2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2012,2013,2014,2015,2017, 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_single kernel generator.
44 #include "../nb_kernel.h"
45 #include "gromacs/gmxlib/nrnb.h"
47 #include "kernelutil_x86_sse2_single.h"
50 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW4W4_VF_sse2_single
51 * Electrostatics interaction: CubicSplineTable
52 * VdW interaction: None
53 * Geometry: Water4-Water4
54 * Calculate force/pot: PotentialAndForce
57 nb_kernel_ElecCSTab_VdwNone_GeomW4W4_VF_sse2_single
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,C,D refer to j loop unrolling done with SSE, e.g. for the four 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;
73 int jnrA,jnrB,jnrC,jnrD;
74 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
75 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
76 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
78 real *shiftvec,*fshift,*x,*f;
79 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
81 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
83 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
85 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
87 __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
88 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
89 __m128 jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
90 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
91 __m128 jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
92 int vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
93 __m128 jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
94 __m128 dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
95 __m128 dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
96 __m128 dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
97 __m128 dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
98 __m128 dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
99 __m128 dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
100 __m128 dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
101 __m128 dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
102 __m128 dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
103 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
106 __m128i ifour = _mm_set1_epi32(4);
107 __m128 rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
109 __m128 dummy_mask,cutoff_mask;
110 __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
111 __m128 one = _mm_set1_ps(1.0);
112 __m128 two = _mm_set1_ps(2.0);
118 jindex = nlist->jindex;
120 shiftidx = nlist->shift;
122 shiftvec = fr->shift_vec[0];
123 fshift = fr->fshift[0];
124 facel = _mm_set1_ps(fr->ic->epsfac);
125 charge = mdatoms->chargeA;
127 vftab = kernel_data->table_elec->data;
128 vftabscale = _mm_set1_ps(kernel_data->table_elec->scale);
130 /* Setup water-specific parameters */
131 inr = nlist->iinr[0];
132 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
133 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
134 iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
136 jq1 = _mm_set1_ps(charge[inr+1]);
137 jq2 = _mm_set1_ps(charge[inr+2]);
138 jq3 = _mm_set1_ps(charge[inr+3]);
139 qq11 = _mm_mul_ps(iq1,jq1);
140 qq12 = _mm_mul_ps(iq1,jq2);
141 qq13 = _mm_mul_ps(iq1,jq3);
142 qq21 = _mm_mul_ps(iq2,jq1);
143 qq22 = _mm_mul_ps(iq2,jq2);
144 qq23 = _mm_mul_ps(iq2,jq3);
145 qq31 = _mm_mul_ps(iq3,jq1);
146 qq32 = _mm_mul_ps(iq3,jq2);
147 qq33 = _mm_mul_ps(iq3,jq3);
149 /* Avoid stupid compiler warnings */
150 jnrA = jnrB = jnrC = jnrD = 0;
159 for(iidx=0;iidx<4*DIM;iidx++)
164 /* Start outer loop over neighborlists */
165 for(iidx=0; iidx<nri; iidx++)
167 /* Load shift vector for this list */
168 i_shift_offset = DIM*shiftidx[iidx];
170 /* Load limits for loop over neighbors */
171 j_index_start = jindex[iidx];
172 j_index_end = jindex[iidx+1];
174 /* Get outer coordinate index */
176 i_coord_offset = DIM*inr;
178 /* Load i particle coords and add shift vector */
179 gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
180 &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
182 fix1 = _mm_setzero_ps();
183 fiy1 = _mm_setzero_ps();
184 fiz1 = _mm_setzero_ps();
185 fix2 = _mm_setzero_ps();
186 fiy2 = _mm_setzero_ps();
187 fiz2 = _mm_setzero_ps();
188 fix3 = _mm_setzero_ps();
189 fiy3 = _mm_setzero_ps();
190 fiz3 = _mm_setzero_ps();
192 /* Reset potential sums */
193 velecsum = _mm_setzero_ps();
195 /* Start inner kernel loop */
196 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
199 /* Get j neighbor index, and coordinate index */
204 j_coord_offsetA = DIM*jnrA;
205 j_coord_offsetB = DIM*jnrB;
206 j_coord_offsetC = DIM*jnrC;
207 j_coord_offsetD = DIM*jnrD;
209 /* load j atom coordinates */
210 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
211 x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
212 &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
214 /* Calculate displacement vector */
215 dx11 = _mm_sub_ps(ix1,jx1);
216 dy11 = _mm_sub_ps(iy1,jy1);
217 dz11 = _mm_sub_ps(iz1,jz1);
218 dx12 = _mm_sub_ps(ix1,jx2);
219 dy12 = _mm_sub_ps(iy1,jy2);
220 dz12 = _mm_sub_ps(iz1,jz2);
221 dx13 = _mm_sub_ps(ix1,jx3);
222 dy13 = _mm_sub_ps(iy1,jy3);
223 dz13 = _mm_sub_ps(iz1,jz3);
224 dx21 = _mm_sub_ps(ix2,jx1);
225 dy21 = _mm_sub_ps(iy2,jy1);
226 dz21 = _mm_sub_ps(iz2,jz1);
227 dx22 = _mm_sub_ps(ix2,jx2);
228 dy22 = _mm_sub_ps(iy2,jy2);
229 dz22 = _mm_sub_ps(iz2,jz2);
230 dx23 = _mm_sub_ps(ix2,jx3);
231 dy23 = _mm_sub_ps(iy2,jy3);
232 dz23 = _mm_sub_ps(iz2,jz3);
233 dx31 = _mm_sub_ps(ix3,jx1);
234 dy31 = _mm_sub_ps(iy3,jy1);
235 dz31 = _mm_sub_ps(iz3,jz1);
236 dx32 = _mm_sub_ps(ix3,jx2);
237 dy32 = _mm_sub_ps(iy3,jy2);
238 dz32 = _mm_sub_ps(iz3,jz2);
239 dx33 = _mm_sub_ps(ix3,jx3);
240 dy33 = _mm_sub_ps(iy3,jy3);
241 dz33 = _mm_sub_ps(iz3,jz3);
243 /* Calculate squared distance and things based on it */
244 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
245 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
246 rsq13 = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
247 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
248 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
249 rsq23 = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
250 rsq31 = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
251 rsq32 = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
252 rsq33 = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
254 rinv11 = sse2_invsqrt_f(rsq11);
255 rinv12 = sse2_invsqrt_f(rsq12);
256 rinv13 = sse2_invsqrt_f(rsq13);
257 rinv21 = sse2_invsqrt_f(rsq21);
258 rinv22 = sse2_invsqrt_f(rsq22);
259 rinv23 = sse2_invsqrt_f(rsq23);
260 rinv31 = sse2_invsqrt_f(rsq31);
261 rinv32 = sse2_invsqrt_f(rsq32);
262 rinv33 = sse2_invsqrt_f(rsq33);
264 fjx1 = _mm_setzero_ps();
265 fjy1 = _mm_setzero_ps();
266 fjz1 = _mm_setzero_ps();
267 fjx2 = _mm_setzero_ps();
268 fjy2 = _mm_setzero_ps();
269 fjz2 = _mm_setzero_ps();
270 fjx3 = _mm_setzero_ps();
271 fjy3 = _mm_setzero_ps();
272 fjz3 = _mm_setzero_ps();
274 /**************************
275 * CALCULATE INTERACTIONS *
276 **************************/
278 r11 = _mm_mul_ps(rsq11,rinv11);
280 /* Calculate table index by multiplying r with table scale and truncate to integer */
281 rt = _mm_mul_ps(r11,vftabscale);
282 vfitab = _mm_cvttps_epi32(rt);
283 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
284 vfitab = _mm_slli_epi32(vfitab,2);
286 /* CUBIC SPLINE TABLE ELECTROSTATICS */
287 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
288 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
289 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
290 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
291 _MM_TRANSPOSE4_PS(Y,F,G,H);
292 Heps = _mm_mul_ps(vfeps,H);
293 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
294 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
295 velec = _mm_mul_ps(qq11,VV);
296 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
297 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
299 /* Update potential sum for this i atom from the interaction with this j atom. */
300 velecsum = _mm_add_ps(velecsum,velec);
304 /* Calculate temporary vectorial force */
305 tx = _mm_mul_ps(fscal,dx11);
306 ty = _mm_mul_ps(fscal,dy11);
307 tz = _mm_mul_ps(fscal,dz11);
309 /* Update vectorial force */
310 fix1 = _mm_add_ps(fix1,tx);
311 fiy1 = _mm_add_ps(fiy1,ty);
312 fiz1 = _mm_add_ps(fiz1,tz);
314 fjx1 = _mm_add_ps(fjx1,tx);
315 fjy1 = _mm_add_ps(fjy1,ty);
316 fjz1 = _mm_add_ps(fjz1,tz);
318 /**************************
319 * CALCULATE INTERACTIONS *
320 **************************/
322 r12 = _mm_mul_ps(rsq12,rinv12);
324 /* Calculate table index by multiplying r with table scale and truncate to integer */
325 rt = _mm_mul_ps(r12,vftabscale);
326 vfitab = _mm_cvttps_epi32(rt);
327 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
328 vfitab = _mm_slli_epi32(vfitab,2);
330 /* CUBIC SPLINE TABLE ELECTROSTATICS */
331 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
332 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
333 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
334 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
335 _MM_TRANSPOSE4_PS(Y,F,G,H);
336 Heps = _mm_mul_ps(vfeps,H);
337 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
338 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
339 velec = _mm_mul_ps(qq12,VV);
340 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
341 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
343 /* Update potential sum for this i atom from the interaction with this j atom. */
344 velecsum = _mm_add_ps(velecsum,velec);
348 /* Calculate temporary vectorial force */
349 tx = _mm_mul_ps(fscal,dx12);
350 ty = _mm_mul_ps(fscal,dy12);
351 tz = _mm_mul_ps(fscal,dz12);
353 /* Update vectorial force */
354 fix1 = _mm_add_ps(fix1,tx);
355 fiy1 = _mm_add_ps(fiy1,ty);
356 fiz1 = _mm_add_ps(fiz1,tz);
358 fjx2 = _mm_add_ps(fjx2,tx);
359 fjy2 = _mm_add_ps(fjy2,ty);
360 fjz2 = _mm_add_ps(fjz2,tz);
362 /**************************
363 * CALCULATE INTERACTIONS *
364 **************************/
366 r13 = _mm_mul_ps(rsq13,rinv13);
368 /* Calculate table index by multiplying r with table scale and truncate to integer */
369 rt = _mm_mul_ps(r13,vftabscale);
370 vfitab = _mm_cvttps_epi32(rt);
371 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
372 vfitab = _mm_slli_epi32(vfitab,2);
374 /* CUBIC SPLINE TABLE ELECTROSTATICS */
375 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
