?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Removed simple.h from nb_kernel_sse2_XX
[gromacs.git] / src / gromacs / gmxlib / nonbonded / nb_kernel_sse2_single / nb_kernel_ElecEwSh_VdwNone_GeomW3W3_sse2_single.c
blob08d154ef82b988bda8103096529243f80376d51c
1 /*
2 * This file is part of the GROMACS molecular simulation package.
3 *
4 * Copyright (c) 2012,2013,2014,2015, 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.
8 *
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.
13 *
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.
18 *
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.
23 *
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.
31 *
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.
34 */
35 /*
36 * Note: this file was generated by the GROMACS sse2_single kernel generator.
37 */
38 #include "gmxpre.h"
39
40 #include "config.h"
41
42 #include <math.h>
43
44 #include "../nb_kernel.h"
45 #include "gromacs/math/vec.h"
46 #include "gromacs/legacyheaders/nrnb.h"
47
48 #include "gromacs/simd/math_x86_sse2_single.h"
49 #include "kernelutil_x86_sse2_single.h"
50
51 /*
52 * Gromacs nonbonded kernel: nb_kernel_ElecEwSh_VdwNone_GeomW3W3_VF_sse2_single
53 * Electrostatics interaction: Ewald
54 * VdW interaction: None
55 * Geometry: Water3-Water3
56 * Calculate force/pot: PotentialAndForce
57 */
58 void
59 nb_kernel_ElecEwSh_VdwNone_GeomW3W3_VF_sse2_single
60 (t_nblist * gmx_restrict nlist,
61 rvec * gmx_restrict xx,
62 rvec * gmx_restrict ff,
63 t_forcerec * gmx_restrict fr,
64 t_mdatoms * gmx_restrict mdatoms,
65 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
66 t_nrnb * gmx_restrict nrnb)
67 {
68 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
69 * just 0 for non-waters.
70 * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
71 * jnr indices corresponding to data put in the four positions in the SIMD register.
72 */
73 int i_shift_offset,i_coord_offset,outeriter,inneriter;
74 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
75 int jnrA,jnrB,jnrC,jnrD;
76 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
77 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
78 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
79 real rcutoff_scalar;
80 real *shiftvec,*fshift,*x,*f;
81 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
82 real scratch[4*DIM];
83 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
84 int vdwioffset0;
85 __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
86 int vdwioffset1;
87 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
88 int vdwioffset2;
89 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
90 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
91 __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
92 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
93 __m128 jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
94 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
95 __m128 jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
96 __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
97 __m128 dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
98 __m128 dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
99 __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
100 __m128 dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
101 __m128 dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
102 __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
103 __m128 dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
104 __m128 dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
105 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
106 real *charge;
107 __m128i ewitab;
108 __m128 ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
109 real *ewtab;
110 __m128 dummy_mask,cutoff_mask;
111 __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
112 __m128 one = _mm_set1_ps(1.0);
113 __m128 two = _mm_set1_ps(2.0);
114 x = xx[0];
115 f = ff[0];
116
117 nri = nlist->nri;
118 iinr = nlist->iinr;
119 jindex = nlist->jindex;
120 jjnr = nlist->jjnr;
121 shiftidx = nlist->shift;
122 gid = nlist->gid;
123 shiftvec = fr->shift_vec[0];
124 fshift = fr->fshift[0];
125 facel = _mm_set1_ps(fr->epsfac);
126 charge = mdatoms->chargeA;
127
128 sh_ewald = _mm_set1_ps(fr->ic->sh_ewald);
129 ewtab = fr->ic->tabq_coul_FDV0;
130 ewtabscale = _mm_set1_ps(fr->ic->tabq_scale);
131 ewtabhalfspace = _mm_set1_ps(0.5/fr->ic->tabq_scale);
132
133 /* Setup water-specific parameters */
134 inr = nlist->iinr[0];
135 iq0 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+0]));
136 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
137 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
138
139 jq0 = _mm_set1_ps(charge[inr+0]);
140 jq1 = _mm_set1_ps(charge[inr+1]);
141 jq2 = _mm_set1_ps(charge[inr+2]);
142 qq00 = _mm_mul_ps(iq0,jq0);
143 qq01 = _mm_mul_ps(iq0,jq1);
144 qq02 = _mm_mul_ps(iq0,jq2);
145 qq10 = _mm_mul_ps(iq1,jq0);
146 qq11 = _mm_mul_ps(iq1,jq1);
147 qq12 = _mm_mul_ps(iq1,jq2);
148 qq20 = _mm_mul_ps(iq2,jq0);
149 qq21 = _mm_mul_ps(iq2,jq1);
150 qq22 = _mm_mul_ps(iq2,jq2);
151
152 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
153 rcutoff_scalar = fr->rcoulomb;
154 rcutoff = _mm_set1_ps(rcutoff_scalar);
155 rcutoff2 = _mm_mul_ps(rcutoff,rcutoff);
156
157 /* Avoid stupid compiler warnings */
158 jnrA = jnrB = jnrC = jnrD = 0;
159 j_coord_offsetA = 0;
160 j_coord_offsetB = 0;
161 j_coord_offsetC = 0;
162 j_coord_offsetD = 0;
163
164 outeriter = 0;
165 inneriter = 0;
166
167 for(iidx=0;iidx<4*DIM;iidx++)
168 {
169 scratch[iidx] = 0.0;
170 }
171
172 /* Start outer loop over neighborlists */
173 for(iidx=0; iidx<nri; iidx++)
174 {
175 /* Load shift vector for this list */
176 i_shift_offset = DIM*shiftidx[iidx];
177
178 /* Load limits for loop over neighbors */
179 j_index_start = jindex[iidx];
180 j_index_end = jindex[iidx+1];
181
182 /* Get outer coordinate index */
183 inr = iinr[iidx];
184 i_coord_offset = DIM*inr;
185
186 /* Load i particle coords and add shift vector */
187 gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
188 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
189
190 fix0 = _mm_setzero_ps();
191 fiy0 = _mm_setzero_ps();
192 fiz0 = _mm_setzero_ps();
193 fix1 = _mm_setzero_ps();
194 fiy1 = _mm_setzero_ps();
195 fiz1 = _mm_setzero_ps();
196 fix2 = _mm_setzero_ps();
197 fiy2 = _mm_setzero_ps();
198 fiz2 = _mm_setzero_ps();
199
200 /* Reset potential sums */
201 velecsum = _mm_setzero_ps();
202
203 /* Start inner kernel loop */
204 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
205 {
206
207 /* Get j neighbor index, and coordinate index */
208 jnrA = jjnr[jidx];
209 jnrB = jjnr[jidx+1];
210 jnrC = jjnr[jidx+2];
211 jnrD = jjnr[jidx+3];
212 j_coord_offsetA = DIM*jnrA;
213 j_coord_offsetB = DIM*jnrB;
214 j_coord_offsetC = DIM*jnrC;
215 j_coord_offsetD = DIM*jnrD;
216
217 /* load j atom coordinates */
218 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
219 x+j_coord_offsetC,x+j_coord_offsetD,
220 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
221
222 /* Calculate displacement vector */
223 dx00 = _mm_sub_ps(ix0,jx0);
224 dy00 = _mm_sub_ps(iy0,jy0);
225 dz00 = _mm_sub_ps(iz0,jz0);
226 dx01 = _mm_sub_ps(ix0,jx1);
227 dy01 = _mm_sub_ps(iy0,jy1);
228 dz01 = _mm_sub_ps(iz0,jz1);
229 dx02 = _mm_sub_ps(ix0,jx2);
230 dy02 = _mm_sub_ps(iy0,jy2);
231 dz02 = _mm_sub_ps(iz0,jz2);
232 dx10 = _mm_sub_ps(ix1,jx0);
233 dy10 = _mm_sub_ps(iy1,jy0);
234 dz10 = _mm_sub_ps(iz1,jz0);
235 dx11 = _mm_sub_ps(ix1,jx1);
236 dy11 = _mm_sub_ps(iy1,jy1);
237 dz11 = _mm_sub_ps(iz1,jz1);
238 dx12 = _mm_sub_ps(ix1,jx2);
239 dy12 = _mm_sub_ps(iy1,jy2);
240 dz12 = _mm_sub_ps(iz1,jz2);
241 dx20 = _mm_sub_ps(ix2,jx0);
242 dy20 = _mm_sub_ps(iy2,jy0);
243 dz20 = _mm_sub_ps(iz2,jz0);
244 dx21 = _mm_sub_ps(ix2,jx1);
245 dy21 = _mm_sub_ps(iy2,jy1);
246 dz21 = _mm_sub_ps(iz2,jz1);
247 dx22 = _mm_sub_ps(ix2,jx2);
248 dy22 = _mm_sub_ps(iy2,jy2);
249 dz22 = _mm_sub_ps(iz2,jz2);
250
251 /* Calculate squared distance and things based on it */
252 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
253 rsq01 = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
254 rsq02 = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
255 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
256 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
257 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
258 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
259 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
260 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
261
262 rinv00 = gmx_mm_invsqrt_ps(rsq00);
263 rinv01 = gmx_mm_invsqrt_ps(rsq01);
264 rinv02 = gmx_mm_invsqrt_ps(rsq02);
265 rinv10 = gmx_mm_invsqrt_ps(rsq10);
266 rinv11 = gmx_mm_invsqrt_ps(rsq11);
267 rinv12 = gmx_mm_invsqrt_ps(rsq12);
268 rinv20 = gmx_mm_invsqrt_ps(rsq20);
269 rinv21 = gmx_mm_invsqrt_ps(rsq21);
270 rinv22 = gmx_mm_invsqrt_ps(rsq22);
271
272 rinvsq00 = _mm_mul_ps(rinv00,rinv00);
273 rinvsq01 = _mm_mul_ps(rinv01,rinv01);
274 rinvsq02 = _mm_mul_ps(rinv02,rinv02);
275 rinvsq10 = _mm_mul_ps(rinv10,rinv10);
276 rinvsq11 = _mm_mul_ps(rinv11,rinv11);
277 rinvsq12 = _mm_mul_ps(rinv12,rinv12);
278 rinvsq20 = _mm_mul_ps(rinv20,rinv20);
279 rinvsq21 = _mm_mul_ps(rinv21,rinv21);
280 rinvsq22 = _mm_mul_ps(rinv22,rinv22);
281
282 fjx0 = _mm_setzero_ps();
283 fjy0 = _mm_setzero_ps();
284 fjz0 = _mm_setzero_ps();
285 fjx1 = _mm_setzero_ps();
286 fjy1 = _mm_setzero_ps();
287 fjz1 = _mm_setzero_ps();
288 fjx2 = _mm_setzero_ps();
289 fjy2 = _mm_setzero_ps();
290 fjz2 = _mm_setzero_ps();
291
292 /**************************
293 * CALCULATE INTERACTIONS *
294 **************************/
295
296 if (gmx_mm_any_lt(rsq00,rcutoff2))
297 {
298
299 r00 = _mm_mul_ps(rsq00,rinv00);
300
301 /* EWALD ELECTROSTATICS */
302
303 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
304 ewrt = _mm_mul_ps(r00,ewtabscale);
305 ewitab = _mm_cvttps_epi32(ewrt);
306 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
307 ewitab = _mm_slli_epi32(ewitab,2);
308 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
309 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
310 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
311 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
312 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
313 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
314 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
315 velec = _mm_mul_ps(qq00,_mm_sub_ps(_mm_sub_ps(rinv00,sh_ewald),velec));
316 felec = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
317
318 cutoff_mask = _mm_cmplt_ps(rsq00,rcutoff2);
319
320 /* Update potential sum for this i atom from the interaction with this j atom. */
321 velec = _mm_and_ps(velec,cutoff_mask);
322 velecsum = _mm_add_ps(velecsum,velec);
323
324 fscal = felec;
325
326 fscal = _mm_and_ps(fscal,cutoff_mask);
327
328 /* Calculate temporary vectorial force */
329 tx = _mm_mul_ps(fscal,dx00);
330 ty = _mm_mul_ps(fscal,dy00);
331 tz = _mm_mul_ps(fscal,dz00);
332
333 /* Update vectorial force */
334 fix0 = _mm_add_ps(fix0,tx);
335 fiy0 = _mm_add_ps(fiy0,ty);
336 fiz0 = _mm_add_ps(fiz0,tz);
337
338 fjx0 = _mm_add_ps(fjx0,tx);
339 fjy0 = _mm_add_ps(fjy0,ty);
340 fjz0 = _mm_add_ps(fjz0,tz);
341
342 }
343
344 /**************************
345 * CALCULATE INTERACTIONS *
346 **************************/
347
348 if (gmx_mm_any_lt(rsq01,rcutoff2))
349 {
350
351 r01 = _mm_mul_ps(rsq01,rinv01);
352
353 /* EWALD ELECTROSTATICS */
354
355 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
356 ewrt = _mm_mul_ps(r01,ewtabscale);
357 ewitab = _mm_cvttps_epi32(ewrt);
358 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
359 ewitab = _mm_slli_epi32(ewitab,2);
360 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
361 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
362 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
363 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
364 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
365 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
366 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
367 velec = _mm_mul_ps(qq01,_mm_sub_ps(_mm_sub_ps(rinv01,sh_ewald),velec));
368 felec = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
369
370 cutoff_mask = _mm_cmplt_ps(rsq01,rcutoff2);
371
372 /* Update potential sum for this i atom from the interaction with this j atom. */
373 velec = _mm_and_ps(velec,cutoff_mask);
374 velecsum = _mm_add_ps(velecsum,velec);
375
376 fscal = felec;
377
378 fscal = _mm_and_ps(fscal,cutoff_mask);
379
380 /* Calculate temporary vectorial force */
381 tx = _mm_mul_ps(fscal,dx01);
382 ty = _mm_mul_ps(fscal,dy01);
383 tz = _mm_mul_ps(fscal,dz01);
