src/gromacs/gmxlib/nonbonded/nb_kernel_sse4_1_double/nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4P1_sse4_1_double.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2017,2018, 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 sse4_1_double 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/gmxlib/nrnb.h"
  46
  47 #include "kernelutil_x86_sse4_1_double.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4P1_VF_sse4_1_double
  51  * Electrostatics interaction: Ewald
  52  * VdW interaction:            LJEwald
  53  * Geometry:                   Water4-Particle
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4P1_VF_sse4_1_double
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      struct t_forcerec           * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  67      * just 0 for non-waters.
  68      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  69      * jnr indices corresponding to data put in the four positions in the SIMD register.
  70      */
  71     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  72     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  73     int              jnrA,jnrB;
  74     int              j_coord_offsetA,j_coord_offsetB;
  75     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  76     real             rcutoff_scalar;
  77     real             *shiftvec,*fshift,*x,*f;
  78     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  79     int              vdwioffset0;
  80     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  81     int              vdwioffset1;
  82     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  83     int              vdwioffset2;
  84     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  85     int              vdwioffset3;
  86     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  87     int              vdwjidx0A,vdwjidx0B;
  88     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  89     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  90     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  91     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  92     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
  93     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  94     real             *charge;
  95     int              nvdwtype;
  96     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  97     int              *vdwtype;
  98     real             *vdwparam;
  99     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 100     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 101     __m128d           c6grid_00;
 102     __m128d           c6grid_10;
 103     __m128d           c6grid_20;
 104     __m128d           c6grid_30;
 105     __m128d           ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,f6A,f6B,sh_lj_ewald;
 106     real             *vdwgridparam;
 107     __m128d           one_half = _mm_set1_pd(0.5);
 108     __m128d           minus_one = _mm_set1_pd(-1.0);
 109     __m128i          ewitab;
 110     __m128d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 111     real             *ewtab;
 112     __m128d          dummy_mask,cutoff_mask;
 113     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 114     __m128d          one     = _mm_set1_pd(1.0);
 115     __m128d          two     = _mm_set1_pd(2.0);
 116     x                = xx[0];
 117     f                = ff[0];
 118
 119     nri              = nlist->nri;
 120     iinr             = nlist->iinr;
 121     jindex           = nlist->jindex;
 122     jjnr             = nlist->jjnr;
 123     shiftidx         = nlist->shift;
 124     gid              = nlist->gid;
 125     shiftvec         = fr->shift_vec[0];
 126     fshift           = fr->fshift[0];
 127     facel            = _mm_set1_pd(fr->ic->epsfac);
 128     charge           = mdatoms->chargeA;
 129     nvdwtype         = fr->ntype;
 130     vdwparam         = fr->nbfp;
 131     vdwtype          = mdatoms->typeA;
 132     vdwgridparam     = fr->ljpme_c6grid;
 133     sh_lj_ewald      = _mm_set1_pd(fr->ic->sh_lj_ewald);
 134     ewclj            = _mm_set1_pd(fr->ic->ewaldcoeff_lj);
 135     ewclj2           = _mm_mul_pd(minus_one,_mm_mul_pd(ewclj,ewclj));
 136
 137     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 138     ewtab            = fr->ic->tabq_coul_FDV0;
 139     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 140     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 141
 142     /* Setup water-specific parameters */
 143     inr              = nlist->iinr[0];
 144     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 145     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 146     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 147     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 148
 149     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 150     rcutoff_scalar   = fr->ic->rcoulomb;
 151     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 152     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 153
 154     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
 155     rvdw             = _mm_set1_pd(fr->ic->rvdw);
 156
 157     /* Avoid stupid compiler warnings */
 158     jnrA = jnrB = 0;
 159     j_coord_offsetA = 0;
 160     j_coord_offsetB = 0;
 161
 162     outeriter        = 0;
 163     inneriter        = 0;
 164
 165     /* Start outer loop over neighborlists */
 166     for(iidx=0; iidx<nri; iidx++)
 167     {
 168         /* Load shift vector for this list */
 169         i_shift_offset   = DIM*shiftidx[iidx];
 170
 171         /* Load limits for loop over neighbors */
 172         j_index_start    = jindex[iidx];
 173         j_index_end      = jindex[iidx+1];
 174
 175         /* Get outer coordinate index */
 176         inr              = iinr[iidx];
 177         i_coord_offset   = DIM*inr;
 178
 179         /* Load i particle coords and add shift vector */
 180         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 181                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 182
 183         fix0             = _mm_setzero_pd();
 184         fiy0             = _mm_setzero_pd();
 185         fiz0             = _mm_setzero_pd();
 186         fix1             = _mm_setzero_pd();
 187         fiy1             = _mm_setzero_pd();
 188         fiz1             = _mm_setzero_pd();
 189         fix2             = _mm_setzero_pd();
 190         fiy2             = _mm_setzero_pd();
 191         fiz2             = _mm_setzero_pd();
 192         fix3             = _mm_setzero_pd();
 193         fiy3             = _mm_setzero_pd();
 194         fiz3             = _mm_setzero_pd();
 195
 196         /* Reset potential sums */
 197         velecsum         = _mm_setzero_pd();
 198         vvdwsum          = _mm_setzero_pd();
 199
 200         /* Start inner kernel loop */
 201         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 202         {
 203
 204             /* Get j neighbor index, and coordinate index */
 205             jnrA             = jjnr[jidx];
 206             jnrB             = jjnr[jidx+1];
 207             j_coord_offsetA  = DIM*jnrA;
 208             j_coord_offsetB  = DIM*jnrB;
 209
 210             /* load j atom coordinates */
