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 sse4_1_single kernel generator.
42 #include "../nb_kernel.h"
43 #include "types/simple.h"
44 #include "gromacs/math/vec.h"
47 #include "gromacs/simd/math_x86_sse4_1_single.h"
48 #include "kernelutil_x86_sse4_1_single.h"
51 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_VF_sse4_1_single
52 * Electrostatics interaction: CubicSplineTable
53 * VdW interaction: CubicSplineTable
54 * Geometry: Water4-Particle
55 * Calculate force/pot: PotentialAndForce
58 nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_VF_sse4_1_single
59 (t_nblist * gmx_restrict nlist,
60 rvec * gmx_restrict xx,
61 rvec * gmx_restrict ff,
62 t_forcerec * gmx_restrict fr,
63 t_mdatoms * gmx_restrict mdatoms,
64 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
65 t_nrnb * gmx_restrict nrnb)
67 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
68 * just 0 for non-waters.
69 * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
70 * jnr indices corresponding to data put in the four positions in the SIMD register.
72 int i_shift_offset,i_coord_offset,outeriter,inneriter;
73 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
74 int jnrA,jnrB,jnrC,jnrD;
75 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
76 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
77 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
79 real *shiftvec,*fshift,*x,*f;
80 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
82 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
84 __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
86 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
88 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
90 __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
91 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
92 __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
93 __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
94 __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
95 __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
96 __m128 dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
97 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
100 __m128 rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
103 __m128 one_sixth = _mm_set1_ps(1.0/6.0);
104 __m128 one_twelfth = _mm_set1_ps(1.0/12.0);
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->epsfac);
125 charge = mdatoms->chargeA;
126 nvdwtype = fr->ntype;
128 vdwtype = mdatoms->typeA;
130 vftab = kernel_data->table_elec_vdw->data;
131 vftabscale = _mm_set1_ps(kernel_data->table_elec_vdw->scale);
133 /* Setup water-specific parameters */
134 inr = nlist->iinr[0];
135 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
136 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
137 iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
138 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
140 /* Avoid stupid compiler warnings */
141 jnrA = jnrB = jnrC = jnrD = 0;
150 for(iidx=0;iidx<4*DIM;iidx++)
155 /* Start outer loop over neighborlists */
156 for(iidx=0; iidx<nri; iidx++)
158 /* Load shift vector for this list */
159 i_shift_offset = DIM*shiftidx[iidx];
161 /* Load limits for loop over neighbors */
162 j_index_start = jindex[iidx];
163 j_index_end = jindex[iidx+1];
165 /* Get outer coordinate index */
167 i_coord_offset = DIM*inr;
169 /* Load i particle coords and add shift vector */
170 gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
171 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
173 fix0 = _mm_setzero_ps();
174 fiy0 = _mm_setzero_ps();
175 fiz0 = _mm_setzero_ps();
176 fix1 = _mm_setzero_ps();
177 fiy1 = _mm_setzero_ps();
178 fiz1 = _mm_setzero_ps();
179 fix2 = _mm_setzero_ps();
180 fiy2 = _mm_setzero_ps();
181 fiz2 = _mm_setzero_ps();
182 fix3 = _mm_setzero_ps();
183 fiy3 = _mm_setzero_ps();
184 fiz3 = _mm_setzero_ps();
186 /* Reset potential sums */
187 velecsum = _mm_setzero_ps();
188 vvdwsum = _mm_setzero_ps();
190 /* Start inner kernel loop */
191 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
194 /* Get j neighbor index, and coordinate index */
199 j_coord_offsetA = DIM*jnrA;
200 j_coord_offsetB = DIM*jnrB;
201 j_coord_offsetC = DIM*jnrC;
202 j_coord_offsetD = DIM*jnrD;
204 /* load j atom coordinates */
205 gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
206 x+j_coord_offsetC,x+j_coord_offsetD,
209 /* Calculate displacement vector */
210 dx00 = _mm_sub_ps(ix0,jx0);
211 dy00 = _mm_sub_ps(iy0,jy0);
212 dz00 = _mm_sub_ps(iz0,jz0);
213 dx10 = _mm_sub_ps(ix1,jx0);
214 dy10 = _mm_sub_ps(iy1,jy0);
215 dz10 = _mm_sub_ps(iz1,jz0);
216 dx20 = _mm_sub_ps(ix2,jx0);
217 dy20 = _mm_sub_ps(iy2,jy0);
218 dz20 = _mm_sub_ps(iz2,jz0);
219 dx30 = _mm_sub_ps(ix3,jx0);
220 dy30 = _mm_sub_ps(iy3,jy0);
221 dz30 = _mm_sub_ps(iz3,jz0);
223 /* Calculate squared distance and things based on it */
224 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
225 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
226 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
227 rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
229 rinv00 = gmx_mm_invsqrt_ps(rsq00);
230 rinv10 = gmx_mm_invsqrt_ps(rsq10);
231 rinv20 = gmx_mm_invsqrt_ps(rsq20);
232 rinv30 = gmx_mm_invsqrt_ps(rsq30);
234 /* Load parameters for j particles */
235 jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
236 charge+jnrC+0,charge+jnrD+0);
237 vdwjidx0A = 2*vdwtype[jnrA+0];
238 vdwjidx0B = 2*vdwtype[jnrB+0];
239 vdwjidx0C = 2*vdwtype[jnrC+0];
240 vdwjidx0D = 2*vdwtype[jnrD+0];
242 fjx0 = _mm_setzero_ps();
243 fjy0 = _mm_setzero_ps();
244 fjz0 = _mm_setzero_ps();
246 /**************************
247 * CALCULATE INTERACTIONS *
248 **************************/
250 r00 = _mm_mul_ps(rsq00,rinv00);
252 /* Compute parameters for interactions between i and j atoms */
253 gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
254 vdwparam+vdwioffset0+vdwjidx0B,
255 vdwparam+vdwioffset0+vdwjidx0C,
256 vdwparam+vdwioffset0+vdwjidx0D,
259 /* Calculate table index by multiplying r with table scale and truncate to integer */
260 rt = _mm_mul_ps(r00,vftabscale);
261 vfitab = _mm_cvttps_epi32(rt);
262 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
263 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
265 /* CUBIC SPLINE TABLE DISPERSION */
266 vfitab = _mm_add_epi32(vfitab,ifour);
267 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
268 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
269 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
270 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
271 _MM_TRANSPOSE4_PS(Y,F,G,H);
272 Heps = _mm_mul_ps(vfeps,H);
273 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
274 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