384
385 /* Update vectorial force */
386 fix0 = _mm_add_ps(fix0,tx);
387 fiy0 = _mm_add_ps(fiy0,ty);
388 fiz0 = _mm_add_ps(fiz0,tz);
389
390 fjx1 = _mm_add_ps(fjx1,tx);
391 fjy1 = _mm_add_ps(fjy1,ty);
392 fjz1 = _mm_add_ps(fjz1,tz);
393
394 }
395
396 /**************************
397 * CALCULATE INTERACTIONS *
398 **************************/
399
400 if (gmx_mm_any_lt(rsq02,rcutoff2))
401 {
402
403 r02 = _mm_mul_ps(rsq02,rinv02);
404
405 /* EWALD ELECTROSTATICS */
406
407 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
408 ewrt = _mm_mul_ps(r02,ewtabscale);
409 ewitab = _mm_cvttps_epi32(ewrt);
410 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
411 ewitab = _mm_slli_epi32(ewitab,2);
412 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
413 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
414 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
415 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
416 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
417 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
418 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
419 velec = _mm_mul_ps(qq02,_mm_sub_ps(_mm_sub_ps(rinv02,sh_ewald),velec));
420 felec = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
421
422 cutoff_mask = _mm_cmplt_ps(rsq02,rcutoff2);
423
424 /* Update potential sum for this i atom from the interaction with this j atom. */
425 velec = _mm_and_ps(velec,cutoff_mask);
426 velecsum = _mm_add_ps(velecsum,velec);
427
428 fscal = felec;
429
430 fscal = _mm_and_ps(fscal,cutoff_mask);
431
432 /* Calculate temporary vectorial force */
433 tx = _mm_mul_ps(fscal,dx02);
434 ty = _mm_mul_ps(fscal,dy02);
435 tz = _mm_mul_ps(fscal,dz02);
436
437 /* Update vectorial force */
438 fix0 = _mm_add_ps(fix0,tx);
439 fiy0 = _mm_add_ps(fiy0,ty);
440 fiz0 = _mm_add_ps(fiz0,tz);
441
442 fjx2 = _mm_add_ps(fjx2,tx);
443 fjy2 = _mm_add_ps(fjy2,ty);
444 fjz2 = _mm_add_ps(fjz2,tz);
445
446 }
447
448 /**************************
449 * CALCULATE INTERACTIONS *
450 **************************/
451
452 if (gmx_mm_any_lt(rsq10,rcutoff2))
453 {
454
455 r10 = _mm_mul_ps(rsq10,rinv10);
456
457 /* EWALD ELECTROSTATICS */
458
459 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
460 ewrt = _mm_mul_ps(r10,ewtabscale);
461 ewitab = _mm_cvttps_epi32(ewrt);
462 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
463 ewitab = _mm_slli_epi32(ewitab,2);
464 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
465 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
466 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
467 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
468 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
469 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
470 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
471 velec = _mm_mul_ps(qq10,_mm_sub_ps(_mm_sub_ps(rinv10,sh_ewald),velec));
472 felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
473
474 cutoff_mask = _mm_cmplt_ps(rsq10,rcutoff2);
475
476 /* Update potential sum for this i atom from the interaction with this j atom. */
477 velec = _mm_and_ps(velec,cutoff_mask);
478 velecsum = _mm_add_ps(velecsum,velec);
479
480 fscal = felec;
481
482 fscal = _mm_and_ps(fscal,cutoff_mask);
483
484 /* Calculate temporary vectorial force */
485 tx = _mm_mul_ps(fscal,dx10);
486 ty = _mm_mul_ps(fscal,dy10);
487 tz = _mm_mul_ps(fscal,dz10);
488
489 /* Update vectorial force */
490 fix1 = _mm_add_ps(fix1,tx);
491 fiy1 = _mm_add_ps(fiy1,ty);
492 fiz1 = _mm_add_ps(fiz1,tz);
493
494 fjx0 = _mm_add_ps(fjx0,tx);
495 fjy0 = _mm_add_ps(fjy0,ty);
496 fjz0 = _mm_add_ps(fjz0,tz);
497
498 }
499
500 /**************************
501 * CALCULATE INTERACTIONS *
502 **************************/
503
504 if (gmx_mm_any_lt(rsq11,rcutoff2))
505 {
506
507 r11 = _mm_mul_ps(rsq11,rinv11);
508
509 /* EWALD ELECTROSTATICS */
510
511 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
512 ewrt = _mm_mul_ps(r11,ewtabscale);
513 ewitab = _mm_cvttps_epi32(ewrt);
514 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
515 ewitab = _mm_slli_epi32(ewitab,2);
516 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
517 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
518 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
519 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
520 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
521 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
522 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
523 velec = _mm_mul_ps(qq11,_mm_sub_ps(_mm_sub_ps(rinv11,sh_ewald),velec));
524 felec = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
525
526 cutoff_mask = _mm_cmplt_ps(rsq11,rcutoff2);
527
528 /* Update potential sum for this i atom from the interaction with this j atom. */
529 velec = _mm_and_ps(velec,cutoff_mask);
530 velecsum = _mm_add_ps(velecsum,velec);
531
532 fscal = felec;
533
534 fscal = _mm_and_ps(fscal,cutoff_mask);
535
536 /* Calculate temporary vectorial force */
537 tx = _mm_mul_ps(fscal,dx11);
538 ty = _mm_mul_ps(fscal,dy11);
539 tz = _mm_mul_ps(fscal,dz11);
540
541 /* Update vectorial force */
542 fix1 = _mm_add_ps(fix1,tx);
543 fiy1 = _mm_add_ps(fiy1,ty);
544 fiz1 = _mm_add_ps(fiz1,tz);
545
546 fjx1 = _mm_add_ps(fjx1,tx);
547 fjy1 = _mm_add_ps(fjy1,ty);
548 fjz1 = _mm_add_ps(fjz1,tz);
549
550 }
551
552 /**************************
553 * CALCULATE INTERACTIONS *
554 **************************/
555
556 if (gmx_mm_any_lt(rsq12,rcutoff2))
557 {
558
559 r12 = _mm_mul_ps(rsq12,rinv12);
560
561 /* EWALD ELECTROSTATICS */
562
563 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
564 ewrt = _mm_mul_ps(r12,ewtabscale);
565 ewitab = _mm_cvttps_epi32(ewrt);
566 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
567 ewitab = _mm_slli_epi32(ewitab,2);
568 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
569 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
570 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
571 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
572 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
573 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
574 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
575 velec = _mm_mul_ps(qq12,_mm_sub_ps(_mm_sub_ps(rinv12,sh_ewald),velec));
576 felec = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
577
578 cutoff_mask = _mm_cmplt_ps(rsq12,rcutoff2);
579
580 /* Update potential sum for this i atom from the interaction with this j atom. */
581 velec = _mm_and_ps(velec,cutoff_mask);
582 velecsum = _mm_add_ps(velecsum,velec);
583
584 fscal = felec;
585
586 fscal = _mm_and_ps(fscal,cutoff_mask);
587
588 /* Calculate temporary vectorial force */
589 tx = _mm_mul_ps(fscal,dx12);
590 ty = _mm_mul_ps(fscal,dy12);
591 tz = _mm_mul_ps(fscal,dz12);
592
593 /* Update vectorial force */
594 fix1 = _mm_add_ps(fix1,tx);
595 fiy1 = _mm_add_ps(fiy1,ty);
596 fiz1 = _mm_add_ps(fiz1,tz);
597
598 fjx2 = _mm_add_ps(fjx2,tx);
599 fjy2 = _mm_add_ps(fjy2,ty);
600 fjz2 = _mm_add_ps(fjz2,tz);
601
602 }
603
604 /**************************
605 * CALCULATE INTERACTIONS *
606 **************************/
607
608 if (gmx_mm_any_lt(rsq20,rcutoff2))
609 {
610
611 r20 = _mm_mul_ps(rsq20,rinv20);
612
613 /* EWALD ELECTROSTATICS */
614
615 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
616 ewrt = _mm_mul_ps(r20,ewtabscale);
617 ewitab = _mm_cvttps_epi32(ewrt);
618 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
619 ewitab = _mm_slli_epi32(ewitab,2);
620 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
621 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
622 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
623 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
624 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
625 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
626 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
627 velec = _mm_mul_ps(qq20,_mm_sub_ps(_mm_sub_ps(rinv20,sh_ewald),velec));
628 felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
629
630 cutoff_mask = _mm_cmplt_ps(rsq20,rcutoff2);
631
632 /* Update potential sum for this i atom from the interaction with this j atom. */
633 velec = _mm_and_ps(velec,cutoff_mask);
634 velecsum = _mm_add_ps(velecsum,velec);
635
636 fscal = felec;
637
638 fscal = _mm_and_ps(fscal,cutoff_mask);
639
640 /* Calculate temporary vectorial force */
641 tx = _mm_mul_ps(fscal,dx20);
642 ty = _mm_mul_ps(fscal,dy20);
643 tz = _mm_mul_ps(fscal,dz20);
644
645 /* Update vectorial force */
646 fix2 = _mm_add_ps(fix2,tx);
647 fiy2 = _mm_add_ps(fiy2,ty);
648 fiz2 = _mm_add_ps(fiz2,tz);
649
650 fjx0 = _mm_add_ps(fjx0,tx);
651 fjy0 = _mm_add_ps(fjy0,ty);
652 fjz0 = _mm_add_ps(fjz0,tz);
653
654 }
655
656 /**************************
657 * CALCULATE INTERACTIONS *
658 **************************/
659
660 if (gmx_mm_any_lt(rsq21,rcutoff2))
661 {
662
663 r21 = _mm_mul_ps(rsq21,rinv21);
664
665 /* EWALD ELECTROSTATICS */
666
667 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
668 ewrt = _mm_mul_ps(r21,ewtabscale);
669 ewitab = _mm_cvttps_epi32(ewrt);
670 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
671 ewitab = _mm_slli_epi32(ewitab,2);
672 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
673 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
674 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
675 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
676 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
677 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
678 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
679 velec = _mm_mul_ps(qq21,_mm_sub_ps(_mm_sub_ps(rinv21,sh_ewald),velec));
680 felec = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
681
682 cutoff_mask = _mm_cmplt_ps(rsq21,rcutoff2);
683
684 /* Update potential sum for this i atom from the interaction with this j atom. */
685 velec = _mm_and_ps(velec,cutoff_mask);
686 velecsum = _mm_add_ps(velecsum,velec);
687
688 fscal = felec;
689
690 fscal = _mm_and_ps(fscal,cutoff_mask);
691
692 /* Calculate temporary vectorial force */
693 tx = _mm_mul_ps(fscal,dx21);
694 ty = _mm_mul_ps(fscal,dy21);
695 tz = _mm_mul_ps(fscal,dz21);
696
697 /* Update vectorial force */
698 fix2 = _mm_add_ps(fix2,tx);
699 fiy2 = _mm_add_ps(fiy2,ty);
700 fiz2 = _mm_add_ps(fiz2,tz);
701
702 fjx1 = _mm_add_ps(fjx1,tx);
703 fjy1 = _mm_add_ps(fjy1,ty);
704 fjz1 = _mm_add_ps(fjz1,tz);
705
706 }
707
708 /**************************
709 * CALCULATE INTERACTIONS *
710 **************************/
711
712 if (gmx_mm_any_lt(rsq22,rcutoff2))
713 {
714
715 r22 = _mm_mul_ps(rsq22,rinv22);
716
717 /* EWALD ELECTROSTATICS */
718
719 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
720 ewrt = _mm_mul_ps(r22,ewtabscale);
721 ewitab = _mm_cvttps_epi32(ewrt);
722 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
723 ewitab = _mm_slli_epi32(ewitab,2);
724 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
725 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
726 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
727 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
728 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
729 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
730 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
731 velec = _mm_mul_ps(qq22,_mm_sub_ps(_mm_sub_ps(rinv22,sh_ewald),velec));
732 felec = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
733
734 cutoff_mask = _mm_cmplt_ps(rsq22,rcutoff2);
735
736 /* Update potential sum for this i atom from the interaction with this j atom. */
737 velec = _mm_and_ps(velec,cutoff_mask);
738 velecsum = _mm_add_ps(velecsum,velec);
739
740 fscal = felec;
741
742 fscal = _mm_and_ps(fscal,cutoff_mask);
743
744 /* Calculate temporary vectorial force */
745 tx = _mm_mul_ps(fscal,dx22);
746 ty = _mm_mul_ps(fscal,dy22);
747 tz = _mm_mul_ps(fscal,dz22);
748
749 /* Update vectorial force */
750 fix2 = _mm_add_ps(fix2,tx);
751 fiy2 = _mm_add_ps(fiy2,ty);
752 fiz2 = _mm_add_ps(fiz2,tz);
753
754 fjx2 = _mm_add_ps(fjx2,tx);
755 fjy2 = _mm_add_ps(fjy2,ty);
756 fjz2 = _mm_add_ps(fjz2,tz);
757
758 }
759
760 fjptrA = f+j_coord_offsetA;
761 fjptrB = f+j_coord_offsetB;
762 fjptrC = f+j_coord_offsetC;
763 fjptrD = f+j_coord_offsetD;
764
765 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
766 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
767
768 /* Inner loop uses 414 flops */
769 }
770
771 if(jidx<j_index_end)
772 {
773
774 /* Get j neighbor index, and coordinate index */
775 jnrlistA = jjnr[jidx];
776 jnrlistB = jjnr[jidx+1];
777 jnrlistC = jjnr[jidx+2];
778 jnrlistD = jjnr[jidx+3];
779 /* Sign of each element will be negative for non-real atoms.