 211             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 212                                               &jx0,&jy0,&jz0);
 213
 214             /* Calculate displacement vector */
 215             dx00             = _mm_sub_pd(ix0,jx0);
 216             dy00             = _mm_sub_pd(iy0,jy0);
 217             dz00             = _mm_sub_pd(iz0,jz0);
 218             dx10             = _mm_sub_pd(ix1,jx0);
 219             dy10             = _mm_sub_pd(iy1,jy0);
 220             dz10             = _mm_sub_pd(iz1,jz0);
 221             dx20             = _mm_sub_pd(ix2,jx0);
 222             dy20             = _mm_sub_pd(iy2,jy0);
 223             dz20             = _mm_sub_pd(iz2,jz0);
 224             dx30             = _mm_sub_pd(ix3,jx0);
 225             dy30             = _mm_sub_pd(iy3,jy0);
 226             dz30             = _mm_sub_pd(iz3,jz0);
 227
 228             /* Calculate squared distance and things based on it */
 229             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 230             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 231             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 232             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 233
 234             rinv00           = sse41_invsqrt_d(rsq00);
 235             rinv10           = sse41_invsqrt_d(rsq10);
 236             rinv20           = sse41_invsqrt_d(rsq20);
 237             rinv30           = sse41_invsqrt_d(rsq30);
 238
 239             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 240             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 241             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 242             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 243
 244             /* Load parameters for j particles */
 245             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 246             vdwjidx0A        = 2*vdwtype[jnrA+0];
 247             vdwjidx0B        = 2*vdwtype[jnrB+0];
 248
 249             fjx0             = _mm_setzero_pd();
 250             fjy0             = _mm_setzero_pd();
 251             fjz0             = _mm_setzero_pd();
 252
 253             /**************************
 254              * CALCULATE INTERACTIONS *
 255              **************************/
 256
 257             if (gmx_mm_any_lt(rsq00,rcutoff2))
 258             {
 259
 260             r00              = _mm_mul_pd(rsq00,rinv00);
 261
 262             /* Compute parameters for interactions between i and j atoms */
 263             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 264                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 265             c6grid_00       = gmx_mm_load_2real_swizzle_pd(vdwgridparam+vdwioffset0+vdwjidx0A,
 266                                                                vdwgridparam+vdwioffset0+vdwjidx0B);
 267
 268             /* Analytical LJ-PME */
 269             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 270             ewcljrsq         = _mm_mul_pd(ewclj2,rsq00);
 271             ewclj6           = _mm_mul_pd(ewclj2,_mm_mul_pd(ewclj2,ewclj2));
 272             exponent         = sse41_exp_d(ewcljrsq);
 273             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
 274             poly             = _mm_mul_pd(exponent,_mm_add_pd(_mm_sub_pd(one,ewcljrsq),_mm_mul_pd(_mm_mul_pd(ewcljrsq,ewcljrsq),one_half)));
 275             /* vvdw6 = [C6 - C6grid * (1-poly)]/r6 */
 276             vvdw6            = _mm_mul_pd(_mm_sub_pd(c6_00,_mm_mul_pd(c6grid_00,_mm_sub_pd(one,poly))),rinvsix);
 277             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 278             vvdw             = _mm_sub_pd(_mm_mul_pd( _mm_sub_pd(vvdw12 , _mm_mul_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6))),one_twelfth),
 279                                _mm_mul_pd( _mm_sub_pd(vvdw6,_mm_add_pd(_mm_mul_pd(c6_00,sh_vdw_invrcut6),_mm_mul_pd(c6grid_00,sh_lj_ewald))),one_sixth));
 280             /* fvdw = vvdw12/r - (vvdw6/r + (C6grid * exponent * beta^6)/r) */
 281             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,_mm_sub_pd(vvdw6,_mm_mul_pd(_mm_mul_pd(c6grid_00,one_sixth),_mm_mul_pd(exponent,ewclj6)))),rinvsq00);
 282
 283             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 284
 285             /* Update potential sum for this i atom from the interaction with this j atom. */
 286             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 287             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 288
 289             fscal            = fvdw;
 290
 291             fscal            = _mm_and_pd(fscal,cutoff_mask);
 292
 293             /* Calculate temporary vectorial force */
 294             tx               = _mm_mul_pd(fscal,dx00);
 295             ty               = _mm_mul_pd(fscal,dy00);
 296             tz               = _mm_mul_pd(fscal,dz00);
 297
 298             /* Update vectorial force */
 299             fix0             = _mm_add_pd(fix0,tx);
 300             fiy0             = _mm_add_pd(fiy0,ty);
 301             fiz0             = _mm_add_pd(fiz0,tz);
 302
 303             fjx0             = _mm_add_pd(fjx0,tx);
 304             fjy0             = _mm_add_pd(fjy0,ty);
 305             fjz0             = _mm_add_pd(fjz0,tz);
 306
 307             }
 308
 309             /**************************
 310              * CALCULATE INTERACTIONS *
 311              **************************/
 312
 313             if (gmx_mm_any_lt(rsq10,rcutoff2))
 314             {
 315
 316             r10              = _mm_mul_pd(rsq10,rinv10);
 317
 318             /* Compute parameters for interactions between i and j atoms */
 319             qq10             = _mm_mul_pd(iq1,jq0);
 320
 321             /* EWALD ELECTROSTATICS */
 322
 323             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 324             ewrt             = _mm_mul_pd(r10,ewtabscale);
 325             ewitab           = _mm_cvttpd_epi32(ewrt);
 326             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 327             ewitab           = _mm_slli_epi32(ewitab,2);
 328             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 329             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 330             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 331             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 332             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 333             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 334             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 335             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 336             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 337             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 338
 339             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 340
 341             /* Update potential sum for this i atom from the interaction with this j atom. */
 342             velec            = _mm_and_pd(velec,cutoff_mask);
 343             velecsum         = _mm_add_pd(velecsum,velec);
 344
 345             fscal            = felec;
 346
 347             fscal            = _mm_and_pd(fscal,cutoff_mask);
 348
