src/gromacs/gmxlib/nonbonded/nb_kernel_sse4_1_double/nb_kernel_ElecCoul_VdwNone_GeomW4P1_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_ElecCoul_VdwNone_GeomW4P1_VF_sse4_1_double
  38  * Electrostatics interaction: Coulomb
  39  * VdW interaction:            None
  40  * Geometry:                   Water4-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecCoul_VdwNone_GeomW4P1_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              vdwioffset1;
  67     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  68     int              vdwioffset2;
  69     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  70     int              vdwioffset3;
  71     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  72     int              vdwjidx0A,vdwjidx0B;
  73     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  74     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  75     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  76     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
  77     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  78     real             *charge;
  79     __m128d          dummy_mask,cutoff_mask;
  80     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  81     __m128d          one     = _mm_set1_pd(1.0);
  82     __m128d          two     = _mm_set1_pd(2.0);
  83     x                = xx[0];
  84     f                = ff[0];
  85
  86     nri              = nlist->nri;
  87     iinr             = nlist->iinr;
  88     jindex           = nlist->jindex;
  89     jjnr             = nlist->jjnr;
  90     shiftidx         = nlist->shift;
  91     gid              = nlist->gid;
  92     shiftvec         = fr->shift_vec[0];
  93     fshift           = fr->fshift[0];
  94     facel            = _mm_set1_pd(fr->epsfac);
  95     charge           = mdatoms->chargeA;
  96
  97     /* Setup water-specific parameters */
  98     inr              = nlist->iinr[0];
  99     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 100     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 101     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 102
 103     /* Avoid stupid compiler warnings */
 104     jnrA = jnrB = 0;
 105     j_coord_offsetA = 0;
 106     j_coord_offsetB = 0;
 107
 108     outeriter        = 0;
 109     inneriter        = 0;
 110
 111     /* Start outer loop over neighborlists */
 112     for(iidx=0; iidx<nri; iidx++)
 113     {
 114         /* Load shift vector for this list */
 115         i_shift_offset   = DIM*shiftidx[iidx];
 116
 117         /* Load limits for loop over neighbors */
 118         j_index_start    = jindex[iidx];
 119         j_index_end      = jindex[iidx+1];
 120
 121         /* Get outer coordinate index */
 122         inr              = iinr[iidx];
 123         i_coord_offset   = DIM*inr;
 124
 125         /* Load i particle coords and add shift vector */
 126         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
 127                                                  &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 128
 129         fix1             = _mm_setzero_pd();
 130         fiy1             = _mm_setzero_pd();
 131         fiz1             = _mm_setzero_pd();
 132         fix2             = _mm_setzero_pd();
 133         fiy2             = _mm_setzero_pd();
 134         fiz2             = _mm_setzero_pd();
 135         fix3             = _mm_setzero_pd();
 136         fiy3             = _mm_setzero_pd();
 137         fiz3             = _mm_setzero_pd();
 138
 139         /* Reset potential sums */
 140         velecsum         = _mm_setzero_pd();
 141
 142         /* Start inner kernel loop */
 143         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 144         {
 145
 146             /* Get j neighbor index, and coordinate index */
 147             jnrA             = jjnr[jidx];
 148             jnrB             = jjnr[jidx+1];
 149             j_coord_offsetA  = DIM*jnrA;
 150             j_coord_offsetB  = DIM*jnrB;
 151
 152             /* load j atom coordinates */
 153             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 154                                               &jx0,&jy0,&jz0);
 155
 156             /* Calculate displacement vector */
 157             dx10             = _mm_sub_pd(ix1,jx0);
 158             dy10             = _mm_sub_pd(iy1,jy0);
 159             dz10             = _mm_sub_pd(iz1,jz0);
 160             dx20             = _mm_sub_pd(ix2,jx0);
 161             dy20             = _mm_sub_pd(iy2,jy0);