780 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
781 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
782 */
783 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
784 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
785 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
786 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
787 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
788 j_coord_offsetA = DIM*jnrA;
789 j_coord_offsetB = DIM*jnrB;
790 j_coord_offsetC = DIM*jnrC;
791 j_coord_offsetD = DIM*jnrD;
792
793 /* load j atom coordinates */
794 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
795 x+j_coord_offsetC,x+j_coord_offsetD,
796 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
797
798 /* Calculate displacement vector */
799 dx00 = _mm_sub_ps(ix0,jx0);
800 dy00 = _mm_sub_ps(iy0,jy0);
801 dz00 = _mm_sub_ps(iz0,jz0);
802 dx01 = _mm_sub_ps(ix0,jx1);
803 dy01 = _mm_sub_ps(iy0,jy1);
804 dz01 = _mm_sub_ps(iz0,jz1);
805 dx02 = _mm_sub_ps(ix0,jx2);
806 dy02 = _mm_sub_ps(iy0,jy2);
807 dz02 = _mm_sub_ps(iz0,jz2);
808 dx10 = _mm_sub_ps(ix1,jx0);
809 dy10 = _mm_sub_ps(iy1,jy0);
810 dz10 = _mm_sub_ps(iz1,jz0);
811 dx11 = _mm_sub_ps(ix1,jx1);
812 dy11 = _mm_sub_ps(iy1,jy1);
813 dz11 = _mm_sub_ps(iz1,jz1);
814 dx12 = _mm_sub_ps(ix1,jx2);
815 dy12 = _mm_sub_ps(iy1,jy2);
816 dz12 = _mm_sub_ps(iz1,jz2);
817 dx20 = _mm_sub_ps(ix2,jx0);
818 dy20 = _mm_sub_ps(iy2,jy0);
819 dz20 = _mm_sub_ps(iz2,jz0);
820 dx21 = _mm_sub_ps(ix2,jx1);
821 dy21 = _mm_sub_ps(iy2,jy1);
822 dz21 = _mm_sub_ps(iz2,jz1);
823 dx22 = _mm_sub_ps(ix2,jx2);
824 dy22 = _mm_sub_ps(iy2,jy2);
825 dz22 = _mm_sub_ps(iz2,jz2);
826
827 /* Calculate squared distance and things based on it */
828 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
829 rsq01 = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
830 rsq02 = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
831 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
832 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
833 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
834 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
835 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
836 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
837
838 rinv00 = gmx_mm_invsqrt_ps(rsq00);
839 rinv01 = gmx_mm_invsqrt_ps(rsq01);
840 rinv02 = gmx_mm_invsqrt_ps(rsq02);
841 rinv10 = gmx_mm_invsqrt_ps(rsq10);
842 rinv11 = gmx_mm_invsqrt_ps(rsq11);
843 rinv12 = gmx_mm_invsqrt_ps(rsq12);
844 rinv20 = gmx_mm_invsqrt_ps(rsq20);
845 rinv21 = gmx_mm_invsqrt_ps(rsq21);
846 rinv22 = gmx_mm_invsqrt_ps(rsq22);
847
848 rinvsq00 = _mm_mul_ps(rinv00,rinv00);
849 rinvsq01 = _mm_mul_ps(rinv01,rinv01);
850 rinvsq02 = _mm_mul_ps(rinv02,rinv02);
851 rinvsq10 = _mm_mul_ps(rinv10,rinv10);
852 rinvsq11 = _mm_mul_ps(rinv11,rinv11);
853 rinvsq12 = _mm_mul_ps(rinv12,rinv12);
854 rinvsq20 = _mm_mul_ps(rinv20,rinv20);
855 rinvsq21 = _mm_mul_ps(rinv21,rinv21);
856 rinvsq22 = _mm_mul_ps(rinv22,rinv22);
857
858 fjx0 = _mm_setzero_ps();
859 fjy0 = _mm_setzero_ps();
860 fjz0 = _mm_setzero_ps();
861 fjx1 = _mm_setzero_ps();
862 fjy1 = _mm_setzero_ps();
863 fjz1 = _mm_setzero_ps();
864 fjx2 = _mm_setzero_ps();
865 fjy2 = _mm_setzero_ps();
866 fjz2 = _mm_setzero_ps();
867
868 /**************************
869 * CALCULATE INTERACTIONS *
870 **************************/
871
872 if (gmx_mm_any_lt(rsq00,rcutoff2))
873 {
874
875 r00 = _mm_mul_ps(rsq00,rinv00);
876 r00 = _mm_andnot_ps(dummy_mask,r00);
877
878 /* EWALD ELECTROSTATICS */
879
880 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
881 ewrt = _mm_mul_ps(r00,ewtabscale);
882 ewitab = _mm_cvttps_epi32(ewrt);
883 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
884 ewitab = _mm_slli_epi32(ewitab,2);
885 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
886 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
887 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
888 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
889 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
890 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
891 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
892 velec = _mm_mul_ps(qq00,_mm_sub_ps(_mm_sub_ps(rinv00,sh_ewald),velec));
893 felec = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
894
895 cutoff_mask = _mm_cmplt_ps(rsq00,rcutoff2);
896
897 /* Update potential sum for this i atom from the interaction with this j atom. */
898 velec = _mm_and_ps(velec,cutoff_mask);
899 velec = _mm_andnot_ps(dummy_mask,velec);
900 velecsum = _mm_add_ps(velecsum,velec);
901
902 fscal = felec;
903
904 fscal = _mm_and_ps(fscal,cutoff_mask);
905
906 fscal = _mm_andnot_ps(dummy_mask,fscal);
907
908 /* Calculate temporary vectorial force */
909 tx = _mm_mul_ps(fscal,dx00);
910 ty = _mm_mul_ps(fscal,dy00);
911 tz = _mm_mul_ps(fscal,dz00);
912
913 /* Update vectorial force */
914 fix0 = _mm_add_ps(fix0,tx);
915 fiy0 = _mm_add_ps(fiy0,ty);
916 fiz0 = _mm_add_ps(fiz0,tz);
917
918 fjx0 = _mm_add_ps(fjx0,tx);
919 fjy0 = _mm_add_ps(fjy0,ty);
920 fjz0 = _mm_add_ps(fjz0,tz);
921
922 }
923
924 /**************************
925 * CALCULATE INTERACTIONS *
926 **************************/
927
928 if (gmx_mm_any_lt(rsq01,rcutoff2))
929 {
930
931 r01 = _mm_mul_ps(rsq01,rinv01);
932 r01 = _mm_andnot_ps(dummy_mask,r01);
933
934 /* EWALD ELECTROSTATICS */
935
936 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
937 ewrt = _mm_mul_ps(r01,ewtabscale);
938 ewitab = _mm_cvttps_epi32(ewrt);
939 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
940 ewitab = _mm_slli_epi32(ewitab,2);
941 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
942 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
943 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
944 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
945 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
946 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
947 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
948 velec = _mm_mul_ps(qq01,_mm_sub_ps(_mm_sub_ps(rinv01,sh_ewald),velec));
949 felec = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
950
951 cutoff_mask = _mm_cmplt_ps(rsq01,rcutoff2);
952
953 /* Update potential sum for this i atom from the interaction with this j atom. */
954 velec = _mm_and_ps(velec,cutoff_mask);
955 velec = _mm_andnot_ps(dummy_mask,velec);
956 velecsum = _mm_add_ps(velecsum,velec);
957
958 fscal = felec;
959
960 fscal = _mm_and_ps(fscal,cutoff_mask);
961
962 fscal = _mm_andnot_ps(dummy_mask,fscal);
963
964 /* Calculate temporary vectorial force */
965 tx = _mm_mul_ps(fscal,dx01);
966 ty = _mm_mul_ps(fscal,dy01);
967 tz = _mm_mul_ps(fscal,dz01);
968
969 /* Update vectorial force */
970 fix0 = _mm_add_ps(fix0,tx);
971 fiy0 = _mm_add_ps(fiy0,ty);
972 fiz0 = _mm_add_ps(fiz0,tz);
973
974 fjx1 = _mm_add_ps(fjx1,tx);
975 fjy1 = _mm_add_ps(fjy1,ty);
976 fjz1 = _mm_add_ps(fjz1,tz);
977
978 }
979
980 /**************************
981 * CALCULATE INTERACTIONS *
982 **************************/
983
984 if (gmx_mm_any_lt(rsq02,rcutoff2))
985 {
986
987 r02 = _mm_mul_ps(rsq02,rinv02);
988 r02 = _mm_andnot_ps(dummy_mask,r02);
989
990 /* EWALD ELECTROSTATICS */
991
992 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
993 ewrt = _mm_mul_ps(r02,ewtabscale);
994 ewitab = _mm_cvttps_epi32(ewrt);
995 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
996 ewitab = _mm_slli_epi32(ewitab,2);
997 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
998 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
999 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1000 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1001 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1002 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1003 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1004 velec = _mm_mul_ps(qq02,_mm_sub_ps(_mm_sub_ps(rinv02,sh_ewald),velec));
1005 felec = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
1006
1007 cutoff_mask = _mm_cmplt_ps(rsq02,rcutoff2);
1008
1009 /* Update potential sum for this i atom from the interaction with this j atom. */
1010 velec = _mm_and_ps(velec,cutoff_mask);
1011 velec = _mm_andnot_ps(dummy_mask,velec);
1012 velecsum = _mm_add_ps(velecsum,velec);
1013
1014 fscal = felec;
1015
1016 fscal = _mm_and_ps(fscal,cutoff_mask);
1017
1018 fscal = _mm_andnot_ps(dummy_mask,fscal);
1019
1020 /* Calculate temporary vectorial force */
1021 tx = _mm_mul_ps(fscal,dx02);
1022 ty = _mm_mul_ps(fscal,dy02);
1023 tz = _mm_mul_ps(fscal,dz02);
1024
1025 /* Update vectorial force */
1026 fix0 = _mm_add_ps(fix0,tx);
1027 fiy0 = _mm_add_ps(fiy0,ty);
1028 fiz0 = _mm_add_ps(fiz0,tz);
1029
1030 fjx2 = _mm_add_ps(fjx2,tx);
1031 fjy2 = _mm_add_ps(fjy2,ty);
1032 fjz2 = _mm_add_ps(fjz2,tz);
1033
1034 }
1035
1036 /**************************
1037 * CALCULATE INTERACTIONS *
1038 **************************/
1039
1040 if (gmx_mm_any_lt(rsq10,rcutoff2))
1041 {
1042
1043 r10 = _mm_mul_ps(rsq10,rinv10);
1044 r10 = _mm_andnot_ps(dummy_mask,r10);
1045
1046 /* EWALD ELECTROSTATICS */
1047
1048 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1049 ewrt = _mm_mul_ps(r10,ewtabscale);
1050 ewitab = _mm_cvttps_epi32(ewrt);
1051 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1052 ewitab = _mm_slli_epi32(ewitab,2);
1053 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1054 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1055 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1056 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1057 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1058 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1059 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1060 velec = _mm_mul_ps(qq10,_mm_sub_ps(_mm_sub_ps(rinv10,sh_ewald),velec));
