src/gmxlib/nonbonded/nb_kernel_sse4_1_double/nb_kernel_ElecEwSh_VdwNone_GeomW3P1_sse4_1_double.c

   1 /*
   2  * Note: this file was generated by the Gromacs sse4_1_double kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_sse4_1_double.h"
  34 #include "kernelutil_x86_sse4_1_double.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwNone_GeomW3P1_VF_sse4_1_double
  38  * Electrostatics interaction: Ewald
  39  * VdW interaction:            None
  40  * Geometry:                   Water3-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecEwSh_VdwNone_GeomW3P1_VF_sse4_1_double
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB;
  61     int              j_coord_offsetA,j_coord_offsetB;
  62     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  63     real             rcutoff_scalar;
  64     real             *shiftvec,*fshift,*x,*f;
  65     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  66     int              vdwioffset0;
  67     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  68     int              vdwioffset1;
  69     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  70     int              vdwioffset2;
  71     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  72     int              vdwjidx0A,vdwjidx0B;
  73     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  74     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  75     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  76     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  77     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  78     real             *charge;
  79     __m128i          ewitab;
  80     __m128d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
  81     real             *ewtab;
  82     __m128d          dummy_mask,cutoff_mask;
  83     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  84     __m128d          one     = _mm_set1_pd(1.0);
  85     __m128d          two     = _mm_set1_pd(2.0);
  86     x                = xx[0];
  87     f                = ff[0];
  88
  89     nri              = nlist->nri;
  90     iinr             = nlist->iinr;
  91     jindex           = nlist->jindex;
  92     jjnr             = nlist->jjnr;
  93     shiftidx         = nlist->shift;
  94     gid              = nlist->gid;
  95     shiftvec         = fr->shift_vec[0];
  96     fshift           = fr->fshift[0];
  97     facel            = _mm_set1_pd(fr->epsfac);
  98     charge           = mdatoms->chargeA;
  99
 100     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 101     ewtab            = fr->ic->tabq_coul_FDV0;
 102     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 103     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 104
 105     /* Setup water-specific parameters */
 106     inr              = nlist->iinr[0];
 107     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 108     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 109     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 110
 111     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 112     rcutoff_scalar   = fr->rcoulomb;
 113     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 114     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 115
 116     /* Avoid stupid compiler warnings */
 117     jnrA = jnrB = 0;
 118     j_coord_offsetA = 0;
 119     j_coord_offsetB = 0;
 120
 121     outeriter        = 0;
 122     inneriter        = 0;
 123
 124     /* Start outer loop over neighborlists */
 125     for(iidx=0; iidx<nri; iidx++)
 126     {
 127         /* Load shift vector for this list */
 128         i_shift_offset   = DIM*shiftidx[iidx];
 129
 130         /* Load limits for loop over neighbors */
 131         j_index_start    = jindex[iidx];
 132         j_index_end      = jindex[iidx+1];
 133
 134         /* Get outer coordinate index */
 135         inr              = iinr[iidx];
 136         i_coord_offset   = DIM*inr;
 137
 138         /* Load i particle coords and add shift vector */
 139         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 140                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 141
 142         fix0             = _mm_setzero_pd();
 143         fiy0             = _mm_setzero_pd();
 144         fiz0             = _mm_setzero_pd();
 145         fix1             = _mm_setzero_pd();
 146         fiy1             = _mm_setzero_pd();
 147         fiz1             = _mm_setzero_pd();
 148         fix2             = _mm_setzero_pd();
 149         fiy2             = _mm_setzero_pd();
 150         fiz2             = _mm_setzero_pd();
 151
 152         /* Reset potential sums */
 153         velecsum         = _mm_setzero_pd();
 154
 155         /* Start inner kernel loop */
 156         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 157         {
 158
 159             /* Get j neighbor index, and coordinate index */
 160             jnrA             = jjnr[jidx];
 161             jnrB             = jjnr[jidx+1];
 162             j_coord_offsetA  = DIM*jnrA;
 163             j_coord_offsetB  = DIM*jnrB;
 164
 165             /* load j atom coordinates */
 166             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 167                                               &jx0,&jy0,&jz0);
 168
 169             /* Calculate displacement vector */
 170             dx00             = _mm_sub_pd(ix0,jx0);
 171             dy00             = _mm_sub_pd(iy0,jy0);
 172             dz00             = _mm_sub_pd(iz0,jz0);
 173             dx10             = _mm_sub_pd(ix1,jx0);
 174             dy10             = _mm_sub_pd(iy1,jy0);
 175             dz10             = _mm_sub_pd(iz1,jz0);
 176             dx20             = _mm_sub_pd(ix2,jx0);
 177             dy20             = _mm_sub_pd(iy2,jy0);
 178             dz20             = _mm_sub_pd(iz2,jz0);
 179
 180             /* Calculate squared distance and things based on it */