 349             /* Calculate temporary vectorial force */
 350             tx               = _mm_mul_pd(fscal,dx10);
 351             ty               = _mm_mul_pd(fscal,dy10);
 352             tz               = _mm_mul_pd(fscal,dz10);
 353
 354             /* Update vectorial force */
 355             fix1             = _mm_add_pd(fix1,tx);
 356             fiy1             = _mm_add_pd(fiy1,ty);
 357             fiz1             = _mm_add_pd(fiz1,tz);
 358
 359             fjx0             = _mm_add_pd(fjx0,tx);
 360             fjy0             = _mm_add_pd(fjy0,ty);
 361             fjz0             = _mm_add_pd(fjz0,tz);
 362
 363             }
 364
 365             /**************************
 366              * CALCULATE INTERACTIONS *
 367              **************************/
 368
 369             if (gmx_mm_any_lt(rsq20,rcutoff2))
 370             {
 371
 372             r20              = _mm_mul_pd(rsq20,rinv20);
 373
 374             /* Compute parameters for interactions between i and j atoms */
 375             qq20             = _mm_mul_pd(iq2,jq0);
 376
 377             /* EWALD ELECTROSTATICS */
 378
 379             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 380             ewrt             = _mm_mul_pd(r20,ewtabscale);
 381             ewitab           = _mm_cvttpd_epi32(ewrt);
 382             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 383             ewitab           = _mm_slli_epi32(ewitab,2);
 384             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 385             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 386             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 387             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 388             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 389             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 390             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 391             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 392             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 393             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 394
 395             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 396
 397             /* Update potential sum for this i atom from the interaction with this j atom. */
 398             velec            = _mm_and_pd(velec,cutoff_mask);
 399             velecsum         = _mm_add_pd(velecsum,velec);
 400
 401             fscal            = felec;
 402
 403             fscal            = _mm_and_pd(fscal,cutoff_mask);
 404
 405             /* Calculate temporary vectorial force */
 406             tx               = _mm_mul_pd(fscal,dx20);
 407             ty               = _mm_mul_pd(fscal,dy20);
 408             tz               = _mm_mul_pd(fscal,dz20);
 409
 410             /* Update vectorial force */
 411             fix2             = _mm_add_pd(fix2,tx);
 412             fiy2             = _mm_add_pd(fiy2,ty);
 413             fiz2             = _mm_add_pd(fiz2,tz);
 414
 415             fjx0             = _mm_add_pd(fjx0,tx);
 416             fjy0             = _mm_add_pd(fjy0,ty);
 417             fjz0             = _mm_add_pd(fjz0,tz);
 418
 419             }
 420
 421             /**************************
 422              * CALCULATE INTERACTIONS *
 423              **************************/
 424
 425             if (gmx_mm_any_lt(rsq30,rcutoff2))
 426             {
 427
 428             r30              = _mm_mul_pd(rsq30,rinv30);
 429
 430             /* Compute parameters for interactions between i and j atoms */
 431             qq30             = _mm_mul_pd(iq3,jq0);
 432
 433             /* EWALD ELECTROSTATICS */
 434
 435             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 436             ewrt             = _mm_mul_pd(r30,ewtabscale);
 437             ewitab           = _mm_cvttpd_epi32(ewrt);
 438             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 439             ewitab           = _mm_slli_epi32(ewitab,2);
 440             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 441             ewtabD           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 442             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 443             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 444             ewtabFn          = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,1) +2);
 445             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 446             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 447             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 448             velec            = _mm_mul_pd(qq30,_mm_sub_pd(_mm_sub_pd(rinv30,sh_ewald),velec));
 449             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
 450
 451             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
 452
 453             /* Update potential sum for this i atom from the interaction with this j atom. */
 454             velec            = _mm_and_pd(velec,cutoff_mask);
 455             velecsum         = _mm_add_pd(velecsum,velec);
 456
 457             fscal            = felec;
 458
 459             fscal            = _mm_and_pd(fscal,cutoff_mask);
 460
 461             /* Calculate temporary vectorial force */
 462             tx               = _mm_mul_pd(fscal,dx30);
 463             ty               = _mm_mul_pd(fscal,dy30);
 464             tz               = _mm_mul_pd(fscal,dz30);
 465
 466             /* Update vectorial force */
 467             fix3             = _mm_add_pd(fix3,tx);
 468             fiy3             = _mm_add_pd(fiy3,ty);
 469             fiz3             = _mm_add_pd(fiz3,tz);
 470
 471             fjx0             = _mm_add_pd(fjx0,tx);
 472             fjy0             = _mm_add_pd(fjy0,ty);
 473             fjz0             = _mm_add_pd(fjz0,tz);
 474
 475             }
 476
 477             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 478
 479             /* Inner loop uses 202 flops */
 480         }
 481
 482         if(jidx<j_index_end)
 483         {
 484
 485             jnrA             = jjnr[jidx];
 486             j_coord_offsetA  = DIM*jnrA;
 487
 488             /* load j atom coordinates */
 489             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 490                                               &jx0,&jy0,&jz0);
 491
 492             /* Calculate displacement vector */
 493             dx00             = _mm_sub_pd(ix0,jx0);
 494             dy00             = _mm_sub_pd(iy0,jy0);
 495             dz00             = _mm_sub_pd(iz0,jz0);
 496             dx10             = _mm_sub_pd(ix1,jx0);
 497             dy10             = _mm_sub_pd(iy1,jy0);
 498             dz10             = _mm_sub_pd(iz1,jz0);
 499             dx20             = _mm_sub_pd(ix2,jx0);
 500             dy20             = _mm_sub_pd(iy2,jy0);
 501             dz20             = _mm_sub_pd(iz2,jz0);
 502             dx30             = _mm_sub_pd(ix3,jx0);
 503             dy30             = _mm_sub_pd(iy3,jy0);
 504             dz30             = _mm_sub_pd(iz3,jz0);
 505
 506             /* Calculate squared distance and things based on it */