 162             dz20             = _mm_sub_pd(iz2,jz0);
 163             dx30             = _mm_sub_pd(ix3,jx0);
 164             dy30             = _mm_sub_pd(iy3,jy0);
 165             dz30             = _mm_sub_pd(iz3,jz0);
 166
 167             /* Calculate squared distance and things based on it */
 168             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 169             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 170             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 171
 172             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 173             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 174             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 175
 176             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 177             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 178             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 179
 180             /* Load parameters for j particles */
 181             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 182
 183             fjx0             = _mm_setzero_pd();
 184             fjy0             = _mm_setzero_pd();
 185             fjz0             = _mm_setzero_pd();
 186
 187             /**************************
 188              * CALCULATE INTERACTIONS *
 189              **************************/
 190
 191             /* Compute parameters for interactions between i and j atoms */
 192             qq10             = _mm_mul_pd(iq1,jq0);
 193
 194             /* COULOMB ELECTROSTATICS */
 195             velec            = _mm_mul_pd(qq10,rinv10);
 196             felec            = _mm_mul_pd(velec,rinvsq10);
 197
 198             /* Update potential sum for this i atom from the interaction with this j atom. */
 199             velecsum         = _mm_add_pd(velecsum,velec);
 200
 201             fscal            = felec;
 202
 203             /* Calculate temporary vectorial force */
 204             tx               = _mm_mul_pd(fscal,dx10);
 205             ty               = _mm_mul_pd(fscal,dy10);
 206             tz               = _mm_mul_pd(fscal,dz10);
 207
 208             /* Update vectorial force */
 209             fix1             = _mm_add_pd(fix1,tx);
 210             fiy1             = _mm_add_pd(fiy1,ty);
 211             fiz1             = _mm_add_pd(fiz1,tz);
 212
 213             fjx0             = _mm_add_pd(fjx0,tx);
 214             fjy0             = _mm_add_pd(fjy0,ty);
 215             fjz0             = _mm_add_pd(fjz0,tz);
 216
 217             /**************************
 218              * CALCULATE INTERACTIONS *
 219              **************************/
 220
 221             /* Compute parameters for interactions between i and j atoms */
 222             qq20             = _mm_mul_pd(iq2,jq0);
 223
 224             /* COULOMB ELECTROSTATICS */
 225             velec            = _mm_mul_pd(qq20,rinv20);
 226             felec            = _mm_mul_pd(velec,rinvsq20);
 227
 228             /* Update potential sum for this i atom from the interaction with this j atom. */
 229             velecsum         = _mm_add_pd(velecsum,velec);
 230
 231             fscal            = felec;
 232
 233             /* Calculate temporary vectorial force */
 234             tx               = _mm_mul_pd(fscal,dx20);
 235             ty               = _mm_mul_pd(fscal,dy20);
 236             tz               = _mm_mul_pd(fscal,dz20);
 237
 238             /* Update vectorial force */
 239             fix2             = _mm_add_pd(fix2,tx);
 240             fiy2             = _mm_add_pd(fiy2,ty);
 241             fiz2             = _mm_add_pd(fiz2,tz);
 242
 243             fjx0             = _mm_add_pd(fjx0,tx);
 244             fjy0             = _mm_add_pd(fjy0,ty);
 245             fjz0             = _mm_add_pd(fjz0,tz);
 246
 247             /**************************
 248              * CALCULATE INTERACTIONS *
 249              **************************/
 250
 251             /* Compute parameters for interactions between i and j atoms */
 252             qq30             = _mm_mul_pd(iq3,jq0);
 253
 254             /* COULOMB ELECTROSTATICS */
 255             velec            = _mm_mul_pd(qq30,rinv30);
 256             felec            = _mm_mul_pd(velec,rinvsq30);
 257
 258             /* Update potential sum for this i atom from the interaction with this j atom. */
 259             velecsum         = _mm_add_pd(velecsum,velec);
 260
 261             fscal            = felec;
 262