 181             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 182             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 183             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 184
 185             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 186             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 187             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 188
 189             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 190             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 191             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 192
 193             /* Load parameters for j particles */
 194             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 195
 196             fjx0             = _mm_setzero_pd();
 197             fjy0             = _mm_setzero_pd();
 198             fjz0             = _mm_setzero_pd();
 199
 200             /**************************
 201              * CALCULATE INTERACTIONS *
 202              **************************/
 203
 204             if (gmx_mm_any_lt(rsq00,rcutoff2))
 205             {
 206
 207             r00              = _mm_mul_pd(rsq00,rinv00);
 208
 209             /* Compute parameters for interactions between i and j atoms */
 210             qq00             = _mm_mul_pd(iq0,jq0);
 211
 212             /* EWALD ELECTROSTATICS */
 213
 214             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 215             ewrt             = _mm_mul_pd(r00,ewtabscale);
 216             ewitab           = _mm_cvttpd_epi32(ewrt);
 217             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 218             ewitab           = _mm_slli_epi32(ewitab,2);
 219             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 220             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 221             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 222             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 223             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 224             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 225             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 226             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 227             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_sub_pd(rinv00,sh_ewald),velec));
 228             felec            = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
 229
 230             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 231
 232             /* Update potential sum for this i atom from the interaction with this j atom. */
 233             velec            = _mm_and_pd(velec,cutoff_mask);
 234             velecsum         = _mm_add_pd(velecsum,velec);
 235
 236             fscal            = felec;
 237
 238             fscal            = _mm_and_pd(fscal,cutoff_mask);
 239
 240             /* Calculate temporary vectorial force */
 241             tx               = _mm_mul_pd(fscal,dx00);
 242             ty               = _mm_mul_pd(fscal,dy00);
 243             tz               = _mm_mul_pd(fscal,dz00);
 244
 245             /* Update vectorial force */
 246             fix0             = _mm_add_pd(fix0,tx);
 247             fiy0             = _mm_add_pd(fiy0,ty);
 248             fiz0             = _mm_add_pd(fiz0,tz);
 249
 250             fjx0             = _mm_add_pd(fjx0,tx);
 251             fjy0             = _mm_add_pd(fjy0,ty);
 252             fjz0             = _mm_add_pd(fjz0,tz);
 253
 254             }
 255
 256             /**************************
 257              * CALCULATE INTERACTIONS *
 258              **************************/
 259
 260             if (gmx_mm_any_lt(rsq10,rcutoff2))
 261             {
 262
 263             r10              = _mm_mul_pd(rsq10,rinv10);
 264
 265             /* Compute parameters for interactions between i and j atoms */
 266             qq10             = _mm_mul_pd(iq1,jq0);
 267
 268             /* EWALD ELECTROSTATICS */
 269
 270             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 271             ewrt             = _mm_mul_pd(r10,ewtabscale);
 272             ewitab           = _mm_cvttpd_epi32(ewrt);
 273             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 274             ewitab           = _mm_slli_epi32(ewitab,2);
 275             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 276             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 277             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 278             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 279             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 280             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 281             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 282             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 283             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 284             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 285
 286             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 287
 288             /* Update potential sum for this i atom from the interaction with this j atom. */
 289             velec            = _mm_and_pd(velec,cutoff_mask);
 290             velecsum         = _mm_add_pd(velecsum,velec);
 291
 292             fscal            = felec;
 293
 294             fscal            = _mm_and_pd(fscal,cutoff_mask);
 295
 296             /* Calculate temporary vectorial force */
 297             tx               = _mm_mul_pd(fscal,dx10);
 298             ty               = _mm_mul_pd(fscal,dy10);
 299             tz               = _mm_mul_pd(fscal,dz10);
 300
 301             /* Update vectorial force */
 302             fix1             = _mm_add_pd(fix1,tx);
 303             fiy1             = _mm_add_pd(fiy1,ty);