 507             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 508             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 509             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 510             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 511
 512             rinv00           = sse41_invsqrt_d(rsq00);
 513             rinv10           = sse41_invsqrt_d(rsq10);
 514             rinv20           = sse41_invsqrt_d(rsq20);
 515             rinv30           = sse41_invsqrt_d(rsq30);
 516
 517             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 518             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 519             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 520             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 521
 522             /* Load parameters for j particles */
 523             jq0              = _mm_load_sd(charge+jnrA+0);
 524             vdwjidx0A        = 2*vdwtype[jnrA+0];
 525
 526             fjx0             = _mm_setzero_pd();
 527             fjy0             = _mm_setzero_pd();
 528             fjz0             = _mm_setzero_pd();
 529
 530             /**************************
 531              * CALCULATE INTERACTIONS *
 532              **************************/
 533
 534             if (gmx_mm_any_lt(rsq00,rcutoff2))
 535             {
 536
 537             r00              = _mm_mul_pd(rsq00,rinv00);
 538
 539             /* Compute parameters for interactions between i and j atoms */
 540             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 541
 542             c6grid_00       = gmx_mm_load_1real_pd(vdwgridparam+vdwioffset0+vdwjidx0A);
 543
 544             /* Analytical LJ-PME */
 545             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 546             ewcljrsq         = _mm_mul_pd(ewclj2,rsq00);
 547             ewclj6           = _mm_mul_pd(ewclj2,_mm_mul_pd(ewclj2,ewclj2));
 548             exponent         = sse41_exp_d(ewcljrsq);
 549             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
 550             poly             = _mm_mul_pd(exponent,_mm_add_pd(_mm_sub_pd(one,ewcljrsq),_mm_mul_pd(_mm_mul_pd(ewcljrsq,ewcljrsq),one_half)));
 551             /* vvdw6 = [C6 - C6grid * (1-poly)]/r6 */
 552             vvdw6            = _mm_mul_pd(_mm_sub_pd(c6_00,_mm_mul_pd(c6grid_00,_mm_sub_pd(one,poly))),rinvsix);
 553             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 554             vvdw             = _mm_sub_pd(_mm_mul_pd( _mm_sub_pd(vvdw12 , _mm_mul_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6))),one_twelfth),
 555                                _mm_mul_pd( _mm_sub_pd(vvdw6,_mm_add_pd(_mm_mul_pd(c6_00,sh_vdw_invrcut6),_mm_mul_pd(c6grid_00,sh_lj_ewald))),one_sixth));
 556             /* fvdw = vvdw12/r - (vvdw6/r + (C6grid * exponent * beta^6)/r) */
 557             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,_mm_sub_pd(vvdw6,_mm_mul_pd(_mm_mul_pd(c6grid_00,one_sixth),_mm_mul_pd(exponent,ewclj6)))),rinvsq00);
 558
 559             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 560
 561             /* Update potential sum for this i atom from the interaction with this j atom. */
 562             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 563             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 564             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 565
 566             fscal            = fvdw;
 567
 568             fscal            = _mm_and_pd(fscal,cutoff_mask);
 569
 570             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 571
 572             /* Calculate temporary vectorial force */
 573             tx               = _mm_mul_pd(fscal,dx00);
 574             ty               = _mm_mul_pd(fscal,dy00);
 575             tz               = _mm_mul_pd(fscal,dz00);
 576
 577             /* Update vectorial force */
 578             fix0             = _mm_add_pd(fix0,tx);
 579             fiy0             = _mm_add_pd(fiy0,ty);
 580             fiz0             = _mm_add_pd(fiz0,tz);
 581
 582             fjx0             = _mm_add_pd(fjx0,tx);
 583             fjy0             = _mm_add_pd(fjy0,ty);
 584             fjz0             = _mm_add_pd(fjz0,tz);
 585
 586             }
 587
 588             /**************************
 589              * CALCULATE INTERACTIONS *
 590              **************************/
 591
 592             if (gmx_mm_any_lt(rsq10,rcutoff2))
 593             {
 594
 595             r10              = _mm_mul_pd(rsq10,rinv10);
 596
 597             /* Compute parameters for interactions between i and j atoms */
 598             qq10             = _mm_mul_pd(iq1,jq0);
 599
 600             /* EWALD ELECTROSTATICS */
 601
 602             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 603             ewrt             = _mm_mul_pd(r10,ewtabscale);
 604             ewitab           = _mm_cvttpd_epi32(ewrt);
 605             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 606             ewitab           = _mm_slli_epi32(ewitab,2);
 607             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 608             ewtabD           = _mm_setzero_pd();
 609             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 610             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 611             ewtabFn          = _mm_setzero_pd();
 612             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 613             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 614             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 615             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 616             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 617
 618             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 619
 620             /* Update potential sum for this i atom from the interaction with this j atom. */
 621             velec            = _mm_and_pd(velec,cutoff_mask);
 622             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 623             velecsum         = _mm_add_pd(velecsum,velec);
 624
 625             fscal            = felec;
 626
 627             fscal            = _mm_and_pd(fscal,cutoff_mask);
 628
 629             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 630
 631             /* Calculate temporary vectorial force */
 632             tx               = _mm_mul_pd(fscal,dx10);
 633             ty               = _mm_mul_pd(fscal,dy10);
 634             tz               = _mm_mul_pd(fscal,dz10);
 635
 636             /* Update vectorial force */
 637             fix1             = _mm_add_pd(fix1,tx);
 638             fiy1             = _mm_add_pd(fiy1,ty);
 639             fiz1             = _mm_add_pd(fiz1,tz);
 640
 641             fjx0             = _mm_add_pd(fjx0,tx);
 642             fjy0             = _mm_add_pd(fjy0,ty);
 643             fjz0             = _mm_add_pd(fjz0,tz);
 644
 645             }
 646
 647             /**************************
 648              * CALCULATE INTERACTIONS *
 649              **************************/
 650
 651             if (gmx_mm_any_lt(rsq20,rcutoff2))
 652             {
 653
 654             r20              = _mm_mul_pd(rsq20,rinv20);
 655
 656             /* Compute parameters for interactions between i and j atoms */
