update vectorial force */
1643 fix0 = _mm_add_pd(fix0,tx);
1644 fiy0 = _mm_add_pd(fiy0,ty);
1645 fiz0 = _mm_add_pd(fiz0,tz);
1647 fjx2 = _mm_add_pd(fjx2,tx);
1648 fjy2 = _mm_add_pd(fjy2,ty);
1649 fjz2 = _mm_add_pd(fjz2,tz);
1651 /**************************
1652 * CALCULATE INTERACTIONS *
1653 **************************/
1655 r10 = _mm_mul_pd(rsq10,rinv10);
1657 /* EWALD ELECTROSTATICS */
1659 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1660 ewrt = _mm_mul_pd(r10,ewtabscale);
1661 ewitab = _mm_cvttpd_epi32(ewrt);
1662 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1663 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1665 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1666 felec = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1670 /* Calculate temporary vectorial force */
1671 tx = _mm_mul_pd(fscal,dx10);
1672 ty = _mm_mul_pd(fscal,dy10);
1673 tz = _mm_mul_pd(fscal,dz10);
1675 /* Update vectorial force */
1676 fix1 = _mm_add_pd(fix1,tx);
1677 fiy1 = _mm_add_pd(fiy1,ty);
1678 fiz1 = _mm_add_pd(fiz1,tz);
1680 fjx0 = _mm_add_pd(fjx0,tx);
1681 fjy0 = _mm_add_pd(fjy0,ty);
1682 fjz0 = _mm_add_pd(fjz0,tz);
1684 /**************************
1685 * CALCULATE INTERACTIONS *
1686 **************************/
1688 r11 = _mm_mul_pd(rsq11,rinv11);
1690 /* EWALD ELECTROSTATICS */
1692 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1693 ewrt = _mm_mul_pd(r11,ewtabscale);
1694 ewitab = _mm_cvttpd_epi32(ewrt);
1695 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1696 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1698 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1699 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
1703 /* Calculate temporary vectorial force */
1704 tx = _mm_mul_pd(fscal,dx11);
1705 ty = _mm_mul_pd(fscal,dy11);
1706 tz = _mm_mul_pd(fscal,dz11);
1708 /* Update vectorial force */
1709 fix1 = _mm_add_pd(fix1,tx);
1710 fiy1 = _mm_add_pd(fiy1,ty);
1711 fiz1 = _mm_add_pd(fiz1,tz);
1713 fjx1 = _mm_add_pd(fjx1,tx);
1714 fjy1 = _mm_add_pd(fjy1,ty);
1715 fjz1 = _mm_add_pd(fjz1,tz);
1717 /**************************
1718 * CALCULATE INTERACTIONS *
1719 **************************/
1721 r12 = _mm_mul_pd(rsq12,rinv12);
1723 /* EWALD ELECTROSTATICS */
1725 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1726 ewrt = _mm_mul_pd(r12,ewtabscale);
1727 ewitab = _mm_cvttpd_epi32(ewrt);
1728 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1729 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1731 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1732 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
1736 /* Calculate temporary vectorial force */
1737 tx = _mm_mul_pd(fscal,dx12);
1738 ty = _mm_mul_pd(fscal,dy12);
1739 tz = _mm_mul_pd(fscal,dz12);
1741 /* Update vectorial force */
1742 fix1 = _mm_add_pd(fix1,tx);
1743 fiy1 = _mm_add_pd(fiy1,ty);
1744 fiz1 = _mm_add_pd(fiz1,tz);
1746 fjx2 = _mm_add_pd(fjx2,tx);
1747 fjy2 = _mm_add_pd(fjy2,ty);
1748 fjz2 = _mm_add_pd(fjz2,tz);
1750 /**************************
1751 * CALCULATE INTERACTIONS *
1752 **************************/
1754 r20 = _mm_mul_pd(rsq20,rinv20);
1756 /* EWALD ELECTROSTATICS */
1758 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1759 ewrt = _mm_mul_pd(r20,ewtabscale);