376 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
377 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
378 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
379 _MM_TRANSPOSE4_PS(Y,F,G,H);
380 Heps = _mm_mul_ps(vfeps,H);
381 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
382 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
383 velec = _mm_mul_ps(qq13,VV);
384 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
385 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
387 /* Update potential sum for this i atom from the interaction with this j atom. */
388 velecsum = _mm_add_ps(velecsum,velec);
392 /* Calculate temporary vectorial force */
393 tx = _mm_mul_ps(fscal,dx13);
394 ty = _mm_mul_ps(fscal,dy13);
395 tz = _mm_mul_ps(fscal,dz13);
397 /* Update vectorial force */
398 fix1 = _mm_add_ps(fix1,tx);
399 fiy1 = _mm_add_ps(fiy1,ty);
400 fiz1 = _mm_add_ps(fiz1,tz);
402 fjx3 = _mm_add_ps(fjx3,tx);
403 fjy3 = _mm_add_ps(fjy3,ty);
404 fjz3 = _mm_add_ps(fjz3,tz);
406 /**************************
407 * CALCULATE INTERACTIONS *
408 **************************/
410 r21 = _mm_mul_ps(rsq21,rinv21);
412 /* Calculate table index by multiplying r with table scale and truncate to integer */
413 rt = _mm_mul_ps(r21,vftabscale);
414 vfitab = _mm_cvttps_epi32(rt);
415 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
416 vfitab = _mm_slli_epi32(vfitab,2);
418 /* CUBIC SPLINE TABLE ELECTROSTATICS */
419 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
420 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
421 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
422 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
423 _MM_TRANSPOSE4_PS(Y,F,G,H);
424 Heps = _mm_mul_ps(vfeps,H);
425 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
426 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
427 velec = _mm_mul_ps(qq21,VV);
428 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
429 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
431 /* Update potential sum for this i atom from the interaction with this j atom. */
432 velecsum = _mm_add_ps(velecsum,velec);
436 /* Calculate temporary vectorial force */
437 tx = _mm_mul_ps(fscal,dx21);
438 ty = _mm_mul_ps(fscal,dy21);
439 tz = _mm_mul_ps(fscal,dz21);
441 /* Update vectorial force */
442 fix2 = _mm_add_ps(fix2,tx);
443 fiy2 = _mm_add_ps(fiy2,ty);
444 fiz2 = _mm_add_ps(fiz2,tz);
446 fjx1 = _mm_add_ps(fjx1,tx);
447 fjy1 = _mm_add_ps(fjy1,ty);
448 fjz1 = _mm_add_ps(fjz1,tz);
450 /**************************
451 * CALCULATE INTERACTIONS *
452 **************************/
454 r22 = _mm_mul_ps(rsq22,rinv22);
456 /* Calculate table index by multiplying r with table scale and truncate to integer */
457 rt = _mm_mul_ps(r22,vftabscale);
458 vfitab = _mm_cvttps_epi32(rt);
459 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
460 vfitab = _mm_slli_epi32(vfitab,2);
462 /* CUBIC SPLINE TABLE ELECTROSTATICS */
463 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
464 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
465 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
466 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
467 _MM_TRANSPOSE4_PS(Y,F,G,H);
468 Heps = _mm_mul_ps(vfeps,H);
469 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
470 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
471 velec = _mm_mul_ps(qq22,VV);
472 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
473 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
475 /* Update potential sum for this i atom from the interaction with this j atom. */
476 velecsum = _mm_add_ps(velecsum,velec);
480 /* Calculate temporary vectorial force */
481 tx = _mm_mul_ps(fscal,dx22);
482 ty = _mm_mul_ps(fscal,dy22);
483 tz = _mm_mul_ps(fscal,dz22);
485 /* Update vectorial force */
486 fix2 = _mm_add_ps(fix2,tx);
487 fiy2 = _mm_add_ps(fiy2,ty);
488 fiz2 = _mm_add_ps(fiz2,tz);
490 fjx2 = _mm_add_ps(fjx2,tx);
491 fjy2 = _mm_add_ps(fjy2,ty);
492 fjz2 = _mm_add_ps(fjz2,tz);
494 /**************************
495 * CALCULATE INTERACTIONS *
496 **************************/
498 r23 = _mm_mul_ps(rsq23,rinv23);
500 /* Calculate table index by multiplying r with table scale and truncate to integer */
501 rt = _mm_mul_ps(r23,vftabscale);
502 vfitab = _mm_cvttps_epi32(rt);
503 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
504 vfitab = _mm_slli_epi32(vfitab,2);
506 /* CUBIC SPLINE TABLE ELECTROSTATICS */
507 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
508 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
509 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
510 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
511 _MM_TRANSPOSE4_PS(Y,F,G,H);
512 Heps = _mm_mul_ps(vfeps,H);
513 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
514 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
515 velec = _mm_mul_ps(qq23,VV);
516 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
517 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
519 /* Update potential sum for this i atom from the interaction with this j atom. */
520 velecsum = _mm_add_ps(velecsum,velec);
524 /* Calculate temporary vectorial force */
525 tx = _mm_mul_ps(fscal,dx23);
526 ty = _mm_mul_ps(fscal,dy23);
527 tz = _mm_mul_ps(fscal,dz23);
529 /* Update vectorial force */
530 fix2 = _mm_add_ps(fix2,tx);
531 fiy2 = _mm_add_ps(fiy2,ty);
532 fiz2 = _mm_add_ps(fiz2,tz);
534 fjx3 = _mm_add_ps(fjx3,tx);
535 fjy3 = _mm_add_ps(fjy3,ty);
536 fjz3 = _mm_add_ps(fjz3,tz);
538 /**************************
539 * CALCULATE INTERACTIONS *
540 **************************/
542 r31 = _mm_mul_ps(rsq31,rinv31);
544 /* Calculate table index by multiplying r with table scale and truncate to integer */
545 rt = _mm_mul_ps(r31,vftabscale);
546 vfitab = _mm_cvttps_epi32(rt);
547 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
548 vfitab = _mm_slli_epi32(vfitab,2);
550 /* CUBIC SPLINE TABLE ELECTROSTATICS */
551 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
552 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
553 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
554 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
555 _MM_TRANSPOSE4_PS(Y,F,G,H);
556 Heps = _mm_mul_ps(vfeps,H);
557 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
558 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
559 velec = _mm_mul_ps(qq31,VV);
560 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
561 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
563 /* Update potential sum for this i atom from the interaction with this j atom. */
564 velecsum = _mm_add_ps(velecsum,velec);
568 /* Calculate temporary vectorial force */
569 tx = _mm_mul_ps(fscal,dx31);
570 ty = _mm_mul_ps(fscal,dy31);
571 tz = _mm_mul_ps(fscal,dz31);
573 /* Update vectorial force */
574 fix3 = _mm_add_ps(fix3,tx);
575 fiy3 = _mm_add_ps(fiy3,ty);
576 fiz3 = _mm_add_ps(fiz3,tz);
578 fjx1 = _mm_add_ps(fjx1,tx);
579 fjy1 = _mm_add_ps(fjy1,ty);
580 fjz1 = _mm_add_ps(fjz1,tz);
582 /**************************
583 * CALCULATE INTERACTIONS *
584 **************************/
586 r32 = _mm_mul_ps(rsq32,rinv32);
588 /* Calculate table index by multiplying r with table scale and truncate to integer */
589 rt = _mm_mul_ps(r32,vftabscale);
590 vfitab = _mm_cvttps_epi32(rt);
591 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
592 vfitab = _mm_slli_epi32(vfitab,2);
594 /* CUBIC SPLINE TABLE ELECTROSTATICS */
595 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
596 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
597 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
598 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
599 _MM_TRANSPOSE4_PS(Y,F,G,H);
600 Heps = _mm_mul_ps(vfeps,H);
601 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
602 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
603 velec = _mm_mul_ps(qq32,VV);
604 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
605 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
607 /* Update potential sum for this i atom from the interaction with this j atom. */
608 velecsum = _mm_add_ps(velecsum,velec);
612 /* Calculate temporary vectorial force */
613 tx = _mm_mul_ps(fscal,dx32);
614 ty = _mm_mul_ps(fscal,dy32);
615 tz = _mm_mul_ps(fscal,dz32);
617 /* Update vectorial force */
618 fix3 = _mm_add_ps(fix3,tx);
619 fiy3 = _mm_add_ps(fiy3,ty);
620 fiz3 = _mm_add_ps(fiz3,tz);
622 fjx2 = _mm_add_ps(fjx2,tx);
623 fjy2 = _mm_add_ps(fjy2,ty);
624 fjz2 = _mm_add_ps(fjz2,tz);
626 /**************************
627 * CALCULATE INTERACTIONS *
628 **************************/
630 r33 = _mm_mul_ps(rsq33,rinv33);
632 /* Calculate table index by multiplying r with table scale and truncate to integer */
633 rt = _mm_mul_ps(r33,vftabscale);
634 vfitab = _mm_cvttps_epi32(rt);
635 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
636 vfitab = _mm_slli_epi32(vfitab,2);
638 /* CUBIC SPLINE TABLE ELECTROSTATICS */
639 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
640 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
641 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
642 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
643 _MM_TRANSPOSE4_PS(Y,F,G,H);
644 Heps = _mm_mul_ps(vfeps,H);
645 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
646 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
647 velec = _mm_mul_ps(qq33,VV);
648 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
649 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
651 /* Update potential sum for this i atom from the interaction with this j atom. */
652 velecsum = _mm_add_ps(velecsum,velec);
656 /* Calculate temporary vectorial force */
657 tx = _mm_mul_ps(fscal,dx33);
658 ty = _mm_mul_ps(fscal,dy33);
659 tz = _mm_mul_ps(fscal,dz33);
661 /* Update vectorial force */
662 fix3 = _mm_add_ps(fix3,tx);
663 fiy3 = _mm_add_ps(fiy3,ty);
664 fiz3 = _mm_add_ps(fiz3,tz);
666 fjx3 = _mm_add_ps(fjx3,tx);
667 fjy3 = _mm_add_ps(fjy3,ty);
668 fjz3 = _mm_add_ps(fjz3,tz);
670 fjptrA = f+j_coord_offsetA;
671 fjptrB = f+j_coord_offsetB;
672 fjptrC = f+j_coord_offsetC;
673 fjptrD = f+j_coord_offsetD;
675 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
676 fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
678 /* Inner loop uses 387 flops */
684 /* Get j neighbor index, and coordinate index */
685 jnrlistA = jjnr[jidx];
686 jnrlistB = jjnr[jidx+1];
687 jnrlistC = jjnr[jidx+2];
688 jnrlistD = jjnr[jidx+3];
689 /* Sign of each element will be negative for non-real atoms.