 657             qq20             = _mm_mul_pd(iq2,jq0);
 658
 659             /* EWALD ELECTROSTATICS */
 660
 661             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 662             ewrt             = _mm_mul_pd(r20,ewtabscale);
 663             ewitab           = _mm_cvttpd_epi32(ewrt);
 664             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 665             ewitab           = _mm_slli_epi32(ewitab,2);
 666             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 667             ewtabD           = _mm_setzero_pd();
 668             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 669             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 670             ewtabFn          = _mm_setzero_pd();
 671             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 672             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 673             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 674             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 675             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 676
 677             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 678
 679             /* Update potential sum for this i atom from the interaction with this j atom. */
 680             velec            = _mm_and_pd(velec,cutoff_mask);
 681             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 682             velecsum         = _mm_add_pd(velecsum,velec);
 683
 684             fscal            = felec;
 685
 686             fscal            = _mm_and_pd(fscal,cutoff_mask);
 687
 688             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 689
 690             /* Calculate temporary vectorial force */
 691             tx               = _mm_mul_pd(fscal,dx20);
 692             ty               = _mm_mul_pd(fscal,dy20);
 693             tz               = _mm_mul_pd(fscal,dz20);
 694
 695             /* Update vectorial force */
 696             fix2             = _mm_add_pd(fix2,tx);
 697             fiy2             = _mm_add_pd(fiy2,ty);
 698             fiz2             = _mm_add_pd(fiz2,tz);
 699
 700             fjx0             = _mm_add_pd(fjx0,tx);
 701             fjy0             = _mm_add_pd(fjy0,ty);
 702             fjz0             = _mm_add_pd(fjz0,tz);
 703
 704             }
 705
 706             /**************************
 707              * CALCULATE INTERACTIONS *
 708              **************************/
 709
 710             if (gmx_mm_any_lt(rsq30,rcutoff2))
 711             {
 712
 713             r30              = _mm_mul_pd(rsq30,rinv30);
 714
 715             /* Compute parameters for interactions between i and j atoms */
 716             qq30             = _mm_mul_pd(iq3,jq0);
 717
 718             /* EWALD ELECTROSTATICS */
 719
 720             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 721             ewrt             = _mm_mul_pd(r30,ewtabscale);
 722             ewitab           = _mm_cvttpd_epi32(ewrt);
 723             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 724             ewitab           = _mm_slli_epi32(ewitab,2);
 725             ewtabF           = _mm_load_pd( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 726             ewtabD           = _mm_setzero_pd();
 727             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 728             ewtabV           = _mm_load_sd( ewtab + gmx_mm_extract_epi32(ewitab,0) +2);
 729             ewtabFn          = _mm_setzero_pd();
 730             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 731             felec            = _mm_add_pd(ewtabF,_mm_mul_pd(eweps,ewtabD));
 732             velec            = _mm_sub_pd(ewtabV,_mm_mul_pd(_mm_mul_pd(ewtabhalfspace,eweps),_mm_add_pd(ewtabF,felec)));
 733             velec            = _mm_mul_pd(qq30,_mm_sub_pd(_mm_sub_pd(rinv30,sh_ewald),velec));
 734             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
 735
 736             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
 737
 738             /* Update potential sum for this i atom from the interaction with this j atom. */
 739             velec            = _mm_and_pd(velec,cutoff_mask);
 740             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 741             velecsum         = _mm_add_pd(velecsum,velec);
 742
 743             fscal            = felec;
 744
 745             fscal            = _mm_and_pd(fscal,cutoff_mask);
 746
 747             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 748
 749             /* Calculate temporary vectorial force */
 750             tx               = _mm_mul_pd(fscal,dx30);
 751             ty               = _mm_mul_pd(fscal,dy30);
 752             tz               = _mm_mul_pd(fscal,dz30);
 753
 754             /* Update vectorial force */
 755             fix3             = _mm_add_pd(fix3,tx);
 756             fiy3             = _mm_add_pd(fiy3,ty);
 757             fiz3             = _mm_add_pd(fiz3,tz);
 758
 759             fjx0             = _mm_add_pd(fjx0,tx);
 760             fjy0             = _mm_add_pd(fjy0,ty);
 761             fjz0             = _mm_add_pd(fjz0,tz);
 762
 763             }
 764
 765             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 766
 767             /* Inner loop uses 202 flops */
 768         }
 769
 770         /* End of innermost loop */
 771
 772         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 773                                               f+i_coord_offset,fshift+i_shift_offset);
 774
 775         ggid                        = gid[iidx];
 776         /* Update potential energies */
 777         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 778         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
 779
 780         /* Increment number of inner iterations */
 781         inneriter                  += j_index_end - j_index_start;
 782
 783         /* Outer loop uses 26 flops */
 784     }
 785
 786     /* Increment number of outer iterations */
 787     outeriter        += nri;
 788
 789     /* Update outer/inner flops */
 790
 791     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*26 + inneriter*202);
 792 }
 793 /*
 794  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4P1_F_sse4_1_double
 795  * Electrostatics interaction: Ewald
 796  * VdW interaction:            LJEwald
 797  * Geometry:                   Water4-Particle
 798  * Calculate force/pot:        Force
 799  */
 800 void
 801 nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4P1_F_sse4_1_double
 802                     (t_nblist                    * gmx_restrict       nlist,
 803                      rvec                        * gmx_restrict          xx,
 804                      rvec                        * gmx_restrict          ff,
 805                      struct t_forcerec           * gmx_restrict          fr,
 806                      t_mdatoms                   * gmx_restrict     mdatoms,
 807                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 808                      t_nrnb                      * gmx_restrict        nrnb)
 809 {
 810     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 811      * just 0 for non-waters.
