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_ElecCSTab_VdwCSTab_GeomW4W4_VF_sse2_double
54 * Electrostatics interaction: CubicSplineTable
55 * VdW interaction: CubicSplineTable
56 * Geometry: Water4-Water4
57 * Calculate force/pot: PotentialAndForce
60 nb_kernel_ElecCSTab_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;
120 __m128d dummy_mask,cutoff_mask;
121 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
122 __m128d one = _mm_set1_pd(1.0);
123 __m128d two = _mm_set1_pd(2.0);
129 jindex = nlist->jindex;
131 shiftidx = nlist->shift;
133 shiftvec = fr->shift_vec[0];
134 fshift = fr->fshift[0];
135 facel = _mm_set1_pd(fr->epsfac);
136 charge = mdatoms->chargeA;
137 nvdwtype = fr->ntype;
139 vdwtype = mdatoms->typeA;
141 vftab = kernel_data->table_elec_vdw->data;
142 vftabscale = _mm_set1_pd(kernel_data->table_elec_vdw->scale);
144 /* Setup water-specific parameters */
145 inr = nlist->iinr[0];
146 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
147 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
148 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
149 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
151 jq1 = _mm_set1_pd(charge[inr+1]);
152 jq2 = _mm_set1_pd(charge[inr+2]);
153 jq3 = _mm_set1_pd(charge[inr+3]);
154 vdwjidx0A = 2*vdwtype[inr+0];
155 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
156 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
157 qq11 = _mm_mul_pd(iq1,jq1);
158 qq12 = _mm_mul_pd(iq1,jq2);
159 qq13 = _mm_mul_pd(iq1,jq3);
160 qq21 = _mm_mul_pd(iq2,jq1);
161 qq22 = _mm_mul_pd(iq2,jq2);
162 qq23 = _mm_mul_pd(iq2,jq3);
163 qq31 = _mm_mul_pd(iq3,jq1);
164 qq32 = _mm_mul_pd(iq3,jq2);
165 qq33 = _mm_mul_pd(iq3,jq3);
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_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
191 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
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();
202 fix3 = _mm_setzero_pd();
203 fiy3 = _mm_setzero_pd();
204 fiz3 = _mm_setzero_pd();
206 /* Reset potential sums */
207 velecsum = _mm_setzero_pd();
208 vvdwsum = _mm_setzero_pd();
210 /* Start inner kernel loop */
211 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
214 /* Get j neighbor index, and coordinate index */
217 j_coord_offsetA = DIM*jnrA;
218 j_coord_offsetB = DIM*jnrB;
220 /* load j atom coordinates */
221 gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
222 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
223 &jy2,&jz2,&jx3,&jy3,&jz3);
225 /* Calculate displacement vector */
226 dx00 = _mm_sub_pd(ix0,jx0);
227 dy00 = _mm_sub_pd(iy0,jy0);
228 dz00 = _mm_sub_pd(iz0,jz0);
229 dx11 = _mm_sub_pd(ix1,jx1);
230 dy11 = _mm_sub_pd(iy1,jy1);
231 dz11 = _mm_sub_pd(iz1,jz1);
232 dx12 = _mm_sub_pd(ix1,jx2);
233 dy12 = _mm_sub_pd(iy1,jy2);
234 dz12 = _mm_sub_pd(iz1,jz2);
235 dx13 = _mm_sub_pd(ix1,jx3);
236 dy13 = _mm_sub_pd(iy1,jy3);
237 dz13 = _mm_sub_pd(iz1,jz3);
238 dx21 = _mm_sub_pd(ix2,jx1);
239 dy21 = _mm_sub_pd(iy2,jy1);
240 dz21 = _mm_sub_pd(iz2,jz1);
241 dx22 = _mm_sub_pd(ix2,jx2);
242 dy22 = _mm_sub_pd(iy2,jy2);
243 dz22 = _mm_sub_pd(iz2,jz2);
244 dx23 = _mm_sub_pd(ix2,jx3);
245 dy23 = _mm_sub_pd(iy2,jy3);
246 dz23 = _mm_sub_pd(iz2,jz3);
247 dx31 = _mm_sub_pd(ix3,jx1);
248 dy31 = _mm_sub_pd(iy3,jy1);
249 dz31 = _mm_sub_pd(iz3,jz1);
250 dx32 = _mm_sub_pd(ix3,jx2);
251 dy32 = _mm_sub_pd(iy3,jy2);
252 dz32 = _mm_sub_pd(iz3,jz2);
253 dx33 = _mm_sub_pd(ix3,jx3);
254 dy33 = _mm_sub_pd(iy3,jy3);
255 dz33 = _mm_sub_pd(iz3,jz3);
257 /* Calculate squared distance and things based on it */
258 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
259 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
260 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
261 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
262 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
263 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
264 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
265 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
266 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
267 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
269 rinv00 = gmx_mm_invsqrt_pd(rsq00);
270 rinv11 = gmx_mm_invsqrt_pd(rsq11);
271 rinv12 = gmx_mm_invsqrt_pd(rsq12);
272 rinv13 = gmx_mm_invsqrt_pd(rsq13);
273 rinv21 = gmx_mm_invsqrt_pd(rsq21);
274 rinv22 = gmx_mm_invsqrt_pd(rsq22);
275 rinv23 = gmx_mm_invsqrt_pd(rsq23);
276 rinv31 = gmx_mm_invsqrt_pd(rsq31);
277 rinv32 = gmx_mm_invsqrt_pd(rsq32);
278 rinv33 = gmx_mm_invsqrt_pd(rsq33);
280 fjx0 = _mm_setzero_pd();
281 fjy0 = _mm_setzero_pd();
282 fjz0 = _mm_setzero_pd();
283 fjx1 = _mm_setzero_pd();
284 fjy1 = _mm_setzero_pd();
285 fjz1 = _mm_setzero_pd();
286 fjx2 = _mm_setzero_pd();
287 fjy2 = _mm_setzero_pd();
288 fjz2 = _mm_setzero_pd();
289 fjx3 = _mm_setzero_pd();
290 fjy3 = _mm_setzero_pd();
291 fjz3 = _mm_setzero_pd();
293 /**************************
294 * CALCULATE INTERACTIONS *
295 **************************/
297 r00 = _mm_mul_pd(rsq00,rinv00);
299 /* Calculate table index by multiplying r with table scale and truncate to integer */
300 rt = _mm_mul_pd(r00,vftabscale);
301 vfitab = _mm_cvttpd_epi32(rt);
302 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
303 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
305 /* CUBIC SPLINE TABLE DISPERSION */
306 vfitab = _mm_add_epi32(vfitab,ifour);
307 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
308 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
309 GMX_MM_TRANSPOSE2_PD(Y,F);
310 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
311 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
312 GMX_MM_TRANSPOSE2_PD(G,H);
313 Heps = _mm_mul_pd(vfeps,H);
314 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
315 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
316 vvdw6 = _mm_mul_pd(c6_00,VV);
317 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
318 fvdw6 = _mm_mul_pd(c6_00,FF);
320 /* CUBIC SPLINE TABLE REPULSION */
321 vfitab = _mm_add_epi32(vfitab,ifour);
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 vvdw12 = _mm_mul_pd(c12_00,VV);
332 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
333 fvdw12 = _mm_mul_pd(c12_00,FF);
334 vvdw = _mm_add_pd(vvdw12,vvdw6);
335 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
337 /* Update potential sum for this i atom from the interaction with this j atom. */
338 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
342 /* Calculate temporary vectorial force */
343 tx = _mm_mul_pd(fscal,dx00);
344 ty = _mm_mul_pd(fscal,dy00);
345 tz = _mm_mul_pd(fscal,dz00);
347 /* Update vectorial force */
348 fix0 = _mm_add_pd(fix0,tx);
349 fiy0 = _mm_add_pd(fiy0,ty);
350 fiz0 = _mm_add_pd(fiz0,tz);
352 fjx0 = _mm_add_pd(fjx0,tx);
353 fjy0 = _mm_add_pd(fjy0,ty);
354 fjz0 = _mm_add_pd(fjz0,tz);
356 /**************************
357 * CALCULATE INTERACTIONS *
358 **************************/
360 r11 = _mm_mul_pd(rsq11,rinv11);
362 /* Calculate table index by multiplying r with table scale and truncate to integer */
363 rt = _mm_mul_pd(r11,vftabscale);
364 vfitab = _mm_cvttpd_epi32(rt);
365 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
366 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
368 /* CUBIC SPLINE TABLE ELECTROSTATICS */
369 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
370 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
371 GMX_MM_TRANSPOSE2_PD(Y,F);
372 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
373 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
374 GMX_MM_TRANSPOSE2_PD(G,H);
375 Heps = _mm_mul_pd(vfeps,H);
376 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
377 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
378 velec = _mm_mul_pd(qq11,VV);
379 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
380 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq11,FF),_mm_mul_pd(vftabscale,rinv11)));
382 /* Update potential sum for this i atom from the interaction with this j atom. */
383 velecsum = _mm_add_pd(velecsum,velec);
387 /* Calculate temporary vectorial force */
388 tx = _mm_mul_pd(fscal,dx11);
389 ty = _mm_mul_pd(fscal,dy11);
390 tz = _mm_mul_pd(fscal,dz11);
392 /* Update vectorial force */
393 fix1 = _mm_add_pd(fix1,tx);
394 fiy1 = _mm_add_pd(fiy1,ty);
395 fiz1 = _mm_add_pd(fiz1,tz);
397 fjx1 = _mm_add_pd(fjx1,tx);
398 fjy1 = _mm_add_pd(fjy1,ty);
399 fjz1 = _mm_add_pd(fjz1,tz);
401 /**************************
402 * CALCULATE INTERACTIONS *
403 **************************/
405 r12 = _mm_mul_pd(rsq12,rinv12);
407 /* Calculate table index by multiplying r with table scale and truncate to integer */
408 rt = _mm_mul_pd(r12,vftabscale);
409 vfitab = _mm_cvttpd_epi32(rt);
410 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
411 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
413 /* CUBIC SPLINE TABLE ELECTROSTATICS */
414 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
415 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
416 GMX_MM_TRANSPOSE2_PD(Y,F);
417 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
418 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
419 GMX_MM_TRANSPOSE2_PD(G,H);
420 Heps = _mm_mul_pd(vfeps,H);
421 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
422 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
423 velec = _mm_mul_pd(qq12,VV);
424 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
425 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq12,FF),_mm_mul_pd(vftabscale,rinv12)));
427 /* Update potential sum for this i atom from the interaction with this j atom. */
428 velecsum = _mm_add_pd(velecsum,velec);
432 /* Calculate temporary vectorial force */
433 tx = _mm_mul_pd(fscal,dx12);
434 ty = _mm_mul_pd(fscal,dy12);
435 tz = _mm_mul_pd(fscal,dz12);
437 /* Update vectorial force */
438 fix1 = _mm_add_pd(fix1,tx);
439 fiy1 = _mm_add_pd(fiy1,ty);
440 fiz1 = _mm_add_pd(fiz1,tz);
442 fjx2 = _mm_add_pd(fjx2,tx);
443 fjy2 = _mm_add_pd(fjy2,ty);
444 fjz2 = _mm_add_pd(fjz2,tz);
446 /**************************
447 * CALCULATE INTERACTIONS *
448 **************************/
450 r13 = _mm_mul_pd(rsq13,rinv13);
452 /* Calculate table index by multiplying r with table scale and truncate to integer */
453 rt = _mm_mul_pd(r13,vftabscale);
454 vfitab = _mm_cvttpd_epi32(rt);
455 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
456 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