690 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
691 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
693 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
694 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
695 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
696 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
697 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
698 j_coord_offsetA = DIM*jnrA;
699 j_coord_offsetB = DIM*jnrB;
700 j_coord_offsetC = DIM*jnrC;
701 j_coord_offsetD = DIM*jnrD;
703 /* load j atom coordinates */
704 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
705 x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
706 &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
708 /* Calculate displacement vector */
709 dx11 = _mm_sub_ps(ix1,jx1);
710 dy11 = _mm_sub_ps(iy1,jy1);
711 dz11 = _mm_sub_ps(iz1,jz1);
712 dx12 = _mm_sub_ps(ix1,jx2);
713 dy12 = _mm_sub_ps(iy1,jy2);
714 dz12 = _mm_sub_ps(iz1,jz2);
715 dx13 = _mm_sub_ps(ix1,jx3);
716 dy13 = _mm_sub_ps(iy1,jy3);
717 dz13 = _mm_sub_ps(iz1,jz3);
718 dx21 = _mm_sub_ps(ix2,jx1);
719 dy21 = _mm_sub_ps(iy2,jy1);
720 dz21 = _mm_sub_ps(iz2,jz1);
721 dx22 = _mm_sub_ps(ix2,jx2);
722 dy22 = _mm_sub_ps(iy2,jy2);
723 dz22 = _mm_sub_ps(iz2,jz2);
724 dx23 = _mm_sub_ps(ix2,jx3);
725 dy23 = _mm_sub_ps(iy2,jy3);
726 dz23 = _mm_sub_ps(iz2,jz3);
727 dx31 = _mm_sub_ps(ix3,jx1);
728 dy31 = _mm_sub_ps(iy3,jy1);
729 dz31 = _mm_sub_ps(iz3,jz1);
730 dx32 = _mm_sub_ps(ix3,jx2);
731 dy32 = _mm_sub_ps(iy3,jy2);
732 dz32 = _mm_sub_ps(iz3,jz2);
733 dx33 = _mm_sub_ps(ix3,jx3);
734 dy33 = _mm_sub_ps(iy3,jy3);
735 dz33 = _mm_sub_ps(iz3,jz3);
737 /* Calculate squared distance and things based on it */
738 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
739 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
740 rsq13 = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
741 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
742 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
743 rsq23 = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
744 rsq31 = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
745 rsq32 = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
746 rsq33 = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
748 rinv11 = sse2_invsqrt_f(rsq11);
749 rinv12 = sse2_invsqrt_f(rsq12);
750 rinv13 = sse2_invsqrt_f(rsq13);
751 rinv21 = sse2_invsqrt_f(rsq21);
752 rinv22 = sse2_invsqrt_f(rsq22);
753 rinv23 = sse2_invsqrt_f(rsq23);
754 rinv31 = sse2_invsqrt_f(rsq31);
755 rinv32 = sse2_invsqrt_f(rsq32);
756 rinv33 = sse2_invsqrt_f(rsq33);
758 fjx1 = _mm_setzero_ps();
759 fjy1 = _mm_setzero_ps();
760 fjz1 = _mm_setzero_ps();
761 fjx2 = _mm_setzero_ps();
762 fjy2 = _mm_setzero_ps();
763 fjz2 = _mm_setzero_ps();
764 fjx3 = _mm_setzero_ps();
765 fjy3 = _mm_setzero_ps();
766 fjz3 = _mm_setzero_ps();
768 /**************************
769 * CALCULATE INTERACTIONS *
770 **************************/
772 r11 = _mm_mul_ps(rsq11,rinv11);
773 r11 = _mm_andnot_ps(dummy_mask,r11);
775 /* Calculate table index by multiplying r with table scale and truncate to integer */
776 rt = _mm_mul_ps(r11,vftabscale);
777 vfitab = _mm_cvttps_epi32(rt);
778 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
779 vfitab = _mm_slli_epi32(vfitab,2);
781 /* CUBIC SPLINE TABLE ELECTROSTATICS */
782 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
783 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
784 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
785 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
786 _MM_TRANSPOSE4_PS(Y,F,G,H);
787 Heps = _mm_mul_ps(vfeps,H);
788 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
789 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
790 velec = _mm_mul_ps(qq11,VV);
791 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
792 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
794 /* Update potential sum for this i atom from the interaction with this j atom. */
795 velec = _mm_andnot_ps(dummy_mask,velec);
796 velecsum = _mm_add_ps(velecsum,velec);
800 fscal = _mm_andnot_ps(dummy_mask,fscal);
802 /* Calculate temporary vectorial force */
803 tx = _mm_mul_ps(fscal,dx11);
804 ty = _mm_mul_ps(fscal,dy11);
805 tz = _mm_mul_ps(fscal,dz11);
807 /* Update vectorial force */
808 fix1 = _mm_add_ps(fix1,tx);
809 fiy1 = _mm_add_ps(fiy1,ty);
810 fiz1 = _mm_add_ps(fiz1,tz);
812 fjx1 = _mm_add_ps(fjx1,tx);
813 fjy1 = _mm_add_ps(fjy1,ty);
814 fjz1 = _mm_add_ps(fjz1,tz);
816 /**************************
817 * CALCULATE INTERACTIONS *
818 **************************/
820 r12 = _mm_mul_ps(rsq12,rinv12);
821 r12 = _mm_andnot_ps(dummy_mask,r12);
823 /* Calculate table index by multiplying r with table scale and truncate to integer */
824 rt = _mm_mul_ps(r12,vftabscale);
825 vfitab = _mm_cvttps_epi32(rt);
826 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
827 vfitab = _mm_slli_epi32(vfitab,2);
829 /* CUBIC SPLINE TABLE ELECTROSTATICS */
830 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
831 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
832 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
833 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
834 _MM_TRANSPOSE4_PS(Y,F,G,H);
835 Heps = _mm_mul_ps(vfeps,H);
836 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
837 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
838 velec = _mm_mul_ps(qq12,VV);
839 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
840 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
842 /* Update potential sum for this i atom from the interaction with this j atom. */
843 velec = _mm_andnot_ps(dummy_mask,velec);
844 velecsum = _mm_add_ps(velecsum,velec);
848 fscal = _mm_andnot_ps(dummy_mask,fscal);
850 /* Calculate temporary vectorial force */
851 tx = _mm_mul_ps(fscal,dx12);
852 ty = _mm_mul_ps(fscal,dy12);
853 tz = _mm_mul_ps(fscal,dz12);
855 /* Update vectorial force */
856 fix1 = _mm_add_ps(fix1,tx);
857 fiy1 = _mm_add_ps(fiy1,ty);
858 fiz1 = _mm_add_ps(fiz1,tz);
860 fjx2 = _mm_add_ps(fjx2,tx);
861 fjy2 = _mm_add_ps(fjy2,ty);
862 fjz2 = _mm_add_ps(fjz2,tz);
864 /**************************
865 * CALCULATE INTERACTIONS *
866 **************************/
868 r13 = _mm_mul_ps(rsq13,rinv13);
869 r13 = _mm_andnot_ps(dummy_mask,r13);
871 /* Calculate table index by multiplying r with table scale and truncate to integer */
872 rt = _mm_mul_ps(r13,vftabscale);
873 vfitab = _mm_cvttps_epi32(rt);
874 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
875 vfitab = _mm_slli_epi32(vfitab,2);
877 /* CUBIC SPLINE TABLE ELECTROSTATICS */
878 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
879 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
880 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
881 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
882 _MM_TRANSPOSE4_PS(Y,F,G,H);
883 Heps = _mm_mul_ps(vfeps,H);
884 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
885 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
886 velec = _mm_mul_ps(qq13,VV);
887 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
888 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
890 /* Update potential sum for this i atom from the interaction with this j atom. */
891 velec = _mm_andnot_ps(dummy_mask,velec);
892 velecsum = _mm_add_ps(velecsum,velec);
896 fscal = _mm_andnot_ps(dummy_mask,fscal);
898 /* Calculate temporary vectorial force */
899 tx = _mm_mul_ps(fscal,dx13);
900 ty = _mm_mul_ps(fscal,dy13);
901 tz = _mm_mul_ps(fscal,dz13);
903 /* Update vectorial force */
904 fix1 = _mm_add_ps(fix1,tx);
905 fiy1 = _mm_add_ps(fiy1,ty);
906 fiz1 = _mm_add_ps(fiz1,tz);
908 fjx3 = _mm_add_ps(fjx3,tx);
909 fjy3 = _mm_add_ps(fjy3,ty);
910 fjz3 = _mm_add_ps(fjz3,tz);
912 /**************************
913 * CALCULATE INTERACTIONS *
914 **************************/
916 r21 = _mm_mul_ps(rsq21,rinv21);
917 r21 = _mm_andnot_ps(dummy_mask,r21);
919 /* Calculate table index by multiplying r with table scale and truncate to integer */
920 rt = _mm_mul_ps(r21,vftabscale);
921 vfitab = _mm_cvttps_epi32(rt);
922 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
923 vfitab = _mm_slli_epi32(vfitab,2);
925 /* CUBIC SPLINE TABLE ELECTROSTATICS */
926 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
927 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
928 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
929 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
930 _MM_TRANSPOSE4_PS(Y,F,G,H);
931 Heps = _mm_mul_ps(vfeps,H);
932 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
933 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
934 velec = _mm_mul_ps(qq21,VV);
935 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
936 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
938 /* Update potential sum for this i atom from the interaction with this j atom. */
939 velec = _mm_andnot_ps(dummy_mask,velec);
940 velecsum = _mm_add_ps(velecsum,velec);
944 fscal = _mm_andnot_ps(dummy_mask,fscal);
946 /* Calculate temporary vectorial force */
947 tx = _mm_mul_ps(fscal,dx21);
948 ty = _mm_mul_ps(fscal,dy21);
949 tz = _mm_mul_ps(fscal,dz21);
951 /* Update vectorial force */
952 fix2 = _mm_add_ps(fix2,tx);
953 fiy2 = _mm_add_ps(fiy2,ty);
954 fiz2 = _mm_add_ps(fiz2,tz);
956 fjx1 = _mm_add_ps(fjx1,tx);
957 fjy1 = _mm_add_ps(fjy1,ty);
958 fjz1 = _mm_add_ps(fjz1,tz);
960 /**************************
961 * CALCULATE INTERACTIONS *
962 **************************/
964 r22 = _mm_mul_ps(rsq22,rinv22);
965 r22 = _mm_andnot_ps(dummy_mask,r22);
967 /* Calculate table index by multiplying r with table scale and truncate to integer */