1061 felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
1062
1063 cutoff_mask = _mm_cmplt_ps(rsq10,rcutoff2);
1064
1065 /* Update potential sum for this i atom from the interaction with this j atom. */
1066 velec = _mm_and_ps(velec,cutoff_mask);
1067 velec = _mm_andnot_ps(dummy_mask,velec);
1068 velecsum = _mm_add_ps(velecsum,velec);
1069
1070 fscal = felec;
1071
1072 fscal = _mm_and_ps(fscal,cutoff_mask);
1073
1074 fscal = _mm_andnot_ps(dummy_mask,fscal);
1075
1076 /* Calculate temporary vectorial force */
1077 tx = _mm_mul_ps(fscal,dx10);
1078 ty = _mm_mul_ps(fscal,dy10);
1079 tz = _mm_mul_ps(fscal,dz10);
1080
1081 /* Update vectorial force */
1082 fix1 = _mm_add_ps(fix1,tx);
1083 fiy1 = _mm_add_ps(fiy1,ty);
1084 fiz1 = _mm_add_ps(fiz1,tz);
1085
1086 fjx0 = _mm_add_ps(fjx0,tx);
1087 fjy0 = _mm_add_ps(fjy0,ty);
1088 fjz0 = _mm_add_ps(fjz0,tz);
1089
1090 }
1091
1092 /**************************
1093 * CALCULATE INTERACTIONS *
1094 **************************/
1095
1096 if (gmx_mm_any_lt(rsq11,rcutoff2))
1097 {
1098
1099 r11 = _mm_mul_ps(rsq11,rinv11);
1100 r11 = _mm_andnot_ps(dummy_mask,r11);
1101
1102 /* EWALD ELECTROSTATICS */
1103
1104 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1105 ewrt = _mm_mul_ps(r11,ewtabscale);
1106 ewitab = _mm_cvttps_epi32(ewrt);
1107 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1108 ewitab = _mm_slli_epi32(ewitab,2);
1109 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1110 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1111 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1112 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1113 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1114 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1115 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1116 velec = _mm_mul_ps(qq11,_mm_sub_ps(_mm_sub_ps(rinv11,sh_ewald),velec));
1117 felec = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
1118
1119 cutoff_mask = _mm_cmplt_ps(rsq11,rcutoff2);
1120
1121 /* Update potential sum for this i atom from the interaction with this j atom. */
1122 velec = _mm_and_ps(velec,cutoff_mask);
1123 velec = _mm_andnot_ps(dummy_mask,velec);
1124 velecsum = _mm_add_ps(velecsum,velec);
1125
1126 fscal = felec;
1127
1128 fscal = _mm_and_ps(fscal,cutoff_mask);
1129
1130 fscal = _mm_andnot_ps(dummy_mask,fscal);
1131
1132 /* Calculate temporary vectorial force */
1133 tx = _mm_mul_ps(fscal,dx11);
1134 ty = _mm_mul_ps(fscal,dy11);
1135 tz = _mm_mul_ps(fscal,dz11);
1136
1137 /* Update vectorial force */
1138 fix1 = _mm_add_ps(fix1,tx);
1139 fiy1 = _mm_add_ps(fiy1,ty);
1140 fiz1 = _mm_add_ps(fiz1,tz);
1141
1142 fjx1 = _mm_add_ps(fjx1,tx);
1143 fjy1 = _mm_add_ps(fjy1,ty);
1144 fjz1 = _mm_add_ps(fjz1,tz);
1145
1146 }
1147
1148 /**************************
1149 * CALCULATE INTERACTIONS *
1150 **************************/
1151
1152 if (gmx_mm_any_lt(rsq12,rcutoff2))
1153 {
1154
1155 r12 = _mm_mul_ps(rsq12,rinv12);
1156 r12 = _mm_andnot_ps(dummy_mask,r12);
1157
1158 /* EWALD ELECTROSTATICS */
1159
1160 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1161 ewrt = _mm_mul_ps(r12,ewtabscale);
1162 ewitab = _mm_cvttps_epi32(ewrt);
1163 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1164 ewitab = _mm_slli_epi32(ewitab,2);
1165 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1166 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1167 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1168 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1169 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1170 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1171 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1172 velec = _mm_mul_ps(qq12,_mm_sub_ps(_mm_sub_ps(rinv12,sh_ewald),velec));
1173 felec = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
1174
1175 cutoff_mask = _mm_cmplt_ps(rsq12,rcutoff2);
1176
1177 /* Update potential sum for this i atom from the interaction with this j atom. */
1178 velec = _mm_and_ps(velec,cutoff_mask);
1179 velec = _mm_andnot_ps(dummy_mask,velec);
1180 velecsum = _mm_add_ps(velecsum,velec);
1181
1182 fscal = felec;
1183
1184 fscal = _mm_and_ps(fscal,cutoff_mask);
1185
1186 fscal = _mm_andnot_ps(dummy_mask,fscal);
1187
1188 /* Calculate temporary vectorial force */
1189 tx = _mm_mul_ps(fscal,dx12);
1190 ty = _mm_mul_ps(fscal,dy12);
1191 tz = _mm_mul_ps(fscal,dz12);
1192
1193 /* Update vectorial force */
1194 fix1 = _mm_add_ps(fix1,tx);
1195 fiy1 = _mm_add_ps(fiy1,ty);
1196 fiz1 = _mm_add_ps(fiz1,tz);
1197
1198 fjx2 = _mm_add_ps(fjx2,tx);
1199 fjy2 = _mm_add_ps(fjy2,ty);
1200 fjz2 = _mm_add_ps(fjz2,tz);
1201
1202 }
1203
1204 /**************************
1205 * CALCULATE INTERACTIONS *
1206 **************************/
1207
1208 if (gmx_mm_any_lt(rsq20,rcutoff2))
1209 {
1210
1211 r20 = _mm_mul_ps(rsq20,rinv20);
1212 r20 = _mm_andnot_ps(dummy_mask,r20);
1213
1214 /* EWALD ELECTROSTATICS */
1215
1216 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1217 ewrt = _mm_mul_ps(r20,ewtabscale);
1218 ewitab = _mm_cvttps_epi32(ewrt);
1219 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1220 ewitab = _mm_slli_epi32(ewitab,2);
1221 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1222 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1223 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1224 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1225 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1226 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1227 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1228 velec = _mm_mul_ps(qq20,_mm_sub_ps(_mm_sub_ps(rinv20,sh_ewald),velec));
1229 felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
1230
1231 cutoff_mask = _mm_cmplt_ps(rsq20,rcutoff2);
1232
1233 /* Update potential sum for this i atom from the interaction with this j atom. */
1234 velec = _mm_and_ps(velec,cutoff_mask);
1235 velec = _mm_andnot_ps(dummy_mask,velec);
1236 velecsum = _mm_add_ps(velecsum,velec);
1237
1238 fscal = felec;
1239
1240 fscal = _mm_and_ps(fscal,cutoff_mask);
1241
1242 fscal = _mm_andnot_ps(dummy_mask,fscal);
1243
1244 /* Calculate temporary vectorial force */
1245 tx = _mm_mul_ps(fscal,dx20);
1246 ty = _mm_mul_ps(fscal,dy20);
1247 tz = _mm_mul_ps(fscal,dz20);
1248
1249 /* Update vectorial force */
1250 fix2 = _mm_add_ps(fix2,tx);
1251 fiy2 = _mm_add_ps(fiy2,ty);
1252 fiz2 = _mm_add_ps(fiz2,tz);
1253
1254 fjx0 = _mm_add_ps(fjx0,tx);
1255 fjy0 = _mm_add_ps(fjy0,ty);
1256 fjz0 = _mm_add_ps(fjz0,tz);
1257
1258 }
1259
1260 /**************************
1261 * CALCULATE INTERACTIONS *
1262 **************************/
1263
1264 if (gmx_mm_any_lt(rsq21,rcutoff2))
1265 {
1266
1267 r21 = _mm_mul_ps(rsq21,rinv21);
1268 r21 = _mm_andnot_ps(dummy_mask,r21);
1269
1270 /* EWALD ELECTROSTATICS */
1271
1272 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1273 ewrt = _mm_mul_ps(r21,ewtabscale);
1274 ewitab = _mm_cvttps_epi32(ewrt);
1275 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1276 ewitab = _mm_slli_epi32(ewitab,2);
1277 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1278 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1279 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1280 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1281 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1282 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1283 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1284 velec = _mm_mul_ps(qq21,_mm_sub_ps(_mm_sub_ps(rinv21,sh_ewald),velec));
1285 felec = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
1286
1287 cutoff_mask = _mm_cmplt_ps(rsq21,rcutoff2);
1288
1289 /* Update potential sum for this i atom from the interaction with this j atom. */
1290 velec = _mm_and_ps(velec,cutoff_mask);
1291 velec = _mm_andnot_ps(dummy_mask,velec);
1292 velecsum = _mm_add_ps(velecsum,velec);
1293
1294 fscal = felec;
1295
1296 fscal = _mm_and_ps(fscal,cutoff_mask);
1297
1298 fscal = _mm_andnot_ps(dummy_mask,fscal);
1299
1300 /* Calculate temporary vectorial force */
1301 tx = _mm_mul_ps(fscal,dx21);
1302 ty = _mm_mul_ps(fscal,dy21);
1303 tz = _mm_mul_ps(fscal,dz21);
1304
1305 /* Update vectorial force */
1306 fix2 = _mm_add_ps(fix2,tx);
1307 fiy2 = _mm_add_ps(fiy2,ty);
1308 fiz2 = _mm_add_ps(fiz2,tz);
1309
1310 fjx1 = _mm_add_ps(fjx1,tx);
1311 fjy1 = _mm_add_ps(fjy1,ty);
1312 fjz1 = _mm_add_ps(fjz1,tz);
1313
1314 }
1315
1316 /**************************
1317 * CALCULATE INTERACTIONS *
1318 **************************/
1319
1320 if (gmx_mm_any_lt(rsq22,rcutoff2))
1321 {
1322
1323 r22 = _mm_mul_ps(rsq22,rinv22);
1324 r22 = _mm_andnot_ps(dummy_mask,r22);
1325
1326 /* EWALD ELECTROSTATICS */
1327
1328 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1329 ewrt = _mm_mul_ps(r22,ewtabscale);
1330 ewitab = _mm_cvttps_epi32(ewrt);
1331 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1332 ewitab = _mm_slli_epi32(ewitab,2);
1333 ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1334 ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1335 ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1336 ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1337 _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1338 felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1339 velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1340 velec = _mm_mul_ps(qq22,_mm_sub_ps(_mm_sub_ps(rinv22,sh_ewald),velec));
1341 felec = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