1760 ewitab = _mm_cvttpd_epi32(ewrt);
1761 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1762 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1764 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1765 felec = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1769 /* Calculate temporary vectorial force */
1770 tx = _mm_mul_pd(fscal,dx20);
1771 ty = _mm_mul_pd(fscal,dy20);
1772 tz = _mm_mul_pd(fscal,dz20);
1774 /* Update vectorial force */
1775 fix2 = _mm_add_pd(fix2,tx);
1776 fiy2 = _mm_add_pd(fiy2,ty);
1777 fiz2 = _mm_add_pd(fiz2,tz);
1779 fjx0 = _mm_add_pd(fjx0,tx);
1780 fjy0 = _mm_add_pd(fjy0,ty);
1781 fjz0 = _mm_add_pd(fjz0,tz);
1783 /**************************
1784 * CALCULATE INTERACTIONS *
1785 **************************/
1787 r21 = _mm_mul_pd(rsq21,rinv21);
1789 /* EWALD ELECTROSTATICS */
1791 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1792 ewrt = _mm_mul_pd(r21,ewtabscale);
1793 ewitab = _mm_cvttpd_epi32(ewrt);
1794 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1795 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1797 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1798 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1802 /* Calculate temporary vectorial force */
1803 tx = _mm_mul_pd(fscal,dx21);
1804 ty = _mm_mul_pd(fscal,dy21);
1805 tz = _mm_mul_pd(fscal,dz21);
1807 /* Update vectorial force */
1808 fix2 = _mm_add_pd(fix2,tx);
1809 fiy2 = _mm_add_pd(fiy2,ty);
1810 fiz2 = _mm_add_pd(fiz2,tz);
1812 fjx1 = _mm_add_pd(fjx1,tx);
1813 fjy1 = _mm_add_pd(fjy1,ty);
1814 fjz1 = _mm_add_pd(fjz1,tz);
1816 /**************************
1817 * CALCULATE INTERACTIONS *
1818 **************************/
1820 r22 = _mm_mul_pd(rsq22,rinv22);
1822 /* EWALD ELECTROSTATICS */
1824 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1825 ewrt = _mm_mul_pd(r22,ewtabscale);
1826 ewitab = _mm_cvttpd_epi32(ewrt);
1827 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1828 gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1830 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1831 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1835 /* Calculate temporary vectorial force */
1836 tx = _mm_mul_pd(fscal,dx22);
1837 ty = _mm_mul_pd(fscal,dy22);
1838 tz = _mm_mul_pd(fscal,dz22);
1840 /* Update vectorial force */
1841 fix2 = _mm_add_pd(fix2,tx);
1842 fiy2 = _mm_add_pd(fiy2,ty);
1843 fiz2 = _mm_add_pd(fiz2,tz);
1845 fjx2 = _mm_add_pd(fjx2,tx);
1846 fjy2 = _mm_add_pd(fjy2,ty);
1847 fjz2 = _mm_add_pd(fjz2,tz);
1849 gmx_mm_decrement_3rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1851 /* Inner loop uses 350 flops */
1854 if(jidx<j_index_end)
1858 j_coord_offsetA = DIM*jnrA;
1860 /* load j atom coordinates */
1861 gmx_mm_load_3rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
1862 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1864 /* Calculate displacement vector */
1865 dx00 = _mm_sub_pd(ix0,jx0);
1866 dy00 = _mm_sub_pd(iy0,jy0);
1867 dz00 = _mm_sub_pd(iz0,jz0);
1868 dx01 = _mm_sub_pd(ix0,jx1);
1869 dy01 = _mm_sub_pd(iy0,jy1);
1870 dz01 = _mm_sub_pd(iz0,jz1);
1871 dx02 = _mm_sub_pd(ix0,jx2);
1872 dy02 = _mm_sub_pd(iy0,jy2);
1873 dz02 = _mm_sub_pd(iz0,jz2);
1874 dx10 = _mm_sub_pd(ix1,jx0);