275 vvdw6 = _mm_mul_ps(c6_00,VV);
276 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
277 fvdw6 = _mm_mul_ps(c6_00,FF);
279 /* CUBIC SPLINE TABLE REPULSION */
280 vfitab = _mm_add_epi32(vfitab,ifour);
281 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
282 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
283 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
284 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
285 _MM_TRANSPOSE4_PS(Y,F,G,H);
286 Heps = _mm_mul_ps(vfeps,H);
287 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
288 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
289 vvdw12 = _mm_mul_ps(c12_00,VV);
290 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
291 fvdw12 = _mm_mul_ps(c12_00,FF);
292 vvdw = _mm_add_ps(vvdw12,vvdw6);
293 fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00)));
295 /* Update potential sum for this i atom from the interaction with this j atom. */
296 vvdwsum = _mm_add_ps(vvdwsum,vvdw);
300 /* Calculate temporary vectorial force */
301 tx = _mm_mul_ps(fscal,dx00);
302 ty = _mm_mul_ps(fscal,dy00);
303 tz = _mm_mul_ps(fscal,dz00);
305 /* Update vectorial force */
306 fix0 = _mm_add_ps(fix0,tx);
307 fiy0 = _mm_add_ps(fiy0,ty);
308 fiz0 = _mm_add_ps(fiz0,tz);
310 fjx0 = _mm_add_ps(fjx0,tx);
311 fjy0 = _mm_add_ps(fjy0,ty);
312 fjz0 = _mm_add_ps(fjz0,tz);
314 /**************************
315 * CALCULATE INTERACTIONS *
316 **************************/
318 r10 = _mm_mul_ps(rsq10,rinv10);
320 /* Compute parameters for interactions between i and j atoms */
321 qq10 = _mm_mul_ps(iq1,jq0);
323 /* Calculate table index by multiplying r with table scale and truncate to integer */
324 rt = _mm_mul_ps(r10,vftabscale);
325 vfitab = _mm_cvttps_epi32(rt);
326 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
327 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
329 /* CUBIC SPLINE TABLE ELECTROSTATICS */
330 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
331 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
332 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
333 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
334 _MM_TRANSPOSE4_PS(Y,F,G,H);
335 Heps = _mm_mul_ps(vfeps,H);
336 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
337 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
338 velec = _mm_mul_ps(qq10,VV);
339 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
340 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq10,FF),_mm_mul_ps(vftabscale,rinv10)));
342 /* Update potential sum for this i atom from the interaction with this j atom. */
343 velecsum = _mm_add_ps(velecsum,velec);
347 /* Calculate temporary vectorial force */
348 tx = _mm_mul_ps(fscal,dx10);
349 ty = _mm_mul_ps(fscal,dy10);
350 tz = _mm_mul_ps(fscal,dz10);
352 /* Update vectorial force */
353 fix1 = _mm_add_ps(fix1,tx);
354 fiy1 = _mm_add_ps(fiy1,ty);
355 fiz1 = _mm_add_ps(fiz1,tz);
357 fjx0 = _mm_add_ps(fjx0,tx);
358 fjy0 = _mm_add_ps(fjy0,ty);
359 fjz0 = _mm_add_ps(fjz0,tz);
361 /**************************
362 * CALCULATE INTERACTIONS *
363 **************************/
365 r20 = _mm_mul_ps(rsq20,rinv20);
367 /* Compute parameters for interactions between i and j atoms */
368 qq20 = _mm_mul_ps(iq2,jq0);
370 /* Calculate table index by multiplying r with table scale and truncate to integer */
371 rt = _mm_mul_ps(r20,vftabscale);
372 vfitab = _mm_cvttps_epi32(rt);
373 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
374 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
376 /* CUBIC SPLINE TABLE ELECTROSTATICS */
377 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
378 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
379 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
380 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
381 _MM_TRANSPOSE4_PS(Y,F,G,H);
382 Heps = _mm_mul_ps(vfeps,H);
383 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
384 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
385 velec = _mm_mul_ps(qq20,VV);
386 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
387 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq20,FF),_mm_mul_ps(vftabscale,rinv20)));
389 /* Update potential sum for this i atom from the interaction with this j atom. */
390 velecsum = _mm_add_ps(velecsum,velec);
394 /* Calculate temporary vectorial force */
395 tx = _mm_mul_ps(fscal,dx20);
396 ty = _mm_mul_ps(fscal,dy20);
397 tz = _mm_mul_ps(fscal,dz20);
399 /* Update vectorial force */
400 fix2 = _mm_add_ps(fix2,tx);
401 fiy2 = _mm_add_ps(fiy2,ty);
402 fiz2 = _mm_add_ps(fiz2,tz);
404 fjx0 = _mm_add_ps(fjx0,tx);
405 fjy0 = _mm_add_ps(fjy0,ty);
406 fjz0 = _mm_add_ps(fjz0,tz);
408 /**************************
409 * CALCULATE INTERACTIONS *
410 **************************/
412 r30 = _mm_mul_ps(rsq30,rinv30);
414 /* Compute parameters for interactions between i and j atoms */
415 qq30 = _mm_mul_ps(iq3,jq0);
417 /* Calculate table index by multiplying r with table scale and truncate to integer */
418 rt = _mm_mul_ps(r30,vftabscale);
419 vfitab = _mm_cvttps_epi32(rt);
420 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
421 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
423 /* CUBIC SPLINE TABLE ELECTROSTATICS */
424 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
425 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
426 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
427 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
428 _MM_TRANSPOSE4_PS(Y,F,G,H);
429 Heps = _mm_mul_ps(vfeps,H);
430 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
431 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
432 velec = _mm_mul_ps(qq30,VV);
433 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
434 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq30,FF),_mm_mul_ps(vftabscale,rinv30)));
436 /* Update potential sum for this i atom from the interaction with this j atom. */
437 velecsum = _mm_add_ps(velecsum,velec);
441 /* Calculate temporary vectorial force */
442 tx = _mm_mul_ps(fscal,dx30);
443 ty = _mm_mul_ps(fscal,dy30);
444 tz = _mm_mul_ps(fscal,dz30);
446 /* Update vectorial force */
447 fix3 = _mm_add_ps(fix3,tx);
448 fiy3 = _mm_add_ps(fiy3,ty);
449 fiz3 = _mm_add_ps(fiz3,tz);
451 fjx0 = _mm_add_ps(fjx0,tx);
452 fjy0 = _mm_add_ps(fjy0,ty);
453 fjz0 = _mm_add_ps(fjz0,tz);
455 fjptrA = f+j_coord_offsetA;
456 fjptrB = f+j_coord_offsetB;
457 fjptrC = f+j_coord_offsetC;
458 fjptrD = f+j_coord_offsetD;
460 gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
462 /* Inner loop uses 185 flops */
468 /* Get j neighbor index, and coordinate index */
469 jnrlistA = jjnr[jidx];
470 jnrlistB = jjnr[jidx+1];
471 jnrlistC = jjnr[jidx+2];
472 jnrlistD = jjnr[jidx+3];
473 /* Sign of each element will be negative for non-real atoms.