 263             /* Calculate temporary vectorial force */
 264             tx               = _mm_mul_pd(fscal,dx30);
 265             ty               = _mm_mul_pd(fscal,dy30);
 266             tz               = _mm_mul_pd(fscal,dz30);
 267
 268             /* Update vectorial force */
 269             fix3             = _mm_add_pd(fix3,tx);
 270             fiy3             = _mm_add_pd(fiy3,ty);
 271             fiz3             = _mm_add_pd(fiz3,tz);
 272
 273             fjx0             = _mm_add_pd(fjx0,tx);
 274             fjy0             = _mm_add_pd(fjy0,ty);
 275             fjz0             = _mm_add_pd(fjz0,tz);
 276
 277             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 278
 279             /* Inner loop uses 87 flops */
 280         }
 281
 282         if(jidx<j_index_end)
 283         {
 284
 285             jnrA             = jjnr[jidx];
 286             j_coord_offsetA  = DIM*jnrA;
 287
 288             /* load j atom coordinates */
 289             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 290                                               &jx0,&jy0,&jz0);
 291
 292             /* Calculate displacement vector */
 293             dx10             = _mm_sub_pd(ix1,jx0);
 294             dy10             = _mm_sub_pd(iy1,jy0);
 295             dz10             = _mm_sub_pd(iz1,jz0);
 296             dx20             = _mm_sub_pd(ix2,jx0);
 297             dy20             = _mm_sub_pd(iy2,jy0);
 298             dz20             = _mm_sub_pd(iz2,jz0);
 299             dx30             = _mm_sub_pd(ix3,jx0);
 300             dy30             = _mm_sub_pd(iy3,jy0);
 301             dz30             = _mm_sub_pd(iz3,jz0);
 302
 303             /* Calculate squared distance and things based on it */
 304             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 305             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 306             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 307
 308             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 309             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 310             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 311
 312             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 313             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 314             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 315
 316             /* Load parameters for j particles */
 317             jq0              = _mm_load_sd(charge+jnrA+0);
 318
 319             fjx0             = _mm_setzero_pd();
 320             fjy0             = _mm_setzero_pd();
 321             fjz0             = _mm_setzero_pd();
 322
 323             /**************************
 324              * CALCULATE INTERACTIONS *
 325              **************************/
 326
 327             /* Compute parameters for interactions between i and j atoms */
 328             qq10             = _mm_mul_pd(iq1,jq0);
 329
 330             /* COULOMB ELECTROSTATICS */
 331             velec            = _mm_mul_pd(qq10,rinv10);
 332             felec            = _mm_mul_pd(velec,rinvsq10);
 333
 334             /* Update potential sum for this i atom from the interaction with this j atom. */
 335             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 336             velecsum         = _mm_add_pd(velecsum,velec);
 337
 338             fscal            = felec;
 339
 340             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 341
 342             /* Calculate temporary vectorial force */
 343             tx               = _mm_mul_pd(fscal,dx10);
 344             ty               = _mm_mul_pd(fscal,dy10);
 345             tz               = _mm_mul_pd(fscal,dz10);
 346
 347             /* Update vectorial force */
 348             fix1             = _mm_add_pd(fix1,tx);
 349             fiy1             = _mm_add_pd(fiy1,ty);
 350             fiz1             = _mm_add_pd(fiz1,tz);
 351
 352             fjx0             = _mm_add_pd(fjx0,tx);
 353             fjy0             = _mm_add_pd(fjy0,ty);
 354             fjz0             = _mm_add_pd(fjz0,tz);
 355
 356             /**************************
 357              * CALCULATE INTERACTIONS *
 358              **************************/
 359
 360             /* Compute parameters for interactions between i and j atoms */
 361             qq20             = _mm_mul_pd(iq2,jq0);
 362
 363             /* COULOMB ELECTROSTATICS */
 364             velec            = _mm_mul_pd(qq20,rinv20);