458 /* CUBIC SPLINE TABLE ELECTROSTATICS */
459 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
460 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
461 GMX_MM_TRANSPOSE2_PD(Y,F);
462 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
463 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
464 GMX_MM_TRANSPOSE2_PD(G,H);
465 Heps = _mm_mul_pd(vfeps,H);
466 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
467 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
468 velec = _mm_mul_pd(qq13,VV);
469 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
470 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq13,FF),_mm_mul_pd(vftabscale,rinv13)));
472 /* Update potential sum for this i atom from the interaction with this j atom. */
473 velecsum = _mm_add_pd(velecsum,velec);
477 /* Calculate temporary vectorial force */
478 tx = _mm_mul_pd(fscal,dx13);
479 ty = _mm_mul_pd(fscal,dy13);
480 tz = _mm_mul_pd(fscal,dz13);
482 /* Update vectorial force */
483 fix1 = _mm_add_pd(fix1,tx);
484 fiy1 = _mm_add_pd(fiy1,ty);
485 fiz1 = _mm_add_pd(fiz1,tz);
487 fjx3 = _mm_add_pd(fjx3,tx);
488 fjy3 = _mm_add_pd(fjy3,ty);
489 fjz3 = _mm_add_pd(fjz3,tz);
491 /**************************
492 * CALCULATE INTERACTIONS *
493 **************************/
495 r21 = _mm_mul_pd(rsq21,rinv21);
497 /* Calculate table index by multiplying r with table scale and truncate to integer */
498 rt = _mm_mul_pd(r21,vftabscale);
499 vfitab = _mm_cvttpd_epi32(rt);
500 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
501 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
503 /* CUBIC SPLINE TABLE ELECTROSTATICS */
504 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
505 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
506 GMX_MM_TRANSPOSE2_PD(Y,F);
507 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
508 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
509 GMX_MM_TRANSPOSE2_PD(G,H);
510 Heps = _mm_mul_pd(vfeps,H);
511 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
512 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
513 velec = _mm_mul_pd(qq21,VV);
514 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
515 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq21,FF),_mm_mul_pd(vftabscale,rinv21)));
517 /* Update potential sum for this i atom from the interaction with this j atom. */
518 velecsum = _mm_add_pd(velecsum,velec);
522 /* Calculate temporary vectorial force */
523 tx = _mm_mul_pd(fscal,dx21);
524 ty = _mm_mul_pd(fscal,dy21);
525 tz = _mm_mul_pd(fscal,dz21);
527 /* Update vectorial force */
528 fix2 = _mm_add_pd(fix2,tx);
529 fiy2 = _mm_add_pd(fiy2,ty);
530 fiz2 = _mm_add_pd(fiz2,tz);
532 fjx1 = _mm_add_pd(fjx1,tx);
533 fjy1 = _mm_add_pd(fjy1,ty);
534 fjz1 = _mm_add_pd(fjz1,tz);
536 /**************************
537 * CALCULATE INTERACTIONS *
538 **************************/
540 r22 = _mm_mul_pd(rsq22,rinv22);
542 /* Calculate table index by multiplying r with table scale and truncate to integer */
543 rt = _mm_mul_pd(r22,vftabscale);
544 vfitab = _mm_cvttpd_epi32(rt);
545 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
546 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
548 /* CUBIC SPLINE TABLE ELECTROSTATICS */
549 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
550 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
551 GMX_MM_TRANSPOSE2_PD(Y,F);
552 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
553 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
554 GMX_MM_TRANSPOSE2_PD(G,H);
555 Heps = _mm_mul_pd(vfeps,H);
556 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
557 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
558 velec = _mm_mul_pd(qq22,VV);
559 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
560 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq22,FF),_mm_mul_pd(vftabscale,rinv22)));
562 /* Update potential sum for this i atom from the interaction with this j atom. */
563 velecsum = _mm_add_pd(velecsum,velec);
567 /* Calculate temporary vectorial force */
568 tx = _mm_mul_pd(fscal,dx22);
569 ty = _mm_mul_pd(fscal,dy22);
570 tz = _mm_mul_pd(fscal,dz22);
572 /* Update vectorial force */
573 fix2 = _mm_add_pd(fix2,tx);
574 fiy2 = _mm_add_pd(fiy2,ty);
575 fiz2 = _mm_add_pd(fiz2,tz);
577 fjx2 = _mm_add_pd(fjx2,tx);
578 fjy2 = _mm_add_pd(fjy2,ty);
579 fjz2 = _mm_add_pd(fjz2,tz);
581 /**************************
582 * CALCULATE INTERACTIONS *
583 **************************/
585 r23 = _mm_mul_pd(rsq23,rinv23);
587 /* Calculate table index by multiplying r with table scale and truncate to integer */
588 rt = _mm_mul_pd(r23,vftabscale);
589 vfitab = _mm_cvttpd_epi32(rt);
590 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
591 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
593 /* CUBIC SPLINE TABLE ELECTROSTATICS */
594 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
595 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
596 GMX_MM_TRANSPOSE2_PD(Y,F);
597 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
598 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
599 GMX_MM_TRANSPOSE2_PD(G,H);
600 Heps = _mm_mul_pd(vfeps,H);
601 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
602 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
603 velec = _mm_mul_pd(qq23,VV);
604 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
605 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq23,FF),_mm_mul_pd(vftabscale,rinv23)));
607 /* Update potential sum for this i atom from the interaction with this j atom. */
608 velecsum = _mm_add_pd(velecsum,velec);
612 /* Calculate temporary vectorial force */
613 tx = _mm_mul_pd(fscal,dx23);
614 ty = _mm_mul_pd(fscal,dy23);
615 tz = _mm_mul_pd(fscal,dz23);
617 /* Update vectorial force */
618 fix2 = _mm_add_pd(fix2,tx);
619 fiy2 = _mm_add_pd(fiy2,ty);
620 fiz2 = _mm_add_pd(fiz2,tz);
622 fjx3 = _mm_add_pd(fjx3,tx);
623 fjy3 = _mm_add_pd(fjy3,ty);
624 fjz3 = _mm_add_pd(fjz3,tz);
626 /**************************
627 * CALCULATE INTERACTIONS *
628 **************************/
630 r31 = _mm_mul_pd(rsq31,rinv31);
632 /* Calculate table index by multiplying r with table scale and truncate to integer */
633 rt = _mm_mul_pd(r31,vftabscale);
634 vfitab = _mm_cvttpd_epi32(rt);
635 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
636 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
638 /* CUBIC SPLINE TABLE ELECTROSTATICS */
639 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
640 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
641 GMX_MM_TRANSPOSE2_PD(Y,F);
642 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
643 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
644 GMX_MM_TRANSPOSE2_PD(G,H);
645 Heps = _mm_mul_pd(vfeps,H);
646 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
647 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
648 velec = _mm_mul_pd(qq31,VV);
649 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
650 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq31,FF),_mm_mul_pd(vftabscale,rinv31)));
652 /* Update potential sum for this i atom from the interaction with this j atom. */
653 velecsum = _mm_add_pd(velecsum,velec);
657 /* Calculate temporary vectorial force */
658 tx = _mm_mul_pd(fscal,dx31);
659 ty = _mm_mul_pd(fscal,dy31);
660 tz = _mm_mul_pd(fscal,dz31);
662 /* Update vectorial force */
663 fix3 = _mm_add_pd(fix3,tx);
664 fiy3 = _mm_add_pd(fiy3,ty);
665 fiz3 = _mm_add_pd(fiz3,tz);
667 fjx1 = _mm_add_pd(fjx1,tx);
668 fjy1 = _mm_add_pd(fjy1,ty);
669 fjz1 = _mm_add_pd(fjz1,tz);
671 /**************************
672 * CALCULATE INTERACTIONS *
673 **************************/
675 r32 = _mm_mul_pd(rsq32,rinv32);
677 /* Calculate table index by multiplying r with table scale and truncate to integer */
678 rt = _mm_mul_pd(r32,vftabscale);
679 vfitab = _mm_cvttpd_epi32(rt);
680 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
681 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
683 /* CUBIC SPLINE TABLE ELECTROSTATICS */
684 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
685 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
686 GMX_MM_TRANSPOSE2_PD(Y,F);
687 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
688 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
689 GMX_MM_TRANSPOSE2_PD(G,H);
690 Heps = _mm_mul_pd(vfeps,H);
691 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
692 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
693 velec = _mm_mul_pd(qq32,VV);
694 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
695 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq32,FF),_mm_mul_pd(vftabscale,rinv32)));
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,dx32);
704 ty = _mm_mul_pd(fscal,dy32);
705 tz = _mm_mul_pd(fscal,dz32);
707 /* Update vectorial force */
708 fix3 = _mm_add_pd(fix3,tx);
709 fiy3 = _mm_add_pd(fiy3,ty);
710 fiz3 = _mm_add_pd(fiz3,tz);
712 fjx2 = _mm_add_pd(fjx2,tx);
713 fjy2 = _mm_add_pd(fjy2,ty);
714 fjz2 = _mm_add_pd(fjz2,tz);
716 /**************************
717 * CALCULATE INTERACTIONS *
718 **************************/
720 r33 = _mm_mul_pd(rsq33,rinv33);
722 /* Calculate table index by multiplying r with table scale and truncate to integer */
723 rt = _mm_mul_pd(r33,vftabscale);
724 vfitab = _mm_cvttpd_epi32(rt);
725 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
726 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
728 /* CUBIC SPLINE TABLE ELECTROSTATICS */
729 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
730 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
731 GMX_MM_TRANSPOSE2_PD(Y,F);
732 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
733 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
734 GMX_MM_TRANSPOSE2_PD(G,H);
735 Heps = _mm_mul_pd(vfeps,H);
736 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
737 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
738 velec = _mm_mul_pd(qq33,VV);
739 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
740 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq33,FF),_mm_mul_pd(vftabscale,rinv33)));
742 /* Update potential sum for this i atom from the interaction with this j atom. */
743 velecsum = _mm_add_pd(velecsum,velec);
747 /* Calculate temporary vectorial force */
748 tx = _mm_mul_pd(fscal,dx33);
749 ty = _mm_mul_pd(fscal,dy33);
750 tz = _mm_mul_pd(fscal,dz33);
752 /* Update vectorial force */
753 fix3 = _mm_add_pd(fix3,tx);