474 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
475 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
477 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
478 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
479 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
480 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
481 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
482 j_coord_offsetA = DIM*jnrA;
483 j_coord_offsetB = DIM*jnrB;
484 j_coord_offsetC = DIM*jnrC;
485 j_coord_offsetD = DIM*jnrD;
487 /* load j atom coordinates */
488 gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
489 x+j_coord_offsetC,x+j_coord_offsetD,
492 /* Calculate displacement vector */
493 dx00 = _mm_sub_ps(ix0,jx0);
494 dy00 = _mm_sub_ps(iy0,jy0);
495 dz00 = _mm_sub_ps(iz0,jz0);
496 dx10 = _mm_sub_ps(ix1,jx0);
497 dy10 = _mm_sub_ps(iy1,jy0);
498 dz10 = _mm_sub_ps(iz1,jz0);
499 dx20 = _mm_sub_ps(ix2,jx0);
500 dy20 = _mm_sub_ps(iy2,jy0);
501 dz20 = _mm_sub_ps(iz2,jz0);
502 dx30 = _mm_sub_ps(ix3,jx0);
503 dy30 = _mm_sub_ps(iy3,jy0);
504 dz30 = _mm_sub_ps(iz3,jz0);
506 /* Calculate squared distance and things based on it */
507 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
508 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
509 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
510 rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
512 rinv00 = gmx_mm_invsqrt_ps(rsq00);
513 rinv10 = gmx_mm_invsqrt_ps(rsq10);
514 rinv20 = gmx_mm_invsqrt_ps(rsq20);
515 rinv30 = gmx_mm_invsqrt_ps(rsq30);
517 /* Load parameters for j particles */
518 jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
519 charge+jnrC+0,charge+jnrD+0);
520 vdwjidx0A = 2*vdwtype[jnrA+0];
521 vdwjidx0B = 2*vdwtype[jnrB+0];
522 vdwjidx0C = 2*vdwtype[jnrC+0];
523 vdwjidx0D = 2*vdwtype[jnrD+0];
525 fjx0 = _mm_setzero_ps();
526 fjy0 = _mm_setzero_ps();
527 fjz0 = _mm_setzero_ps();
529 /**************************
530 * CALCULATE INTERACTIONS *
531 **************************/
533 r00 = _mm_mul_ps(rsq00,rinv00);
534 r00 = _mm_andnot_ps(dummy_mask,r00);
536 /* Compute parameters for interactions between i and j atoms */
537 gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
538 vdwparam+vdwioffset0+vdwjidx0B,
539 vdwparam+vdwioffset0+vdwjidx0C,
540 vdwparam+vdwioffset0+vdwjidx0D,
543 /* Calculate table index by multiplying r with table scale and truncate to integer */
544 rt = _mm_mul_ps(r00,vftabscale);
545 vfitab = _mm_cvttps_epi32(rt);
546 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
547 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
549 /* CUBIC SPLINE TABLE DISPERSION */
550 vfitab = _mm_add_epi32(vfitab,ifour);
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 vvdw6 = _mm_mul_ps(c6_00,VV);
560 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
561 fvdw6 = _mm_mul_ps(c6_00,FF);
563 /* CUBIC SPLINE TABLE REPULSION */
564 vfitab = _mm_add_epi32(vfitab,ifour);
565 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
566 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
567 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
568 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
569 _MM_TRANSPOSE4_PS(Y,F,G,H);
570 Heps = _mm_mul_ps(vfeps,H);
571 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
572 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
573 vvdw12 = _mm_mul_ps(c12_00,VV);
574 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
575 fvdw12 = _mm_mul_ps(c12_00,FF);
576 vvdw = _mm_add_ps(vvdw12,vvdw6);
577 fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00)));
579 /* Update potential sum for this i atom from the interaction with this j atom. */
580 vvdw = _mm_andnot_ps(dummy_mask,vvdw);
581 vvdwsum = _mm_add_ps(vvdwsum,vvdw);
585 fscal = _mm_andnot_ps(dummy_mask,fscal);
587 /* Calculate temporary vectorial force */
588 tx = _mm_mul_ps(fscal,dx00);
589 ty = _mm_mul_ps(fscal,dy00);
590 tz = _mm_mul_ps(fscal,dz00);
592 /* Update vectorial force */
593 fix0 = _mm_add_ps(fix0,tx);
594 fiy0 = _mm_add_ps(fiy0,ty);
595 fiz0 = _mm_add_ps(fiz0,tz);
597 fjx0 = _mm_add_ps(fjx0,tx);
598 fjy0 = _mm_add_ps(fjy0,ty);
599 fjz0 = _mm_add_ps(fjz0,tz);
601 /**************************
602 * CALCULATE INTERACTIONS *
603 **************************/
605 r10 = _mm_mul_ps(rsq10,rinv10);
606 r10 = _mm_andnot_ps(dummy_mask,r10);
608 /* Compute parameters for interactions between i and j atoms */
609 qq10 = _mm_mul_ps(iq1,jq0);
611 /* Calculate table index by multiplying r with table scale and truncate to integer */
612 rt = _mm_mul_ps(r10,vftabscale);
613 vfitab = _mm_cvttps_epi32(rt);
614 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
615 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
617 /* CUBIC SPLINE TABLE ELECTROSTATICS */
618 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
619 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
620 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
621 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
622 _MM_TRANSPOSE4_PS(Y,F,G,H);
623 Heps = _mm_mul_ps(vfeps,H);
624 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
625 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
626 velec = _mm_mul_ps(qq10,VV);
627 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
628 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq10,FF),_mm_mul_ps(vftabscale,rinv10)));
630 /* Update potential sum for this i atom from the interaction with this j atom. */
631 velec = _mm_andnot_ps(dummy_mask,velec);
632 velecsum = _mm_add_ps(velecsum,velec);