 365             felec            = _mm_mul_pd(velec,rinvsq20);
 366
 367             /* Update potential sum for this i atom from the interaction with this j atom. */
 368             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 369             velecsum         = _mm_add_pd(velecsum,velec);
 370
 371             fscal            = felec;
 372
 373             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 374
 375             /* Calculate temporary vectorial force */
 376             tx               = _mm_mul_pd(fscal,dx20);
 377             ty               = _mm_mul_pd(fscal,dy20);
 378             tz               = _mm_mul_pd(fscal,dz20);
 379
 380             /* Update vectorial force */
 381             fix2             = _mm_add_pd(fix2,tx);
 382             fiy2             = _mm_add_pd(fiy2,ty);
 383             fiz2             = _mm_add_pd(fiz2,tz);
 384
 385             fjx0             = _mm_add_pd(fjx0,tx);
 386             fjy0             = _mm_add_pd(fjy0,ty);
 387             fjz0             = _mm_add_pd(fjz0,tz);
 388
 389             /**************************
 390              * CALCULATE INTERACTIONS *
 391              **************************/
 392
 393             /* Compute parameters for interactions between i and j atoms */
 394             qq30             = _mm_mul_pd(iq3,jq0);
 395
 396             /* COULOMB ELECTROSTATICS */
 397             velec            = _mm_mul_pd(qq30,rinv30);
 398             felec            = _mm_mul_pd(velec,rinvsq30);
 399
 400             /* Update potential sum for this i atom from the interaction with this j atom. */
 401             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 402             velecsum         = _mm_add_pd(velecsum,velec);
 403
 404             fscal            = felec;
 405
 406             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 407
 408             /* Calculate temporary vectorial force */
 409             tx               = _mm_mul_pd(fscal,dx30);
 410             ty               = _mm_mul_pd(fscal,dy30);
 411             tz               = _mm_mul_pd(fscal,dz30);
 412
 413             /* Update vectorial force */
 414             fix3             = _mm_add_pd(fix3,tx);
 415             fiy3             = _mm_add_pd(fiy3,ty);
 416             fiz3             = _mm_add_pd(fiz3,tz);
 417
 418             fjx0             = _mm_add_pd(fjx0,tx);
 419             fjy0             = _mm_add_pd(fjy0,ty);
 420             fjz0             = _mm_add_pd(fjz0,tz);
 421
 422             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 423
 424             /* Inner loop uses 87 flops */
 425         }
 426
 427         /* End of innermost loop */
 428
 429         gmx_mm_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 430                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
 431
 432         ggid                        = gid[iidx];
 433         /* Update potential energies */
 434         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 435
 436         /* Increment number of inner iterations */
 437         inneriter                  += j_index_end - j_index_start;
 438
 439         /* Outer loop uses 19 flops */
 440     }
 441
 442     /* Increment number of outer iterations */
 443     outeriter        += nri;
 444
 445     /* Update outer/inner flops */
 446
 447     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_VF,outeriter*19 + inneriter*87);
 448 }
 449 /*
 450  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwNone_GeomW4P1_F_sse4_1_double
 451  * Electrostatics interaction: Coulomb
 452  * VdW interaction:            None
 453  * Geometry:                   Water4-Particle
 454  * Calculate force/pot:        Force
 455  */
 456 void
 457 nb_kernel_ElecCoul_VdwNone_GeomW4P1_F_sse4_1_double
 458                     (t_nblist * gmx_restrict                nlist,
 459                      rvec * gmx_restrict                    xx,
 460                      rvec * gmx_restrict                    ff,
 461                      t_forcerec * gmx_restrict              fr,
 462                      t_mdatoms * gmx_restrict               mdatoms,
 463                      nb_kernel_data_t * gmx_restrict        kernel_data,
 464                      t_nrnb * gmx_restrict                  nrnb)
 465 {
 466     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 467      * just 0 for non-waters.
 468      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 469      * jnr indices corresponding to data put in the four positions in the SIMD register.