1342
1343 cutoff_mask = _mm_cmplt_ps(rsq22,rcutoff2);
1344
1345 /* Update potential sum for this i atom from the interaction with this j atom. */
1346 velec = _mm_and_ps(velec,cutoff_mask);
1347 velec = _mm_andnot_ps(dummy_mask,velec);
1348 velecsum = _mm_add_ps(velecsum,velec);
1349
1350 fscal = felec;
1351
1352 fscal = _mm_and_ps(fscal,cutoff_mask);
1353
1354 fscal = _mm_andnot_ps(dummy_mask,fscal);
1355
1356 /* Calculate temporary vectorial force */
1357 tx = _mm_mul_ps(fscal,dx22);
1358 ty = _mm_mul_ps(fscal,dy22);
1359 tz = _mm_mul_ps(fscal,dz22);
1360
1361 /* Update vectorial force */
1362 fix2 = _mm_add_ps(fix2,tx);
1363 fiy2 = _mm_add_ps(fiy2,ty);
1364 fiz2 = _mm_add_ps(fiz2,tz);
1365
1366 fjx2 = _mm_add_ps(fjx2,tx);
1367 fjy2 = _mm_add_ps(fjy2,ty);
1368 fjz2 = _mm_add_ps(fjz2,tz);
1369
1370 }
1371
1372 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1373 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1374 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1375 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1376
1377 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
1378 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1379
1380 /* Inner loop uses 423 flops */
1381 }
1382
1383 /* End of innermost loop */
1384
1385 gmx_mm_update_iforce_3atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1386 f+i_coord_offset,fshift+i_shift_offset);
1387
1388 ggid = gid[iidx];
1389 /* Update potential energies */
1390 gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1391
1392 /* Increment number of inner iterations */
1393 inneriter += j_index_end - j_index_start;
1394
1395 /* Outer loop uses 19 flops */
1396 }
1397
1398 /* Increment number of outer iterations */
1399 outeriter += nri;
1400
1401 /* Update outer/inner flops */
1402
1403 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3W3_VF,outeriter*19 + inneriter*423);
1404 }
1405 /*
1406 * Gromacs nonbonded kernel: nb_kernel_ElecEwSh_VdwNone_GeomW3W3_F_sse2_single
1407 * Electrostatics interaction: Ewald
1408 * VdW interaction: None
1409 * Geometry: Water3-Water3
1410 * Calculate force/pot: Force
1411 */
1412 void
1413 nb_kernel_ElecEwSh_VdwNone_GeomW3W3_F_sse2_single
1414 (t_nblist * gmx_restrict nlist,
1415 rvec * gmx_restrict xx,
1416 rvec * gmx_restrict ff,
1417 t_forcerec * gmx_restrict fr,
1418 t_mdatoms * gmx_restrict mdatoms,
1419 nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1420 t_nrnb * gmx_restrict nrnb)
1421 {
1422 /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1423 * just 0 for non-waters.
1424 * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
1425 * jnr indices corresponding to data put in the four positions in the SIMD register.
1426 */
1427 int i_shift_offset,i_coord_offset,outeriter,inneriter;
1428 int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1429 int jnrA,jnrB,jnrC,jnrD;
1430 int jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1431 int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1432 int *iinr,*jindex,*jjnr,*shiftidx,*gid;
1433 real rcutoff_scalar;
1434 real *shiftvec,*fshift,*x,*f;
1435 real *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1436 real scratch[4*DIM];
1437 __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1438 int vdwioffset0;
1439 __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1440 int vdwioffset1;
1441 __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1442 int vdwioffset2;
1443 __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1444 int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1445 __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1446 int vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1447 __m128 jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1448 int vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1449 __m128 jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1450 __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1451 __m128 dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1452 __m128 dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1453 __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1454 __m128 dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1455 __m128 dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1456 __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1457 __m128 dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1458 __m128 dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1459 __m128 velec,felec,velecsum,facel,crf,krf,krf2;
1460 real *charge;
1461 __m128i ewitab;
1462 __m128 ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1463 real *ewtab;
1464 __m128 dummy_mask,cutoff_mask;
1465 __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1466 __m128 one = _mm_set1_ps(1.0);
1467 __m128 two = _mm_set1_ps(2.0);
1468 x = xx[0];
1469 f = ff[0];
1470
1471 nri = nlist->nri;
1472 iinr = nlist->iinr;
1473 jindex = nlist->jindex;
1474 jjnr = nlist->jjnr;
1475 shiftidx = nlist->shift;
1476 gid = nlist->gid;
1477 shiftvec = fr->shift_vec[0];
1478 fshift = fr->fshift[0];
1479 facel = _mm_set1_ps(fr->epsfac);
1480 charge = mdatoms->chargeA;
1481
1482 sh_ewald = _mm_set1_ps(fr->ic->sh_ewald);
1483 ewtab = fr->ic->tabq_coul_F;
1484 ewtabscale = _mm_set1_ps(fr->ic->tabq_scale);
1485 ewtabhalfspace = _mm_set1_ps(0.5/fr->ic->tabq_scale);
1486
1487 /* Setup water-specific parameters */
1488 inr = nlist->iinr[0];
1489 iq0 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+0]));
1490 iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1491 iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1492
1493 jq0 = _mm_set1_ps(charge[inr+0]);
1494 jq1 = _mm_set1_ps(charge[inr+1]);
1495 jq2 = _mm_set1_ps(charge[inr+2]);
1496 qq00 = _mm_mul_ps(iq0,jq0);
1497 qq01 = _mm_mul_ps(iq0,jq1);
1498 qq02 = _mm_mul_ps(iq0,jq2);
1499 qq10 = _mm_mul_ps(iq1,jq0);
1500 qq11 = _mm_mul_ps(iq1,jq1);
1501 qq12 = _mm_mul_ps(iq1,jq2);
1502 qq20 = _mm_mul_ps(iq2,jq0);
1503 qq21 = _mm_mul_ps(iq2,jq1);
1504 qq22 = _mm_mul_ps(iq2,jq2);
1505
1506 /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1507 rcutoff_scalar = fr->rcoulomb;
1508 rcutoff = _mm_set1_ps(rcutoff_scalar);
1509 rcutoff2 = _mm_mul_ps(rcutoff,rcutoff);
1510
1511 /* Avoid stupid compiler warnings */
1512 jnrA = jnrB = jnrC = jnrD = 0;
1513 j_coord_offsetA = 0;
1514 j_coord_offsetB = 0;
1515 j_coord_offsetC = 0;
1516 j_coord_offsetD = 0;
1517
1518 outeriter = 0;
1519 inneriter = 0;
1520
1521 for(iidx=0;iidx<4*DIM;iidx++)
1522 {
1523 scratch[iidx] = 0.0;
1524 }
1525
1526 /* Start outer loop over neighborlists */
1527 for(iidx=0; iidx<nri; iidx++)
1528 {
1529 /* Load shift vector for this list */
1530 i_shift_offset = DIM*shiftidx[iidx];
1531
1532 /* Load limits for loop over neighbors */
1533 j_index_start = jindex[iidx];
1534 j_index_end = jindex[iidx+1];
1535
1536 /* Get outer coordinate index */
1537 inr = iinr[iidx];
1538 i_coord_offset = DIM*inr;
1539
1540 /* Load i particle coords and add shift vector */
1541 gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
1542 &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1543
1544 fix0 = _mm_setzero_ps();
1545 fiy0 = _mm_setzero_ps();
1546 fiz0 = _mm_setzero_ps();
1547 fix1 = _mm_setzero_ps();
1548 fiy1 = _mm_setzero_ps();
1549 fiz1 = _mm_setzero_ps();
1550 fix2 = _mm_setzero_ps();
1551 fiy2 = _mm_setzero_ps();
1552 fiz2 = _mm_setzero_ps();
1553
1554 /* Start inner kernel loop */
1555 for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1556 {
1557
1558 /* Get j neighbor index, and coordinate index */
1559 jnrA = jjnr[jidx];
1560 jnrB = jjnr[jidx+1];
1561 jnrC = jjnr[jidx+2];
1562 jnrD = jjnr[jidx+3];
1563 j_coord_offsetA = DIM*jnrA;
1564 j_coord_offsetB = DIM*jnrB;
1565 j_coord_offsetC = DIM*jnrC;
1566 j_coord_offsetD = DIM*jnrD;
1567
1568 /* load j atom coordinates */
1569 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1570 x+j_coord_offsetC,x+j_coord_offsetD,
1571 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1572
1573 /* Calculate displacement vector */
1574 dx00 = _mm_sub_ps(ix0,jx0);
1575 dy00 = _mm_sub_ps(iy0,jy0);
1576 dz00 = _mm_sub_ps(iz0,jz0);
1577 dx01 = _mm_sub_ps(ix0,jx1);
1578 dy01 = _mm_sub_ps(iy0,jy1);
1579 dz01 = _mm_sub_ps(iz0,jz1);
1580 dx02 = _mm_sub_ps(ix0,jx2);
1581 dy02 = _mm_sub_ps(iy0,jy2);
1582 dz02 = _mm_sub_ps(iz0,jz2);
1583 dx10 = _mm_sub_ps(ix1,jx0);
1584 dy10 = _mm_sub_ps(iy1,jy0);
1585 dz10 = _mm_sub_ps(iz1,jz0);
1586 dx11 = _mm_sub_ps(ix1,jx1);
1587 dy11 = _mm_sub_ps(iy1,jy1);
1588 dz11 = _mm_sub_ps(iz1,jz1);
1589 dx12 = _mm_sub_ps(ix1,jx2);
1590 dy12 = _mm_sub_ps(iy1,jy2);
1591 dz12 = _mm_sub_ps(iz1,jz2);
1592 dx20 = _mm_sub_ps(ix2,jx0);
1593 dy20 = _mm_sub_ps(iy2,jy0);
1594 dz20 = _mm_sub_ps(iz2,jz0);
1595 dx21 = _mm_sub_ps(ix2,jx1);
1596 dy21 = _mm_sub_ps(iy2,jy1);
1597 dz21 = _mm_sub_ps(iz2,jz1);
1598 dx22 = _mm_sub_ps(ix2,jx2);
1599 dy22 = _mm_sub_ps(iy2,jy2);
1600 dz22 = _mm_sub_ps(iz2,jz2);
1601
1602 /* Calculate squared distance and things based on it */
1603 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
1604 rsq01 = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
1605 rsq02 = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
1606 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
1607 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1608 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1609 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
1610 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1611 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1612
1613 rinv00 = gmx_mm_invsqrt_ps(rsq00);
1614 rinv01 = gmx_mm_invsqrt_ps(rsq01);
1615 rinv02 = gmx_mm_invsqrt_ps(rsq02);