754 fiy3 = _mm_add_pd(fiy3,ty);
755 fiz3 = _mm_add_pd(fiz3,tz);
757 fjx3 = _mm_add_pd(fjx3,tx);
758 fjy3 = _mm_add_pd(fjy3,ty);
759 fjz3 = _mm_add_pd(fjz3,tz);
761 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);
763 /* Inner loop uses 446 flops */
770 j_coord_offsetA = DIM*jnrA;
772 /* load j atom coordinates */
773 gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
774 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
775 &jy2,&jz2,&jx3,&jy3,&jz3);
777 /* Calculate displacement vector */
778 dx00 = _mm_sub_pd(ix0,jx0);
779 dy00 = _mm_sub_pd(iy0,jy0);
780 dz00 = _mm_sub_pd(iz0,jz0);
781 dx11 = _mm_sub_pd(ix1,jx1);
782 dy11 = _mm_sub_pd(iy1,jy1);
783 dz11 = _mm_sub_pd(iz1,jz1);
784 dx12 = _mm_sub_pd(ix1,jx2);
785 dy12 = _mm_sub_pd(iy1,jy2);
786 dz12 = _mm_sub_pd(iz1,jz2);
787 dx13 = _mm_sub_pd(ix1,jx3);
788 dy13 = _mm_sub_pd(iy1,jy3);
789 dz13 = _mm_sub_pd(iz1,jz3);
790 dx21 = _mm_sub_pd(ix2,jx1);
791 dy21 = _mm_sub_pd(iy2,jy1);
792 dz21 = _mm_sub_pd(iz2,jz1);
793 dx22 = _mm_sub_pd(ix2,jx2);
794 dy22 = _mm_sub_pd(iy2,jy2);
795 dz22 = _mm_sub_pd(iz2,jz2);
796 dx23 = _mm_sub_pd(ix2,jx3);
797 dy23 = _mm_sub_pd(iy2,jy3);
798 dz23 = _mm_sub_pd(iz2,jz3);
799 dx31 = _mm_sub_pd(ix3,jx1);
800 dy31 = _mm_sub_pd(iy3,jy1);
801 dz31 = _mm_sub_pd(iz3,jz1);
802 dx32 = _mm_sub_pd(ix3,jx2);
803 dy32 = _mm_sub_pd(iy3,jy2);
804 dz32 = _mm_sub_pd(iz3,jz2);
805 dx33 = _mm_sub_pd(ix3,jx3);
806 dy33 = _mm_sub_pd(iy3,jy3);
807 dz33 = _mm_sub_pd(iz3,jz3);
809 /* Calculate squared distance and things based on it */
810 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
811 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
812 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
813 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
814 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
815 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
816 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
817 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
818 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
819 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
821 rinv00 = gmx_mm_invsqrt_pd(rsq00);
822 rinv11 = gmx_mm_invsqrt_pd(rsq11);
823 rinv12 = gmx_mm_invsqrt_pd(rsq12);
824 rinv13 = gmx_mm_invsqrt_pd(rsq13);
825 rinv21 = gmx_mm_invsqrt_pd(rsq21);
826 rinv22 = gmx_mm_invsqrt_pd(rsq22);
827 rinv23 = gmx_mm_invsqrt_pd(rsq23);
828 rinv31 = gmx_mm_invsqrt_pd(rsq31);
829 rinv32 = gmx_mm_invsqrt_pd(rsq32);
830 rinv33 = gmx_mm_invsqrt_pd(rsq33);
832 fjx0 = _mm_setzero_pd();
833 fjy0 = _mm_setzero_pd();
834 fjz0 = _mm_setzero_pd();
835 fjx1 = _mm_setzero_pd();
836 fjy1 = _mm_setzero_pd();
837 fjz1 = _mm_setzero_pd();
838 fjx2 = _mm_setzero_pd();
839 fjy2 = _mm_setzero_pd();
840 fjz2 = _mm_setzero_pd();
841 fjx3 = _mm_setzero_pd();
842 fjy3 = _mm_setzero_pd();
843 fjz3 = _mm_setzero_pd();
845 /**************************
846 * CALCULATE INTERACTIONS *
847 **************************/
849 r00 = _mm_mul_pd(rsq00,rinv00);
851 /* Calculate table index by multiplying r with table scale and truncate to integer */
852 rt = _mm_mul_pd(r00,vftabscale);
853 vfitab = _mm_cvttpd_epi32(rt);
854 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
855 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
857 /* CUBIC SPLINE TABLE DISPERSION */
858 vfitab = _mm_add_epi32(vfitab,ifour);
859 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
860 F = _mm_setzero_pd();
861 GMX_MM_TRANSPOSE2_PD(Y,F);
862 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
863 H = _mm_setzero_pd();
864 GMX_MM_TRANSPOSE2_PD(G,H);
865 Heps = _mm_mul_pd(vfeps,H);
866 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
867 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
868 vvdw6 = _mm_mul_pd(c6_00,VV);
869 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
870 fvdw6 = _mm_mul_pd(c6_00,FF);
872 /* CUBIC SPLINE TABLE REPULSION */
873 vfitab = _mm_add_epi32(vfitab,ifour);
874 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
875 F = _mm_setzero_pd();
876 GMX_MM_TRANSPOSE2_PD(Y,F);
877 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
878 H = _mm_setzero_pd();
879 GMX_MM_TRANSPOSE2_PD(G,H);
880 Heps = _mm_mul_pd(vfeps,H);
881 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
882 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
883 vvdw12 = _mm_mul_pd(c12_00,VV);
884 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
885 fvdw12 = _mm_mul_pd(c12_00,FF);
886 vvdw = _mm_add_pd(vvdw12,vvdw6);
887 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
889 /* Update potential sum for this i atom from the interaction with this j atom. */
890 vvdw = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
891 vvdwsum = _mm_add_pd(vvdwsum,vvdw);
895 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
897 /* Calculate temporary vectorial force */
898 tx = _mm_mul_pd(fscal,dx00);
899 ty = _mm_mul_pd(fscal,dy00);
900 tz = _mm_mul_pd(fscal,dz00);
902 /* Update vectorial force */
903 fix0 = _mm_add_pd(fix0,tx);
904 fiy0 = _mm_add_pd(fiy0,ty);
905 fiz0 = _mm_add_pd(fiz0,tz);
907 fjx0 = _mm_add_pd(fjx0,tx);
908 fjy0 = _mm_add_pd(fjy0,ty);
909 fjz0 = _mm_add_pd(fjz0,tz);
911 /**************************
912 * CALCULATE INTERACTIONS *
913 **************************/
915 r11 = _mm_mul_pd(rsq11,rinv11);
917 /* Calculate table index by multiplying r with table scale and truncate to integer */
918 rt = _mm_mul_pd(r11,vftabscale);
919 vfitab = _mm_cvttpd_epi32(rt);
920 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
921 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
923 /* CUBIC SPLINE TABLE ELECTROSTATICS */
924 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
925 F = _mm_setzero_pd();
926 GMX_MM_TRANSPOSE2_PD(Y,F);
927 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
928 H = _mm_setzero_pd();
929 GMX_MM_TRANSPOSE2_PD(G,H);
930 Heps = _mm_mul_pd(vfeps,H);
931 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
932 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
933 velec = _mm_mul_pd(qq11,VV);
934 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
935 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq11,FF),_mm_mul_pd(vftabscale,rinv11)));
937 /* Update potential sum for this i atom from the interaction with this j atom. */
938 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
939 velecsum = _mm_add_pd(velecsum,velec);
943 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
945 /* Calculate temporary vectorial force */
946 tx = _mm_mul_pd(fscal,dx11);
947 ty = _mm_mul_pd(fscal,dy11);
948 tz = _mm_mul_pd(fscal,dz11);
950 /* Update vectorial force */
951 fix1 = _mm_add_pd(fix1,tx);
952 fiy1 = _mm_add_pd(fiy1,ty);
953 fiz1 = _mm_add_pd(fiz1,tz);
955 fjx1 = _mm_add_pd(fjx1,tx);
956 fjy1 = _mm_add_pd(fjy1,ty);
957 fjz1 = _mm_add_pd(fjz1,tz);
959 /**************************
960 * CALCULATE INTERACTIONS *
961 **************************/
963 r12 = _mm_mul_pd(rsq12,rinv12);
965 /* Calculate table index by multiplying r with table scale and truncate to integer */
966 rt = _mm_mul_pd(r12,vftabscale);
967 vfitab = _mm_cvttpd_epi32(rt);
968 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
969 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
971 /* CUBIC SPLINE TABLE ELECTROSTATICS */
972 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
973 F = _mm_setzero_pd();
974 GMX_MM_TRANSPOSE2_PD(Y,F);
975 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
976 H = _mm_setzero_pd();
977 GMX_MM_TRANSPOSE2_PD(G,H);
978 Heps = _mm_mul_pd(vfeps,H);
979 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
980 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
981 velec = _mm_mul_pd(qq12,VV);
982 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
983 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq12,FF),_mm_mul_pd(vftabscale,rinv12)));
985 /* Update potential sum for this i atom from the interaction with this j atom. */
986 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
987 velecsum = _mm_add_pd(velecsum,velec);
991 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
993 /* Calculate temporary vectorial force */
994 tx = _mm_mul_pd(fscal,dx12);
995 ty = _mm_mul_pd(fscal,dy12);
996 tz = _mm_mul_pd(fscal,dz12);
998 /* Update vectorial force */
999 fix1 = _mm_add_pd(fix1,tx);
1000 fiy1 = _mm_add_pd(fiy1,ty);
1001 fiz1 = _mm_add_pd(fiz1,tz);
1003 fjx2 = _mm_add_pd(fjx2,tx);
1004 fjy2 = _mm_add_pd(fjy2,ty);
1005 fjz2 = _mm_add_pd(fjz2,tz);
1007 /**************************
1008 * CALCULATE INTERACTIONS *
1009 **************************/
1011 r13 = _mm_mul_pd(rsq13,rinv13);
1013 /* Calculate table index by multiplying r with table scale and truncate to integer */
1014 rt = _mm_mul_pd(r13,vftabscale);
1015 vfitab = _mm_cvttpd_epi32(rt);
1016 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1017 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1019 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1020 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1021 F = _mm_setzero_pd();
1022 GMX_MM_TRANSPOSE2_PD(Y,F);
1023 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1024 H = _mm_setzero_pd();
1025 GMX_MM_TRANSPOSE2_PD(G,H);
1026 Heps = _mm_mul_pd(vfeps,H);
1027 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1028 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1029 velec = _mm_mul_pd(qq13,VV);
1030 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1031 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq13,FF),_mm_mul_pd(vftabscale,rinv13)));
1033 /* Update potential sum for this i atom from the interaction with this j atom. */
1034 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1035 velecsum = _mm_add_pd(velecsum,velec);
1039 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1041 /* Calculate temporary vectorial force */
1042 tx = _mm_mul_pd(fscal,dx13);
1043 ty = _mm_mul_pd(fscal,dy13);
1044 tz = _mm_mul_pd(fscal,dz13);
1046 /* Update vectorial force */
1047 fix1 = _mm_add_pd(fix1,tx);
1048 fiy1 = _mm_add_pd(fiy1,ty);
1049 fiz1 = _mm_add_pd(fiz1,tz);
1051 fjx3 = _mm_add_pd(fjx3,tx);
1052 fjy3 = _mm_add_pd(fjy3,ty);
1053 fjz3 = _mm_add_pd(fjz3,tz);
1055 /**************************
1056 * CALCULATE INTERACTIONS *
1057 **************************/
1059 r21 = _mm_mul_pd(rsq21,rinv21);
1061 /* Calculate table index by multiplying r with table scale and truncate to integer */
1062 rt = _mm_mul_pd(r21,vftabscale);
1063 vfitab = _mm_cvttpd_epi32(rt);