 812      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 813      * jnr indices corresponding to data put in the four positions in the SIMD register.
 814      */
 815     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 816     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 817     int              jnrA,jnrB;
 818     int              j_coord_offsetA,j_coord_offsetB;
 819     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 820     real             rcutoff_scalar;
 821     real             *shiftvec,*fshift,*x,*f;
 822     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 823     int              vdwioffset0;
 824     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 825     int              vdwioffset1;
 826     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 827     int              vdwioffset2;
 828     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 829     int              vdwioffset3;
 830     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 831     int              vdwjidx0A,vdwjidx0B;
 832     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 833     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 834     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 835     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 836     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
 837     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 838     real             *charge;
 839     int              nvdwtype;
 840     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 841     int              *vdwtype;
 842     real             *vdwparam;
 843     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 844     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 845     __m128d           c6grid_00;
 846     __m128d           c6grid_10;
 847     __m128d           c6grid_20;
 848     __m128d           c6grid_30;
 849     __m128d           ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,f6A,f6B,sh_lj_ewald;
 850     real             *vdwgridparam;
 851     __m128d           one_half = _mm_set1_pd(0.5);
 852     __m128d           minus_one = _mm_set1_pd(-1.0);
 853     __m128i          ewitab;
 854     __m128d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 855     real             *ewtab;
 856     __m128d          dummy_mask,cutoff_mask;
 857     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 858     __m128d          one     = _mm_set1_pd(1.0);
 859     __m128d          two     = _mm_set1_pd(2.0);
 860     x                = xx[0];
 861     f                = ff[0];
 862
 863     nri              = nlist->nri;
 864     iinr             = nlist->iinr;
 865     jindex           = nlist->jindex;
 866     jjnr             = nlist->jjnr;
 867     shiftidx         = nlist->shift;
 868     gid              = nlist->gid;
 869     shiftvec         = fr->shift_vec[0];
 870     fshift           = fr->fshift[0];
 871     facel            = _mm_set1_pd(fr->ic->epsfac);
 872     charge           = mdatoms->chargeA;
 873     nvdwtype         = fr->ntype;
 874     vdwparam         = fr->nbfp;
 875     vdwtype          = mdatoms->typeA;
 876     vdwgridparam     = fr->ljpme_c6grid;
 877     sh_lj_ewald      = _mm_set1_pd(fr->ic->sh_lj_ewald);
 878     ewclj            = _mm_set1_pd(fr->ic->ewaldcoeff_lj);
 879     ewclj2           = _mm_mul_pd(minus_one,_mm_mul_pd(ewclj,ewclj));
 880
 881     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 882     ewtab            = fr->ic->tabq_coul_F;
 883     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 884     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 885
 886     /* Setup water-specific parameters */
 887     inr              = nlist->iinr[0];
 888     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 889     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 890     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 891     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 892
 893     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 894     rcutoff_scalar   = fr->ic->rcoulomb;
 895     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 896     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 897
 898     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
 899     rvdw             = _mm_set1_pd(fr->ic->rvdw);
 900
 901     /* Avoid stupid compiler warnings */
 902     jnrA = jnrB = 0;
 903     j_coord_offsetA = 0;
 904     j_coord_offsetB = 0;
 905
 906     outeriter        = 0;
 907     inneriter        = 0;
 908
 909     /* Start outer loop over neighborlists */
 910     for(iidx=0; iidx<nri; iidx++)
 911     {
 912         /* Load shift vector for this list */
 913         i_shift_offset   = DIM*shiftidx[iidx];
 914
 915         /* Load limits for loop over neighbors */
 916         j_index_start    = jindex[iidx];
 917         j_index_end      = jindex[iidx+1];
 918
 919         /* Get outer coordinate index */
 920         inr              = iinr[iidx];
 921         i_coord_offset   = DIM*inr;
 922
 923         /* Load i particle coords and add shift vector */
 924         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 925                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 926
 927         fix0             = _mm_setzero_pd();
 928         fiy0             = _mm_setzero_pd();
 929         fiz0             = _mm_setzero_pd();
 930         fix1             = _mm_setzero_pd();
 931         fiy1             = _mm_setzero_pd();
 932         fiz1             = _mm_setzero_pd();
 933         fix2             = _mm_setzero_pd();
 934         fiy2             = _mm_setzero_pd();
 935         fiz2             = _mm_setzero_pd();
 936         fix3             = _mm_setzero_pd();
 937         fiy3             = _mm_setzero_pd();
 938         fiz3             = _mm_setzero_pd();
 939
 940         /* Start inner kernel loop */
 941         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 942         {
 943
 944             /* Get j neighbor index, and coordinate index */
 945             jnrA             = jjnr[jidx];
 946             jnrB             = jjnr[jidx+1];
 947             j_coord_offsetA  = DIM*jnrA;
 948             j_coord_offsetB  = DIM*jnrB;
 949
 950             /* load j atom coordinates */
 951             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 952                                               &jx0,&jy0,&jz0);
 953
 954             /* Calculate displacement vector */
 955             dx00             = _mm_sub_pd(ix0,jx0);
 956             dy00             = _mm_sub_pd(iy0,jy0);