1875 dy10 = _mm_sub_pd(iy1,jy0);
1876 dz10 = _mm_sub_pd(iz1,jz0);
1877 dx11 = _mm_sub_pd(ix1,jx1);
1878 dy11 = _mm_sub_pd(iy1,jy1);
1879 dz11 = _mm_sub_pd(iz1,jz1);
1880 dx12 = _mm_sub_pd(ix1,jx2);
1881 dy12 = _mm_sub_pd(iy1,jy2);
1882 dz12 = _mm_sub_pd(iz1,jz2);
1883 dx20 = _mm_sub_pd(ix2,jx0);
1884 dy20 = _mm_sub_pd(iy2,jy0);
1885 dz20 = _mm_sub_pd(iz2,jz0);
1886 dx21 = _mm_sub_pd(ix2,jx1);
1887 dy21 = _mm_sub_pd(iy2,jy1);
1888 dz21 = _mm_sub_pd(iz2,jz1);
1889 dx22 = _mm_sub_pd(ix2,jx2);
1890 dy22 = _mm_sub_pd(iy2,jy2);
1891 dz22 = _mm_sub_pd(iz2,jz2);
1893 /* Calculate squared distance and things based on it */
1894 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1895 rsq01 = gmx_mm_calc_rsq_pd(dx01,dy01,dz01);
1896 rsq02 = gmx_mm_calc_rsq_pd(dx02,dy02,dz02);
1897 rsq10 = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
1898 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
1899 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
1900 rsq20 = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
1901 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
1902 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
1904 rinv00 = sse2_invsqrt_d(rsq00);
1905 rinv01 = sse2_invsqrt_d(rsq01);
1906 rinv02 = sse2_invsqrt_d(rsq02);
1907 rinv10 = sse2_invsqrt_d(rsq10);
1908 rinv11 = sse2_invsqrt_d(rsq11);
1909 rinv12 = sse2_invsqrt_d(rsq12);
1910 rinv20 = sse2_invsqrt_d(rsq20);
1911 rinv21 = sse2_invsqrt_d(rsq21);
1912 rinv22 = sse2_invsqrt_d(rsq22);
1914 rinvsq00 = _mm_mul_pd(rinv00,rinv00);
1915 rinvsq01 = _mm_mul_pd(rinv01,rinv01);
1916 rinvsq02 = _mm_mul_pd(rinv02,rinv02);
1917 rinvsq10 = _mm_mul_pd(rinv10,rinv10);
1918 rinvsq11 = _mm_mul_pd(rinv11,rinv11);
1919 rinvsq12 = _mm_mul_pd(rinv12,rinv12);
1920 rinvsq20 = _mm_mul_pd(rinv20,rinv20);
1921 rinvsq21 = _mm_mul_pd(rinv21,rinv21);
1922 rinvsq22 = _mm_mul_pd(rinv22,rinv22);
1924 fjx0 = _mm_setzero_pd();
1925 fjy0 = _mm_setzero_pd();
1926 fjz0 = _mm_setzero_pd();
1927 fjx1 = _mm_setzero_pd();
1928 fjy1 = _mm_setzero_pd();
1929 fjz1 = _mm_setzero_pd();
1930 fjx2 = _mm_setzero_pd();
1931 fjy2 = _mm_setzero_pd();
1932 fjz2 = _mm_setzero_pd();
1934 /**************************
1935 * CALCULATE INTERACTIONS *
1936 **************************/
1938 r00 = _mm_mul_pd(rsq00,rinv00);
1940 /* Calculate table index by multiplying r with table scale and truncate to integer */
1941 rt = _mm_mul_pd(r00,vftabscale);
1942 vfitab = _mm_cvttpd_epi32(rt);
1943 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1944 vfitab = _mm_slli_epi32(vfitab,3);
1946 /* EWALD ELECTROSTATICS */
1948 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1949 ewrt = _mm_mul_pd(r00,ewtabscale);
1950 ewitab = _mm_cvttpd_epi32(ewrt);
1951 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
1952 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1953 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1954 felec = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
1956 /* CUBIC SPLINE TABLE DISPERSION */
1957 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1958 F = _mm_setzero_pd();
1959 GMX_MM_TRANSPOSE2_PD(Y,F);
1960 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1961 H = _mm_setzero_pd();
1962 GMX_MM_TRANSPOSE2_PD(G,H);