1616 rinv10 = gmx_mm_invsqrt_ps(rsq10);
1617 rinv11 = gmx_mm_invsqrt_ps(rsq11);
1618 rinv12 = gmx_mm_invsqrt_ps(rsq12);
1619 rinv20 = gmx_mm_invsqrt_ps(rsq20);
1620 rinv21 = gmx_mm_invsqrt_ps(rsq21);
1621 rinv22 = gmx_mm_invsqrt_ps(rsq22);
1622
1623 rinvsq00 = _mm_mul_ps(rinv00,rinv00);
1624 rinvsq01 = _mm_mul_ps(rinv01,rinv01);
1625 rinvsq02 = _mm_mul_ps(rinv02,rinv02);
1626 rinvsq10 = _mm_mul_ps(rinv10,rinv10);
1627 rinvsq11 = _mm_mul_ps(rinv11,rinv11);
1628 rinvsq12 = _mm_mul_ps(rinv12,rinv12);
1629 rinvsq20 = _mm_mul_ps(rinv20,rinv20);
1630 rinvsq21 = _mm_mul_ps(rinv21,rinv21);
1631 rinvsq22 = _mm_mul_ps(rinv22,rinv22);
1632
1633 fjx0 = _mm_setzero_ps();
1634 fjy0 = _mm_setzero_ps();
1635 fjz0 = _mm_setzero_ps();
1636 fjx1 = _mm_setzero_ps();
1637 fjy1 = _mm_setzero_ps();
1638 fjz1 = _mm_setzero_ps();
1639 fjx2 = _mm_setzero_ps();
1640 fjy2 = _mm_setzero_ps();
1641 fjz2 = _mm_setzero_ps();
1642
1643 /**************************
1644 * CALCULATE INTERACTIONS *
1645 **************************/
1646
1647 if (gmx_mm_any_lt(rsq00,rcutoff2))
1648 {
1649
1650 r00 = _mm_mul_ps(rsq00,rinv00);
1651
1652 /* EWALD ELECTROSTATICS */
1653
1654 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1655 ewrt = _mm_mul_ps(r00,ewtabscale);
1656 ewitab = _mm_cvttps_epi32(ewrt);
1657 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1658 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1659 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1660 &ewtabF,&ewtabFn);
1661 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1662 felec = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
1663
1664 cutoff_mask = _mm_cmplt_ps(rsq00,rcutoff2);
1665
1666 fscal = felec;
1667
1668 fscal = _mm_and_ps(fscal,cutoff_mask);
1669
1670 /* Calculate temporary vectorial force */
1671 tx = _mm_mul_ps(fscal,dx00);
1672 ty = _mm_mul_ps(fscal,dy00);
1673 tz = _mm_mul_ps(fscal,dz00);
1674
1675 /* Update vectorial force */
1676 fix0 = _mm_add_ps(fix0,tx);
1677 fiy0 = _mm_add_ps(fiy0,ty);
1678 fiz0 = _mm_add_ps(fiz0,tz);
1679
1680 fjx0 = _mm_add_ps(fjx0,tx);
1681 fjy0 = _mm_add_ps(fjy0,ty);
1682 fjz0 = _mm_add_ps(fjz0,tz);
1683
1684 }
1685
1686 /**************************
1687 * CALCULATE INTERACTIONS *
1688 **************************/
1689
1690 if (gmx_mm_any_lt(rsq01,rcutoff2))
1691 {
1692
1693 r01 = _mm_mul_ps(rsq01,rinv01);
1694
1695 /* EWALD ELECTROSTATICS */
1696
1697 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1698 ewrt = _mm_mul_ps(r01,ewtabscale);
1699 ewitab = _mm_cvttps_epi32(ewrt);
1700 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1701 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1702 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1703 &ewtabF,&ewtabFn);
1704 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1705 felec = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
1706
1707 cutoff_mask = _mm_cmplt_ps(rsq01,rcutoff2);
1708
1709 fscal = felec;
1710
1711 fscal = _mm_and_ps(fscal,cutoff_mask);
1712
1713 /* Calculate temporary vectorial force */
1714 tx = _mm_mul_ps(fscal,dx01);
1715 ty = _mm_mul_ps(fscal,dy01);
1716 tz = _mm_mul_ps(fscal,dz01);
1717
1718 /* Update vectorial force */
1719 fix0 = _mm_add_ps(fix0,tx);
1720 fiy0 = _mm_add_ps(fiy0,ty);
1721 fiz0 = _mm_add_ps(fiz0,tz);
1722
1723 fjx1 = _mm_add_ps(fjx1,tx);
1724 fjy1 = _mm_add_ps(fjy1,ty);
1725 fjz1 = _mm_add_ps(fjz1,tz);
1726
1727 }
1728
1729 /**************************
1730 * CALCULATE INTERACTIONS *
1731 **************************/
1732
1733 if (gmx_mm_any_lt(rsq02,rcutoff2))
1734 {
1735
1736 r02 = _mm_mul_ps(rsq02,rinv02);
1737
1738 /* EWALD ELECTROSTATICS */
1739
1740 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1741 ewrt = _mm_mul_ps(r02,ewtabscale);
1742 ewitab = _mm_cvttps_epi32(ewrt);
1743 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1744 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1745 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1746 &ewtabF,&ewtabFn);
1747 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1748 felec = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
1749
1750 cutoff_mask = _mm_cmplt_ps(rsq02,rcutoff2);
1751
1752 fscal = felec;
1753
1754 fscal = _mm_and_ps(fscal,cutoff_mask);
1755
1756 /* Calculate temporary vectorial force */
1757 tx = _mm_mul_ps(fscal,dx02);
1758 ty = _mm_mul_ps(fscal,dy02);
1759 tz = _mm_mul_ps(fscal,dz02);
1760
1761 /* Update vectorial force */
1762 fix0 = _mm_add_ps(fix0,tx);
1763 fiy0 = _mm_add_ps(fiy0,ty);
1764 fiz0 = _mm_add_ps(fiz0,tz);
1765
1766 fjx2 = _mm_add_ps(fjx2,tx);
1767 fjy2 = _mm_add_ps(fjy2,ty);
1768 fjz2 = _mm_add_ps(fjz2,tz);
1769
1770 }
1771
1772 /**************************
1773 * CALCULATE INTERACTIONS *
1774 **************************/
1775
1776 if (gmx_mm_any_lt(rsq10,rcutoff2))
1777 {
1778
1779 r10 = _mm_mul_ps(rsq10,rinv10);
1780
1781 /* EWALD ELECTROSTATICS */
1782
1783 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1784 ewrt = _mm_mul_ps(r10,ewtabscale);
1785 ewitab = _mm_cvttps_epi32(ewrt);
1786 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1787 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1788 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1789 &ewtabF,&ewtabFn);
1790 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1791 felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
1792
1793 cutoff_mask = _mm_cmplt_ps(rsq10,rcutoff2);
1794
1795 fscal = felec;
1796
1797 fscal = _mm_and_ps(fscal,cutoff_mask);
1798
1799 /* Calculate temporary vectorial force */
1800 tx = _mm_mul_ps(fscal,dx10);
1801 ty = _mm_mul_ps(fscal,dy10);
1802 tz = _mm_mul_ps(fscal,dz10);
1803
1804 /* Update vectorial force */
1805 fix1 = _mm_add_ps(fix1,tx);
1806 fiy1 = _mm_add_ps(fiy1,ty);
1807 fiz1 = _mm_add_ps(fiz1,tz);
1808
1809 fjx0 = _mm_add_ps(fjx0,tx);
1810 fjy0 = _mm_add_ps(fjy0,ty);
1811 fjz0 = _mm_add_ps(fjz0,tz);
1812
1813 }
1814
1815 /**************************
1816 * CALCULATE INTERACTIONS *
1817 **************************/
1818
1819 if (gmx_mm_any_lt(rsq11,rcutoff2))
1820 {
1821
1822 r11 = _mm_mul_ps(rsq11,rinv11);
1823
1824 /* EWALD ELECTROSTATICS */
1825
1826 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1827 ewrt = _mm_mul_ps(r11,ewtabscale);
1828 ewitab = _mm_cvttps_epi32(ewrt);
1829 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1830 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1831 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1832 &ewtabF,&ewtabFn);
1833 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1834 felec = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
1835
1836 cutoff_mask = _mm_cmplt_ps(rsq11,rcutoff2);
1837
1838 fscal = felec;
1839
1840 fscal = _mm_and_ps(fscal,cutoff_mask);
1841
1842 /* Calculate temporary vectorial force */
1843 tx = _mm_mul_ps(fscal,dx11);
1844 ty = _mm_mul_ps(fscal,dy11);
1845 tz = _mm_mul_ps(fscal,dz11);
1846
1847 /* Update vectorial force */
1848 fix1 = _mm_add_ps(fix1,tx);
1849 fiy1 = _mm_add_ps(fiy1,ty);
1850 fiz1 = _mm_add_ps(fiz1,tz);
1851
1852 fjx1 = _mm_add_ps(fjx1,tx);
1853 fjy1 = _mm_add_ps(fjy1,ty);
1854 fjz1 = _mm_add_ps(fjz1,tz);
1855
1856 }
1857
1858 /**************************
1859 * CALCULATE INTERACTIONS *
1860 **************************/
1861
1862 if (gmx_mm_any_lt(rsq12,rcutoff2))
1863 {
1864
1865 r12 = _mm_mul_ps(rsq12,rinv12);
1866
1867 /* EWALD ELECTROSTATICS */
1868
1869 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1870 ewrt = _mm_mul_ps(r12,ewtabscale);
1871 ewitab = _mm_cvttps_epi32(ewrt);
1872 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1873 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1874 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1875 &ewtabF,&ewtabFn);
1876 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1877 felec = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
1878
1879 cutoff_mask = _mm_cmplt_ps(rsq12,rcutoff2);
1880
1881 fscal = felec;
1882
1883 fscal = _mm_and_ps(fscal,cutoff_mask);
1884
1885 /* Calculate temporary vectorial force */
1886 tx = _mm_mul_ps(fscal,dx12);
1887 ty = _mm_mul_ps(fscal,dy12);
1888 tz = _mm_mul_ps(fscal,dz12);
1889
1890 /* Update vectorial force */
1891 fix1 = _mm_add_ps(fix1,tx);
1892 fiy1 = _mm_add_ps(fiy1,ty);
1893 fiz1 = _mm_add_ps(fiz1,tz);
1894
1895 fjx2 = _mm_add_ps(fjx2,tx);
1896 fjy2 = _mm_add_ps(fjy2,ty);
1897 fjz2 = _mm_add_ps(fjz2,tz);
1898
1899 }
1900
1901 /**************************
1902 * CALCULATE INTERACTIONS *
1903 **************************/
1904
1905 if (gmx_mm_any_lt(rsq20,rcutoff2))
1906 {
1907
1908 r20 = _mm_mul_ps(rsq20,rinv20);
1909
1910 /* EWALD ELECTROSTATICS */
1911
1912 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1913 ewrt = _mm_mul_ps(r20,ewtabscale);
1914 ewitab = _mm_cvttps_epi32(ewrt);
1915 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1916 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1917 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1918 &ewtabF,&ewtabFn);
1919 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1920 felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
1921
1922 cutoff_mask = _mm_cmplt_ps(rsq20,rcutoff2);
1923
1924 fscal = felec;
1925
1926 fscal = _mm_and_ps(fscal,cutoff_mask);
1927
1928 /* Calculate temporary vectorial force */
1929 tx = _mm_mul_ps(fscal,dx20);
1930 ty = _mm_mul_ps(fscal,dy20);
1931 tz = _mm_mul_ps(fscal,dz20);
1932
1933 /* Update vectorial force */
1934 fix2 = _mm_add_ps(fix2,tx);
1935 fiy2 = _mm_add_ps(fiy2,ty);
1936 fiz2 = _mm_add_ps(fiz2,tz);
1937
1938 fjx0 = _mm_add_ps(fjx0,tx);
1939 fjy0 = _mm_add_ps(fjy0,ty);
1940 fjz0 = _mm_add_ps(fjz0,tz);
1941
1942 }
1943
1944 /**************************
1945 * CALCULATE INTERACTIONS *