 957             dz00             = _mm_sub_pd(iz0,jz0);
 958             dx10             = _mm_sub_pd(ix1,jx0);
 959             dy10             = _mm_sub_pd(iy1,jy0);
 960             dz10             = _mm_sub_pd(iz1,jz0);
 961             dx20             = _mm_sub_pd(ix2,jx0);
 962             dy20             = _mm_sub_pd(iy2,jy0);
 963             dz20             = _mm_sub_pd(iz2,jz0);
 964             dx30             = _mm_sub_pd(ix3,jx0);
 965             dy30             = _mm_sub_pd(iy3,jy0);
 966             dz30             = _mm_sub_pd(iz3,jz0);
 967
 968             /* Calculate squared distance and things based on it */
 969             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 970             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 971             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 972             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 973
 974             rinv00           = sse41_invsqrt_d(rsq00);
 975             rinv10           = sse41_invsqrt_d(rsq10);
 976             rinv20           = sse41_invsqrt_d(rsq20);
 977             rinv30           = sse41_invsqrt_d(rsq30);
 978
 979             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 980             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 981             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 982             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 983
 984             /* Load parameters for j particles */
 985             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 986             vdwjidx0A        = 2*vdwtype[jnrA+0];
 987             vdwjidx0B        = 2*vdwtype[jnrB+0];
 988
 989             fjx0             = _mm_setzero_pd();
 990             fjy0             = _mm_setzero_pd();
 991             fjz0             = _mm_setzero_pd();
 992
 993             /**************************
 994              * CALCULATE INTERACTIONS *
 995              **************************/
 996
 997             if (gmx_mm_any_lt(rsq00,rcutoff2))
 998             {
 999
1000             r00              = _mm_mul_pd(rsq00,rinv00);
1001
1002             /* Compute parameters for interactions between i and j atoms */
1003             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
1004                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
1005             c6grid_00       = gmx_mm_load_2real_swizzle_pd(vdwgridparam+vdwioffset0+vdwjidx0A,
1006                                                                vdwgridparam+vdwioffset0+vdwjidx0B);
1007
1008             /* Analytical LJ-PME */
1009             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1010             ewcljrsq         = _mm_mul_pd(ewclj2,rsq00);
1011             ewclj6           = _mm_mul_pd(ewclj2,_mm_mul_pd(ewclj2,ewclj2));
1012             exponent         = sse41_exp_d(ewcljrsq);
1013             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
1014             poly             = _mm_mul_pd(exponent,_mm_add_pd(_mm_sub_pd(one,ewcljrsq),_mm_mul_pd(_mm_mul_pd(ewcljrsq,ewcljrsq),one_half)));
1015             /* f6A = 6 * C6grid * (1 - poly) */
1016             f6A              = _mm_mul_pd(c6grid_00,_mm_sub_pd(one,poly));
1017             /* f6B = C6grid * exponent * beta^6 */
1018             f6B              = _mm_mul_pd(_mm_mul_pd(c6grid_00,one_sixth),_mm_mul_pd(exponent,ewclj6));
1019             /* fvdw = 12*C12/r13 - ((6*C6 - f6A)/r6 + f6B)/r */
1020             fvdw              = _mm_mul_pd(_mm_add_pd(_mm_mul_pd(_mm_sub_pd(_mm_mul_pd(c12_00,rinvsix),_mm_sub_pd(c6_00,f6A)),rinvsix),f6B),rinvsq00);
1021
1022             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
1023
1024             fscal            = fvdw;
1025
1026             fscal            = _mm_and_pd(fscal,cutoff_mask);
1027
1028             /* Calculate temporary vectorial force */
1029             tx               = _mm_mul_pd(fscal,dx00);
1030             ty               = _mm_mul_pd(fscal,dy00);
1031             tz               = _mm_mul_pd(fscal,dz00);
1032
1033             /* Update vectorial force */
1034             fix0             = _mm_add_pd(fix0,tx);
1035             fiy0             = _mm_add_pd(fiy0,ty);
1036             fiz0             = _mm_add_pd(fiz0,tz);
1037
1038             fjx0             = _mm_add_pd(fjx0,tx);
1039             fjy0             = _mm_add_pd(fjy0,ty);
1040             fjz0             = _mm_add_pd(fjz0,tz);
1041
1042             }
1043
1044             /**************************
1045              * CALCULATE INTERACTIONS *
1046              **************************/
1047
1048             if (gmx_mm_any_lt(rsq10,rcutoff2))
1049             {
1050
1051             r10              = _mm_mul_pd(rsq10,rinv10);
1052
1053             /* Compute parameters for interactions between i and j atoms */
1054             qq10             = _mm_mul_pd(iq1,jq0);
1055
1056             /* EWALD ELECTROSTATICS */
1057
1058             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1059             ewrt             = _mm_mul_pd(r10,ewtabscale);
1060             ewitab           = _mm_cvttpd_epi32(ewrt);
1061             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1062             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1063                                          &ewtabF,&ewtabFn);
1064             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1065             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1066
1067             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
1068
1069             fscal            = felec;
1070
1071             fscal            = _mm_and_pd(fscal,cutoff_mask);
1072
1073             /* Calculate temporary vectorial force */
1074             tx               = _mm_mul_pd(fscal,dx10);
1075             ty               = _mm_mul_pd(fscal,dy10);
1076             tz               = _mm_mul_pd(fscal,dz10);
1077
1078             /* Update vectorial force */
1079             fix1             = _mm_add_pd(fix1,tx);
1080             fiy1             = _mm_add_pd(fiy1,ty);
1081             fiz1             = _mm_add_pd(fiz1,tz);
1082
1083             fjx0             = _mm_add_pd(fjx0,tx);
1084             fjy0             = _mm_add_pd(fjy0,ty);
1085             fjz0             = _mm_add_pd(fjz0,tz);
1086
1087             }
1088
1089             /**************************
1090              * CALCULATE INTERACTIONS *
1091              **************************/
1092
1093             if (gmx_mm_any_lt(rsq20,rcutoff2))
1094             {
1095
1096             r20              = _mm_mul_pd(rsq20,rinv20);