636 fscal = _mm_andnot_ps(dummy_mask,fscal);
638 /* Calculate temporary vectorial force */
639 tx = _mm_mul_ps(fscal,dx10);
640 ty = _mm_mul_ps(fscal,dy10);
641 tz = _mm_mul_ps(fscal,dz10);
643 /* Update vectorial force */
644 fix1 = _mm_add_ps(fix1,tx);
645 fiy1 = _mm_add_ps(fiy1,ty);
646 fiz1 = _mm_add_ps(fiz1,tz);
648 fjx0 = _mm_add_ps(fjx0,tx);
649 fjy0 = _mm_add_ps(fjy0,ty);
650 fjz0 = _mm_add_ps(fjz0,tz);
652 /**************************
653 * CALCULATE INTERACTIONS *
654 **************************/
656 r20 = _mm_mul_ps(rsq20,rinv20);
657 r20 = _mm_andnot_ps(dummy_mask,r20);
659 /* Compute parameters for interactions between i and j atoms */
660 qq20 = _mm_mul_ps(iq2,jq0);
662 /* Calculate table index by multiplying r with table scale and truncate to integer */
663 rt = _mm_mul_ps(r20,vftabscale);
664 vfitab = _mm_cvttps_epi32(rt);
665 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
666 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
668 /* CUBIC SPLINE TABLE ELECTROSTATICS */
669 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
670 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
671 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
672 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
673 _MM_TRANSPOSE4_PS(Y,F,G,H);
674 Heps = _mm_mul_ps(vfeps,H);
675 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
676 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
677 velec = _mm_mul_ps(qq20,VV);
678 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
679 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq20,FF),_mm_mul_ps(vftabscale,rinv20)));
681 /* Update potential sum for this i atom from the interaction with this j atom. */
682 velec = _mm_andnot_ps(dummy_mask,velec);
683 velecsum = _mm_add_ps(velecsum,velec);
687 fscal = _mm_andnot_ps(dummy_mask,fscal);
689 /* Calculate temporary vectorial force */
690 tx = _mm_mul_ps(fscal,dx20);
691 ty = _mm_mul_ps(fscal,dy20);
692 tz = _mm_mul_ps(fscal,dz20);
694 /* Update vectorial force */
695 fix2 = _mm_add_ps(fix2,tx);
696 fiy2 = _mm_add_ps(fiy2,ty);
697 fiz2 = _mm_add_ps(fiz2,tz);
699 fjx0 = _mm_add_ps(fjx0,tx);
700 fjy0 = _mm_add_ps(fjy0,ty);
701 fjz0 = _mm_add_ps(fjz0,tz);
703 /**************************
704 * CALCULATE INTERACTIONS *
705 **************************/
707 r30 = _mm_mul_ps(rsq30,rinv30);
708 r30 = _mm_andnot_ps(dummy_mask,r30);
710 /* Compute parameters for interactions between i and j atoms */
711 qq30 = _mm_mul_ps(iq3,jq0);
713 /* Calculate table index by multiplying r with table scale and truncate to integer */
714 rt = _mm_mul_ps(r30,vftabscale);
715 vfitab = _mm_cvttps_epi32(rt);
716 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
717 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
719 /* CUBIC SPLINE TABLE ELECTROSTATICS */
720 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
721 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
722 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
723 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
724 _MM_TRANSPOSE4_PS(Y,F,G,H);
725 Heps = _mm_mul_ps(vfeps,H);
726 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
727 VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
728 velec = _mm_mul_ps(qq30,VV);
729 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
730 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq30,FF),_mm_mul_ps(vftabscale,rinv30)));
732 /* Update potential sum for this i atom from the interaction with this j atom. */
733 velec = _mm_andnot_ps(dummy_mask,velec);
734 velecsum = _mm_add_ps(velecsum,velec);
738 fscal = _mm_andnot_ps(dummy_mask,fscal);
740 /* Calculate temporary vectorial force */
741 tx = _mm_mul_ps(fscal,dx30);
742 ty = _mm_mul_ps(fscal,dy30);
743 tz = _mm_mul_ps(fscal,dz30);
745 /* Update vectorial force */
746 fix3 = _mm_add_ps(fix3,tx);
747 fiy3 = _mm_add_ps(fiy3,ty);
748 fiz3 = _mm_add_ps(fiz3,tz);
750 fjx0 = _mm_add_ps(fjx0,tx);
751 fjy0 = _mm_add_ps(fjy0,ty);
752 fjz0 = _mm_add_ps(fjz0,tz);
754 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
755 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
756 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
757 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
759 gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
761 /* Inner loop uses 189 flops */
764 /* End of innermost loop */
766 gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
767 f+i_coord_offset,fshift+i_shift_offset);
770 /* Update potential energies */
771 gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
772 gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
774 /* Increment number of inner iterations */
775 inneriter += j_index_end - j_index_start;
777 /* Outer loop uses 26 flops */
780 /* Increment number of outer iterations */
783 /* Update outer/inner flops */
785 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*26 + inneriter*189);
788 * Gromacs nonbonded kernel: nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_F_sse4_1_single
789 * Electrostatics interaction: CubicSplineTable
790 * VdW interaction: CubicSplineTable
791 * Geometry: Water4-Particle
792 * Calculate force/pot: Force
795 nb_kernel_ElecCSTab_VdwCSTab_GeomW4P1_F_sse4_1_single
796 (t_nblist * gmx_restrict nlist,
797 rvec * gmx_restrict xx,
798 rvec * gmx_restrict ff,
799 t_forcerec * gmx_restrict fr,
800 t_mdatoms * gmx_restrict mdatoms,
801 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
802 t_nrnb * gmx_restrict nrnb)
804 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
805 * just 0 for non-waters.