1064 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1065 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1067 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1068 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1069 F = _mm_setzero_pd();
1070 GMX_MM_TRANSPOSE2_PD(Y,F);
1071 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1072 H = _mm_setzero_pd();
1073 GMX_MM_TRANSPOSE2_PD(G,H);
1074 Heps = _mm_mul_pd(vfeps,H);
1075 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1076 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1077 velec = _mm_mul_pd(qq21,VV);
1078 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1079 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq21,FF),_mm_mul_pd(vftabscale,rinv21)));
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 /* Calculate table index by multiplying r with table scale and truncate to integer */
1110 rt = _mm_mul_pd(r22,vftabscale);
1111 vfitab = _mm_cvttpd_epi32(rt);
1112 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1113 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1115 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1116 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1117 F = _mm_setzero_pd();
1118 GMX_MM_TRANSPOSE2_PD(Y,F);
1119 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1120 H = _mm_setzero_pd();
1121 GMX_MM_TRANSPOSE2_PD(G,H);
1122 Heps = _mm_mul_pd(vfeps,H);
1123 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1124 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1125 velec = _mm_mul_pd(qq22,VV);
1126 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1127 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq22,FF),_mm_mul_pd(vftabscale,rinv22)));
1129 /* Update potential sum for this i atom from the interaction with this j atom. */
1130 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1131 velecsum = _mm_add_pd(velecsum,velec);
1135 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1137 /* Calculate temporary vectorial force */
1138 tx = _mm_mul_pd(fscal,dx22);
1139 ty = _mm_mul_pd(fscal,dy22);
1140 tz = _mm_mul_pd(fscal,dz22);
1142 /* Update vectorial force */
1143 fix2 = _mm_add_pd(fix2,tx);
1144 fiy2 = _mm_add_pd(fiy2,ty);
1145 fiz2 = _mm_add_pd(fiz2,tz);
1147 fjx2 = _mm_add_pd(fjx2,tx);
1148 fjy2 = _mm_add_pd(fjy2,ty);
1149 fjz2 = _mm_add_pd(fjz2,tz);
1151 /**************************
1152 * CALCULATE INTERACTIONS *
1153 **************************/
1155 r23 = _mm_mul_pd(rsq23,rinv23);
1157 /* Calculate table index by multiplying r with table scale and truncate to integer */
1158 rt = _mm_mul_pd(r23,vftabscale);
1159 vfitab = _mm_cvttpd_epi32(rt);
1160 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1161 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1163 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1164 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1165 F = _mm_setzero_pd();
1166 GMX_MM_TRANSPOSE2_PD(Y,F);
1167 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1168 H = _mm_setzero_pd();
1169 GMX_MM_TRANSPOSE2_PD(G,H);
1170 Heps = _mm_mul_pd(vfeps,H);
1171 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1172 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1173 velec = _mm_mul_pd(qq23,VV);
1174 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1175 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq23,FF),_mm_mul_pd(vftabscale,rinv23)));
1177 /* Update potential sum for this i atom from the interaction with this j atom. */
1178 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1179 velecsum = _mm_add_pd(velecsum,velec);
1183 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1185 /* Calculate temporary vectorial force */
1186 tx = _mm_mul_pd(fscal,dx23);
1187 ty = _mm_mul_pd(fscal,dy23);
1188 tz = _mm_mul_pd(fscal,dz23);
1190 /* Update vectorial force */
1191 fix2 = _mm_add_pd(fix2,tx);
1192 fiy2 = _mm_add_pd(fiy2,ty);
1193 fiz2 = _mm_add_pd(fiz2,tz);
1195 fjx3 = _mm_add_pd(fjx3,tx);
1196 fjy3 = _mm_add_pd(fjy3,ty);
1197 fjz3 = _mm_add_pd(fjz3,tz);
1199 /**************************
1200 * CALCULATE INTERACTIONS *
1201 **************************/
1203 r31 = _mm_mul_pd(rsq31,rinv31);
1205 /* Calculate table index by multiplying r with table scale and truncate to integer */
1206 rt = _mm_mul_pd(r31,vftabscale);
1207 vfitab = _mm_cvttpd_epi32(rt);
1208 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1209 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1211 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1212 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1213 F = _mm_setzero_pd();
1214 GMX_MM_TRANSPOSE2_PD(Y,F);
1215 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1216 H = _mm_setzero_pd();
1217 GMX_MM_TRANSPOSE2_PD(G,H);
1218 Heps = _mm_mul_pd(vfeps,H);
1219 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1220 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1221 velec = _mm_mul_pd(qq31,VV);
1222 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1223 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq31,FF),_mm_mul_pd(vftabscale,rinv31)));
1225 /* Update potential sum for this i atom from the interaction with this j atom. */
1226 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1227 velecsum = _mm_add_pd(velecsum,velec);
1231 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1233 /* Calculate temporary vectorial force */
1234 tx = _mm_mul_pd(fscal,dx31);
1235 ty = _mm_mul_pd(fscal,dy31);
1236 tz = _mm_mul_pd(fscal,dz31);
1238 /* Update vectorial force */
1239 fix3 = _mm_add_pd(fix3,tx);
1240 fiy3 = _mm_add_pd(fiy3,ty);
1241 fiz3 = _mm_add_pd(fiz3,tz);
1243 fjx1 = _mm_add_pd(fjx1,tx);
1244 fjy1 = _mm_add_pd(fjy1,ty);
1245 fjz1 = _mm_add_pd(fjz1,tz);
1247 /**************************
1248 * CALCULATE INTERACTIONS *
1249 **************************/
1251 r32 = _mm_mul_pd(rsq32,rinv32);
1253 /* Calculate table index by multiplying r with table scale and truncate to integer */
1254 rt = _mm_mul_pd(r32,vftabscale);
1255 vfitab = _mm_cvttpd_epi32(rt);
1256 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1257 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1259 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1260 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1261 F = _mm_setzero_pd();
1262 GMX_MM_TRANSPOSE2_PD(Y,F);
1263 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1264 H = _mm_setzero_pd();
1265 GMX_MM_TRANSPOSE2_PD(G,H);
1266 Heps = _mm_mul_pd(vfeps,H);
1267 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1268 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1269 velec = _mm_mul_pd(qq32,VV);
1270 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1271 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq32,FF),_mm_mul_pd(vftabscale,rinv32)));
1273 /* Update potential sum for this i atom from the interaction with this j atom. */
1274 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1275 velecsum = _mm_add_pd(velecsum,velec);
1279 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1281 /* Calculate temporary vectorial force */
1282 tx = _mm_mul_pd(fscal,dx32);
1283 ty = _mm_mul_pd(fscal,dy32);
1284 tz = _mm_mul_pd(fscal,dz32);
1286 /* Update vectorial force */
1287 fix3 = _mm_add_pd(fix3,tx);
1288 fiy3 = _mm_add_pd(fiy3,ty);
1289 fiz3 = _mm_add_pd(fiz3,tz);
1291 fjx2 = _mm_add_pd(fjx2,tx);
1292 fjy2 = _mm_add_pd(fjy2,ty);
1293 fjz2 = _mm_add_pd(fjz2,tz);
1295 /**************************
1296 * CALCULATE INTERACTIONS *
1297 **************************/
1299 r33 = _mm_mul_pd(rsq33,rinv33);
1301 /* Calculate table index by multiplying r with table scale and truncate to integer */
1302 rt = _mm_mul_pd(r33,vftabscale);
1303 vfitab = _mm_cvttpd_epi32(rt);
1304 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1305 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1307 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1308 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1309 F = _mm_setzero_pd();
1310 GMX_MM_TRANSPOSE2_PD(Y,F);
1311 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1312 H = _mm_setzero_pd();
1313 GMX_MM_TRANSPOSE2_PD(G,H);
1314 Heps = _mm_mul_pd(vfeps,H);
1315 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1316 VV = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
1317 velec = _mm_mul_pd(qq33,VV);
1318 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1319 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq33,FF),_mm_mul_pd(vftabscale,rinv33)));
1321 /* Update potential sum for this i atom from the interaction with this j atom. */
1322 velec = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1323 velecsum = _mm_add_pd(velecsum,velec);
1327 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1329 /* Calculate temporary vectorial force */
1330 tx = _mm_mul_pd(fscal,dx33);
1331 ty = _mm_mul_pd(fscal,dy33);
1332 tz = _mm_mul_pd(fscal,dz33);
1334 /* Update vectorial force */
1335 fix3 = _mm_add_pd(fix3,tx);
1336 fiy3 = _mm_add_pd(fiy3,ty);
1337 fiz3 = _mm_add_pd(fiz3,tz);
1339 fjx3 = _mm_add_pd(fjx3,tx);
1340 fjy3 = _mm_add_pd(fjy3,ty);
1341 fjz3 = _mm_add_pd(fjz3,tz);
1343 gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1345 /* Inner loop uses 446 flops */
1348 /* End of innermost loop */
1350 gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1351 f+i_coord_offset,fshift+i_shift_offset);
1354 /* Update potential energies */
1355 gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1356 gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1358 /* Increment number of inner iterations */
1359 inneriter += j_index_end - j_index_start;
1361 /* Outer loop uses 26 flops */
1364 /* Increment number of outer iterations */
1367 /* Update outer/inner flops */
1369 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*446);
1372 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW4W4_F_sse2_double
1373 * Electrostatics interaction: CubicSplineTable
1374 * VdW interaction: CubicSplineTable
1375 * Geometry: Water4-Water4
1376 * Calculate force/pot: Force
1379 nb_kernel_ElecCSTab_VdwCSTab_GeomW4W4_F_sse2_double
1380 (t_nblist * gmx_restrict nlist,
1381 rvec * gmx_restrict xx,
1382 rvec * gmx_restrict ff,
1383 t_forcerec * gmx_restrict fr,
1384 t_mdatoms * gmx_restrict mdatoms,
1385 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1386 t_nrnb * gmx_restrict nrnb)
1388 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1389 * just 0 for non-waters.