 304             fiz1             = _mm_add_pd(fiz1,tz);
 305
 306             fjx0             = _mm_add_pd(fjx0,tx);
 307             fjy0             = _mm_add_pd(fjy0,ty);
 308             fjz0             = _mm_add_pd(fjz0,tz);
 309
 310             }
 311
 312             /**************************
 313              * CALCULATE INTERACTIONS *
 314              **************************/
 315
 316             if (gmx_mm_any_lt(rsq20,rcutoff2))
 317             {
 318
 319             r20              = _mm_mul_pd(rsq20,rinv20);
 320
 321             /* Compute parameters for interactions between i and j atoms */
 322             qq20             = _mm_mul_pd(iq2,jq0);
 323
 324             /* EWALD ELECTROSTATICS */
 325
 326             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 327             ewrt             = _mm_mul_pd(r20,ewtabscale);
 328             ewitab           = _mm_cvttpd_epi32(ewrt);
 329             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 330             ewitab           = _mm_slli_epi32(ewitab,2);
 331             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 332             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 333             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 334             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 335             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 336             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 337             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 338             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 339             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 340             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 341
 342             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 343
 344             /* Update potential sum for this i atom from the interaction with this j atom. */
 345             velec            = _mm_and_pd(velec,cutoff_mask);
 346             velecsum         = _mm_add_pd(velecsum,velec);
 347
 348             fscal            = felec;
 349
 350             fscal            = _mm_and_pd(fscal,cutoff_mask);
 351
 352             /* Calculate temporary vectorial force */
 353             tx               = _mm_mul_pd(fscal,dx20);
 354             ty               = _mm_mul_pd(fscal,dy20);
 355             tz               = _mm_mul_pd(fscal,dz20);
 356
 357             /* Update vectorial force */
 358             fix2             = _mm_add_pd(fix2,tx);
 359             fiy2             = _mm_add_pd(fiy2,ty);
 360             fiz2             = _mm_add_pd(fiz2,tz);
 361
 362             fjx0             = _mm_add_pd(fjx0,tx);
 363             fjy0             = _mm_add_pd(fjy0,ty);
 364             fjz0             = _mm_add_pd(fjz0,tz);
 365
 366             }
 367
 368             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 369
 370             /* Inner loop uses 141 flops */
 371         }
 372
 373         if(jidx<j_index_end)
 374         {
 375
 376             jnrA             = jjnr[jidx];
 377             j_coord_offsetA  = DIM*jnrA;
 378
 379             /* load j atom coordinates */
 380             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 381                                               &jx0,&jy0,&jz0);
 382
 383             /* Calculate displacement vector */
 384             dx00             = _mm_sub_pd(ix0,jx0);
 385             dy00             = _mm_sub_pd(iy0,jy0);
 386             dz00             = _mm_sub_pd(iz0,jz0);
 387             dx10             = _mm_sub_pd(ix1,jx0);
 388             dy10             = _mm_sub_pd(iy1,jy0);
 389             dz10             = _mm_sub_pd(iz1,jz0);
 390             dx20             = _mm_sub_pd(ix2,jx0);
 391             dy20             = _mm_sub_pd(iy2,jy0);
 392             dz20             = _mm_sub_pd(iz2,jz0);
 393
 394             /* Calculate squared distance and things based on it */
 395             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 396             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 397             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 398
 399             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 400             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 401             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 402
 403             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 404             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 405             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 406
 407             /* Load parameters for j particles */
 408             jq0              = _mm_load_sd(charge+jnrA+0);
 409
 410             fjx0             = _mm_setzero_pd();
 411             fjy0             = _mm_setzero_pd();
 412             fjz0             = _mm_setzero_pd();
 413
 414             /**************************
 415              * CALCULATE INTERACTIONS *
 416              **************************/
 417
 418             if (gmx_mm_any_lt(rsq00,rcutoff2))
 419             {
 420
 421             r00              = _mm_mul_pd(rsq00,rinv00);
 422
 423             /* Compute parameters for interactions between i and j atoms */
 424             qq00             = _mm_mul_pd(iq0,jq0);
 425
 426             /* EWALD ELECTROSTATICS */
 427
 428             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 429             ewrt             = _mm_mul_pd(r00,ewtabscale);
 430             ewitab           = _mm_cvttpd_epi32(ewrt);
 431             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 432             ewitab           = _mm_slli_epi32(ewitab,2);
 433             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 434             ewtabD           = _mm_setzero_pd();
 435             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 436             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 437             ewtabFn          = _mm_setzero_pd();