1946 **************************/
1947
1948 if (gmx_mm_any_lt(rsq21,rcutoff2))
1949 {
1950
1951 r21 = _mm_mul_ps(rsq21,rinv21);
1952
1953 /* EWALD ELECTROSTATICS */
1954
1955 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1956 ewrt = _mm_mul_ps(r21,ewtabscale);
1957 ewitab = _mm_cvttps_epi32(ewrt);
1958 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1959 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1960 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1961 &ewtabF,&ewtabFn);
1962 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1963 felec = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
1964
1965 cutoff_mask = _mm_cmplt_ps(rsq21,rcutoff2);
1966
1967 fscal = felec;
1968
1969 fscal = _mm_and_ps(fscal,cutoff_mask);
1970
1971 /* Calculate temporary vectorial force */
1972 tx = _mm_mul_ps(fscal,dx21);
1973 ty = _mm_mul_ps(fscal,dy21);
1974 tz = _mm_mul_ps(fscal,dz21);
1975
1976 /* Update vectorial force */
1977 fix2 = _mm_add_ps(fix2,tx);
1978 fiy2 = _mm_add_ps(fiy2,ty);
1979 fiz2 = _mm_add_ps(fiz2,tz);
1980
1981 fjx1 = _mm_add_ps(fjx1,tx);
1982 fjy1 = _mm_add_ps(fjy1,ty);
1983 fjz1 = _mm_add_ps(fjz1,tz);
1984
1985 }
1986
1987 /**************************
1988 * CALCULATE INTERACTIONS *
1989 **************************/
1990
1991 if (gmx_mm_any_lt(rsq22,rcutoff2))
1992 {
1993
1994 r22 = _mm_mul_ps(rsq22,rinv22);
1995
1996 /* EWALD ELECTROSTATICS */
1997
1998 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1999 ewrt = _mm_mul_ps(r22,ewtabscale);
2000 ewitab = _mm_cvttps_epi32(ewrt);
2001 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2002 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2003 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2004 &ewtabF,&ewtabFn);
2005 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2006 felec = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
2007
2008 cutoff_mask = _mm_cmplt_ps(rsq22,rcutoff2);
2009
2010 fscal = felec;
2011
2012 fscal = _mm_and_ps(fscal,cutoff_mask);
2013
2014 /* Calculate temporary vectorial force */
2015 tx = _mm_mul_ps(fscal,dx22);
2016 ty = _mm_mul_ps(fscal,dy22);
2017 tz = _mm_mul_ps(fscal,dz22);
2018
2019 /* Update vectorial force */
2020 fix2 = _mm_add_ps(fix2,tx);
2021 fiy2 = _mm_add_ps(fiy2,ty);
2022 fiz2 = _mm_add_ps(fiz2,tz);
2023
2024 fjx2 = _mm_add_ps(fjx2,tx);
2025 fjy2 = _mm_add_ps(fjy2,ty);
2026 fjz2 = _mm_add_ps(fjz2,tz);
2027
2028 }
2029
2030 fjptrA = f+j_coord_offsetA;
2031 fjptrB = f+j_coord_offsetB;
2032 fjptrC = f+j_coord_offsetC;
2033 fjptrD = f+j_coord_offsetD;
2034
2035 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2036 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2037
2038 /* Inner loop uses 351 flops */
2039 }
2040
2041 if(jidx<j_index_end)
2042 {
2043
2044 /* Get j neighbor index, and coordinate index */
2045 jnrlistA = jjnr[jidx];
2046 jnrlistB = jjnr[jidx+1];
2047 jnrlistC = jjnr[jidx+2];
2048 jnrlistD = jjnr[jidx+3];
2049 /* Sign of each element will be negative for non-real atoms.
2050 * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
2051 * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
2052 */
2053 dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
2054 jnrA = (jnrlistA>=0) ? jnrlistA : 0;
2055 jnrB = (jnrlistB>=0) ? jnrlistB : 0;
2056 jnrC = (jnrlistC>=0) ? jnrlistC : 0;
2057 jnrD = (jnrlistD>=0) ? jnrlistD : 0;
2058 j_coord_offsetA = DIM*jnrA;
2059 j_coord_offsetB = DIM*jnrB;
2060 j_coord_offsetC = DIM*jnrC;
2061 j_coord_offsetD = DIM*jnrD;
2062
2063 /* load j atom coordinates */
2064 gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
2065 x+j_coord_offsetC,x+j_coord_offsetD,
2066 &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
2067
2068 /* Calculate displacement vector */
2069 dx00 = _mm_sub_ps(ix0,jx0);
2070 dy00 = _mm_sub_ps(iy0,jy0);
2071 dz00 = _mm_sub_ps(iz0,jz0);
2072 dx01 = _mm_sub_ps(ix0,jx1);
2073 dy01 = _mm_sub_ps(iy0,jy1);
2074 dz01 = _mm_sub_ps(iz0,jz1);
2075 dx02 = _mm_sub_ps(ix0,jx2);
2076 dy02 = _mm_sub_ps(iy0,jy2);
2077 dz02 = _mm_sub_ps(iz0,jz2);
2078 dx10 = _mm_sub_ps(ix1,jx0);
2079 dy10 = _mm_sub_ps(iy1,jy0);
2080 dz10 = _mm_sub_ps(iz1,jz0);
2081 dx11 = _mm_sub_ps(ix1,jx1);
2082 dy11 = _mm_sub_ps(iy1,jy1);
2083 dz11 = _mm_sub_ps(iz1,jz1);
2084 dx12 = _mm_sub_ps(ix1,jx2);
2085 dy12 = _mm_sub_ps(iy1,jy2);
2086 dz12 = _mm_sub_ps(iz1,jz2);
2087 dx20 = _mm_sub_ps(ix2,jx0);
2088 dy20 = _mm_sub_ps(iy2,jy0);
2089 dz20 = _mm_sub_ps(iz2,jz0);
2090 dx21 = _mm_sub_ps(ix2,jx1);
2091 dy21 = _mm_sub_ps(iy2,jy1);
2092 dz21 = _mm_sub_ps(iz2,jz1);
2093 dx22 = _mm_sub_ps(ix2,jx2);
2094 dy22 = _mm_sub_ps(iy2,jy2);
2095 dz22 = _mm_sub_ps(iz2,jz2);
2096
2097 /* Calculate squared distance and things based on it */
2098 rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
2099 rsq01 = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
2100 rsq02 = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
2101 rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
2102 rsq11 = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
2103 rsq12 = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
2104 rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
2105 rsq21 = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
2106 rsq22 = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
2107
2108 rinv00 = gmx_mm_invsqrt_ps(rsq00);
2109 rinv01 = gmx_mm_invsqrt_ps(rsq01);
2110 rinv02 = gmx_mm_invsqrt_ps(rsq02);
2111 rinv10 = gmx_mm_invsqrt_ps(rsq10);
2112 rinv11 = gmx_mm_invsqrt_ps(rsq11);
2113 rinv12 = gmx_mm_invsqrt_ps(rsq12);
2114 rinv20 = gmx_mm_invsqrt_ps(rsq20);
2115 rinv21 = gmx_mm_invsqrt_ps(rsq21);
2116 rinv22 = gmx_mm_invsqrt_ps(rsq22);
2117
2118 rinvsq00 = _mm_mul_ps(rinv00,rinv00);
2119 rinvsq01 = _mm_mul_ps(rinv01,rinv01);
2120 rinvsq02 = _mm_mul_ps(rinv02,rinv02);
2121 rinvsq10 = _mm_mul_ps(rinv10,rinv10);
2122 rinvsq11 = _mm_mul_ps(rinv11,rinv11);
2123 rinvsq12 = _mm_mul_ps(rinv12,rinv12);
2124 rinvsq20 = _mm_mul_ps(rinv20,rinv20);
2125 rinvsq21 = _mm_mul_ps(rinv21,rinv21);
2126 rinvsq22 = _mm_mul_ps(rinv22,rinv22);
2127
2128 fjx0 = _mm_setzero_ps();
2129 fjy0 = _mm_setzero_ps();
2130 fjz0 = _mm_setzero_ps();
2131 fjx1 = _mm_setzero_ps();
2132 fjy1 = _mm_setzero_ps();
2133 fjz1 = _mm_setzero_ps();
2134 fjx2 = _mm_setzero_ps();
2135 fjy2 = _mm_setzero_ps();
2136 fjz2 = _mm_setzero_ps();
2137
2138 /**************************
2139 * CALCULATE INTERACTIONS *
2140 **************************/
2141
2142 if (gmx_mm_any_lt(rsq00,rcutoff2))
2143 {
2144
2145 r00 = _mm_mul_ps(rsq00,rinv00);
2146 r00 = _mm_andnot_ps(dummy_mask,r00);
2147
2148 /* EWALD ELECTROSTATICS */
2149
2150 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2151 ewrt = _mm_mul_ps(r00,ewtabscale);
2152 ewitab = _mm_cvttps_epi32(ewrt);
2153 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2154 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2155 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2156 &ewtabF,&ewtabFn);
2157 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2158 felec = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
2159
2160 cutoff_mask = _mm_cmplt_ps(rsq00,rcutoff2);
2161
2162 fscal = felec;
2163
2164 fscal = _mm_and_ps(fscal,cutoff_mask);
2165
2166 fscal = _mm_andnot_ps(dummy_mask,fscal);
2167
2168 /* Calculate temporary vectorial force */
2169 tx = _mm_mul_ps(fscal,dx00);
2170 ty = _mm_mul_ps(fscal,dy00);
2171 tz = _mm_mul_ps(fscal,dz00);
2172
2173 /* Update vectorial force */
2174 fix0 = _mm_add_ps(fix0,tx);
2175 fiy0 = _mm_add_ps(fiy0,ty);
2176 fiz0 = _mm_add_ps(fiz0,tz);
2177
2178 fjx0 = _mm_add_ps(fjx0,tx);
2179 fjy0 = _mm_add_ps(fjy0,ty);
2180 fjz0 = _mm_add_ps(fjz0,tz);
2181
2182 }
2183
2184 /**************************
2185 * CALCULATE INTERACTIONS *
2186 **************************/
2187
2188 if (gmx_mm_any_lt(rsq01,rcutoff2))
2189 {
2190
2191 r01 = _mm_mul_ps(rsq01,rinv01);
2192 r01 = _mm_andnot_ps(dummy_mask,r01);
2193
2194 /* EWALD ELECTROSTATICS */
2195
2196 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2197 ewrt = _mm_mul_ps(r01,ewtabscale);
2198 ewitab = _mm_cvttps_epi32(ewrt);
2199 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2200 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2201 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2202 &ewtabF,&ewtabFn);
2203 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2204 felec = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
2205
2206 cutoff_mask = _mm_cmplt_ps(rsq01,rcutoff2);
2207
2208 fscal = felec;
2209
2210 fscal = _mm_and_ps(fscal,cutoff_mask);
2211
2212 fscal = _mm_andnot_ps(dummy_mask,fscal);
2213
2214 /* Calculate temporary vectorial force */
2215 tx = _mm_mul_ps(fscal,dx01);
2216 ty = _mm_mul_ps(fscal,dy01);
2217 tz = _mm_mul_ps(fscal,dz01);
2218
2219 /* Update vectorial force */
2220 fix0 = _mm_add_ps(fix0,tx);
2221 fiy0 = _mm_add_ps(fiy0,ty);
2222 fiz0 = _mm_add_ps(fiz0,tz);
2223
2224 fjx1 = _mm_add_ps(fjx1,tx);
2225 fjy1 = _mm_add_ps(fjy1,ty);
2226 fjz1 = _mm_add_ps(fjz1,tz);
2227
2228 }
2229
2230 /**************************
2231 * CALCULATE INTERACTIONS *
2232 **************************/
2233
2234 if (gmx_mm_any_lt(rsq02,rcutoff2))
2235 {
2236
2237 r02 = _mm_mul_ps(rsq02,rinv02);
2238 r02 = _mm_andnot_ps(dummy_mask,r02);
2239
2240 /* EWALD ELECTROSTATICS */
2241
2242 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2243 ewrt = _mm_mul_ps(r02,ewtabscale);
2244 ewitab = _mm_cvttps_epi32(ewrt);
2245 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2246 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2247 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2248 &ewtabF,&ewtabFn);
2249 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2250 felec = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