806 * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
807 * jnr indices corresponding to data put in the four positions in the SIMD register.
809 int i_shift_offset,i_coord_offset,outeriter,inneriter;
810 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
811 int jnrA,jnrB,jnrC,jnrD;
812 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
813 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
814 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
816 real *shiftvec,*fshift,*x,*f;
817 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
819 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
821 __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
823 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
825 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
827 __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
828 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
829 __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
830 __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
831 __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
832 __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
833 __m128 dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
834 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
837 __m128 rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
840 __m128 one_sixth = _mm_set1_ps(1.0/6.0);
841 __m128 one_twelfth = _mm_set1_ps(1.0/12.0);
843 __m128i ifour = _mm_set1_epi32(4);
844 __m128 rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
846 __m128 dummy_mask,cutoff_mask;
847 __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
848 __m128 one = _mm_set1_ps(1.0);
849 __m128 two = _mm_set1_ps(2.0);
855 jindex = nlist->jindex;
857 shiftidx = nlist->shift;
859 shiftvec = fr->shift_vec[0];
860 fshift = fr->fshift[0];
861 facel = _mm_set1_ps(fr->epsfac);
862 charge = mdatoms->chargeA;
863 nvdwtype = fr->ntype;
865 vdwtype = mdatoms->typeA;
867 vftab = kernel_data->table_elec_vdw->data;
868 vftabscale = _mm_set1_ps(kernel_data->table_elec_vdw->scale);
870 /* Setup water-specific parameters */
871 inr = nlist->iinr[0];
872 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
873 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
874 iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
875 vdwioffset0 = 2*nvdwtype*vdwtype[inr+0];
877 /* Avoid stupid compiler warnings */
878 jnrA = jnrB = jnrC = jnrD = 0;
887 for(iidx=0;iidx<4*DIM;iidx++)
892 /* Start outer loop over neighborlists */
893 for(iidx=0; iidx<nri; iidx++)
895 /* Load shift vector for this list */
896 i_shift_offset = DIM*shiftidx[iidx];
898 /* Load limits for loop over neighbors */
899 j_index_start = jindex[iidx];
900 j_index_end = jindex[iidx+1];
902 /* Get outer coordinate index */
904 i_coord_offset = DIM*inr;
906 /* Load i particle coords and add shift vector */
907 gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
908 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
910 fix0 = _mm_setzero_ps();
911 fiy0 = _mm_setzero_ps();
912 fiz0 = _mm_setzero_ps();
913 fix1 = _mm_setzero_ps();
914 fiy1 = _mm_setzero_ps();
915 fiz1 = _mm_setzero_ps();
916 fix2 = _mm_setzero_ps();
917 fiy2 = _mm_setzero_ps();
918 fiz2 = _mm_setzero_ps();
919 fix3 = _mm_setzero_ps();
920 fiy3 = _mm_setzero_ps();
921 fiz3 = _mm_setzero_ps();
923 /* Start inner kernel loop */
924 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
927 /* Get j neighbor index, and coordinate index */
932 j_coord_offsetA = DIM*jnrA;
933 j_coord_offsetB = DIM*jnrB;
934 j_coord_offsetC = DIM*jnrC;
935 j_coord_offsetD = DIM*jnrD;
937 /* load j atom coordinates */
938 gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
939 x+j_coord_offsetC,x+j_coord_offsetD,
942 /* Calculate displacement vector */
943 dx00 = _mm_sub_ps(ix0,jx0);
944 dy00 = _mm_sub_ps(iy0,jy0);
945 dz00 = _mm_sub_ps(iz0,jz0);
946 dx10 = _mm_sub_ps(ix1,jx0);
947 dy10 = _mm_sub_ps(iy1,jy0);
948 dz10 = _mm_sub_ps(iz1,jz0);
949 dx20 = _mm_sub_ps(ix2,jx0);
950 dy20 = _mm_sub_ps(iy2,jy0);
951 dz20 = _mm_sub_ps(iz2,jz0);
952 dx30 = _mm_sub_ps(ix3,jx0);
953 dy30 = _mm_sub_ps(iy3,jy0);
954 dz30 = _mm_sub_ps(iz3,jz0);
956 /* Calculate squared distance and things based on it */
957 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
958 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
959 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
960 rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
962 rinv00 = gmx_mm_invsqrt_ps(rsq00);
963 rinv10 = gmx_mm_invsqrt_ps(rsq10);
964 rinv20 = gmx_mm_invsqrt_ps(rsq20);
965 rinv30 = gmx_mm_invsqrt_ps(rsq30);
967 /* Load parameters for j particles */
968 jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
969 charge+jnrC+0,charge+jnrD+0);
970 vdwjidx0A = 2*vdwtype[jnrA+0];
971 vdwjidx0B = 2*vdwtype[jnrB+0];
972 vdwjidx0C = 2*vdwtype[jnrC+0];
973 vdwjidx0D = 2*vdwtype[jnrD+0];
975 fjx0 = _mm_setzero_ps();
976 fjy0 = _mm_setzero_ps();
977 fjz0 = _mm_setzero_ps();
979 /**************************
980 * CALCULATE INTERACTIONS *
981 **************************/
983 r00 = _mm_mul_ps(rsq00,rinv00);
985 /* Compute parameters for interactions between i and j atoms */
986 gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
987 vdwparam+vdwioffset0+vdwjidx0B,
988 vdwparam+vdwioffset0+vdwjidx0C,
989 vdwparam+vdwioffset0+vdwjidx0D,
992 /* Calculate table index by multiplying r with table scale and truncate to integer */
993 rt = _mm_mul_ps(r00,vftabscale);
994 vfitab = _mm_cvttps_epi32(rt);
995 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
996 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
998 /* CUBIC SPLINE TABLE DISPERSION */
999 vfitab = _mm_add_epi32(vfitab,ifour);
1000 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1001 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1002 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1003 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1004 _MM_TRANSPOSE4_PS(Y,F,G,H);
1005 Heps = _mm_mul_ps(vfeps,H);