 438             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 439             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 440             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 441             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_sub_pd(rinv00,sh_ewald),velec));
 442             felec            = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
 443
 444             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 445
 446             /* Update potential sum for this i atom from the interaction with this j atom. */
 447             velec            = _mm_and_pd(velec,cutoff_mask);
 448             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 449             velecsum         = _mm_add_pd(velecsum,velec);
 450
 451             fscal            = felec;
 452
 453             fscal            = _mm_and_pd(fscal,cutoff_mask);
 454
 455             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 456
 457             /* Calculate temporary vectorial force */
 458             tx               = _mm_mul_pd(fscal,dx00);
 459             ty               = _mm_mul_pd(fscal,dy00);
 460             tz               = _mm_mul_pd(fscal,dz00);
 461
 462             /* Update vectorial force */
 463             fix0             = _mm_add_pd(fix0,tx);
 464             fiy0             = _mm_add_pd(fiy0,ty);
 465             fiz0             = _mm_add_pd(fiz0,tz);
 466
 467             fjx0             = _mm_add_pd(fjx0,tx);
 468             fjy0             = _mm_add_pd(fjy0,ty);
 469             fjz0             = _mm_add_pd(fjz0,tz);
 470
 471             }
 472
 473             /**************************
 474              * CALCULATE INTERACTIONS *
 475              **************************/
 476
 477             if (gmx_mm_any_lt(rsq10,rcutoff2))
 478             {
 479
 480             r10              = _mm_mul_pd(rsq10,rinv10);
 481
 482             /* Compute parameters for interactions between i and j atoms */
 483             qq10             = _mm_mul_pd(iq1,jq0);
 484
 485             /* EWALD ELECTROSTATICS */
 486
 487             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 488             ewrt             = _mm_mul_pd(r10,ewtabscale);
 489             ewitab           = _mm_cvttpd_epi32(ewrt);
 490             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 491             ewitab           = _mm_slli_epi32(ewitab,2);
 492             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 493             ewtabD           = _mm_setzero_pd();
 494             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 495             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 496             ewtabFn          = _mm_setzero_pd();
 497             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 498             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 499             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 500             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 501             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 502
 503             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 504
 505             /* Update potential sum for this i atom from the interaction with this j atom. */
 506             velec            = _mm_and_pd(velec,cutoff_mask);
 507             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 508             velecsum         = _mm_add_pd(velecsum,velec);
 509
 510             fscal            = felec;
 511
 512             fscal            = _mm_and_pd(fscal,cutoff_mask);
 513
 514             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 515
 516             /* Calculate temporary vectorial force */
 517             tx               = _mm_mul_pd(fscal,dx10);
 518             ty               = _mm_mul_pd(fscal,dy10);
 519             tz               = _mm_mul_pd(fscal,dz10);
 520
 521             /* Update vectorial force */
 522             fix1             = _mm_add_pd(fix1,tx);
 523             fiy1             = _mm_add_pd(fiy1,ty);
 524             fiz1             = _mm_add_pd(fiz1,tz);
 525
 526             fjx0             = _mm_add_pd(fjx0,tx);
 527             fjy0             = _mm_add_pd(fjy0,ty);
 528             fjz0             = _mm_add_pd(fjz0,tz);
 529
 530             }
 531
 532             /**************************
 533              * CALCULATE INTERACTIONS *
 534              **************************/
 535
 536             if (gmx_mm_any_lt(rsq20,rcutoff2))
 537             {
 538
 539             r20              = _mm_mul_pd(rsq20,rinv20);
 540
 541             /* Compute parameters for interactions between i and j atoms */
 542             qq20             = _mm_mul_pd(iq2,jq0);
 543
 544             /* EWALD ELECTROSTATICS */
 545
 546             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 547             ewrt             = _mm_mul_pd(r20,ewtabscale);
 548             ewitab           = _mm_cvttpd_epi32(ewrt);
 549             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 550             ewitab           = _mm_slli_epi32(ewitab,2);
 551             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 552             ewtabD           = _mm_setzero_pd();
 553             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 554             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 555             ewtabFn          = _mm_setzero_pd();
 556             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 557             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 558             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 559             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 560             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 561
 562             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 563