1390 * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
1391 * jnr indices corresponding to data put in the four positions in the SIMD register.
1393 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1394 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1396 int j_coord_offsetA,j_coord_offsetB;
1397 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1398 real rcutoff_scalar;
1399 real *shiftvec,*fshift,*x,*f;
1400 __m128d tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1402 __m128d ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1404 __m128d ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1406 __m128d ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1408 __m128d ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1409 int vdwjidx0A,vdwjidx0B;
1410 __m128d jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1411 int vdwjidx1A,vdwjidx1B;
1412 __m128d jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1413 int vdwjidx2A,vdwjidx2B;
1414 __m128d jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1415 int vdwjidx3A,vdwjidx3B;
1416 __m128d jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1417 __m128d dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1418 __m128d dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1419 __m128d dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1420 __m128d dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1421 __m128d dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1422 __m128d dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1423 __m128d dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1424 __m128d dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1425 __m128d dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1426 __m128d dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1427 __m128d velec,felec,velecsum,facel,crf,krf,krf2;
1430 __m128d rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1433 __m128d one_sixth = _mm_set1_pd(1.0/6.0);
1434 __m128d one_twelfth = _mm_set1_pd(1.0/12.0);
1436 __m128i ifour = _mm_set1_epi32(4);
1437 __m128d rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1439 __m128d dummy_mask,cutoff_mask;
1440 __m128d signbit = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
1441 __m128d one = _mm_set1_pd(1.0);
1442 __m128d two = _mm_set1_pd(2.0);
1448 jindex = nlist->jindex;
1450 shiftidx = nlist->shift;
1452 shiftvec = fr->shift_vec[0];
1453 fshift = fr->fshift[0];
1454 facel = _mm_set1_pd(fr->epsfac);
1455 charge = mdatoms->chargeA;
1456 nvdwtype = fr->ntype;
1457 vdwparam = fr->nbfp;
1458 vdwtype = mdatoms->typeA;
1460 vftab = kernel_data->table_elec_vdw->data;
1461 vftabscale = _mm_set1_pd(kernel_data->table_elec_vdw->scale);
1463 /* Setup water-specific parameters */
1464 inr = nlist->iinr[0];
1465 iq1 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
1466 iq2 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
1467 iq3 = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
1468 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
1470 jq1 = _mm_set1_pd(charge[inr+1]);
1471 jq2 = _mm_set1_pd(charge[inr+2]);
1472 jq3 = _mm_set1_pd(charge[inr+3]);
1473 vdwjidx0A = 2*vdwtype[inr+0];
1474 c6_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
1475 c12_00 = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
1476 qq11 = _mm_mul_pd(iq1,jq1);
1477 qq12 = _mm_mul_pd(iq1,jq2);
1478 qq13 = _mm_mul_pd(iq1,jq3);
1479 qq21 = _mm_mul_pd(iq2,jq1);
1480 qq22 = _mm_mul_pd(iq2,jq2);
1481 qq23 = _mm_mul_pd(iq2,jq3);
1482 qq31 = _mm_mul_pd(iq3,jq1);
1483 qq32 = _mm_mul_pd(iq3,jq2);
1484 qq33 = _mm_mul_pd(iq3,jq3);
1486 /* Avoid stupid compiler warnings */
1488 j_coord_offsetA = 0;
1489 j_coord_offsetB = 0;
1494 /* Start outer loop over neighborlists */
1495 for(iidx=0; iidx<nri; iidx++)
1497 /* Load shift vector for this list */
1498 i_shift_offset = DIM*shiftidx[iidx];
1500 /* Load limits for loop over neighbors */
1501 j_index_start = jindex[iidx];
1502 j_index_end = jindex[iidx+1];
1504 /* Get outer coordinate index */
1506 i_coord_offset = DIM*inr;
1508 /* Load i particle coords and add shift vector */
1509 gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1510 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1512 fix0 = _mm_setzero_pd();
1513 fiy0 = _mm_setzero_pd();
1514 fiz0 = _mm_setzero_pd();
1515 fix1 = _mm_setzero_pd();
1516 fiy1 = _mm_setzero_pd();
1517 fiz1 = _mm_setzero_pd();
1518 fix2 = _mm_setzero_pd();
1519 fiy2 = _mm_setzero_pd();
1520 fiz2 = _mm_setzero_pd();
1521 fix3 = _mm_setzero_pd();
1522 fiy3 = _mm_setzero_pd();
1523 fiz3 = _mm_setzero_pd();
1525 /* Start inner kernel loop */
1526 for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
1529 /* Get j neighbor index, and coordinate index */
1531 jnrB = jjnr[jidx+1];
1532 j_coord_offsetA = DIM*jnrA;
1533 j_coord_offsetB = DIM*jnrB;
1535 /* load j atom coordinates */
1536 gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1537 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1538 &jy2,&jz2,&jx3,&jy3,&jz3);
1540 /* Calculate displacement vector */
1541 dx00 = _mm_sub_pd(ix0,jx0);
1542 dy00 = _mm_sub_pd(iy0,jy0);
1543 dz00 = _mm_sub_pd(iz0,jz0);
1544 dx11 = _mm_sub_pd(ix1,jx1);
1545 dy11 = _mm_sub_pd(iy1,jy1);
1546 dz11 = _mm_sub_pd(iz1,jz1);
1547 dx12 = _mm_sub_pd(ix1,jx2);
1548 dy12 = _mm_sub_pd(iy1,jy2);
1549 dz12 = _mm_sub_pd(iz1,jz2);
1550 dx13 = _mm_sub_pd(ix1,jx3);
1551 dy13 = _mm_sub_pd(iy1,jy3);
1552 dz13 = _mm_sub_pd(iz1,jz3);
1553 dx21 = _mm_sub_pd(ix2,jx1);
1554 dy21 = _mm_sub_pd(iy2,jy1);
1555 dz21 = _mm_sub_pd(iz2,jz1);
1556 dx22 = _mm_sub_pd(ix2,jx2);
1557 dy22 = _mm_sub_pd(iy2,jy2);
1558 dz22 = _mm_sub_pd(iz2,jz2);
1559 dx23 = _mm_sub_pd(ix2,jx3);
1560 dy23 = _mm_sub_pd(iy2,jy3);
1561 dz23 = _mm_sub_pd(iz2,jz3);
1562 dx31 = _mm_sub_pd(ix3,jx1);
1563 dy31 = _mm_sub_pd(iy3,jy1);
1564 dz31 = _mm_sub_pd(iz3,jz1);
1565 dx32 = _mm_sub_pd(ix3,jx2);
1566 dy32 = _mm_sub_pd(iy3,jy2);
1567 dz32 = _mm_sub_pd(iz3,jz2);
1568 dx33 = _mm_sub_pd(ix3,jx3);
1569 dy33 = _mm_sub_pd(iy3,jy3);
1570 dz33 = _mm_sub_pd(iz3,jz3);
1572 /* Calculate squared distance and things based on it */
1573 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1574 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
1575 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
1576 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
1577 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
1578 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
1579 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
1580 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
1581 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
1582 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
1584 rinv00 = gmx_mm_invsqrt_pd(rsq00);
1585 rinv11 = gmx_mm_invsqrt_pd(rsq11);
1586 rinv12 = gmx_mm_invsqrt_pd(rsq12);
1587 rinv13 = gmx_mm_invsqrt_pd(rsq13);
1588 rinv21 = gmx_mm_invsqrt_pd(rsq21);
1589 rinv22 = gmx_mm_invsqrt_pd(rsq22);
1590 rinv23 = gmx_mm_invsqrt_pd(rsq23);
1591 rinv31 = gmx_mm_invsqrt_pd(rsq31);
1592 rinv32 = gmx_mm_invsqrt_pd(rsq32);
1593 rinv33 = gmx_mm_invsqrt_pd(rsq33);
1595 fjx0 = _mm_setzero_pd();
1596 fjy0 = _mm_setzero_pd();
1597 fjz0 = _mm_setzero_pd();
1598 fjx1 = _mm_setzero_pd();
1599 fjy1 = _mm_setzero_pd();
1600 fjz1 = _mm_setzero_pd();
1601 fjx2 = _mm_setzero_pd();
1602 fjy2 = _mm_setzero_pd();
1603 fjz2 = _mm_setzero_pd();
1604 fjx3 = _mm_setzero_pd();
1605 fjy3 = _mm_setzero_pd();
1606 fjz3 = _mm_setzero_pd();
1608 /**************************
1609 * CALCULATE INTERACTIONS *
1610 **************************/
1612 r00 = _mm_mul_pd(rsq00,rinv00);
1614 /* Calculate table index by multiplying r with table scale and truncate to integer */
1615 rt = _mm_mul_pd(r00,vftabscale);
1616 vfitab = _mm_cvttpd_epi32(rt);
1617 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1618 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1620 /* CUBIC SPLINE TABLE DISPERSION */
1621 vfitab = _mm_add_epi32(vfitab,ifour);
1622 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1623 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1624 GMX_MM_TRANSPOSE2_PD(Y,F);
1625 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1626 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1627 GMX_MM_TRANSPOSE2_PD(G,H);
1628 Heps = _mm_mul_pd(vfeps,H);
1629 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1630 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1631 fvdw6 = _mm_mul_pd(c6_00,FF);
1633 /* CUBIC SPLINE TABLE REPULSION */
1634 vfitab = _mm_add_epi32(vfitab,ifour);
1635 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1636 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1637 GMX_MM_TRANSPOSE2_PD(Y,F);
1638 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1639 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1640 GMX_MM_TRANSPOSE2_PD(G,H);
1641 Heps = _mm_mul_pd(vfeps,H);
1642 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1643 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1644 fvdw12 = _mm_mul_pd(c12_00,FF);
1645 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
1649 /* Calculate temporary vectorial force */
1650 tx = _mm_mul_pd(fscal,dx00);
1651 ty = _mm_mul_pd(fscal,dy00);
1652 tz = _mm_mul_pd(fscal,dz00);
1654 /* Update vectorial force */
1655 fix0 = _mm_add_pd(fix0,tx);
1656 fiy0 = _mm_add_pd(fiy0,ty);
1657 fiz0 = _mm_add_pd(fiz0,tz);
1659 fjx0 = _mm_add_pd(fjx0,tx);
1660 fjy0 = _mm_add_pd(fjy0,ty);
1661 fjz0 = _mm_add_pd(fjz0,tz);
1663 /**************************
1664 * CALCULATE INTERACTIONS *
1665 **************************/
1667 r11 = _mm_mul_pd(rsq11,rinv11);
1669 /* Calculate table index by multiplying r with table scale and truncate to integer */
1670 rt = _mm_mul_pd(r11,vftabscale);
1671 vfitab = _mm_cvttpd_epi32(rt);