968 rt = _mm_mul_ps(r22,vftabscale);
969 vfitab = _mm_cvttps_epi32(rt);
970 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
971 vfitab = _mm_slli_epi32(vfitab,2);
973 /* CUBIC SPLINE TABLE ELECTROSTATICS */
974 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
975 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
976 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
977 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
978 _MM_TRANSPOSE4_PS(Y,F,G,H);
979 Heps = _mm_mul_ps(vfeps,H);
980 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
981 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
982 velec = _mm_mul_ps(qq22,VV);
983 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
984 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
986 /* Update potential sum for this i atom from the interaction with this j atom. */
987 velec = _mm_andnot_ps(dummy_mask,velec);
988 velecsum = _mm_add_ps(velecsum,velec);
992 fscal = _mm_andnot_ps(dummy_mask,fscal);
994 /* Calculate temporary vectorial force */
995 tx = _mm_mul_ps(fscal,dx22);
996 ty = _mm_mul_ps(fscal,dy22);
997 tz = _mm_mul_ps(fscal,dz22);
999 /* Update vectorial force */
1000 fix2 = _mm_add_ps(fix2,tx);
1001 fiy2 = _mm_add_ps(fiy2,ty);
1002 fiz2 = _mm_add_ps(fiz2,tz);
1004 fjx2 = _mm_add_ps(fjx2,tx);
1005 fjy2 = _mm_add_ps(fjy2,ty);
1006 fjz2 = _mm_add_ps(fjz2,tz);
1008 /**************************
1009 * CALCULATE INTERACTIONS *
1010 **************************/
1012 r23 = _mm_mul_ps(rsq23,rinv23);
1013 r23 = _mm_andnot_ps(dummy_mask,r23);
1015 /* Calculate table index by multiplying r with table scale and truncate to integer */
1016 rt = _mm_mul_ps(r23,vftabscale);
1017 vfitab = _mm_cvttps_epi32(rt);
1018 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1019 vfitab = _mm_slli_epi32(vfitab,2);
1021 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1022 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1023 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1024 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1025 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1026 _MM_TRANSPOSE4_PS(Y,F,G,H);
1027 Heps = _mm_mul_ps(vfeps,H);
1028 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1029 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1030 velec = _mm_mul_ps(qq23,VV);
1031 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1032 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1034 /* Update potential sum for this i atom from the interaction with this j atom. */
1035 velec = _mm_andnot_ps(dummy_mask,velec);
1036 velecsum = _mm_add_ps(velecsum,velec);
1040 fscal = _mm_andnot_ps(dummy_mask,fscal);
1042 /* Calculate temporary vectorial force */
1043 tx = _mm_mul_ps(fscal,dx23);
1044 ty = _mm_mul_ps(fscal,dy23);
1045 tz = _mm_mul_ps(fscal,dz23);
1047 /* Update vectorial force */
1048 fix2 = _mm_add_ps(fix2,tx);
1049 fiy2 = _mm_add_ps(fiy2,ty);
1050 fiz2 = _mm_add_ps(fiz2,tz);
1052 fjx3 = _mm_add_ps(fjx3,tx);
1053 fjy3 = _mm_add_ps(fjy3,ty);
1054 fjz3 = _mm_add_ps(fjz3,tz);
1056 /**************************
1057 * CALCULATE INTERACTIONS *
1058 **************************/
1060 r31 = _mm_mul_ps(rsq31,rinv31);
1061 r31 = _mm_andnot_ps(dummy_mask,r31);
1063 /* Calculate table index by multiplying r with table scale and truncate to integer */
1064 rt = _mm_mul_ps(r31,vftabscale);
1065 vfitab = _mm_cvttps_epi32(rt);
1066 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1067 vfitab = _mm_slli_epi32(vfitab,2);
1069 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1070 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1071 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1072 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1073 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1074 _MM_TRANSPOSE4_PS(Y,F,G,H);
1075 Heps = _mm_mul_ps(vfeps,H);
1076 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1077 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1078 velec = _mm_mul_ps(qq31,VV);
1079 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1080 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1082 /* Update potential sum for this i atom from the interaction with this j atom. */
1083 velec = _mm_andnot_ps(dummy_mask,velec);
1084 velecsum = _mm_add_ps(velecsum,velec);
1088 fscal = _mm_andnot_ps(dummy_mask,fscal);
1090 /* Calculate temporary vectorial force */
1091 tx = _mm_mul_ps(fscal,dx31);
1092 ty = _mm_mul_ps(fscal,dy31);
1093 tz = _mm_mul_ps(fscal,dz31);
1095 /* Update vectorial force */
1096 fix3 = _mm_add_ps(fix3,tx);
1097 fiy3 = _mm_add_ps(fiy3,ty);
1098 fiz3 = _mm_add_ps(fiz3,tz);
1100 fjx1 = _mm_add_ps(fjx1,tx);
1101 fjy1 = _mm_add_ps(fjy1,ty);
1102 fjz1 = _mm_add_ps(fjz1,tz);
1104 /**************************
1105 * CALCULATE INTERACTIONS *
1106 **************************/
1108 r32 = _mm_mul_ps(rsq32,rinv32);
1109 r32 = _mm_andnot_ps(dummy_mask,r32);
1111 /* Calculate table index by multiplying r with table scale and truncate to integer */
1112 rt = _mm_mul_ps(r32,vftabscale);
1113 vfitab = _mm_cvttps_epi32(rt);
1114 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1115 vfitab = _mm_slli_epi32(vfitab,2);
1117 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1118 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1119 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1120 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1121 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1122 _MM_TRANSPOSE4_PS(Y,F,G,H);
1123 Heps = _mm_mul_ps(vfeps,H);
1124 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1125 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1126 velec = _mm_mul_ps(qq32,VV);
1127 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1128 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1130 /* Update potential sum for this i atom from the interaction with this j atom. */
1131 velec = _mm_andnot_ps(dummy_mask,velec);
1132 velecsum = _mm_add_ps(velecsum,velec);
1136 fscal = _mm_andnot_ps(dummy_mask,fscal);
1138 /* Calculate temporary vectorial force */
1139 tx = _mm_mul_ps(fscal,dx32);
1140 ty = _mm_mul_ps(fscal,dy32);
1141 tz = _mm_mul_ps(fscal,dz32);
1143 /* Update vectorial force */
1144 fix3 = _mm_add_ps(fix3,tx);
1145 fiy3 = _mm_add_ps(fiy3,ty);
1146 fiz3 = _mm_add_ps(fiz3,tz);
1148 fjx2 = _mm_add_ps(fjx2,tx);
1149 fjy2 = _mm_add_ps(fjy2,ty);
1150 fjz2 = _mm_add_ps(fjz2,tz);
1152 /**************************
1153 * CALCULATE INTERACTIONS *
1154 **************************/
1156 r33 = _mm_mul_ps(rsq33,rinv33);
1157 r33 = _mm_andnot_ps(dummy_mask,r33);
1159 /* Calculate table index by multiplying r with table scale and truncate to integer */
1160 rt = _mm_mul_ps(r33,vftabscale);
1161 vfitab = _mm_cvttps_epi32(rt);
1162 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1163 vfitab = _mm_slli_epi32(vfitab,2);
1165 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1166 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1167 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1168 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1169 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1170 _MM_TRANSPOSE4_PS(Y,F,G,H);
1171 Heps = _mm_mul_ps(vfeps,H);
1172 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1173 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1174 velec = _mm_mul_ps(qq33,VV);
1175 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1176 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1178 /* Update potential sum for this i atom from the interaction with this j atom. */
1179 velec = _mm_andnot_ps(dummy_mask,velec);
1180 velecsum = _mm_add_ps(velecsum,velec);
1184 fscal = _mm_andnot_ps(dummy_mask,fscal);
1186 /* Calculate temporary vectorial force */
1187 tx = _mm_mul_ps(fscal,dx33);
1188 ty = _mm_mul_ps(fscal,dy33);
1189 tz = _mm_mul_ps(fscal,dz33);
1191 /* Update vectorial force */
1192 fix3 = _mm_add_ps(fix3,tx);
1193 fiy3 = _mm_add_ps(fiy3,ty);
1194 fiz3 = _mm_add_ps(fiz3,tz);
1196 fjx3 = _mm_add_ps(fjx3,tx);
1197 fjy3 = _mm_add_ps(fjy3,ty);
1198 fjz3 = _mm_add_ps(fjz3,tz);
1200 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1201 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1202 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1203 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1205 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
1206 fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1208 /* Inner loop uses 396 flops */
1211 /* End of innermost loop */
1213 gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1214 f+i_coord_offset+DIM,fshift+i_shift_offset);
1217 /* Update potential energies */
1218 gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1220 /* Increment number of inner iterations */
1221 inneriter += j_index_end - j_index_start;
1223 /* Outer loop uses 19 flops */
1226 /* Increment number of outer iterations */
1229 /* Update outer/inner flops */
1231 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_VF,outeriter*19 + inneriter*396);
1234 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwNone_GeomW4W4_F_sse2_single
1235 * Electrostatics interaction: CubicSplineTable
1236 * VdW interaction: None
1237 * Geometry: Water4-Water4
1238 * Calculate force/pot: Force
1241 nb_kernel_ElecCSTab_VdwNone_GeomW4W4_F_sse2_single
1242 (t_nblist * gmx_restrict nlist,
1243 rvec * gmx_restrict xx,
1244 rvec * gmx_restrict ff,
1245 struct t_forcerec * gmx_restrict fr,
1246 t_mdatoms * gmx_restrict mdatoms,
1247 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1248 t_nrnb * gmx_restrict nrnb)
1250 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1251 * just 0 for non-waters.