 564             /* Update potential sum for this i atom from the interaction with this j atom. */
 565             velec            = _mm_and_pd(velec,cutoff_mask);
 566             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 567             velecsum         = _mm_add_pd(velecsum,velec);
 568
 569             fscal            = felec;
 570
 571             fscal            = _mm_and_pd(fscal,cutoff_mask);
 572
 573             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 574
 575             /* Calculate temporary vectorial force */
 576             tx               = _mm_mul_pd(fscal,dx20);
 577             ty               = _mm_mul_pd(fscal,dy20);
 578             tz               = _mm_mul_pd(fscal,dz20);
 579
 580             /* Update vectorial force */
 581             fix2             = _mm_add_pd(fix2,tx);
 582             fiy2             = _mm_add_pd(fiy2,ty);
 583             fiz2             = _mm_add_pd(fiz2,tz);
 584
 585             fjx0             = _mm_add_pd(fjx0,tx);
 586             fjy0             = _mm_add_pd(fjy0,ty);
 587             fjz0             = _mm_add_pd(fjz0,tz);
 588
 589             }
 590
 591             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 592
 593             /* Inner loop uses 141 flops */
 594         }
 595
 596         /* End of innermost loop */
 597
 598         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
 599                                               f+i_coord_offset,fshift+i_shift_offset);
 600
 601         ggid                        = gid[iidx];
 602         /* Update potential energies */
 603         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 604
 605         /* Increment number of inner iterations */
 606         inneriter                  += j_index_end - j_index_start;
 607
 608         /* Outer loop uses 19 flops */
 609     }
 610
 611     /* Increment number of outer iterations */
 612     outeriter        += nri;
 613
 614     /* Update outer/inner flops */
 615
 616     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3_VF,outeriter*19 + inneriter*141);
 617 }
 618 /*
 619  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwNone_GeomW3P1_F_sse4_1_double
 620  * Electrostatics interaction: Ewald
 621  * VdW interaction:            None
 622  * Geometry:                   Water3-Particle
 623  * Calculate force/pot:        Force
 624  */
 625 void
 626 nb_kernel_ElecEwSh_VdwNone_GeomW3P1_F_sse4_1_double
 627                     (t_nblist * gmx_restrict                nlist,
 628                      rvec * gmx_restrict                    xx,
 629                      rvec * gmx_restrict                    ff,
 630                      t_forcerec * gmx_restrict              fr,
 631                      t_mdatoms * gmx_restrict               mdatoms,
 632                      nb_kernel_data_t * gmx_restrict        kernel_data,
 633                      t_nrnb * gmx_restrict                  nrnb)
 634 {
 635     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 636      * just 0 for non-waters.
 637      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 638      * jnr indices corresponding to data put in the four positions in the SIMD register.
 639      */
 640     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 641     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 642     int              jnrA,jnrB;
 643     int              j_coord_offsetA,j_coord_offsetB;
 644     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 645     real             rcutoff_scalar;
 646     real             *shiftvec,*fshift,*x,*f;
 647     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 648     int              vdwioffset0;
 649     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 650     int              vdwioffset1;
 651     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 652     int              vdwioffset2;
 653     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 654     int              vdwjidx0A,vdwjidx0B;
 655     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 656     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 657     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 658     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 659     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 660     real             *charge;
 661     __m128i          ewitab;
 662     __m128d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 663     real             *ewtab;
 664     __m128d          dummy_mask,cutoff_mask;
 665     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 666     __m128d          one     = _mm_set1_pd(1.0);
 667     __m128d          two     = _mm_set1_pd(2.0);
 668     x                = xx[0];
 669     f                = ff[0];
 670
 671     nri              = nlist->nri;
 672     iinr             = nlist->iinr;
 673     jindex           = nlist->jindex;
 674     jjnr             = nlist->jjnr;
 675     shiftidx         = nlist->shift;
 676     gid              = nlist->gid;
 677     shiftvec         = fr->shift_vec[0];
 678     fshift           = fr->fshift[0];
 679     facel            = _mm_set1_pd(fr->epsfac);
 680     charge           = mdatoms->chargeA;
 681
 682     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 683     ewtab            = fr->ic->tabq_coul_F;
 684     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 685     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 686
 687     /* Setup water-specific parameters */
 688     inr              = nlist->iinr[0];
 689     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 690     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 691     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 692
 693     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 694     rcutoff_scalar   = fr->rcoulomb;