1963 Heps = _mm_mul_pd(vfeps,H);
1964 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1965 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1966 fvdw6 = _mm_mul_pd(c6_00,FF);
1968 /* CUBIC SPLINE TABLE REPULSION */
1969 vfitab = _mm_add_epi32(vfitab,ifour);
1970 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1971 F = _mm_setzero_pd();
1972 GMX_MM_TRANSPOSE2_PD(Y,F);
1973 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1974 H = _mm_setzero_pd();
1975 GMX_MM_TRANSPOSE2_PD(G,H);
1976 Heps = _mm_mul_pd(vfeps,H);
1977 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1978 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1979 fvdw12 = _mm_mul_pd(c12_00,FF);
1980 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
1982 fscal = _mm_add_pd(felec,fvdw);
1984 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1986 /* Calculate temporary vectorial force */
1987 tx = _mm_mul_pd(fscal,dx00);
1988 ty = _mm_mul_pd(fscal,dy00);
1989 tz = _mm_mul_pd(fscal,dz00);
1991 /* Update vectorial force */
1992 fix0 = _mm_add_pd(fix0,tx);
1993 fiy0 = _mm_add_pd(fiy0,ty);
1994 fiz0 = _mm_add_pd(fiz0,tz);
1996 fjx0 = _mm_add_pd(fjx0,tx);
1997 fjy0 = _mm_add_pd(fjy0,ty);
1998 fjz0 = _mm_add_pd(fjz0,tz);
2000 /**************************
2001 * CALCULATE INTERACTIONS *
2002 **************************/
2004 r01 = _mm_mul_pd(rsq01,rinv01);
2006 /* EWALD ELECTROSTATICS */
2008 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2009 ewrt = _mm_mul_pd(r01,ewtabscale);
2010 ewitab = _mm_cvttpd_epi32(ewrt);
2011 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2012 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2013 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2014 felec = _mm_mul_pd(_mm_mul_pd(qq01,rinv01),_mm_sub_pd(rinvsq01,felec));
2018 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2020 /* Calculate temporary vectorial force */
2021 tx = _mm_mul_pd(fscal,dx01);
2022 ty = _mm_mul_pd(fscal,dy01);
2023 tz = _mm_mul_pd(fscal,dz01);
2025 /* Update vectorial force */
2026 fix0 = _mm_add_pd(fix0,tx);
2027 fiy0 = _mm_add_pd(fiy0,ty);
2028 fiz0 = _mm_add_pd(fiz0,tz);
2030 fjx1 = _mm_add_pd(fjx1,tx);
2031 fjy1 = _mm_add_pd(fjy1,ty);
2032 fjz1 = _mm_add_pd(fjz1,tz);
2034 /**************************
2035 * CALCULATE INTERACTIONS *
2036 **************************/
2038 r02 = _mm_mul_pd(rsq02,rinv02);
2040 /* EWALD ELECTROSTATICS */
2042 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2043 ewrt = _mm_mul_pd(r02,ewtabscale);
2044 ewitab = _mm_cvttpd_epi32(ewrt);
2045 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2046 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2047 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2048 felec = _mm_mul_pd(_mm_mul_pd(qq02,rinv02),_mm_sub_pd(rinvsq02,felec));
2052 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2054 /* Calculate temporary vectorial force */
2055 tx = _mm_mul_pd(fscal,dx02);
2056 ty = _mm_mul_pd(fscal,dy02);
2057 tz = _mm_mul_pd(fscal,dz02);
2059 /* Update vectorial force */
2060 fix0 = _mm_add_pd(fix0,tx);
2061 fiy0 = _mm_add_pd(fiy0,ty);
2062 fiz0 = _mm_add_pd(fiz0,tz);
2064 fjx2 = _mm_add_pd(fjx2,tx);
2065 fjy2 = _mm_add_pd(fjy2,ty);
2066 fjz2 = _mm_add_pd(fjz2,tz);