1672 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1673 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1675 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1676 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1677 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1678 GMX_MM_TRANSPOSE2_PD(Y,F);
1679 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1680 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1681 GMX_MM_TRANSPOSE2_PD(G,H);
1682 Heps = _mm_mul_pd(vfeps,H);
1683 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1684 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1685 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq11,FF),_mm_mul_pd(vftabscale,rinv11)));
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 /* Calculate table index by multiplying r with table scale and truncate to integer */
1710 rt = _mm_mul_pd(r12,vftabscale);
1711 vfitab = _mm_cvttpd_epi32(rt);
1712 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1713 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1715 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1716 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1717 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1718 GMX_MM_TRANSPOSE2_PD(Y,F);
1719 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1720 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1721 GMX_MM_TRANSPOSE2_PD(G,H);
1722 Heps = _mm_mul_pd(vfeps,H);
1723 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1724 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1725 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq12,FF),_mm_mul_pd(vftabscale,rinv12)));
1729 /* Calculate temporary vectorial force */
1730 tx = _mm_mul_pd(fscal,dx12);
1731 ty = _mm_mul_pd(fscal,dy12);
1732 tz = _mm_mul_pd(fscal,dz12);
1734 /* Update vectorial force */
1735 fix1 = _mm_add_pd(fix1,tx);
1736 fiy1 = _mm_add_pd(fiy1,ty);
1737 fiz1 = _mm_add_pd(fiz1,tz);
1739 fjx2 = _mm_add_pd(fjx2,tx);
1740 fjy2 = _mm_add_pd(fjy2,ty);
1741 fjz2 = _mm_add_pd(fjz2,tz);
1743 /**************************
1744 * CALCULATE INTERACTIONS *
1745 **************************/
1747 r13 = _mm_mul_pd(rsq13,rinv13);
1749 /* Calculate table index by multiplying r with table scale and truncate to integer */
1750 rt = _mm_mul_pd(r13,vftabscale);
1751 vfitab = _mm_cvttpd_epi32(rt);
1752 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1753 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1755 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1756 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1757 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1758 GMX_MM_TRANSPOSE2_PD(Y,F);
1759 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1760 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1761 GMX_MM_TRANSPOSE2_PD(G,H);
1762 Heps = _mm_mul_pd(vfeps,H);
1763 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1764 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1765 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq13,FF),_mm_mul_pd(vftabscale,rinv13)));
1769 /* Calculate temporary vectorial force */
1770 tx = _mm_mul_pd(fscal,dx13);
1771 ty = _mm_mul_pd(fscal,dy13);
1772 tz = _mm_mul_pd(fscal,dz13);
1774 /* Update vectorial force */
1775 fix1 = _mm_add_pd(fix1,tx);
1776 fiy1 = _mm_add_pd(fiy1,ty);
1777 fiz1 = _mm_add_pd(fiz1,tz);
1779 fjx3 = _mm_add_pd(fjx3,tx);
1780 fjy3 = _mm_add_pd(fjy3,ty);
1781 fjz3 = _mm_add_pd(fjz3,tz);
1783 /**************************
1784 * CALCULATE INTERACTIONS *
1785 **************************/
1787 r21 = _mm_mul_pd(rsq21,rinv21);
1789 /* Calculate table index by multiplying r with table scale and truncate to integer */
1790 rt = _mm_mul_pd(r21,vftabscale);
1791 vfitab = _mm_cvttpd_epi32(rt);
1792 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1793 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1795 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1796 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1797 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1798 GMX_MM_TRANSPOSE2_PD(Y,F);
1799 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1800 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1801 GMX_MM_TRANSPOSE2_PD(G,H);
1802 Heps = _mm_mul_pd(vfeps,H);
1803 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1804 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1805 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq21,FF),_mm_mul_pd(vftabscale,rinv21)));
1809 /* Calculate temporary vectorial force */
1810 tx = _mm_mul_pd(fscal,dx21);
1811 ty = _mm_mul_pd(fscal,dy21);
1812 tz = _mm_mul_pd(fscal,dz21);
1814 /* Update vectorial force */
1815 fix2 = _mm_add_pd(fix2,tx);
1816 fiy2 = _mm_add_pd(fiy2,ty);
1817 fiz2 = _mm_add_pd(fiz2,tz);
1819 fjx1 = _mm_add_pd(fjx1,tx);
1820 fjy1 = _mm_add_pd(fjy1,ty);
1821 fjz1 = _mm_add_pd(fjz1,tz);
1823 /**************************
1824 * CALCULATE INTERACTIONS *
1825 **************************/
1827 r22 = _mm_mul_pd(rsq22,rinv22);
1829 /* Calculate table index by multiplying r with table scale and truncate to integer */
1830 rt = _mm_mul_pd(r22,vftabscale);
1831 vfitab = _mm_cvttpd_epi32(rt);
1832 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1833 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1835 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1836 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1837 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1838 GMX_MM_TRANSPOSE2_PD(Y,F);
1839 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1840 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1841 GMX_MM_TRANSPOSE2_PD(G,H);
1842 Heps = _mm_mul_pd(vfeps,H);
1843 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1844 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1845 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq22,FF),_mm_mul_pd(vftabscale,rinv22)));
1849 /* Calculate temporary vectorial force */
1850 tx = _mm_mul_pd(fscal,dx22);
1851 ty = _mm_mul_pd(fscal,dy22);
1852 tz = _mm_mul_pd(fscal,dz22);
1854 /* Update vectorial force */
1855 fix2 = _mm_add_pd(fix2,tx);
1856 fiy2 = _mm_add_pd(fiy2,ty);
1857 fiz2 = _mm_add_pd(fiz2,tz);
1859 fjx2 = _mm_add_pd(fjx2,tx);
1860 fjy2 = _mm_add_pd(fjy2,ty);
1861 fjz2 = _mm_add_pd(fjz2,tz);
1863 /**************************
1864 * CALCULATE INTERACTIONS *
1865 **************************/
1867 r23 = _mm_mul_pd(rsq23,rinv23);
1869 /* Calculate table index by multiplying r with table scale and truncate to integer */
1870 rt = _mm_mul_pd(r23,vftabscale);
1871 vfitab = _mm_cvttpd_epi32(rt);
1872 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1873 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1875 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1876 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1877 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1878 GMX_MM_TRANSPOSE2_PD(Y,F);
1879 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1880 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1881 GMX_MM_TRANSPOSE2_PD(G,H);
1882 Heps = _mm_mul_pd(vfeps,H);
1883 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1884 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1885 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq23,FF),_mm_mul_pd(vftabscale,rinv23)));
1889 /* Calculate temporary vectorial force */
1890 tx = _mm_mul_pd(fscal,dx23);
1891 ty = _mm_mul_pd(fscal,dy23);
1892 tz = _mm_mul_pd(fscal,dz23);
1894 /* Update vectorial force */
1895 fix2 = _mm_add_pd(fix2,tx);
1896 fiy2 = _mm_add_pd(fiy2,ty);
1897 fiz2 = _mm_add_pd(fiz2,tz);
1899 fjx3 = _mm_add_pd(fjx3,tx);
1900 fjy3 = _mm_add_pd(fjy3,ty);
1901 fjz3 = _mm_add_pd(fjz3,tz);
1903 /**************************
1904 * CALCULATE INTERACTIONS *
1905 **************************/
1907 r31 = _mm_mul_pd(rsq31,rinv31);
1909 /* Calculate table index by multiplying r with table scale and truncate to integer */
1910 rt = _mm_mul_pd(r31,vftabscale);
1911 vfitab = _mm_cvttpd_epi32(rt);
1912 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1913 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1915 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1916 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1917 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1918 GMX_MM_TRANSPOSE2_PD(Y,F);
1919 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1920 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1921 GMX_MM_TRANSPOSE2_PD(G,H);
1922 Heps = _mm_mul_pd(vfeps,H);
1923 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1924 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1925 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq31,FF),_mm_mul_pd(vftabscale,rinv31)));
1929 /* Calculate temporary vectorial force */
1930 tx = _mm_mul_pd(fscal,dx31);
1931 ty = _mm_mul_pd(fscal,dy31);
1932 tz = _mm_mul_pd(fscal,dz31);
1934 /* Update vectorial force */
1935 fix3 = _mm_add_pd(fix3,tx);
1936 fiy3 = _mm_add_pd(fiy3,ty);
1937 fiz3 = _mm_add_pd(fiz3,tz);
1939 fjx1 = _mm_add_pd(fjx1,tx);
1940 fjy1 = _mm_add_pd(fjy1,ty);
1941 fjz1 = _mm_add_pd(fjz1,tz);
1943 /**************************
1944 * CALCULATE INTERACTIONS *
1945 **************************/
1947 r32 = _mm_mul_pd(rsq32,rinv32);
1949 /* Calculate table index by multiplying r with table scale and truncate to integer */
1950 rt = _mm_mul_pd(r32,vftabscale);
1951 vfitab = _mm_cvttpd_epi32(rt);
1952 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1953 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1955 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1956 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1957 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1958 GMX_MM_TRANSPOSE2_PD(Y,F);
1959 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
1960 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
1961 GMX_MM_TRANSPOSE2_PD(G,H);
1962 Heps = _mm_mul_pd(vfeps,H);
1963 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
1964 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