1252 * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
1253 * jnr indices corresponding to data put in the four positions in the SIMD register.
1255 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1256 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1257 int jnrA,jnrB,jnrC,jnrD;
1258 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1259 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1260 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1261 real rcutoff_scalar;
1262 real *shiftvec,*fshift,*x,*f;
1263 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1264 real scratch[4*DIM];
1265 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1267 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1269 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1271 __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1272 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1273 __m128 jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1274 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1275 __m128 jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1276 int vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1277 __m128 jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1278 __m128 dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1279 __m128 dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1280 __m128 dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1281 __m128 dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1282 __m128 dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1283 __m128 dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1284 __m128 dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1285 __m128 dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1286 __m128 dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1287 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
1290 __m128i ifour = _mm_set1_epi32(4);
1291 __m128 rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1293 __m128 dummy_mask,cutoff_mask;
1294 __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1295 __m128 one = _mm_set1_ps(1.0);
1296 __m128 two = _mm_set1_ps(2.0);
1302 jindex = nlist->jindex;
1304 shiftidx = nlist->shift;
1306 shiftvec = fr->shift_vec[0];
1307 fshift = fr->fshift[0];
1308 facel = _mm_set1_ps(fr->ic->epsfac);
1309 charge = mdatoms->chargeA;
1311 vftab = kernel_data->table_elec->data;
1312 vftabscale = _mm_set1_ps(kernel_data->table_elec->scale);
1314 /* Setup water-specific parameters */
1315 inr = nlist->iinr[0];
1316 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1317 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1318 iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
1320 jq1 = _mm_set1_ps(charge[inr+1]);
1321 jq2 = _mm_set1_ps(charge[inr+2]);
1322 jq3 = _mm_set1_ps(charge[inr+3]);
1323 qq11 = _mm_mul_ps(iq1,jq1);
1324 qq12 = _mm_mul_ps(iq1,jq2);
1325 qq13 = _mm_mul_ps(iq1,jq3);
1326 qq21 = _mm_mul_ps(iq2,jq1);
1327 qq22 = _mm_mul_ps(iq2,jq2);
1328 qq23 = _mm_mul_ps(iq2,jq3);
1329 qq31 = _mm_mul_ps(iq3,jq1);
1330 qq32 = _mm_mul_ps(iq3,jq2);
1331 qq33 = _mm_mul_ps(iq3,jq3);
1333 /* Avoid stupid compiler warnings */
1334 jnrA = jnrB = jnrC = jnrD = 0;
1335 j_coord_offsetA = 0;
1336 j_coord_offsetB = 0;
1337 j_coord_offsetC = 0;
1338 j_coord_offsetD = 0;
1343 for(iidx=0;iidx<4*DIM;iidx++)
1345 scratch[iidx] = 0.0;
1348 /* Start outer loop over neighborlists */
1349 for(iidx=0; iidx<nri; iidx++)
1351 /* Load shift vector for this list */
1352 i_shift_offset = DIM*shiftidx[iidx];
1354 /* Load limits for loop over neighbors */
1355 j_index_start = jindex[iidx];
1356 j_index_end = jindex[iidx+1];
1358 /* Get outer coordinate index */
1360 i_coord_offset = DIM*inr;
1362 /* Load i particle coords and add shift vector */
1363 gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
1364 &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1366 fix1 = _mm_setzero_ps();
1367 fiy1 = _mm_setzero_ps();
1368 fiz1 = _mm_setzero_ps();
1369 fix2 = _mm_setzero_ps();
1370 fiy2 = _mm_setzero_ps();
1371 fiz2 = _mm_setzero_ps();
1372 fix3 = _mm_setzero_ps();
1373 fiy3 = _mm_setzero_ps();
1374 fiz3 = _mm_setzero_ps();
1376 /* Start inner kernel loop */
1377 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1380 /* Get j neighbor index, and coordinate index */
1382 jnrB = jjnr[jidx+1];
1383 jnrC = jjnr[jidx+2];
1384 jnrD = jjnr[jidx+3];
1385 j_coord_offsetA = DIM*jnrA;
1386 j_coord_offsetB = DIM*jnrB;
1387 j_coord_offsetC = DIM*jnrC;
1388 j_coord_offsetD = DIM*jnrD;
1390 /* load j atom coordinates */
1391 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1392 x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1393 &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1395 /* Calculate displacement vector */
1396 dx11 = _mm_sub_ps(ix1,jx1);
1397 dy11 = _mm_sub_ps(iy1,jy1);
1398 dz11 = _mm_sub_ps(iz1,jz1);
1399 dx12 = _mm_sub_ps(ix1,jx2);
1400 dy12 = _mm_sub_ps(iy1,jy2);
1401 dz12 = _mm_sub_ps(iz1,jz2);
1402 dx13 = _mm_sub_ps(ix1,jx3);
1403 dy13 = _mm_sub_ps(iy1,jy3);
1404 dz13 = _mm_sub_ps(iz1,jz3);
1405 dx21 = _mm_sub_ps(ix2,jx1);
1406 dy21 = _mm_sub_ps(iy2,jy1);
1407 dz21 = _mm_sub_ps(iz2,jz1);
1408 dx22 = _mm_sub_ps(ix2,jx2);
1409 dy22 = _mm_sub_ps(iy2,jy2);
1410 dz22 = _mm_sub_ps(iz2,jz2);
1411 dx23 = _mm_sub_ps(ix2,jx3);
1412 dy23 = _mm_sub_ps(iy2,jy3);
1413 dz23 = _mm_sub_ps(iz2,jz3);
1414 dx31 = _mm_sub_ps(ix3,jx1);
1415 dy31 = _mm_sub_ps(iy3,jy1);
1416 dz31 = _mm_sub_ps(iz3,jz1);
1417 dx32 = _mm_sub_ps(ix3,jx2);
1418 dy32 = _mm_sub_ps(iy3,jy2);
1419 dz32 = _mm_sub_ps(iz3,jz2);
1420 dx33 = _mm_sub_ps(ix3,jx3);
1421 dy33 = _mm_sub_ps(iy3,jy3);
1422 dz33 = _mm_sub_ps(iz3,jz3);
1424 /* Calculate squared distance and things based on it */
1425 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1426 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1427 rsq13 = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1428 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1429 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1430 rsq23 = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1431 rsq31 = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1432 rsq32 = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1433 rsq33 = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1435 rinv11 = sse2_invsqrt_f(rsq11);
1436 rinv12 = sse2_invsqrt_f(rsq12);
1437 rinv13 = sse2_invsqrt_f(rsq13);
1438 rinv21 = sse2_invsqrt_f(rsq21);
1439 rinv22 = sse2_invsqrt_f(rsq22);
1440 rinv23 = sse2_invsqrt_f(rsq23);
1441 rinv31 = sse2_invsqrt_f(rsq31);
1442 rinv32 = sse2_invsqrt_f(rsq32);
1443 rinv33 = sse2_invsqrt_f(rsq33);
1445 fjx1 = _mm_setzero_ps();
1446 fjy1 = _mm_setzero_ps();
1447 fjz1 = _mm_setzero_ps();
1448 fjx2 = _mm_setzero_ps();
1449 fjy2 = _mm_setzero_ps();
1450 fjz2 = _mm_setzero_ps();
1451 fjx3 = _mm_setzero_ps();
1452 fjy3 = _mm_setzero_ps();
1453 fjz3 = _mm_setzero_ps();
1455 /**************************
1456 * CALCULATE INTERACTIONS *
1457 **************************/
1459 r11 = _mm_mul_ps(rsq11,rinv11);
1461 /* Calculate table index by multiplying r with table scale and truncate to integer */
1462 rt = _mm_mul_ps(r11,vftabscale);
1463 vfitab = _mm_cvttps_epi32(rt);
1464 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1465 vfitab = _mm_slli_epi32(vfitab,2);
1467 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1468 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1469 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1470 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1471 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1472 _MM_TRANSPOSE4_PS(Y,F,G,H);
1473 Heps = _mm_mul_ps(vfeps,H);
1474 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1475 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1476 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
1480 /* Calculate temporary vectorial force */
1481 tx = _mm_mul_ps(fscal,dx11);
1482 ty = _mm_mul_ps(fscal,dy11);
1483 tz = _mm_mul_ps(fscal,dz11);
1485 /* Update vectorial force */
1486 fix1 = _mm_add_ps(fix1,tx);
1487 fiy1 = _mm_add_ps(fiy1,ty);
1488 fiz1 = _mm_add_ps(fiz1,tz);
1490 fjx1 = _mm_add_ps(fjx1,tx);
1491 fjy1 = _mm_add_ps(fjy1,ty);
1492 fjz1 = _mm_add_ps(fjz1,tz);
1494 /**************************
1495 * CALCULATE INTERACTIONS *
1496 **************************/
1498 r12 = _mm_mul_ps(rsq12,rinv12);
1500 /* Calculate table index by multiplying r with table scale and truncate to integer */
1501 rt = _mm_mul_ps(r12,vftabscale);
1502 vfitab = _mm_cvttps_epi32(rt);
1503 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1504 vfitab = _mm_slli_epi32(vfitab,2);
1506 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1507 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1508 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1509 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1510 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1511 _MM_TRANSPOSE4_PS(Y,F,G,H);
1512 Heps = _mm_mul_ps(vfeps,H);
1513 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1514 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1515 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
1519 /* Calculate temporary vectorial force */
1520 tx = _mm_mul_ps(fscal,dx12);
1521 ty = _mm_mul_ps(fscal,dy12);
1522 tz = _mm_mul_ps(fscal,dz12);
1524 /* Update vectorial force */
1525 fix1 = _mm_add_ps(fix1,tx);
1526 fiy1 = _mm_add_ps(fiy1,ty);
1527 fiz1 = _mm_add_ps(fiz1,tz);
1529 fjx2 = _mm_add_ps(fjx2,tx);
1530 fjy2 = _mm_add_ps(fjy2,ty);
1531 fjz2 = _mm_add_ps(fjz2,tz);
1533 /**************************
1534 * CALCULATE INTERACTIONS *
1535 **************************/