 695     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 696     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 697
 698     /* Avoid stupid compiler warnings */
 699     jnrA = jnrB = 0;
 700     j_coord_offsetA = 0;
 701     j_coord_offsetB = 0;
 702
 703     outeriter        = 0;
 704     inneriter        = 0;
 705
 706     /* Start outer loop over neighborlists */
 707     for(iidx=0; iidx<nri; iidx++)
 708     {
 709         /* Load shift vector for this list */
 710         i_shift_offset   = DIM*shiftidx[iidx];
 711
 712         /* Load limits for loop over neighbors */
 713         j_index_start    = jindex[iidx];
 714         j_index_end      = jindex[iidx+1];
 715
 716         /* Get outer coordinate index */
 717         inr              = iinr[iidx];
 718         i_coord_offset   = DIM*inr;
 719
 720         /* Load i particle coords and add shift vector */
 721         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 722                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 723
 724         fix0             = _mm_setzero_pd();
 725         fiy0             = _mm_setzero_pd();
 726         fiz0             = _mm_setzero_pd();
 727         fix1             = _mm_setzero_pd();
 728         fiy1             = _mm_setzero_pd();
 729         fiz1             = _mm_setzero_pd();
 730         fix2             = _mm_setzero_pd();
 731         fiy2             = _mm_setzero_pd();
 732         fiz2             = _mm_setzero_pd();
 733
 734         /* Start inner kernel loop */
 735         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 736         {
 737
 738             /* Get j neighbor index, and coordinate index */
 739             jnrA             = jjnr[jidx];
 740             jnrB             = jjnr[jidx+1];
 741             j_coord_offsetA  = DIM*jnrA;
 742             j_coord_offsetB  = DIM*jnrB;
 743
 744             /* load j atom coordinates */
 745             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 746                                               &jx0,&jy0,&jz0);
 747
 748             /* Calculate displacement vector */
 749             dx00             = _mm_sub_pd(ix0,jx0);
 750             dy00             = _mm_sub_pd(iy0,jy0);
 751             dz00             = _mm_sub_pd(iz0,jz0);
 752             dx10             = _mm_sub_pd(ix1,jx0);
 753             dy10             = _mm_sub_pd(iy1,jy0);
 754             dz10             = _mm_sub_pd(iz1,jz0);
 755             dx20             = _mm_sub_pd(ix2,jx0);
 756             dy20             = _mm_sub_pd(iy2,jy0);
 757             dz20             = _mm_sub_pd(iz2,jz0);
 758
 759             /* Calculate squared distance and things based on it */
 760             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 761             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 762             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 763
 764             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 765             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 766             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 767
 768             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 769             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 770             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 771
 772             /* Load parameters for j particles */
 773             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 774
 775             fjx0             = _mm_setzero_pd();
 776             fjy0             = _mm_setzero_pd();
 777             fjz0             = _mm_setzero_pd();
 778
 779             /**************************
 780              * CALCULATE INTERACTIONS *
 781              **************************/
 782
 783             if (gmx_mm_any_lt(rsq00,rcutoff2))
 784             {
 785
 786             r00              = _mm_mul_pd(rsq00,rinv00);
 787
 788             /* Compute parameters for interactions between i and j atoms */
 789             qq00             = _mm_mul_pd(iq0,jq0);
 790
 791             /* EWALD ELECTROSTATICS */
 792
 793             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 794             ewrt             = _mm_mul_pd(r00,ewtabscale);
 795             ewitab           = _mm_cvttpd_epi32(ewrt);
 796             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 797             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
 798                                          &ewtabF,&ewtabFn);
 799             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
 800             felec            = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
 801
 802             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 803
 804             fscal            = felec;
 805
 806             fscal            = _mm_and_pd(fscal,cutoff_mask);
 807
 808             /* Calculate temporary vectorial force */
 809             tx               = _mm_mul_pd(fscal,dx00);
 810             ty               = _mm_mul_pd(fscal,dy00);
 811             tz               = _mm_mul_pd(fscal,dz00);
 812
 813             /* Update vectorial force */
 814             fix0             = _mm_add_pd(fix0,tx);
 815             fiy0             = _mm_add_pd(fiy0,ty);
 816             fiz0             = _mm_add_pd(fiz0,tz);
 817
 818             fjx0             = _mm_add_pd(fjx0,tx);
 819             fjy0             = _mm_add_pd(fjy0,ty);
 820             fjz0             = _mm_add_pd(fjz0,tz);
 821
 822             }
 823
 824             /**************************
 825              * CALCULATE INTERACTIONS *
 826              **************************/
 827
 828             if (gmx_mm_any_lt(rsq10,rcutoff2))
 829             {
 830
 831             r10              = _mm_mul_pd(rsq10,rinv10);
 832
 833             /* Compute parameters for interactions between i and j atoms */
 834             qq10             = _mm_mul_pd(iq1,jq0);