2068 /**************************
2069 * CALCULATE INTERACTIONS *
2070 **************************/
2072 r10 = _mm_mul_pd(rsq10,rinv10);
2074 /* EWALD ELECTROSTATICS */
2076 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2077 ewrt = _mm_mul_pd(r10,ewtabscale);
2078 ewitab = _mm_cvttpd_epi32(ewrt);
2079 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2080 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2081 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2082 felec = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
2086 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2088 /* Calculate temporary vectorial force */
2089 tx = _mm_mul_pd(fscal,dx10);
2090 ty = _mm_mul_pd(fscal,dy10);
2091 tz = _mm_mul_pd(fscal,dz10);
2093 /* Update vectorial force */
2094 fix1 = _mm_add_pd(fix1,tx);
2095 fiy1 = _mm_add_pd(fiy1,ty);
2096 fiz1 = _mm_add_pd(fiz1,tz);
2098 fjx0 = _mm_add_pd(fjx0,tx);
2099 fjy0 = _mm_add_pd(fjy0,ty);
2100 fjz0 = _mm_add_pd(fjz0,tz);
2102 /**************************
2103 * CALCULATE INTERACTIONS *
2104 **************************/
2106 r11 = _mm_mul_pd(rsq11,rinv11);
2108 /* EWALD ELECTROSTATICS */
2110 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2111 ewrt = _mm_mul_pd(r11,ewtabscale);
2112 ewitab = _mm_cvttpd_epi32(ewrt);
2113 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2114 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2115 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2116 felec = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
2120 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2122 /* Calculate temporary vectorial force */
2123 tx = _mm_mul_pd(fscal,dx11);
2124 ty = _mm_mul_pd(fscal,dy11);
2125 tz = _mm_mul_pd(fscal,dz11);
2127 /* Update vectorial force */
2128 fix1 = _mm_add_pd(fix1,tx);
2129 fiy1 = _mm_add_pd(fiy1,ty);
2130 fiz1 = _mm_add_pd(fiz1,tz);
2132 fjx1 = _mm_add_pd(fjx1,tx);
2133 fjy1 = _mm_add_pd(fjy1,ty);
2134 fjz1 = _mm_add_pd(fjz1,tz);
2136 /**************************
2137 * CALCULATE INTERACTIONS *
2138 **************************/
2140 r12 = _mm_mul_pd(rsq12,rinv12);
2142 /* EWALD ELECTROSTATICS */
2144 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2145 ewrt = _mm_mul_pd(r12,ewtabscale);
2146 ewitab = _mm_cvttpd_epi32(ewrt);
2147 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2148 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2149 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2150 felec = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
2154 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2156 /* Calculate temporary vectorial force */
2157 tx = _mm_mul_pd(fscal,dx12);
2158 ty = _mm_mul_pd(fscal,dy12);
2159 tz = _mm_mul_pd(fscal,dz12);
2161 /* Update vectorial force */
2162 fix1 = _mm_add_pd(fix1,tx);
2163 fiy1 = _mm_add_pd(fiy1,ty);
2164 fiz1 = _mm_add_pd(fiz1,tz);
2166 fjx2 = _mm_add_pd(fjx2,tx);
2167 fjy2 = _mm_add_pd(fjy2,ty);
2168 fjz2 = _mm_add_pd(fjz2,tz);
2170 /**************************
2171 * CALCULATE INTERACTIONS *
2172 **************************/
2174 r20 = _mm_mul_pd(rsq20,rinv20);
2176 /* EWALD ELECTROSTATICS */
2178 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2179 ewrt = _mm_mul_pd(r20,ewtabscale);
2180 ewitab = _mm_cvttpd_epi32(ewrt);