1006 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1007 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1008 fvdw6 = _mm_mul_ps(c6_00,FF);
1010 /* CUBIC SPLINE TABLE REPULSION */
1011 vfitab = _mm_add_epi32(vfitab,ifour);
1012 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1013 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1014 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1015 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1016 _MM_TRANSPOSE4_PS(Y,F,G,H);
1017 Heps = _mm_mul_ps(vfeps,H);
1018 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1019 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1020 fvdw12 = _mm_mul_ps(c12_00,FF);
1021 fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00)));
1025 /* Calculate temporary vectorial force */
1026 tx = _mm_mul_ps(fscal,dx00);
1027 ty = _mm_mul_ps(fscal,dy00);
1028 tz = _mm_mul_ps(fscal,dz00);
1030 /* Update vectorial force */
1031 fix0 = _mm_add_ps(fix0,tx);
1032 fiy0 = _mm_add_ps(fiy0,ty);
1033 fiz0 = _mm_add_ps(fiz0,tz);
1035 fjx0 = _mm_add_ps(fjx0,tx);
1036 fjy0 = _mm_add_ps(fjy0,ty);
1037 fjz0 = _mm_add_ps(fjz0,tz);
1039 /**************************
1040 * CALCULATE INTERACTIONS *
1041 **************************/
1043 r10 = _mm_mul_ps(rsq10,rinv10);
1045 /* Compute parameters for interactions between i and j atoms */
1046 qq10 = _mm_mul_ps(iq1,jq0);
1048 /* Calculate table index by multiplying r with table scale and truncate to integer */
1049 rt = _mm_mul_ps(r10,vftabscale);
1050 vfitab = _mm_cvttps_epi32(rt);
1051 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1052 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1054 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1055 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1056 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1057 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1058 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1059 _MM_TRANSPOSE4_PS(Y,F,G,H);
1060 Heps = _mm_mul_ps(vfeps,H);
1061 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1062 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1063 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq10,FF),_mm_mul_ps(vftabscale,rinv10)));
1067 /* Calculate temporary vectorial force */
1068 tx = _mm_mul_ps(fscal,dx10);
1069 ty = _mm_mul_ps(fscal,dy10);
1070 tz = _mm_mul_ps(fscal,dz10);
1072 /* Update vectorial force */
1073 fix1 = _mm_add_ps(fix1,tx);
1074 fiy1 = _mm_add_ps(fiy1,ty);
1075 fiz1 = _mm_add_ps(fiz1,tz);
1077 fjx0 = _mm_add_ps(fjx0,tx);
1078 fjy0 = _mm_add_ps(fjy0,ty);
1079 fjz0 = _mm_add_ps(fjz0,tz);
1081 /**************************
1082 * CALCULATE INTERACTIONS *
1083 **************************/
1085 r20 = _mm_mul_ps(rsq20,rinv20);
1087 /* Compute parameters for interactions between i and j atoms */
1088 qq20 = _mm_mul_ps(iq2,jq0);
1090 /* Calculate table index by multiplying r with table scale and truncate to integer */
1091 rt = _mm_mul_ps(r20,vftabscale);
1092 vfitab = _mm_cvttps_epi32(rt);
1093 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1094 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1096 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1097 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1098 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1099 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1100 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1101 _MM_TRANSPOSE4_PS(Y,F,G,H);
1102 Heps = _mm_mul_ps(vfeps,H);
1103 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1104 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1105 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq20,FF),_mm_mul_ps(vftabscale,rinv20)));
1109 /* Calculate temporary vectorial force */
1110 tx = _mm_mul_ps(fscal,dx20);
1111 ty = _mm_mul_ps(fscal,dy20);
1112 tz = _mm_mul_ps(fscal,dz20);
1114 /* Update vectorial force */
1115 fix2 = _mm_add_ps(fix2,tx);
1116 fiy2 = _mm_add_ps(fiy2,ty);
1117 fiz2 = _mm_add_ps(fiz2,tz);
1119 fjx0 = _mm_add_ps(fjx0,tx);
1120 fjy0 = _mm_add_ps(fjy0,ty);
1121 fjz0 = _mm_add_ps(fjz0,tz);
1123 /**************************
1124 * CALCULATE INTERACTIONS *
1125 **************************/
1127 r30 = _mm_mul_ps(rsq30,rinv30);
1129 /* Compute parameters for interactions between i and j atoms */
1130 qq30 = _mm_mul_ps(iq3,jq0);
1132 /* Calculate table index by multiplying r with table scale and truncate to integer */
1133 rt = _mm_mul_ps(r30,vftabscale);
1134 vfitab = _mm_cvttps_epi32(rt);
1135 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1136 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1138 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1139 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1140 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1141 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1142 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1143 _MM_TRANSPOSE4_PS(Y,F,G,H);
1144 Heps = _mm_mul_ps(vfeps,H);
1145 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1146 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1147 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq30,FF),_mm_mul_ps(vftabscale,rinv30)));
1151 /* Calculate temporary vectorial force */
1152 tx = _mm_mul_ps(fscal,dx30);
1153 ty = _mm_mul_ps(fscal,dy30);
1154 tz = _mm_mul_ps(fscal,dz30);
1156 /* Update vectorial force */
1157 fix3 = _mm_add_ps(fix3,tx);
1158 fiy3 = _mm_add_ps(fiy3,ty);
1159 fiz3 = _mm_add_ps(fiz3,tz);
1161 fjx0 = _mm_add_ps(fjx0,tx);
1162 fjy0 = _mm_add_ps(fjy0,ty);
1163 fjz0 = _mm_add_ps(fjz0,tz);
1165 fjptrA = f+j_coord_offsetA;
1166 fjptrB = f+j_coord_offsetB;
1167 fjptrC = f+j_coord_offsetC;
1168 fjptrD = f+j_coord_offsetD;
1170 gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
1172 /* Inner loop uses 165 flops */
1175 if(jidx<j_index_end)
1178 /* Get j neighbor index, and coordinate index */
1179 jnrlistA = jjnr[jidx];
1180 jnrlistB = jjnr[jidx+1];
1181 jnrlistC = jjnr[jidx+2];
1182 jnrlistD = jjnr[jidx+3];
1183 /* Sign of each element will be negative for non-real atoms.