 470      */
 471     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 472     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 473     int              jnrA,jnrB;
 474     int              j_coord_offsetA,j_coord_offsetB;
 475     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 476     real             rcutoff_scalar;
 477     real             *shiftvec,*fshift,*x,*f;
 478     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 479     int              vdwioffset1;
 480     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 481     int              vdwioffset2;
 482     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 483     int              vdwioffset3;
 484     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 485     int              vdwjidx0A,vdwjidx0B;
 486     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 487     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 488     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 489     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
 490     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 491     real             *charge;
 492     __m128d          dummy_mask,cutoff_mask;
 493     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 494     __m128d          one     = _mm_set1_pd(1.0);
 495     __m128d          two     = _mm_set1_pd(2.0);
 496     x                = xx[0];
 497     f                = ff[0];
 498
 499     nri              = nlist->nri;
 500     iinr             = nlist->iinr;
 501     jindex           = nlist->jindex;
 502     jjnr             = nlist->jjnr;
 503     shiftidx         = nlist->shift;
 504     gid              = nlist->gid;
 505     shiftvec         = fr->shift_vec[0];
 506     fshift           = fr->fshift[0];
 507     facel            = _mm_set1_pd(fr->epsfac);
 508     charge           = mdatoms->chargeA;
 509
 510     /* Setup water-specific parameters */
 511     inr              = nlist->iinr[0];
 512     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 513     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 514     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 515
 516     /* Avoid stupid compiler warnings */
 517     jnrA = jnrB = 0;
 518     j_coord_offsetA = 0;
 519     j_coord_offsetB = 0;
 520
 521     outeriter        = 0;
 522     inneriter        = 0;
 523
 524     /* Start outer loop over neighborlists */
 525     for(iidx=0; iidx<nri; iidx++)
 526     {
 527         /* Load shift vector for this list */
 528         i_shift_offset   = DIM*shiftidx[iidx];
 529
 530         /* Load limits for loop over neighbors */
 531         j_index_start    = jindex[iidx];
 532         j_index_end      = jindex[iidx+1];
 533
 534         /* Get outer coordinate index */
 535         inr              = iinr[iidx];
 536         i_coord_offset   = DIM*inr;
 537
 538         /* Load i particle coords and add shift vector */
 539         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
 540                                                  &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 541
 542         fix1             = _mm_setzero_pd();
 543         fiy1             = _mm_setzero_pd();
 544         fiz1             = _mm_setzero_pd();
 545         fix2             = _mm_setzero_pd();
 546         fiy2             = _mm_setzero_pd();
 547         fiz2             = _mm_setzero_pd();
 548         fix3             = _mm_setzero_pd();
 549         fiy3             = _mm_setzero_pd();
 550         fiz3             = _mm_setzero_pd();
 551
 552         /* Start inner kernel loop */
 553         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 554         {
 555
 556             /* Get j neighbor index, and coordinate index */
 557             jnrA             = jjnr[jidx];
 558             jnrB             = jjnr[jidx+1];
 559             j_coord_offsetA  = DIM*jnrA;
 560             j_coord_offsetB  = DIM*jnrB;
 561
 562             /* load j atom coordinates */
 563             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 564                                               &jx0,&jy0,&jz0);
 565
 566             /* Calculate displacement vector */
 567             dx10             = _mm_sub_pd(ix1,jx0);
 568             dy10             = _mm_sub_pd(iy1,jy0);
 569             dz10             = _mm_sub_pd(iz1,jz0);
 570             dx20             = _mm_sub_pd(ix2,jx0);
 571             dy20             = _mm_sub_pd(iy2,jy0);
 572             dz20             = _mm_sub_pd(iz2,jz0);
 573             dx30             = _mm_sub_pd(ix3,jx0);
 574             dy30             = _mm_sub_pd(iy3,jy0);
 575             dz30             = _mm_sub_pd(iz3,jz0);
 576
 577             /* Calculate squared distance and things based on it */
 578             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 579             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 580             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 581
 582             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 583             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 584             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 585