 835
 836             /* EWALD ELECTROSTATICS */
 837
 838             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 839             ewrt             = _mm_mul_pd(r10,ewtabscale);
 840             ewitab           = _mm_cvttpd_epi32(ewrt);
 841             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 842             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
 843                                          &ewtabF,&ewtabFn);
 844             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
 845             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 846
 847             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 848
 849             fscal            = felec;
 850
 851             fscal            = _mm_and_pd(fscal,cutoff_mask);
 852
 853             /* Calculate temporary vectorial force */
 854             tx               = _mm_mul_pd(fscal,dx10);
 855             ty               = _mm_mul_pd(fscal,dy10);
 856             tz               = _mm_mul_pd(fscal,dz10);
 857
 858             /* Update vectorial force */
 859             fix1             = _mm_add_pd(fix1,tx);
 860             fiy1             = _mm_add_pd(fiy1,ty);
 861             fiz1             = _mm_add_pd(fiz1,tz);
 862
 863             fjx0             = _mm_add_pd(fjx0,tx);
 864             fjy0             = _mm_add_pd(fjy0,ty);
 865             fjz0             = _mm_add_pd(fjz0,tz);
 866
 867             }
 868
 869             /**************************
 870              * CALCULATE INTERACTIONS *
 871              **************************/
 872
 873             if (gmx_mm_any_lt(rsq20,rcutoff2))
 874             {
 875
 876             r20              = _mm_mul_pd(rsq20,rinv20);
 877
 878             /* Compute parameters for interactions between i and j atoms */
 879             qq20             = _mm_mul_pd(iq2,jq0);
 880
 881             /* EWALD ELECTROSTATICS */
 882
 883             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 884             ewrt             = _mm_mul_pd(r20,ewtabscale);
 885             ewitab           = _mm_cvttpd_epi32(ewrt);
 886             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 887             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
 888                                          &ewtabF,&ewtabFn);
 889             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
 890             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 891
 892             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 893
 894             fscal            = felec;
 895
 896             fscal            = _mm_and_pd(fscal,cutoff_mask);
 897
 898             /* Calculate temporary vectorial force */
 899             tx               = _mm_mul_pd(fscal,dx20);
 900             ty               = _mm_mul_pd(fscal,dy20);
 901             tz               = _mm_mul_pd(fscal,dz20);
 902
 903             /* Update vectorial force */
 904             fix2             = _mm_add_pd(fix2,tx);
 905             fiy2             = _mm_add_pd(fiy2,ty);
 906             fiz2             = _mm_add_pd(fiz2,tz);
 907
 908             fjx0             = _mm_add_pd(fjx0,tx);
 909             fjy0             = _mm_add_pd(fjy0,ty);
 910             fjz0             = _mm_add_pd(fjz0,tz);
 911
 912             }
 913
 914             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 915
 916             /* Inner loop uses 120 flops */
 917         }
 918
 919         if(jidx<j_index_end)
 920         {
 921
 922             jnrA             = jjnr[jidx];
 923             j_coord_offsetA  = DIM*jnrA;
 924
 925             /* load j atom coordinates */
 926             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 927                                               &jx0,&jy0,&jz0);
 928
 929             /* Calculate displacement vector */
 930             dx00             = _mm_sub_pd(ix0,jx0);
 931             dy00             = _mm_sub_pd(iy0,jy0);
 932             dz00             = _mm_sub_pd(iz0,jz0);
 933             dx10             = _mm_sub_pd(ix1,jx0);
 934             dy10             = _mm_sub_pd(iy1,jy0);
 935             dz10             = _mm_sub_pd(iz1,jz0);
 936             dx20             = _mm_sub_pd(ix2,jx0);
 937             dy20             = _mm_sub_pd(iy2,jy0);
 938             dz20             = _mm_sub_pd(iz2,jz0);
 939
 940             /* Calculate squared distance and things based on it */
 941             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 942             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 943             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 944
 945             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 946             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 947             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 948
 949             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 950             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 951             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 952
 953             /* Load parameters for j particles */
 954             jq0              = _mm_load_sd(charge+jnrA+0);
 955
 956             fjx0             = _mm_setzero_pd();
 957             fjy0             = _mm_setzero_pd();
 958             fjz0             = _mm_setzero_pd();
 959
 960             /**************************
 961              * CALCULATE INTERACTIONS *
 962              **************************/
 963
 964             if (gmx_mm_any_lt(rsq00,rcutoff2))
 965             {
 966
 967             r00              = _mm_mul_pd(rsq00,rinv00);
 968
 969             /* Compute parameters for interactions between i and j atoms */
 970             qq00             = _mm_mul_pd(iq0,jq0);
 971
 972             /* EWALD ELECTROSTATICS */
 973