1184 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1185 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
1187 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1188 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
1189 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
1190 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
1191 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
1192 j_coord_offsetA = DIM*jnrA;
1193 j_coord_offsetB = DIM*jnrB;
1194 j_coord_offsetC = DIM*jnrC;
1195 j_coord_offsetD = DIM*jnrD;
1197 /* load j atom coordinates */
1198 gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1199 x+j_coord_offsetC,x+j_coord_offsetD,
1202 /* Calculate displacement vector */
1203 dx00 = _mm_sub_ps(ix0,jx0);
1204 dy00 = _mm_sub_ps(iy0,jy0);
1205 dz00 = _mm_sub_ps(iz0,jz0);
1206 dx10 = _mm_sub_ps(ix1,jx0);
1207 dy10 = _mm_sub_ps(iy1,jy0);
1208 dz10 = _mm_sub_ps(iz1,jz0);
1209 dx20 = _mm_sub_ps(ix2,jx0);
1210 dy20 = _mm_sub_ps(iy2,jy0);
1211 dz20 = _mm_sub_ps(iz2,jz0);
1212 dx30 = _mm_sub_ps(ix3,jx0);
1213 dy30 = _mm_sub_ps(iy3,jy0);
1214 dz30 = _mm_sub_ps(iz3,jz0);
1216 /* Calculate squared distance and things based on it */
1217 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
1218 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
1219 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
1220 rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
1222 rinv00 = gmx_mm_invsqrt_ps(rsq00);
1223 rinv10 = gmx_mm_invsqrt_ps(rsq10);
1224 rinv20 = gmx_mm_invsqrt_ps(rsq20);
1225 rinv30 = gmx_mm_invsqrt_ps(rsq30);
1227 /* Load parameters for j particles */
1228 jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
1229 charge+jnrC+0,charge+jnrD+0);
1230 vdwjidx0A = 2*vdwtype[jnrA+0];
1231 vdwjidx0B = 2*vdwtype[jnrB+0];
1232 vdwjidx0C = 2*vdwtype[jnrC+0];
1233 vdwjidx0D = 2*vdwtype[jnrD+0];
1235 fjx0 = _mm_setzero_ps();
1236 fjy0 = _mm_setzero_ps();
1237 fjz0 = _mm_setzero_ps();
1239 /**************************
1240 * CALCULATE INTERACTIONS *
1241 **************************/
1243 r00 = _mm_mul_ps(rsq00,rinv00);
1244 r00 = _mm_andnot_ps(dummy_mask,r00);
1246 /* Compute parameters for interactions between i and j atoms */
1247 gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
1248 vdwparam+vdwioffset0+vdwjidx0B,
1249 vdwparam+vdwioffset0+vdwjidx0C,
1250 vdwparam+vdwioffset0+vdwjidx0D,
1253 /* Calculate table index by multiplying r with table scale and truncate to integer */
1254 rt = _mm_mul_ps(r00,vftabscale);
1255 vfitab = _mm_cvttps_epi32(rt);
1256 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1257 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1259 /* CUBIC SPLINE TABLE DISPERSION */
1260 vfitab = _mm_add_epi32(vfitab,ifour);
1261 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1262 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1263 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1264 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1265 _MM_TRANSPOSE4_PS(Y,F,G,H);
1266 Heps = _mm_mul_ps(vfeps,H);
1267 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1268 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1269 fvdw6 = _mm_mul_ps(c6_00,FF);
1271 /* CUBIC SPLINE TABLE REPULSION */
1272 vfitab = _mm_add_epi32(vfitab,ifour);
1273 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1274 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1275 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1276 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1277 _MM_TRANSPOSE4_PS(Y,F,G,H);
1278 Heps = _mm_mul_ps(vfeps,H);
1279 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1280 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1281 fvdw12 = _mm_mul_ps(c12_00,FF);
1282 fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00)));
1286 fscal = _mm_andnot_ps(dummy_mask,fscal);
1288 /* Calculate temporary vectorial force */
1289 tx = _mm_mul_ps(fscal,dx00);
1290 ty = _mm_mul_ps(fscal,dy00);
1291 tz = _mm_mul_ps(fscal,dz00);
1293 /* Update vectorial force */
1294 fix0 = _mm_add_ps(fix0,tx);
1295 fiy0 = _mm_add_ps(fiy0,ty);
1296 fiz0 = _mm_add_ps(fiz0,tz);
1298 fjx0 = _mm_add_ps(fjx0,tx);
1299 fjy0 = _mm_add_ps(fjy0,ty);
1300 fjz0 = _mm_add_ps(fjz0,tz);
1302 /**************************
1303 * CALCULATE INTERACTIONS *
1304 **************************/
1306 r10 = _mm_mul_ps(rsq10,rinv10);
1307 r10 = _mm_andnot_ps(dummy_mask,r10);
1309 /* Compute parameters for interactions between i and j atoms */
1310 qq10 = _mm_mul_ps(iq1,jq0);
1312 /* Calculate table index by multiplying r with table scale and truncate to integer */
1313 rt = _mm_mul_ps(r10,vftabscale);
1314 vfitab = _mm_cvttps_epi32(rt);