2181 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2182 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2183 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2184 felec = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
2188 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2190 /* Calculate temporary vectorial force */
2191 tx = _mm_mul_pd(fscal,dx20);
2192 ty = _mm_mul_pd(fscal,dy20);
2193 tz = _mm_mul_pd(fscal,dz20);
2195 /* Update vectorial force */
2196 fix2 = _mm_add_pd(fix2,tx);
2197 fiy2 = _mm_add_pd(fiy2,ty);
2198 fiz2 = _mm_add_pd(fiz2,tz);
2200 fjx0 = _mm_add_pd(fjx0,tx);
2201 fjy0 = _mm_add_pd(fjy0,ty);
2202 fjz0 = _mm_add_pd(fjz0,tz);
2204 /**************************
2205 * CALCULATE INTERACTIONS *
2206 **************************/
2208 r21 = _mm_mul_pd(rsq21,rinv21);
2210 /* EWALD ELECTROSTATICS */
2212 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2213 ewrt = _mm_mul_pd(r21,ewtabscale);
2214 ewitab = _mm_cvttpd_epi32(ewrt);
2215 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2216 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2217 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2218 felec = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
2222 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2224 /* Calculate temporary vectorial force */
2225 tx = _mm_mul_pd(fscal,dx21);
2226 ty = _mm_mul_pd(fscal,dy21);
2227 tz = _mm_mul_pd(fscal,dz21);
2229 /* Update vectorial force */
2230 fix2 = _mm_add_pd(fix2,tx);
2231 fiy2 = _mm_add_pd(fiy2,ty);
2232 fiz2 = _mm_add_pd(fiz2,tz);
2234 fjx1 = _mm_add_pd(fjx1,tx);
2235 fjy1 = _mm_add_pd(fjy1,ty);
2236 fjz1 = _mm_add_pd(fjz1,tz);
2238 /**************************
2239 * CALCULATE INTERACTIONS *
2240 **************************/
2242 r22 = _mm_mul_pd(rsq22,rinv22);
2244 /* EWALD ELECTROSTATICS */
2246 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2247 ewrt = _mm_mul_pd(r22,ewtabscale);
2248 ewitab = _mm_cvttpd_epi32(ewrt);
2249 eweps = _mm_sub_pd(ewrt,_mm_cvtepi32_pd(ewitab));
2250 gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2251 felec = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
2252 felec = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
2256 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2258 /* Calculate temporary vectorial force */
2259 tx = _mm_mul_pd(fscal,dx22);
2260 ty = _mm_mul_pd(fscal,dy22);
2261 tz = _mm_mul_pd(fscal,dz22);
2263 /* Update vectorial force */
2264 fix2 = _mm_add_pd(fix2,tx);
2265 fiy2 = _mm_add_pd(fiy2,ty);
2266 fiz2 = _mm_add_pd(fiz2,tz);
2268 fjx2 = _mm_add_pd(fjx2,tx);
2269 fjy2 = _mm_add_pd(fjy2,ty);
2270 fjz2 = _mm_add_pd(fjz2,tz);
2272 gmx_mm_decrement_3rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2274 /* Inner loop uses 350 flops */
2277 /* End of innermost loop */
2279 gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
2280 f+i_coord_offset,fshift+i_shift_offset);
2282 /* Increment number of inner iterations */
2283 inneriter += j_index_end - j_index_start;
2285 /* Outer loop uses 18 flops */
2288 /* Increment number of outer iterations */
2291 /* Update outer/inner flops */
2293 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_F,outeriter*18 + inneriter*350);