 586             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 587             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 588             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 589
 590             /* Load parameters for j particles */
 591             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 592
 593             fjx0             = _mm_setzero_pd();
 594             fjy0             = _mm_setzero_pd();
 595             fjz0             = _mm_setzero_pd();
 596
 597             /**************************
 598              * CALCULATE INTERACTIONS *
 599              **************************/
 600
 601             /* Compute parameters for interactions between i and j atoms */
 602             qq10             = _mm_mul_pd(iq1,jq0);
 603
 604             /* COULOMB ELECTROSTATICS */
 605             velec            = _mm_mul_pd(qq10,rinv10);
 606             felec            = _mm_mul_pd(velec,rinvsq10);
 607
 608             fscal            = felec;
 609
 610             /* Calculate temporary vectorial force */
 611             tx               = _mm_mul_pd(fscal,dx10);
 612             ty               = _mm_mul_pd(fscal,dy10);
 613             tz               = _mm_mul_pd(fscal,dz10);
 614
 615             /* Update vectorial force */
 616             fix1             = _mm_add_pd(fix1,tx);
 617             fiy1             = _mm_add_pd(fiy1,ty);
 618             fiz1             = _mm_add_pd(fiz1,tz);
 619
 620             fjx0             = _mm_add_pd(fjx0,tx);
 621             fjy0             = _mm_add_pd(fjy0,ty);
 622             fjz0             = _mm_add_pd(fjz0,tz);
 623
 624             /**************************
 625              * CALCULATE INTERACTIONS *
 626              **************************/
 627
 628             /* Compute parameters for interactions between i and j atoms */
 629             qq20             = _mm_mul_pd(iq2,jq0);
 630
 631             /* COULOMB ELECTROSTATICS */
 632             velec            = _mm_mul_pd(qq20,rinv20);
 633             felec            = _mm_mul_pd(velec,rinvsq20);
 634
 635             fscal            = felec;
 636
 637             /* Calculate temporary vectorial force */
 638             tx               = _mm_mul_pd(fscal,dx20);
 639             ty               = _mm_mul_pd(fscal,dy20);
 640             tz               = _mm_mul_pd(fscal,dz20);
 641
 642             /* Update vectorial force */
 643             fix2             = _mm_add_pd(fix2,tx);
 644             fiy2             = _mm_add_pd(fiy2,ty);
 645             fiz2             = _mm_add_pd(fiz2,tz);
 646
 647             fjx0             = _mm_add_pd(fjx0,tx);
 648             fjy0             = _mm_add_pd(fjy0,ty);
 649             fjz0             = _mm_add_pd(fjz0,tz);
 650
 651             /**************************
 652              * CALCULATE INTERACTIONS *
 653              **************************/
 654
 655             /* Compute parameters for interactions between i and j atoms */
 656             qq30             = _mm_mul_pd(iq3,jq0);
 657
 658             /* COULOMB ELECTROSTATICS */
 659             velec            = _mm_mul_pd(qq30,rinv30);
 660             felec            = _mm_mul_pd(velec,rinvsq30);
 661
 662             fscal            = felec;
 663
 664             /* Calculate temporary vectorial force */
 665             tx               = _mm_mul_pd(fscal,dx30);
 666             ty               = _mm_mul_pd(fscal,dy30);
 667             tz               = _mm_mul_pd(fscal,dz30);
 668
 669             /* Update vectorial force */
 670             fix3             = _mm_add_pd(fix3,tx);
 671             fiy3             = _mm_add_pd(fiy3,ty);
 672             fiz3             = _mm_add_pd(fiz3,tz);
 673
 674             fjx0             = _mm_add_pd(fjx0,tx);
 675             fjy0             = _mm_add_pd(fjy0,ty);
 676             fjz0             = _mm_add_pd(fjz0,tz);
 677
 678             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 679
 680             /* Inner loop uses 84 flops */
 681         }
 682
 683         if(jidx<j_index_end)
 684         {
 685
 686             jnrA             = jjnr[jidx];
 687             j_coord_offsetA  = DIM*jnrA;
 688
 689             /* load j atom coordinates */
 690             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 691                                               &jx0,&jy0,&jz0);
 692
 693             /* Calculate displacement vector */
 694             dx10             = _mm_sub_pd(ix1,jx0);
 695             dy10             = _mm_sub_pd(iy1,jy0);
 696             dz10             = _mm_sub_pd(iz1,jz0);
 697             dx20             = _mm_sub_pd(ix2,jx0);
 698             dy20             = _mm_sub_pd(iy2,jy0);
 699             dz20             = _mm_sub_pd(iz2,jz0);
 700             dx30             = _mm_sub_pd(ix3,jx0);
 701             dy30             = _mm_sub_pd(iy3,jy0);
 702             dz30             = _mm_sub_pd(iz3,jz0);
 703
 704             /* Calculate squared distance and things based on it */
 705             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 706             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 707             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 708