1965 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq32,FF),_mm_mul_pd(vftabscale,rinv32)));
1969 /* Calculate temporary vectorial force */
1970 tx = _mm_mul_pd(fscal,dx32);
1971 ty = _mm_mul_pd(fscal,dy32);
1972 tz = _mm_mul_pd(fscal,dz32);
1974 /* Update vectorial force */
1975 fix3 = _mm_add_pd(fix3,tx);
1976 fiy3 = _mm_add_pd(fiy3,ty);
1977 fiz3 = _mm_add_pd(fiz3,tz);
1979 fjx2 = _mm_add_pd(fjx2,tx);
1980 fjy2 = _mm_add_pd(fjy2,ty);
1981 fjz2 = _mm_add_pd(fjz2,tz);
1983 /**************************
1984 * CALCULATE INTERACTIONS *
1985 **************************/
1987 r33 = _mm_mul_pd(rsq33,rinv33);
1989 /* Calculate table index by multiplying r with table scale and truncate to integer */
1990 rt = _mm_mul_pd(r33,vftabscale);
1991 vfitab = _mm_cvttpd_epi32(rt);
1992 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
1993 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1995 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1996 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
1997 F = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
1998 GMX_MM_TRANSPOSE2_PD(Y,F);
1999 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2000 H = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
2001 GMX_MM_TRANSPOSE2_PD(G,H);
2002 Heps = _mm_mul_pd(vfeps,H);
2003 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2004 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2005 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq33,FF),_mm_mul_pd(vftabscale,rinv33)));
2009 /* Calculate temporary vectorial force */
2010 tx = _mm_mul_pd(fscal,dx33);
2011 ty = _mm_mul_pd(fscal,dy33);
2012 tz = _mm_mul_pd(fscal,dz33);
2014 /* Update vectorial force */
2015 fix3 = _mm_add_pd(fix3,tx);
2016 fiy3 = _mm_add_pd(fiy3,ty);
2017 fiz3 = _mm_add_pd(fiz3,tz);
2019 fjx3 = _mm_add_pd(fjx3,tx);
2020 fjy3 = _mm_add_pd(fjy3,ty);
2021 fjz3 = _mm_add_pd(fjz3,tz);
2023 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);
2025 /* Inner loop uses 402 flops */
2028 if(jidx<j_index_end)
2032 j_coord_offsetA = DIM*jnrA;
2034 /* load j atom coordinates */
2035 gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
2036 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
2037 &jy2,&jz2,&jx3,&jy3,&jz3);
2039 /* Calculate displacement vector */
2040 dx00 = _mm_sub_pd(ix0,jx0);
2041 dy00 = _mm_sub_pd(iy0,jy0);
2042 dz00 = _mm_sub_pd(iz0,jz0);
2043 dx11 = _mm_sub_pd(ix1,jx1);
2044 dy11 = _mm_sub_pd(iy1,jy1);
2045 dz11 = _mm_sub_pd(iz1,jz1);
2046 dx12 = _mm_sub_pd(ix1,jx2);
2047 dy12 = _mm_sub_pd(iy1,jy2);
2048 dz12 = _mm_sub_pd(iz1,jz2);
2049 dx13 = _mm_sub_pd(ix1,jx3);
2050 dy13 = _mm_sub_pd(iy1,jy3);
2051 dz13 = _mm_sub_pd(iz1,jz3);
2052 dx21 = _mm_sub_pd(ix2,jx1);
2053 dy21 = _mm_sub_pd(iy2,jy1);
2054 dz21 = _mm_sub_pd(iz2,jz1);
2055 dx22 = _mm_sub_pd(ix2,jx2);
2056 dy22 = _mm_sub_pd(iy2,jy2);
2057 dz22 = _mm_sub_pd(iz2,jz2);
2058 dx23 = _mm_sub_pd(ix2,jx3);
2059 dy23 = _mm_sub_pd(iy2,jy3);
2060 dz23 = _mm_sub_pd(iz2,jz3);
2061 dx31 = _mm_sub_pd(ix3,jx1);
2062 dy31 = _mm_sub_pd(iy3,jy1);
2063 dz31 = _mm_sub_pd(iz3,jz1);
2064 dx32 = _mm_sub_pd(ix3,jx2);
2065 dy32 = _mm_sub_pd(iy3,jy2);
2066 dz32 = _mm_sub_pd(iz3,jz2);
2067 dx33 = _mm_sub_pd(ix3,jx3);
2068 dy33 = _mm_sub_pd(iy3,jy3);
2069 dz33 = _mm_sub_pd(iz3,jz3);
2071 /* Calculate squared distance and things based on it */
2072 rsq00 = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
2073 rsq11 = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
2074 rsq12 = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
2075 rsq13 = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
2076 rsq21 = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
2077 rsq22 = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
2078 rsq23 = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
2079 rsq31 = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
2080 rsq32 = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
2081 rsq33 = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
2083 rinv00 = gmx_mm_invsqrt_pd(rsq00);
2084 rinv11 = gmx_mm_invsqrt_pd(rsq11);
2085 rinv12 = gmx_mm_invsqrt_pd(rsq12);
2086 rinv13 = gmx_mm_invsqrt_pd(rsq13);
2087 rinv21 = gmx_mm_invsqrt_pd(rsq21);
2088 rinv22 = gmx_mm_invsqrt_pd(rsq22);
2089 rinv23 = gmx_mm_invsqrt_pd(rsq23);
2090 rinv31 = gmx_mm_invsqrt_pd(rsq31);
2091 rinv32 = gmx_mm_invsqrt_pd(rsq32);
2092 rinv33 = gmx_mm_invsqrt_pd(rsq33);
2094 fjx0 = _mm_setzero_pd();
2095 fjy0 = _mm_setzero_pd();
2096 fjz0 = _mm_setzero_pd();
2097 fjx1 = _mm_setzero_pd();
2098 fjy1 = _mm_setzero_pd();
2099 fjz1 = _mm_setzero_pd();
2100 fjx2 = _mm_setzero_pd();
2101 fjy2 = _mm_setzero_pd();
2102 fjz2 = _mm_setzero_pd();
2103 fjx3 = _mm_setzero_pd();
2104 fjy3 = _mm_setzero_pd();
2105 fjz3 = _mm_setzero_pd();
2107 /**************************
2108 * CALCULATE INTERACTIONS *
2109 **************************/
2111 r00 = _mm_mul_pd(rsq00,rinv00);
2113 /* Calculate table index by multiplying r with table scale and truncate to integer */
2114 rt = _mm_mul_pd(r00,vftabscale);
2115 vfitab = _mm_cvttpd_epi32(rt);
2116 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2117 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2119 /* CUBIC SPLINE TABLE DISPERSION */
2120 vfitab = _mm_add_epi32(vfitab,ifour);
2121 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2122 F = _mm_setzero_pd();
2123 GMX_MM_TRANSPOSE2_PD(Y,F);
2124 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2125 H = _mm_setzero_pd();
2126 GMX_MM_TRANSPOSE2_PD(G,H);
2127 Heps = _mm_mul_pd(vfeps,H);
2128 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2129 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2130 fvdw6 = _mm_mul_pd(c6_00,FF);
2132 /* CUBIC SPLINE TABLE REPULSION */
2133 vfitab = _mm_add_epi32(vfitab,ifour);
2134 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2135 F = _mm_setzero_pd();
2136 GMX_MM_TRANSPOSE2_PD(Y,F);
2137 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2138 H = _mm_setzero_pd();
2139 GMX_MM_TRANSPOSE2_PD(G,H);
2140 Heps = _mm_mul_pd(vfeps,H);
2141 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2142 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2143 fvdw12 = _mm_mul_pd(c12_00,FF);
2144 fvdw = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
2148 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2150 /* Calculate temporary vectorial force */
2151 tx = _mm_mul_pd(fscal,dx00);
2152 ty = _mm_mul_pd(fscal,dy00);
2153 tz = _mm_mul_pd(fscal,dz00);
2155 /* Update vectorial force */
2156 fix0 = _mm_add_pd(fix0,tx);
2157 fiy0 = _mm_add_pd(fiy0,ty);
2158 fiz0 = _mm_add_pd(fiz0,tz);
2160 fjx0 = _mm_add_pd(fjx0,tx);
2161 fjy0 = _mm_add_pd(fjy0,ty);
2162 fjz0 = _mm_add_pd(fjz0,tz);
2164 /**************************
2165 * CALCULATE INTERACTIONS *
2166 **************************/
2168 r11 = _mm_mul_pd(rsq11,rinv11);
2170 /* Calculate table index by multiplying r with table scale and truncate to integer */
2171 rt = _mm_mul_pd(r11,vftabscale);
2172 vfitab = _mm_cvttpd_epi32(rt);
2173 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2174 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2176 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2177 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2178 F = _mm_setzero_pd();
2179 GMX_MM_TRANSPOSE2_PD(Y,F);
2180 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2181 H = _mm_setzero_pd();
2182 GMX_MM_TRANSPOSE2_PD(G,H);
2183 Heps = _mm_mul_pd(vfeps,H);
2184 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2185 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2186 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq11,FF),_mm_mul_pd(vftabscale,rinv11)));
2190 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2192 /* Calculate temporary vectorial force */
2193 tx = _mm_mul_pd(fscal,dx11);
2194 ty = _mm_mul_pd(fscal,dy11);
2195 tz = _mm_mul_pd(fscal,dz11);
2197 /* Update vectorial force */
2198 fix1 = _mm_add_pd(fix1,tx);
2199 fiy1 = _mm_add_pd(fiy1,ty);
2200 fiz1 = _mm_add_pd(fiz1,tz);
2202 fjx1 = _mm_add_pd(fjx1,tx);
2203 fjy1 = _mm_add_pd(fjy1,ty);
2204 fjz1 = _mm_add_pd(fjz1,tz);
2206 /**************************
2207 * CALCULATE INTERACTIONS *
2208 **************************/
2210 r12 = _mm_mul_pd(rsq12,rinv12);
2212 /* Calculate table index by multiplying r with table scale and truncate to integer */
2213 rt = _mm_mul_pd(r12,vftabscale);
2214 vfitab = _mm_cvttpd_epi32(rt);
2215 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2216 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2218 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2219 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2220 F = _mm_setzero_pd();
2221 GMX_MM_TRANSPOSE2_PD(Y,F);
2222 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2223 H = _mm_setzero_pd();
2224 GMX_MM_TRANSPOSE2_PD(G,H);
2225 Heps = _mm_mul_pd(vfeps,H);
2226 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2227 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2228 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq12,FF),_mm_mul_pd(vftabscale,rinv12)));
2232 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2234 /* Calculate temporary vectorial force */
2235 tx = _mm_mul_pd(fscal,dx12);
2236 ty = _mm_mul_pd(fscal,dy12);
2237 tz = _mm_mul_pd(fscal,dz12);
2239 /* Update vectorial force */
2240 fix1 = _mm_add_pd(fix1,tx);
2241 fiy1 = _mm_add_pd(fiy1,ty);
2242 fiz1 = _mm_add_pd(fiz1,tz);
2244 fjx2 = _mm_add_pd(fjx2,tx);
2245 fjy2 = _mm_add_pd(fjy2,ty);
2246 fjz2 = _mm_add_pd(fjz2,tz);
2248 /**************************
2249 * CALCULATE INTERACTIONS *
2250 **************************/
2252 r13 = _mm_mul_pd(rsq13,rinv13);
2254 /* Calculate table index by multiplying r with table scale and truncate to integer */
2255 rt = _mm_mul_pd(r13,vftabscale);
2256 vfitab = _mm_cvttpd_epi32(rt);
2257 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2258 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2260 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2261 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2262 F = _mm_setzero_pd();