2251
2252 cutoff_mask = _mm_cmplt_ps(rsq02,rcutoff2);
2253
2254 fscal = felec;
2255
2256 fscal = _mm_and_ps(fscal,cutoff_mask);
2257
2258 fscal = _mm_andnot_ps(dummy_mask,fscal);
2259
2260 /* Calculate temporary vectorial force */
2261 tx = _mm_mul_ps(fscal,dx02);
2262 ty = _mm_mul_ps(fscal,dy02);
2263 tz = _mm_mul_ps(fscal,dz02);
2264
2265 /* Update vectorial force */
2266 fix0 = _mm_add_ps(fix0,tx);
2267 fiy0 = _mm_add_ps(fiy0,ty);
2268 fiz0 = _mm_add_ps(fiz0,tz);
2269
2270 fjx2 = _mm_add_ps(fjx2,tx);
2271 fjy2 = _mm_add_ps(fjy2,ty);
2272 fjz2 = _mm_add_ps(fjz2,tz);
2273
2274 }
2275
2276 /**************************
2277 * CALCULATE INTERACTIONS *
2278 **************************/
2279
2280 if (gmx_mm_any_lt(rsq10,rcutoff2))
2281 {
2282
2283 r10 = _mm_mul_ps(rsq10,rinv10);
2284 r10 = _mm_andnot_ps(dummy_mask,r10);
2285
2286 /* EWALD ELECTROSTATICS */
2287
2288 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2289 ewrt = _mm_mul_ps(r10,ewtabscale);
2290 ewitab = _mm_cvttps_epi32(ewrt);
2291 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2292 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2293 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2294 &ewtabF,&ewtabFn);
2295 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2296 felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
2297
2298 cutoff_mask = _mm_cmplt_ps(rsq10,rcutoff2);
2299
2300 fscal = felec;
2301
2302 fscal = _mm_and_ps(fscal,cutoff_mask);
2303
2304 fscal = _mm_andnot_ps(dummy_mask,fscal);
2305
2306 /* Calculate temporary vectorial force */
2307 tx = _mm_mul_ps(fscal,dx10);
2308 ty = _mm_mul_ps(fscal,dy10);
2309 tz = _mm_mul_ps(fscal,dz10);
2310
2311 /* Update vectorial force */
2312 fix1 = _mm_add_ps(fix1,tx);
2313 fiy1 = _mm_add_ps(fiy1,ty);
2314 fiz1 = _mm_add_ps(fiz1,tz);
2315
2316 fjx0 = _mm_add_ps(fjx0,tx);
2317 fjy0 = _mm_add_ps(fjy0,ty);
2318 fjz0 = _mm_add_ps(fjz0,tz);
2319
2320 }
2321
2322 /**************************
2323 * CALCULATE INTERACTIONS *
2324 **************************/
2325
2326 if (gmx_mm_any_lt(rsq11,rcutoff2))
2327 {
2328
2329 r11 = _mm_mul_ps(rsq11,rinv11);
2330 r11 = _mm_andnot_ps(dummy_mask,r11);
2331
2332 /* EWALD ELECTROSTATICS */
2333
2334 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2335 ewrt = _mm_mul_ps(r11,ewtabscale);
2336 ewitab = _mm_cvttps_epi32(ewrt);
2337 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2338 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2339 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2340 &ewtabF,&ewtabFn);
2341 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2342 felec = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
2343
2344 cutoff_mask = _mm_cmplt_ps(rsq11,rcutoff2);
2345
2346 fscal = felec;
2347
2348 fscal = _mm_and_ps(fscal,cutoff_mask);
2349
2350 fscal = _mm_andnot_ps(dummy_mask,fscal);
2351
2352 /* Calculate temporary vectorial force */
2353 tx = _mm_mul_ps(fscal,dx11);
2354 ty = _mm_mul_ps(fscal,dy11);
2355 tz = _mm_mul_ps(fscal,dz11);
2356
2357 /* Update vectorial force */
2358 fix1 = _mm_add_ps(fix1,tx);
2359 fiy1 = _mm_add_ps(fiy1,ty);
2360 fiz1 = _mm_add_ps(fiz1,tz);
2361
2362 fjx1 = _mm_add_ps(fjx1,tx);
2363 fjy1 = _mm_add_ps(fjy1,ty);
2364 fjz1 = _mm_add_ps(fjz1,tz);
2365
2366 }
2367
2368 /**************************
2369 * CALCULATE INTERACTIONS *
2370 **************************/
2371
2372 if (gmx_mm_any_lt(rsq12,rcutoff2))
2373 {
2374
2375 r12 = _mm_mul_ps(rsq12,rinv12);
2376 r12 = _mm_andnot_ps(dummy_mask,r12);
2377
2378 /* EWALD ELECTROSTATICS */
2379
2380 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2381 ewrt = _mm_mul_ps(r12,ewtabscale);
2382 ewitab = _mm_cvttps_epi32(ewrt);
2383 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2384 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2385 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2386 &ewtabF,&ewtabFn);
2387 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2388 felec = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
2389
2390 cutoff_mask = _mm_cmplt_ps(rsq12,rcutoff2);
2391
2392 fscal = felec;
2393
2394 fscal = _mm_and_ps(fscal,cutoff_mask);
2395
2396 fscal = _mm_andnot_ps(dummy_mask,fscal);
2397
2398 /* Calculate temporary vectorial force */
2399 tx = _mm_mul_ps(fscal,dx12);
2400 ty = _mm_mul_ps(fscal,dy12);
2401 tz = _mm_mul_ps(fscal,dz12);
2402
2403 /* Update vectorial force */
2404 fix1 = _mm_add_ps(fix1,tx);
2405 fiy1 = _mm_add_ps(fiy1,ty);
2406 fiz1 = _mm_add_ps(fiz1,tz);
2407
2408 fjx2 = _mm_add_ps(fjx2,tx);
2409 fjy2 = _mm_add_ps(fjy2,ty);
2410 fjz2 = _mm_add_ps(fjz2,tz);
2411
2412 }
2413
2414 /**************************
2415 * CALCULATE INTERACTIONS *
2416 **************************/
2417
2418 if (gmx_mm_any_lt(rsq20,rcutoff2))
2419 {
2420
2421 r20 = _mm_mul_ps(rsq20,rinv20);
2422 r20 = _mm_andnot_ps(dummy_mask,r20);
2423
2424 /* EWALD ELECTROSTATICS */
2425
2426 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2427 ewrt = _mm_mul_ps(r20,ewtabscale);
2428 ewitab = _mm_cvttps_epi32(ewrt);
2429 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2430 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2431 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2432 &ewtabF,&ewtabFn);
2433 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2434 felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
2435
2436 cutoff_mask = _mm_cmplt_ps(rsq20,rcutoff2);
2437
2438 fscal = felec;
2439
2440 fscal = _mm_and_ps(fscal,cutoff_mask);
2441
2442 fscal = _mm_andnot_ps(dummy_mask,fscal);
2443
2444 /* Calculate temporary vectorial force */
2445 tx = _mm_mul_ps(fscal,dx20);
2446 ty = _mm_mul_ps(fscal,dy20);
2447 tz = _mm_mul_ps(fscal,dz20);
2448
2449 /* Update vectorial force */
2450 fix2 = _mm_add_ps(fix2,tx);
2451 fiy2 = _mm_add_ps(fiy2,ty);
2452 fiz2 = _mm_add_ps(fiz2,tz);
2453
2454 fjx0 = _mm_add_ps(fjx0,tx);
2455 fjy0 = _mm_add_ps(fjy0,ty);
2456 fjz0 = _mm_add_ps(fjz0,tz);
2457
2458 }
2459
2460 /**************************
2461 * CALCULATE INTERACTIONS *
2462 **************************/
2463
2464 if (gmx_mm_any_lt(rsq21,rcutoff2))
2465 {
2466
2467 r21 = _mm_mul_ps(rsq21,rinv21);
2468 r21 = _mm_andnot_ps(dummy_mask,r21);
2469
2470 /* EWALD ELECTROSTATICS */
2471
2472 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2473 ewrt = _mm_mul_ps(r21,ewtabscale);
2474 ewitab = _mm_cvttps_epi32(ewrt);
2475 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2476 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2477 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2478 &ewtabF,&ewtabFn);
2479 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2480 felec = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
2481
2482 cutoff_mask = _mm_cmplt_ps(rsq21,rcutoff2);
2483
2484 fscal = felec;
2485
2486 fscal = _mm_and_ps(fscal,cutoff_mask);
2487
2488 fscal = _mm_andnot_ps(dummy_mask,fscal);
2489
2490 /* Calculate temporary vectorial force */
2491 tx = _mm_mul_ps(fscal,dx21);
2492 ty = _mm_mul_ps(fscal,dy21);
2493 tz = _mm_mul_ps(fscal,dz21);
2494
2495 /* Update vectorial force */
2496 fix2 = _mm_add_ps(fix2,tx);
2497 fiy2 = _mm_add_ps(fiy2,ty);
2498 fiz2 = _mm_add_ps(fiz2,tz);
2499
2500 fjx1 = _mm_add_ps(fjx1,tx);
2501 fjy1 = _mm_add_ps(fjy1,ty);
2502 fjz1 = _mm_add_ps(fjz1,tz);
2503
2504 }
2505
2506 /**************************
2507 * CALCULATE INTERACTIONS *
2508 **************************/
2509
2510 if (gmx_mm_any_lt(rsq22,rcutoff2))
2511 {
2512
2513 r22 = _mm_mul_ps(rsq22,rinv22);
2514 r22 = _mm_andnot_ps(dummy_mask,r22);
2515
2516 /* EWALD ELECTROSTATICS */
2517
2518 /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2519 ewrt = _mm_mul_ps(r22,ewtabscale);
2520 ewitab = _mm_cvttps_epi32(ewrt);
2521 eweps = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2522 gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2523 ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2524 &ewtabF,&ewtabFn);
2525 felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2526 felec = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
2527
2528 cutoff_mask = _mm_cmplt_ps(rsq22,rcutoff2);
2529
2530 fscal = felec;
2531
2532 fscal = _mm_and_ps(fscal,cutoff_mask);
2533
2534 fscal = _mm_andnot_ps(dummy_mask,fscal);
2535
2536 /* Calculate temporary vectorial force */
2537 tx = _mm_mul_ps(fscal,dx22);
2538 ty = _mm_mul_ps(fscal,dy22);
2539 tz = _mm_mul_ps(fscal,dz22);
2540
2541 /* Update vectorial force */
2542 fix2 = _mm_add_ps(fix2,tx);
2543 fiy2 = _mm_add_ps(fiy2,ty);
2544 fiz2 = _mm_add_ps(fiz2,tz);
2545
2546 fjx2 = _mm_add_ps(fjx2,tx);
2547 fjy2 = _mm_add_ps(fjy2,ty);
2548 fjz2 = _mm_add_ps(fjz2,tz);
2549
2550 }
2551
2552 fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2553 fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2554 fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2555 fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2556
2557 gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2558 fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2559
2560 /* Inner loop uses 360 flops */
2561 }
2562
2563 /* End of innermost loop */
2564
2565 gmx_mm_update_iforce_3atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
2566 f+i_coord_offset,fshift+i_shift_offset);
2567
2568 /* Increment number of inner iterations */
2569 inneriter += j_index_end - j_index_start;
2570
2571 /* Outer loop uses 18 flops */
2572 }
2573
2574 /* Increment number of outer iterations */
2575 outeriter += nri;
2576
2577 /* Update outer/inner flops */
2578
2579 inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3W3_F,outeriter*18 + inneriter*360);
2580 }
The ultimate molecular dynamics simulation package