 974             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 975             ewrt             = _mm_mul_pd(r00,ewtabscale);
 976             ewitab           = _mm_cvttpd_epi32(ewrt);
 977             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 978             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
 979             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
 980             felec            = _mm_mul_pd(_mm_mul_pd(qq00,rinv00),_mm_sub_pd(rinvsq00,felec));
 981
 982             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 983
 984             fscal            = felec;
 985
 986             fscal            = _mm_and_pd(fscal,cutoff_mask);
 987
 988             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 989
 990             /* Calculate temporary vectorial force */
 991             tx               = _mm_mul_pd(fscal,dx00);
 992             ty               = _mm_mul_pd(fscal,dy00);
 993             tz               = _mm_mul_pd(fscal,dz00);
 994
 995             /* Update vectorial force */
 996             fix0             = _mm_add_pd(fix0,tx);
 997             fiy0             = _mm_add_pd(fiy0,ty);
 998             fiz0             = _mm_add_pd(fiz0,tz);
 999
1000             fjx0             = _mm_add_pd(fjx0,tx);
1001             fjy0             = _mm_add_pd(fjy0,ty);
1002             fjz0             = _mm_add_pd(fjz0,tz);
1003
1004             }
1005
1006             /**************************
1007              * CALCULATE INTERACTIONS *
1008              **************************/
1009
1010             if (gmx_mm_any_lt(rsq10,rcutoff2))
1011             {
1012
1013             r10              = _mm_mul_pd(rsq10,rinv10);
1014
1015             /* Compute parameters for interactions between i and j atoms */
1016             qq10             = _mm_mul_pd(iq1,jq0);
1017
1018             /* EWALD ELECTROSTATICS */
1019
1020             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1021             ewrt             = _mm_mul_pd(r10,ewtabscale);
1022             ewitab           = _mm_cvttpd_epi32(ewrt);
1023             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1024             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1025             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1026             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1027
1028             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
1029
1030             fscal            = felec;
1031
1032             fscal            = _mm_and_pd(fscal,cutoff_mask);
1033
1034             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1035
1036             /* Calculate temporary vectorial force */
1037             tx               = _mm_mul_pd(fscal,dx10);
1038             ty               = _mm_mul_pd(fscal,dy10);
1039             tz               = _mm_mul_pd(fscal,dz10);
1040
1041             /* Update vectorial force */
1042             fix1             = _mm_add_pd(fix1,tx);
1043             fiy1             = _mm_add_pd(fiy1,ty);
1044             fiz1             = _mm_add_pd(fiz1,tz);
1045
1046             fjx0             = _mm_add_pd(fjx0,tx);
1047             fjy0             = _mm_add_pd(fjy0,ty);
1048             fjz0             = _mm_add_pd(fjz0,tz);
1049
1050             }
1051
1052             /**************************
1053              * CALCULATE INTERACTIONS *
1054              **************************/
1055
1056             if (gmx_mm_any_lt(rsq20,rcutoff2))
1057             {
1058
1059             r20              = _mm_mul_pd(rsq20,rinv20);
1060
1061             /* Compute parameters for interactions between i and j atoms */
1062             qq20             = _mm_mul_pd(iq2,jq0);
1063
1064             /* EWALD ELECTROSTATICS */
1065
1066             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1067             ewrt             = _mm_mul_pd(r20,ewtabscale);
1068             ewitab           = _mm_cvttpd_epi32(ewrt);
1069             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1070             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1071             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1072             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1073
1074             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
1075
1076             fscal            = felec;
1077
1078             fscal            = _mm_and_pd(fscal,cutoff_mask);
1079
1080             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1081
1082             /* Calculate temporary vectorial force */
1083             tx               = _mm_mul_pd(fscal,dx20);
1084             ty               = _mm_mul_pd(fscal,dy20);
1085             tz               = _mm_mul_pd(fscal,dz20);
1086
1087             /* Update vectorial force */
1088             fix2             = _mm_add_pd(fix2,tx);
1089             fiy2             = _mm_add_pd(fiy2,ty);
1090             fiz2             = _mm_add_pd(fiz2,tz);
1091
1092             fjx0             = _mm_add_pd(fjx0,tx);
1093             fjy0             = _mm_add_pd(fjy0,ty);
1094             fjz0             = _mm_add_pd(fjz0,tz);
1095
1096             }
1097
1098             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1099
1100             /* Inner loop uses 120 flops */
1101         }
1102
1103         /* End of innermost loop */
1104
1105         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1106                                               f+i_coord_offset,fshift+i_shift_offset);
1107
1108         /* Increment number of inner iterations */
1109         inneriter                  += j_index_end - j_index_start;
1110
1111         /* Outer loop uses 18 flops */
1112     }
1113
1114     /* Increment number of outer iterations */
1115     outeriter        += nri;
1116
1117     /* Update outer/inner flops */
1118
1119     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3_F,outeriter*18 + inneriter*120);
1120 }