1537 r13 = _mm_mul_ps(rsq13,rinv13);
1539 /* Calculate table index by multiplying r with table scale and truncate to integer */
1540 rt = _mm_mul_ps(r13,vftabscale);
1541 vfitab = _mm_cvttps_epi32(rt);
1542 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1543 vfitab = _mm_slli_epi32(vfitab,2);
1545 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1546 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1547 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1548 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1549 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1550 _MM_TRANSPOSE4_PS(Y,F,G,H);
1551 Heps = _mm_mul_ps(vfeps,H);
1552 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1553 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1554 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
1558 /* Calculate temporary vectorial force */
1559 tx = _mm_mul_ps(fscal,dx13);
1560 ty = _mm_mul_ps(fscal,dy13);
1561 tz = _mm_mul_ps(fscal,dz13);
1563 /* Update vectorial force */
1564 fix1 = _mm_add_ps(fix1,tx);
1565 fiy1 = _mm_add_ps(fiy1,ty);
1566 fiz1 = _mm_add_ps(fiz1,tz);
1568 fjx3 = _mm_add_ps(fjx3,tx);
1569 fjy3 = _mm_add_ps(fjy3,ty);
1570 fjz3 = _mm_add_ps(fjz3,tz);
1572 /**************************
1573 * CALCULATE INTERACTIONS *
1574 **************************/
1576 r21 = _mm_mul_ps(rsq21,rinv21);
1578 /* Calculate table index by multiplying r with table scale and truncate to integer */
1579 rt = _mm_mul_ps(r21,vftabscale);
1580 vfitab = _mm_cvttps_epi32(rt);
1581 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1582 vfitab = _mm_slli_epi32(vfitab,2);
1584 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1585 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1586 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1587 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1588 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1589 _MM_TRANSPOSE4_PS(Y,F,G,H);
1590 Heps = _mm_mul_ps(vfeps,H);
1591 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1592 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1593 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
1597 /* Calculate temporary vectorial force */
1598 tx = _mm_mul_ps(fscal,dx21);
1599 ty = _mm_mul_ps(fscal,dy21);
1600 tz = _mm_mul_ps(fscal,dz21);
1602 /* Update vectorial force */
1603 fix2 = _mm_add_ps(fix2,tx);
1604 fiy2 = _mm_add_ps(fiy2,ty);
1605 fiz2 = _mm_add_ps(fiz2,tz);
1607 fjx1 = _mm_add_ps(fjx1,tx);
1608 fjy1 = _mm_add_ps(fjy1,ty);
1609 fjz1 = _mm_add_ps(fjz1,tz);
1611 /**************************
1612 * CALCULATE INTERACTIONS *
1613 **************************/
1615 r22 = _mm_mul_ps(rsq22,rinv22);
1617 /* Calculate table index by multiplying r with table scale and truncate to integer */
1618 rt = _mm_mul_ps(r22,vftabscale);
1619 vfitab = _mm_cvttps_epi32(rt);
1620 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1621 vfitab = _mm_slli_epi32(vfitab,2);
1623 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1624 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1625 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1626 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1627 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1628 _MM_TRANSPOSE4_PS(Y,F,G,H);
1629 Heps = _mm_mul_ps(vfeps,H);
1630 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1631 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1632 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
1636 /* Calculate temporary vectorial force */
1637 tx = _mm_mul_ps(fscal,dx22);
1638 ty = _mm_mul_ps(fscal,dy22);
1639 tz = _mm_mul_ps(fscal,dz22);
1641 /* Update vectorial force */
1642 fix2 = _mm_add_ps(fix2,tx);
1643 fiy2 = _mm_add_ps(fiy2,ty);
1644 fiz2 = _mm_add_ps(fiz2,tz);
1646 fjx2 = _mm_add_ps(fjx2,tx);
1647 fjy2 = _mm_add_ps(fjy2,ty);
1648 fjz2 = _mm_add_ps(fjz2,tz);
1650 /**************************
1651 * CALCULATE INTERACTIONS *
1652 **************************/
1654 r23 = _mm_mul_ps(rsq23,rinv23);
1656 /* Calculate table index by multiplying r with table scale and truncate to integer */
1657 rt = _mm_mul_ps(r23,vftabscale);
1658 vfitab = _mm_cvttps_epi32(rt);
1659 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1660 vfitab = _mm_slli_epi32(vfitab,2);
1662 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1663 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1664 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1665 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1666 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1667 _MM_TRANSPOSE4_PS(Y,F,G,H);
1668 Heps = _mm_mul_ps(vfeps,H);
1669 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1670 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1671 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1675 /* Calculate temporary vectorial force */
1676 tx = _mm_mul_ps(fscal,dx23);
1677 ty = _mm_mul_ps(fscal,dy23);
1678 tz = _mm_mul_ps(fscal,dz23);
1680 /* Update vectorial force */
1681 fix2 = _mm_add_ps(fix2,tx);
1682 fiy2 = _mm_add_ps(fiy2,ty);
1683 fiz2 = _mm_add_ps(fiz2,tz);
1685 fjx3 = _mm_add_ps(fjx3,tx);
1686 fjy3 = _mm_add_ps(fjy3,ty);
1687 fjz3 = _mm_add_ps(fjz3,tz);
1689 /**************************
1690 * CALCULATE INTERACTIONS *
1691 **************************/
1693 r31 = _mm_mul_ps(rsq31,rinv31);
1695 /* Calculate table index by multiplying r with table scale and truncate to integer */
1696 rt = _mm_mul_ps(r31,vftabscale);
1697 vfitab = _mm_cvttps_epi32(rt);
1698 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1699 vfitab = _mm_slli_epi32(vfitab,2);
1701 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1702 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1703 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1704 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1705 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1706 _MM_TRANSPOSE4_PS(Y,F,G,H);
1707 Heps = _mm_mul_ps(vfeps,H);
1708 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1709 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1710 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1714 /* Calculate temporary vectorial force */
1715 tx = _mm_mul_ps(fscal,dx31);
1716 ty = _mm_mul_ps(fscal,dy31);
1717 tz = _mm_mul_ps(fscal,dz31);
1719 /* Update vectorial force */
1720 fix3 = _mm_add_ps(fix3,tx);
1721 fiy3 = _mm_add_ps(fiy3,ty);
1722 fiz3 = _mm_add_ps(fiz3,tz);
1724 fjx1 = _mm_add_ps(fjx1,tx);
1725 fjy1 = _mm_add_ps(fjy1,ty);
1726 fjz1 = _mm_add_ps(fjz1,tz);
1728 /**************************
1729 * CALCULATE INTERACTIONS *
1730 **************************/
1732 r32 = _mm_mul_ps(rsq32,rinv32);
1734 /* Calculate table index by multiplying r with table scale and truncate to integer */
1735 rt = _mm_mul_ps(r32,vftabscale);
1736 vfitab = _mm_cvttps_epi32(rt);
1737 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1738 vfitab = _mm_slli_epi32(vfitab,2);
1740 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1741 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1742 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1743 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1744 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1745 _MM_TRANSPOSE4_PS(Y,F,G,H);
1746 Heps = _mm_mul_ps(vfeps,H);
1747 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1748 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1749 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1753 /* Calculate temporary vectorial force */
1754 tx = _mm_mul_ps(fscal,dx32);
1755 ty = _mm_mul_ps(fscal,dy32);
1756 tz = _mm_mul_ps(fscal,dz32);
1758 /* Update vectorial force */
1759 fix3 = _mm_add_ps(fix3,tx);
1760 fiy3 = _mm_add_ps(fiy3,ty);
1761 fiz3 = _mm_add_ps(fiz3,tz);
1763 fjx2 = _mm_add_ps(fjx2,tx);
1764 fjy2 = _mm_add_ps(fjy2,ty);
1765 fjz2 = _mm_add_ps(fjz2,tz);
1767 /**************************
1768 * CALCULATE INTERACTIONS *
1769 **************************/
1771 r33 = _mm_mul_ps(rsq33,rinv33);
1773 /* Calculate table index by multiplying r with table scale and truncate to integer */
1774 rt = _mm_mul_ps(r33,vftabscale);
1775 vfitab = _mm_cvttps_epi32(rt);
1776 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1777 vfitab = _mm_slli_epi32(vfitab,2);
1779 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1780 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1781 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1782 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1783 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1784 _MM_TRANSPOSE4_PS(Y,F,G,H);
1785 Heps = _mm_mul_ps(vfeps,H);
1786 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1787 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1788 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1792 /* Calculate temporary vectorial force */
1793 tx = _mm_mul_ps(fscal,dx33);
1794 ty = _mm_mul_ps(fscal,dy33);
1795 tz = _mm_mul_ps(fscal,dz33);
1797 /* Update vectorial force */
1798 fix3 = _mm_add_ps(fix3,tx);
1799 fiy3 = _mm_add_ps(fiy3,ty);
1800 fiz3 = _mm_add_ps(fiz3,tz);
1802 fjx3 = _mm_add_ps(fjx3,tx);
1803 fjy3 = _mm_add_ps(fjy3,ty);
1804 fjz3 = _mm_add_ps(fjz3,tz);
1806 fjptrA = f+j_coord_offsetA;
1807 fjptrB = f+j_coord_offsetB;
1808 fjptrC = f+j_coord_offsetC;
1809 fjptrD = f+j_coord_offsetD;
1811 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
1812 fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1814 /* Inner loop uses 351 flops */
1817 if(jidx<j_index_end)
1820 /* Get j neighbor index, and coordinate index */
1821 jnrlistA = jjnr[jidx];
1822 jnrlistB = jjnr[jidx+1];
1823 jnrlistC = jjnr[jidx+2];
1824 jnrlistD = jjnr[jidx+3];
1825 /* Sign of each element will be negative for non-real atoms.