1315 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1316 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1318 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1319 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1320 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1321 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1322 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1323 _MM_TRANSPOSE4_PS(Y,F,G,H);
1324 Heps = _mm_mul_ps(vfeps,H);
1325 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1326 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1327 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq10,FF),_mm_mul_ps(vftabscale,rinv10)));
1331 fscal = _mm_andnot_ps(dummy_mask,fscal);
1333 /* Calculate temporary vectorial force */
1334 tx = _mm_mul_ps(fscal,dx10);
1335 ty = _mm_mul_ps(fscal,dy10);
1336 tz = _mm_mul_ps(fscal,dz10);
1338 /* Update vectorial force */
1339 fix1 = _mm_add_ps(fix1,tx);
1340 fiy1 = _mm_add_ps(fiy1,ty);
1341 fiz1 = _mm_add_ps(fiz1,tz);
1343 fjx0 = _mm_add_ps(fjx0,tx);
1344 fjy0 = _mm_add_ps(fjy0,ty);
1345 fjz0 = _mm_add_ps(fjz0,tz);
1347 /**************************
1348 * CALCULATE INTERACTIONS *
1349 **************************/
1351 r20 = _mm_mul_ps(rsq20,rinv20);
1352 r20 = _mm_andnot_ps(dummy_mask,r20);
1354 /* Compute parameters for interactions between i and j atoms */
1355 qq20 = _mm_mul_ps(iq2,jq0);
1357 /* Calculate table index by multiplying r with table scale and truncate to integer */
1358 rt = _mm_mul_ps(r20,vftabscale);
1359 vfitab = _mm_cvttps_epi32(rt);
1360 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1361 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1363 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1364 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1365 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1366 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1367 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1368 _MM_TRANSPOSE4_PS(Y,F,G,H);
1369 Heps = _mm_mul_ps(vfeps,H);
1370 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1371 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1372 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq20,FF),_mm_mul_ps(vftabscale,rinv20)));
1376 fscal = _mm_andnot_ps(dummy_mask,fscal);
1378 /* Calculate temporary vectorial force */
1379 tx = _mm_mul_ps(fscal,dx20);
1380 ty = _mm_mul_ps(fscal,dy20);
1381 tz = _mm_mul_ps(fscal,dz20);
1383 /* Update vectorial force */
1384 fix2 = _mm_add_ps(fix2,tx);
1385 fiy2 = _mm_add_ps(fiy2,ty);
1386 fiz2 = _mm_add_ps(fiz2,tz);
1388 fjx0 = _mm_add_ps(fjx0,tx);
1389 fjy0 = _mm_add_ps(fjy0,ty);
1390 fjz0 = _mm_add_ps(fjz0,tz);
1392 /**************************
1393 * CALCULATE INTERACTIONS *
1394 **************************/
1396 r30 = _mm_mul_ps(rsq30,rinv30);
1397 r30 = _mm_andnot_ps(dummy_mask,r30);
1399 /* Compute parameters for interactions between i and j atoms */
1400 qq30 = _mm_mul_ps(iq3,jq0);
1402 /* Calculate table index by multiplying r with table scale and truncate to integer */
1403 rt = _mm_mul_ps(r30,vftabscale);
1404 vfitab = _mm_cvttps_epi32(rt);
1405 vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1406 vfitab = _mm_slli_epi32(_mm_add_epi32(vfitab,_mm_slli_epi32(vfitab,1)),2);
1408 /* CUBIC SPLINE TABLE ELECTROSTATICS */
1409 Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1410 F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1411 G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1412 H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1413 _MM_TRANSPOSE4_PS(Y,F,G,H);
1414 Heps = _mm_mul_ps(vfeps,H);
1415 Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1416 FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1417 felec = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq30,FF),_mm_mul_ps(vftabscale,rinv30)));
1421 fscal = _mm_andnot_ps(dummy_mask,fscal);
1423 /* Calculate temporary vectorial force */
1424 tx = _mm_mul_ps(fscal,dx30);
1425 ty = _mm_mul_ps(fscal,dy30);
1426 tz = _mm_mul_ps(fscal,dz30);
1428 /* Update vectorial force */
1429 fix3 = _mm_add_ps(fix3,tx);
1430 fiy3 = _mm_add_ps(fiy3,ty);
1431 fiz3 = _mm_add_ps(fiz3,tz);
1433 fjx0 = _mm_add_ps(fjx0,tx);
1434 fjy0 = _mm_add_ps(fjy0,ty);
1435 fjz0 = _mm_add_ps(fjz0,tz);
1437 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1438 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1439 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1440 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1442 gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
1444 /* Inner loop uses 169 flops */
1447 /* End of innermost loop */
1449 gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1450 f+i_coord_offset,fshift+i_shift_offset);
1452 /* Increment number of inner iterations */
1453 inneriter += j_index_end - j_index_start;
1455 /* Outer loop uses 24 flops */
1458 /* Increment number of outer iterations */
1461 /* Update outer/inner flops */
1463 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*24 + inneriter*169);