2263 GMX_MM_TRANSPOSE2_PD(Y,F);
2264 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2265 H = _mm_setzero_pd();
2266 GMX_MM_TRANSPOSE2_PD(G,H);
2267 Heps = _mm_mul_pd(vfeps,H);
2268 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2269 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2270 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq13,FF),_mm_mul_pd(vftabscale,rinv13)));
2274 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2276 /* Calculate temporary vectorial force */
2277 tx = _mm_mul_pd(fscal,dx13);
2278 ty = _mm_mul_pd(fscal,dy13);
2279 tz = _mm_mul_pd(fscal,dz13);
2281 /* Update vectorial force */
2282 fix1 = _mm_add_pd(fix1,tx);
2283 fiy1 = _mm_add_pd(fiy1,ty);
2284 fiz1 = _mm_add_pd(fiz1,tz);
2286 fjx3 = _mm_add_pd(fjx3,tx);
2287 fjy3 = _mm_add_pd(fjy3,ty);
2288 fjz3 = _mm_add_pd(fjz3,tz);
2290 /**************************
2291 * CALCULATE INTERACTIONS *
2292 **************************/
2294 r21 = _mm_mul_pd(rsq21,rinv21);
2296 /* Calculate table index by multiplying r with table scale and truncate to integer */
2297 rt = _mm_mul_pd(r21,vftabscale);
2298 vfitab = _mm_cvttpd_epi32(rt);
2299 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2300 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2302 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2303 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2304 F = _mm_setzero_pd();
2305 GMX_MM_TRANSPOSE2_PD(Y,F);
2306 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2307 H = _mm_setzero_pd();
2308 GMX_MM_TRANSPOSE2_PD(G,H);
2309 Heps = _mm_mul_pd(vfeps,H);
2310 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2311 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2312 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq21,FF),_mm_mul_pd(vftabscale,rinv21)));
2316 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2318 /* Calculate temporary vectorial force */
2319 tx = _mm_mul_pd(fscal,dx21);
2320 ty = _mm_mul_pd(fscal,dy21);
2321 tz = _mm_mul_pd(fscal,dz21);
2323 /* Update vectorial force */
2324 fix2 = _mm_add_pd(fix2,tx);
2325 fiy2 = _mm_add_pd(fiy2,ty);
2326 fiz2 = _mm_add_pd(fiz2,tz);
2328 fjx1 = _mm_add_pd(fjx1,tx);
2329 fjy1 = _mm_add_pd(fjy1,ty);
2330 fjz1 = _mm_add_pd(fjz1,tz);
2332 /**************************
2333 * CALCULATE INTERACTIONS *
2334 **************************/
2336 r22 = _mm_mul_pd(rsq22,rinv22);
2338 /* Calculate table index by multiplying r with table scale and truncate to integer */
2339 rt = _mm_mul_pd(r22,vftabscale);
2340 vfitab = _mm_cvttpd_epi32(rt);
2341 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2342 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2344 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2345 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2346 F = _mm_setzero_pd();
2347 GMX_MM_TRANSPOSE2_PD(Y,F);
2348 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2349 H = _mm_setzero_pd();
2350 GMX_MM_TRANSPOSE2_PD(G,H);
2351 Heps = _mm_mul_pd(vfeps,H);
2352 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2353 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2354 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq22,FF),_mm_mul_pd(vftabscale,rinv22)));
2358 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2360 /* Calculate temporary vectorial force */
2361 tx = _mm_mul_pd(fscal,dx22);
2362 ty = _mm_mul_pd(fscal,dy22);
2363 tz = _mm_mul_pd(fscal,dz22);
2365 /* Update vectorial force */
2366 fix2 = _mm_add_pd(fix2,tx);
2367 fiy2 = _mm_add_pd(fiy2,ty);
2368 fiz2 = _mm_add_pd(fiz2,tz);
2370 fjx2 = _mm_add_pd(fjx2,tx);
2371 fjy2 = _mm_add_pd(fjy2,ty);
2372 fjz2 = _mm_add_pd(fjz2,tz);
2374 /**************************
2375 * CALCULATE INTERACTIONS *
2376 **************************/
2378 r23 = _mm_mul_pd(rsq23,rinv23);
2380 /* Calculate table index by multiplying r with table scale and truncate to integer */
2381 rt = _mm_mul_pd(r23,vftabscale);
2382 vfitab = _mm_cvttpd_epi32(rt);
2383 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2384 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2386 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2387 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2388 F = _mm_setzero_pd();
2389 GMX_MM_TRANSPOSE2_PD(Y,F);
2390 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2391 H = _mm_setzero_pd();
2392 GMX_MM_TRANSPOSE2_PD(G,H);
2393 Heps = _mm_mul_pd(vfeps,H);
2394 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2395 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2396 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq23,FF),_mm_mul_pd(vftabscale,rinv23)));
2400 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2402 /* Calculate temporary vectorial force */
2403 tx = _mm_mul_pd(fscal,dx23);
2404 ty = _mm_mul_pd(fscal,dy23);
2405 tz = _mm_mul_pd(fscal,dz23);
2407 /* Update vectorial force */
2408 fix2 = _mm_add_pd(fix2,tx);
2409 fiy2 = _mm_add_pd(fiy2,ty);
2410 fiz2 = _mm_add_pd(fiz2,tz);
2412 fjx3 = _mm_add_pd(fjx3,tx);
2413 fjy3 = _mm_add_pd(fjy3,ty);
2414 fjz3 = _mm_add_pd(fjz3,tz);
2416 /**************************
2417 * CALCULATE INTERACTIONS *
2418 **************************/
2420 r31 = _mm_mul_pd(rsq31,rinv31);
2422 /* Calculate table index by multiplying r with table scale and truncate to integer */
2423 rt = _mm_mul_pd(r31,vftabscale);
2424 vfitab = _mm_cvttpd_epi32(rt);
2425 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2426 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2428 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2429 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2430 F = _mm_setzero_pd();
2431 GMX_MM_TRANSPOSE2_PD(Y,F);
2432 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2433 H = _mm_setzero_pd();
2434 GMX_MM_TRANSPOSE2_PD(G,H);
2435 Heps = _mm_mul_pd(vfeps,H);
2436 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2437 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2438 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq31,FF),_mm_mul_pd(vftabscale,rinv31)));
2442 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2444 /* Calculate temporary vectorial force */
2445 tx = _mm_mul_pd(fscal,dx31);
2446 ty = _mm_mul_pd(fscal,dy31);
2447 tz = _mm_mul_pd(fscal,dz31);
2449 /* Update vectorial force */
2450 fix3 = _mm_add_pd(fix3,tx);
2451 fiy3 = _mm_add_pd(fiy3,ty);
2452 fiz3 = _mm_add_pd(fiz3,tz);
2454 fjx1 = _mm_add_pd(fjx1,tx);
2455 fjy1 = _mm_add_pd(fjy1,ty);
2456 fjz1 = _mm_add_pd(fjz1,tz);
2458 /**************************
2459 * CALCULATE INTERACTIONS *
2460 **************************/
2462 r32 = _mm_mul_pd(rsq32,rinv32);
2464 /* Calculate table index by multiplying r with table scale and truncate to integer */
2465 rt = _mm_mul_pd(r32,vftabscale);
2466 vfitab = _mm_cvttpd_epi32(rt);
2467 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2468 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2470 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2471 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2472 F = _mm_setzero_pd();
2473 GMX_MM_TRANSPOSE2_PD(Y,F);
2474 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2475 H = _mm_setzero_pd();
2476 GMX_MM_TRANSPOSE2_PD(G,H);
2477 Heps = _mm_mul_pd(vfeps,H);
2478 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2479 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2480 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq32,FF),_mm_mul_pd(vftabscale,rinv32)));
2484 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2486 /* Calculate temporary vectorial force */
2487 tx = _mm_mul_pd(fscal,dx32);
2488 ty = _mm_mul_pd(fscal,dy32);
2489 tz = _mm_mul_pd(fscal,dz32);
2491 /* Update vectorial force */
2492 fix3 = _mm_add_pd(fix3,tx);
2493 fiy3 = _mm_add_pd(fiy3,ty);
2494 fiz3 = _mm_add_pd(fiz3,tz);
2496 fjx2 = _mm_add_pd(fjx2,tx);
2497 fjy2 = _mm_add_pd(fjy2,ty);
2498 fjz2 = _mm_add_pd(fjz2,tz);
2500 /**************************
2501 * CALCULATE INTERACTIONS *
2502 **************************/
2504 r33 = _mm_mul_pd(rsq33,rinv33);
2506 /* Calculate table index by multiplying r with table scale and truncate to integer */
2507 rt = _mm_mul_pd(r33,vftabscale);
2508 vfitab = _mm_cvttpd_epi32(rt);
2509 vfeps = _mm_sub_pd(rt,_mm_cvtepi32_pd(vfitab));
2510 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
2512 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2513 Y = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
2514 F = _mm_setzero_pd();
2515 GMX_MM_TRANSPOSE2_PD(Y,F);
2516 G = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
2517 H = _mm_setzero_pd();
2518 GMX_MM_TRANSPOSE2_PD(G,H);
2519 Heps = _mm_mul_pd(vfeps,H);
2520 Fp = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
2521 FF = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
2522 felec = _mm_xor_pd(signbit,_mm_mul_pd(_mm_mul_pd(qq33,FF),_mm_mul_pd(vftabscale,rinv33)));
2526 fscal = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2528 /* Calculate temporary vectorial force */
2529 tx = _mm_mul_pd(fscal,dx33);
2530 ty = _mm_mul_pd(fscal,dy33);
2531 tz = _mm_mul_pd(fscal,dz33);
2533 /* Update vectorial force */
2534 fix3 = _mm_add_pd(fix3,tx);
2535 fiy3 = _mm_add_pd(fiy3,ty);
2536 fiz3 = _mm_add_pd(fiz3,tz);
2538 fjx3 = _mm_add_pd(fjx3,tx);
2539 fjy3 = _mm_add_pd(fjy3,ty);
2540 fjz3 = _mm_add_pd(fjz3,tz);
2542 gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2544 /* Inner loop uses 402 flops */
2547 /* End of innermost loop */
2549 gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2550 f+i_coord_offset,fshift+i_shift_offset);
2552 /* Increment number of inner iterations */
2553 inneriter += j_index_end - j_index_start;
2555 /* Outer loop uses 24 flops */
2558 /* Increment number of outer iterations */
2561 /* Update outer/inner flops */
2563 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*402);