1826 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1827 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
1829 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1830 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
1831 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
1832 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
1833 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
1834 j_coord_offsetA = DIM*jnrA;
1835 j_coord_offsetB = DIM*jnrB;
1836 j_coord_offsetC = DIM*jnrC;
1837 j_coord_offsetD = DIM*jnrD;
1839 /* load j atom coordinates */
1840 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1841 x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1842 &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1844 /* Calculate displacement vector */
1845 dx11 = _mm_sub_ps(ix1,jx1);
1846 dy11 = _mm_sub_ps(iy1,jy1);
1847 dz11 = _mm_sub_ps(iz1,jz1);
1848 dx12 = _mm_sub_ps(ix1,jx2);
1849 dy12 = _mm_sub_ps(iy1,jy2);
1850 dz12 = _mm_sub_ps(iz1,jz2);
1851 dx13 = _mm_sub_ps(ix1,jx3);
1852 dy13 = _mm_sub_ps(iy1,jy3);
1853 dz13 = _mm_sub_ps(iz1,jz3);
1854 dx21 = _mm_sub_ps(ix2,jx1);
1855 dy21 = _mm_sub_ps(iy2,jy1);
1856 dz21 = _mm_sub_ps(iz2,jz1);
1857 dx22 = _mm_sub_ps(ix2,jx2);
1858 dy22 = _mm_sub_ps(iy2,jy2);
1859 dz22 = _mm_sub_ps(iz2,jz2);
1860 dx23 = _mm_sub_ps(ix2,jx3);
1861 dy23 = _mm_sub_ps(iy2,jy3);
1862 dz23 = _mm_sub_ps(iz2,jz3);
1863 dx31 = _mm_sub_ps(ix3,jx1);
1864 dy31 = _mm_sub_ps(iy3,jy1);
1865 dz31 = _mm_sub_ps(iz3,jz1);
1866 dx32 = _mm_sub_ps(ix3,jx2);
1867 dy32 = _mm_sub_ps(iy3,jy2);
1868 dz32 = _mm_sub_ps(iz3,jz2);
1869 dx33 = _mm_sub_ps(ix3,jx3);
1870 dy33 = _mm_sub_ps(iy3,jy3);
1871 dz33 = _mm_sub_ps(iz3,jz3);
1873 /* Calculate squared distance and things based on it */
1874 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1875 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1876 rsq13 = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1877 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1878 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1879 rsq23 = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1880 rsq31 = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1881 rsq32 = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1882 rsq33 = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1884 rinv11 = sse2_invsqrt_f(rsq11);
1885 rinv12 = sse2_invsqrt_f(rsq12);
1886 rinv13 = sse2_invsqrt_f(rsq13);
1887 rinv21 = sse2_invsqrt_f(rsq21);
1888 rinv22 = sse2_invsqrt_f(rsq22);
1889 rinv23 = sse2_invsqrt_f(rsq23);
1890 rinv31 = sse2_invsqrt_f(rsq31);
1891 rinv32 = sse2_invsqrt_f(rsq32);
1892 rinv33 = sse2_invsqrt_f(rsq33);
1894 fjx1 = _mm_setzero_ps();
1895 fjy1 = _mm_setzero_ps();
1896 fjz1 = _mm_setzero_ps();
1897 fjx2 = _mm_setzero_ps();
1898 fjy2 = _mm_setzero_ps();
1899 fjz2 = _mm_setzero_ps();
1900 fjx3 = _mm_setzero_ps();
1901 fjy3 = _mm_setzero_ps();
1902 fjz3 = _mm_setzero_ps();
1904 /**************************
1905 * CALCULATE INTERACTIONS *
1906 **************************/
1908 r11 = _mm_mul_ps(rsq11,rinv11);
1909 r11 = _mm_andnot_ps(dummy_mask,r11);
1911 /* Calculate table index by multiplying r with table scale and truncate to integer */
1912 rt = _mm_mul_ps(r11,vftabscale);
1913 vfitab = _mm_cvttps_epi32(rt);
1914 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1915 vfitab = _mm_slli_epi32(vfitab,2);
1917 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1918 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1919 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1920 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1921 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1922 _MM_TRANSPOSE4_PS(Y,F,G,H);
1923 Heps = _mm_mul_ps(vfeps,H);
1924 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1925 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1926 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
1930 fscal = _mm_andnot_ps(dummy_mask,fscal);
1932 /* Calculate temporary vectorial force */
1933 tx = _mm_mul_ps(fscal,dx11);
1934 ty = _mm_mul_ps(fscal,dy11);
1935 tz = _mm_mul_ps(fscal,dz11);
1937 /* Update vectorial force */
1938 fix1 = _mm_add_ps(fix1,tx);
1939 fiy1 = _mm_add_ps(fiy1,ty);
1940 fiz1 = _mm_add_ps(fiz1,tz);
1942 fjx1 = _mm_add_ps(fjx1,tx);
1943 fjy1 = _mm_add_ps(fjy1,ty);
1944 fjz1 = _mm_add_ps(fjz1,tz);
1946 /**************************
1947 * CALCULATE INTERACTIONS *
1948 **************************/
1950 r12 = _mm_mul_ps(rsq12,rinv12);
1951 r12 = _mm_andnot_ps(dummy_mask,r12);
1953 /* Calculate table index by multiplying r with table scale and truncate to integer */
1954 rt = _mm_mul_ps(r12,vftabscale);
1955 vfitab = _mm_cvttps_epi32(rt);
1956 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1957 vfitab = _mm_slli_epi32(vfitab,2);
1959 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1960 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1961 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1962 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1963 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1964 _MM_TRANSPOSE4_PS(Y,F,G,H);
1965 Heps = _mm_mul_ps(vfeps,H);
1966 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1967 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1968 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
1972 fscal = _mm_andnot_ps(dummy_mask,fscal);
1974 /* Calculate temporary vectorial force */
1975 tx = _mm_mul_ps(fscal,dx12);
1976 ty = _mm_mul_ps(fscal,dy12);
1977 tz = _mm_mul_ps(fscal,dz12);
1979 /* Update vectorial force */
1980 fix1 = _mm_add_ps(fix1,tx);
1981 fiy1 = _mm_add_ps(fiy1,ty);
1982 fiz1 = _mm_add_ps(fiz1,tz);
1984 fjx2 = _mm_add_ps(fjx2,tx);
1985 fjy2 = _mm_add_ps(fjy2,ty);
1986 fjz2 = _mm_add_ps(fjz2,tz);
1988 /**************************
1989 * CALCULATE INTERACTIONS *
1990 **************************/
1992 r13 = _mm_mul_ps(rsq13,rinv13);
1993 r13 = _mm_andnot_ps(dummy_mask,r13);
1995 /* Calculate table index by multiplying r with table scale and truncate to integer */
1996 rt = _mm_mul_ps(r13,vftabscale);
1997 vfitab = _mm_cvttps_epi32(rt);
1998 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1999 vfitab = _mm_slli_epi32(vfitab,2);
2001 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2002 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2003 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2004 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2005 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2006 _MM_TRANSPOSE4_PS(Y,F,G,H);
2007 Heps = _mm_mul_ps(vfeps,H);
2008 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2009 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2010 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
2014 fscal = _mm_andnot_ps(dummy_mask,fscal);
2016 /* Calculate temporary vectorial force */
2017 tx = _mm_mul_ps(fscal,dx13);
2018 ty = _mm_mul_ps(fscal,dy13);
2019 tz = _mm_mul_ps(fscal,dz13);
2021 /* Update vectorial force */
2022 fix1 = _mm_add_ps(fix1,tx);
2023 fiy1 = _mm_add_ps(fiy1,ty);
2024 fiz1 = _mm_add_ps(fiz1,tz);
2026 fjx3 = _mm_add_ps(fjx3,tx);
2027 fjy3 = _mm_add_ps(fjy3,ty);
2028 fjz3 = _mm_add_ps(fjz3,tz);
2030 /**************************
2031 * CALCULATE INTERACTIONS *
2032 **************************/
2034 r21 = _mm_mul_ps(rsq21,rinv21);
2035 r21 = _mm_andnot_ps(dummy_mask,r21);
2037 /* Calculate table index by multiplying r with table scale and truncate to integer */
2038 rt = _mm_mul_ps(r21,vftabscale);
2039 vfitab = _mm_cvttps_epi32(rt);
2040 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2041 vfitab = _mm_slli_epi32(vfitab,2);
2043 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2044 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2045 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2046 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2047 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2048 _MM_TRANSPOSE4_PS(Y,F,G,H);
2049 Heps = _mm_mul_ps(vfeps,H);
2050 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2051 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2052 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
2056 fscal = _mm_andnot_ps(dummy_mask,fscal);
2058 /* Calculate temporary vectorial force */