 709             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 710             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 711             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 712
 713             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 714             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 715             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 716
 717             /* Load parameters for j particles */
 718             jq0              = _mm_load_sd(charge+jnrA+0);
 719
 720             fjx0             = _mm_setzero_pd();
 721             fjy0             = _mm_setzero_pd();
 722             fjz0             = _mm_setzero_pd();
 723
 724             /**************************
 725              * CALCULATE INTERACTIONS *
 726              **************************/
 727
 728             /* Compute parameters for interactions between i and j atoms */
 729             qq10             = _mm_mul_pd(iq1,jq0);
 730
 731             /* COULOMB ELECTROSTATICS */
 732             velec            = _mm_mul_pd(qq10,rinv10);
 733             felec            = _mm_mul_pd(velec,rinvsq10);
 734
 735             fscal            = felec;
 736
 737             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 738
 739             /* Calculate temporary vectorial force */
 740             tx               = _mm_mul_pd(fscal,dx10);
 741             ty               = _mm_mul_pd(fscal,dy10);
 742             tz               = _mm_mul_pd(fscal,dz10);
 743
 744             /* Update vectorial force */
 745             fix1             = _mm_add_pd(fix1,tx);
 746             fiy1             = _mm_add_pd(fiy1,ty);
 747             fiz1             = _mm_add_pd(fiz1,tz);
 748
 749             fjx0             = _mm_add_pd(fjx0,tx);
 750             fjy0             = _mm_add_pd(fjy0,ty);
 751             fjz0             = _mm_add_pd(fjz0,tz);
 752
 753             /**************************
 754              * CALCULATE INTERACTIONS *
 755              **************************/
 756
 757             /* Compute parameters for interactions between i and j atoms */
 758             qq20             = _mm_mul_pd(iq2,jq0);
 759
 760             /* COULOMB ELECTROSTATICS */
 761             velec            = _mm_mul_pd(qq20,rinv20);
 762             felec            = _mm_mul_pd(velec,rinvsq20);
 763
 764             fscal            = felec;
 765
 766             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 767
 768             /* Calculate temporary vectorial force */
 769             tx               = _mm_mul_pd(fscal,dx20);
 770             ty               = _mm_mul_pd(fscal,dy20);
 771             tz               = _mm_mul_pd(fscal,dz20);
 772
 773             /* Update vectorial force */
 774             fix2             = _mm_add_pd(fix2,tx);
 775             fiy2             = _mm_add_pd(fiy2,ty);
 776             fiz2             = _mm_add_pd(fiz2,tz);
 777
 778             fjx0             = _mm_add_pd(fjx0,tx);
 779             fjy0             = _mm_add_pd(fjy0,ty);
 780             fjz0             = _mm_add_pd(fjz0,tz);
 781
 782             /**************************
 783              * CALCULATE INTERACTIONS *
 784              **************************/
 785
 786             /* Compute parameters for interactions between i and j atoms */
 787             qq30             = _mm_mul_pd(iq3,jq0);
 788
 789             /* COULOMB ELECTROSTATICS */
 790             velec            = _mm_mul_pd(qq30,rinv30);
 791             felec            = _mm_mul_pd(velec,rinvsq30);
 792
 793             fscal            = felec;
 794
 795             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 796
 797             /* Calculate temporary vectorial force */
 798             tx               = _mm_mul_pd(fscal,dx30);
 799             ty               = _mm_mul_pd(fscal,dy30);
 800             tz               = _mm_mul_pd(fscal,dz30);
 801
 802             /* Update vectorial force */
 803             fix3             = _mm_add_pd(fix3,tx);
 804             fiy3             = _mm_add_pd(fiy3,ty);
 805             fiz3             = _mm_add_pd(fiz3,tz);
 806
 807             fjx0             = _mm_add_pd(fjx0,tx);
 808             fjy0             = _mm_add_pd(fjy0,ty);
 809             fjz0             = _mm_add_pd(fjz0,tz);
 810
 811             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 812
 813             /* Inner loop uses 84 flops */
 814         }
 815
 816         /* End of innermost loop */
 817
 818         gmx_mm_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 819                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
 820
 821         /* Increment number of inner iterations */
 822         inneriter                  += j_index_end - j_index_start;
 823
 824         /* Outer loop uses 18 flops */
 825     }
 826
 827     /* Increment number of outer iterations */
 828     outeriter        += nri;
 829
 830     /* Update outer/inner flops */
 831
 832     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_F,outeriter*18 + inneriter*84);
 833 }