2059 tx = _mm_mul_ps(fscal,dx21);
2060 ty = _mm_mul_ps(fscal,dy21);
2061 tz = _mm_mul_ps(fscal,dz21);
2063 /* Update vectorial force */
2064 fix2 = _mm_add_ps(fix2,tx);
2065 fiy2 = _mm_add_ps(fiy2,ty);
2066 fiz2 = _mm_add_ps(fiz2,tz);
2068 fjx1 = _mm_add_ps(fjx1,tx);
2069 fjy1 = _mm_add_ps(fjy1,ty);
2070 fjz1 = _mm_add_ps(fjz1,tz);
2072 /**************************
2073 * CALCULATE INTERACTIONS *
2074 **************************/
2076 r22 = _mm_mul_ps(rsq22,rinv22);
2077 r22 = _mm_andnot_ps(dummy_mask,r22);
2079 /* Calculate table index by multiplying r with table scale and truncate to integer */
2080 rt = _mm_mul_ps(r22,vftabscale);
2081 vfitab = _mm_cvttps_epi32(rt);
2082 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2083 vfitab = _mm_slli_epi32(vfitab,2);
2085 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2086 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2087 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2088 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2089 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2090 _MM_TRANSPOSE4_PS(Y,F,G,H);
2091 Heps = _mm_mul_ps(vfeps,H);
2092 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2093 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2094 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
2098 fscal = _mm_andnot_ps(dummy_mask,fscal);
2100 /* Calculate temporary vectorial force */
2101 tx = _mm_mul_ps(fscal,dx22);
2102 ty = _mm_mul_ps(fscal,dy22);
2103 tz = _mm_mul_ps(fscal,dz22);
2105 /* Update vectorial force */
2106 fix2 = _mm_add_ps(fix2,tx);
2107 fiy2 = _mm_add_ps(fiy2,ty);
2108 fiz2 = _mm_add_ps(fiz2,tz);
2110 fjx2 = _mm_add_ps(fjx2,tx);
2111 fjy2 = _mm_add_ps(fjy2,ty);
2112 fjz2 = _mm_add_ps(fjz2,tz);
2114 /**************************
2115 * CALCULATE INTERACTIONS *
2116 **************************/
2118 r23 = _mm_mul_ps(rsq23,rinv23);
2119 r23 = _mm_andnot_ps(dummy_mask,r23);
2121 /* Calculate table index by multiplying r with table scale and truncate to integer */
2122 rt = _mm_mul_ps(r23,vftabscale);
2123 vfitab = _mm_cvttps_epi32(rt);
2124 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2125 vfitab = _mm_slli_epi32(vfitab,2);
2127 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2128 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2129 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2130 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2131 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2132 _MM_TRANSPOSE4_PS(Y,F,G,H);
2133 Heps = _mm_mul_ps(vfeps,H);
2134 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2135 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2136 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
2140 fscal = _mm_andnot_ps(dummy_mask,fscal);
2142 /* Calculate temporary vectorial force */
2143 tx = _mm_mul_ps(fscal,dx23);
2144 ty = _mm_mul_ps(fscal,dy23);
2145 tz = _mm_mul_ps(fscal,dz23);
2147 /* Update vectorial force */
2148 fix2 = _mm_add_ps(fix2,tx);
2149 fiy2 = _mm_add_ps(fiy2,ty);
2150 fiz2 = _mm_add_ps(fiz2,tz);
2152 fjx3 = _mm_add_ps(fjx3,tx);
2153 fjy3 = _mm_add_ps(fjy3,ty);
2154 fjz3 = _mm_add_ps(fjz3,tz);
2156 /**************************
2157 * CALCULATE INTERACTIONS *
2158 **************************/
2160 r31 = _mm_mul_ps(rsq31,rinv31);
2161 r31 = _mm_andnot_ps(dummy_mask,r31);
2163 /* Calculate table index by multiplying r with table scale and truncate to integer */
2164 rt = _mm_mul_ps(r31,vftabscale);
2165 vfitab = _mm_cvttps_epi32(rt);
2166 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2167 vfitab = _mm_slli_epi32(vfitab,2);
2169 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2170 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2171 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2172 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2173 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2174 _MM_TRANSPOSE4_PS(Y,F,G,H);
2175 Heps = _mm_mul_ps(vfeps,H);
2176 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2177 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2178 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
2182 fscal = _mm_andnot_ps(dummy_mask,fscal);
2184 /* Calculate temporary vectorial force */
2185 tx = _mm_mul_ps(fscal,dx31);
2186 ty = _mm_mul_ps(fscal,dy31);
2187 tz = _mm_mul_ps(fscal,dz31);
2189 /* Update vectorial force */
2190 fix3 = _mm_add_ps(fix3,tx);
2191 fiy3 = _mm_add_ps(fiy3,ty);
2192 fiz3 = _mm_add_ps(fiz3,tz);
2194 fjx1 = _mm_add_ps(fjx1,tx);
2195 fjy1 = _mm_add_ps(fjy1,ty);
2196 fjz1 = _mm_add_ps(fjz1,tz);
2198 /**************************
2199 * CALCULATE INTERACTIONS *
2200 **************************/
2202 r32 = _mm_mul_ps(rsq32,rinv32);
2203 r32 = _mm_andnot_ps(dummy_mask,r32);
2205 /* Calculate table index by multiplying r with table scale and truncate to integer */
2206 rt = _mm_mul_ps(r32,vftabscale);
2207 vfitab = _mm_cvttps_epi32(rt);
2208 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2209 vfitab = _mm_slli_epi32(vfitab,2);
2211 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2212 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2213 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2214 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2215 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2216 _MM_TRANSPOSE4_PS(Y,F,G,H);
2217 Heps = _mm_mul_ps(vfeps,H);
2218 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2219 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2220 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
2224 fscal = _mm_andnot_ps(dummy_mask,fscal);
2226 /* Calculate temporary vectorial force */
2227 tx = _mm_mul_ps(fscal,dx32);
2228 ty = _mm_mul_ps(fscal,dy32);
2229 tz = _mm_mul_ps(fscal,dz32);
2231 /* Update vectorial force */
2232 fix3 = _mm_add_ps(fix3,tx);
2233 fiy3 = _mm_add_ps(fiy3,ty);
2234 fiz3 = _mm_add_ps(fiz3,tz);
2236 fjx2 = _mm_add_ps(fjx2,tx);
2237 fjy2 = _mm_add_ps(fjy2,ty);
2238 fjz2 = _mm_add_ps(fjz2,tz);
2240 /**************************
2241 * CALCULATE INTERACTIONS *
2242 **************************/
2244 r33 = _mm_mul_ps(rsq33,rinv33);
2245 r33 = _mm_andnot_ps(dummy_mask,r33);
2247 /* Calculate table index by multiplying r with table scale and truncate to integer */
2248 rt = _mm_mul_ps(r33,vftabscale);
2249 vfitab = _mm_cvttps_epi32(rt);
2250 vfeps = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2251 vfitab = _mm_slli_epi32(vfitab,2);
2253 /* CUBIC SPLINE TABLE ELECTROSTATICS */
2254 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2255 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2256 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2257 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2258 _MM_TRANSPOSE4_PS(Y,F,G,H);
2259 Heps = _mm_mul_ps(vfeps,H);
2260 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2261 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2262 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
2266 fscal = _mm_andnot_ps(dummy_mask,fscal);
2268 /* Calculate temporary vectorial force */
2269 tx = _mm_mul_ps(fscal,dx33);
2270 ty = _mm_mul_ps(fscal,dy33);
2271 tz = _mm_mul_ps(fscal,dz33);
2273 /* Update vectorial force */
2274 fix3 = _mm_add_ps(fix3,tx);
2275 fiy3 = _mm_add_ps(fiy3,ty);
2276 fiz3 = _mm_add_ps(fiz3,tz);
2278 fjx3 = _mm_add_ps(fjx3,tx);
2279 fjy3 = _mm_add_ps(fjy3,ty);
2280 fjz3 = _mm_add_ps(fjz3,tz);
2282 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2283 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2284 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2285 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2287 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
2288 fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2290 /* Inner loop uses 360 flops */
2293 /* End of innermost loop */
2295 gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2296 f+i_coord_offset+DIM,fshift+i_shift_offset);
2298 /* Increment number of inner iterations */
2299 inneriter += j_index_end - j_index_start;
2301 /* Outer loop uses 18 flops */
2304 /* Increment number of outer iterations */
2307 /* Update outer/inner flops */
2309 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_F,outeriter*18 + inneriter*360);