1097
1098             /* Compute parameters for interactions between i and j atoms */
1099             qq20             = _mm_mul_pd(iq2,jq0);
1100
1101             /* EWALD ELECTROSTATICS */
1102
1103             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1104             ewrt             = _mm_mul_pd(r20,ewtabscale);
1105             ewitab           = _mm_cvttpd_epi32(ewrt);
1106             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1107             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1108                                          &ewtabF,&ewtabFn);
1109             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1110             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1111
1112             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
1113
1114             fscal            = felec;
1115
1116             fscal            = _mm_and_pd(fscal,cutoff_mask);
1117
1118             /* Calculate temporary vectorial force */
1119             tx               = _mm_mul_pd(fscal,dx20);
1120             ty               = _mm_mul_pd(fscal,dy20);
1121             tz               = _mm_mul_pd(fscal,dz20);
1122
1123             /* Update vectorial force */
1124             fix2             = _mm_add_pd(fix2,tx);
1125             fiy2             = _mm_add_pd(fiy2,ty);
1126             fiz2             = _mm_add_pd(fiz2,tz);
1127
1128             fjx0             = _mm_add_pd(fjx0,tx);
1129             fjy0             = _mm_add_pd(fjy0,ty);
1130             fjz0             = _mm_add_pd(fjz0,tz);
1131
1132             }
1133
1134             /**************************
1135              * CALCULATE INTERACTIONS *
1136              **************************/
1137
1138             if (gmx_mm_any_lt(rsq30,rcutoff2))
1139             {
1140
1141             r30              = _mm_mul_pd(rsq30,rinv30);
1142
1143             /* Compute parameters for interactions between i and j atoms */
1144             qq30             = _mm_mul_pd(iq3,jq0);
1145
1146             /* EWALD ELECTROSTATICS */
1147
1148             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1149             ewrt             = _mm_mul_pd(r30,ewtabscale);
1150             ewitab           = _mm_cvttpd_epi32(ewrt);
1151             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1152             gmx_mm_load_2pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1153                                          &ewtabF,&ewtabFn);
1154             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1155             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
1156
1157             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
1158
1159             fscal            = felec;
1160
1161             fscal            = _mm_and_pd(fscal,cutoff_mask);
1162
1163             /* Calculate temporary vectorial force */
1164             tx               = _mm_mul_pd(fscal,dx30);
1165             ty               = _mm_mul_pd(fscal,dy30);
1166             tz               = _mm_mul_pd(fscal,dz30);
1167
1168             /* Update vectorial force */
1169             fix3             = _mm_add_pd(fix3,tx);
1170             fiy3             = _mm_add_pd(fiy3,ty);
1171             fiz3             = _mm_add_pd(fiz3,tz);
1172
1173             fjx0             = _mm_add_pd(fjx0,tx);
1174             fjy0             = _mm_add_pd(fjy0,ty);
1175             fjz0             = _mm_add_pd(fjz0,tz);
1176
1177             }
1178
1179             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
1180
1181             /* Inner loop uses 169 flops */
1182         }
1183
1184         if(jidx<j_index_end)
1185         {
1186
1187             jnrA             = jjnr[jidx];
1188             j_coord_offsetA  = DIM*jnrA;
1189
1190             /* load j atom coordinates */
1191             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
1192                                               &jx0,&jy0,&jz0);
1193
1194             /* Calculate displacement vector */
1195             dx00             = _mm_sub_pd(ix0,jx0);
1196             dy00             = _mm_sub_pd(iy0,jy0);
1197             dz00             = _mm_sub_pd(iz0,jz0);
1198             dx10             = _mm_sub_pd(ix1,jx0);
1199             dy10             = _mm_sub_pd(iy1,jy0);
1200             dz10             = _mm_sub_pd(iz1,jz0);
1201             dx20             = _mm_sub_pd(ix2,jx0);
1202             dy20             = _mm_sub_pd(iy2,jy0);
1203             dz20             = _mm_sub_pd(iz2,jz0);
1204             dx30             = _mm_sub_pd(ix3,jx0);
1205             dy30             = _mm_sub_pd(iy3,jy0);
1206             dz30             = _mm_sub_pd(iz3,jz0);
1207
1208             /* Calculate squared distance and things based on it */
1209             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1210             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
1211             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
1212             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
1213
1214             rinv00           = sse41_invsqrt_d(rsq00);
1215             rinv10           = sse41_invsqrt_d(rsq10);
1216             rinv20           = sse41_invsqrt_d(rsq20);
1217             rinv30           = sse41_invsqrt_d(rsq30);
1218
1219             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
1220             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
1221             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
1222             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
1223
1224             /* Load parameters for j particles */
1225             jq0              = _mm_load_sd(charge+jnrA+0);
1226             vdwjidx0A        = 2*vdwtype[jnrA+0];
1227
1228             fjx0             = _mm_setzero_pd();
1229             fjy0             = _mm_setzero_pd();
1230             fjz0             = _mm_setzero_pd();
1231
1232             /**************************
1233              * CALCULATE INTERACTIONS *
1234              **************************/
1235
1236             if (gmx_mm_any_lt(rsq00,rcutoff2))
1237             {
1238
1239             r00              = _mm_mul_pd(rsq00,rinv00);
1240
1241             /* Compute parameters for interactions between i and j atoms */
1242             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
1243
1244             c6grid_00       = gmx_mm_load_1real_pd(vdwgridparam+vdwioffset0+vdwjidx0A);
1245
1246             /* Analytical LJ-PME */
1247             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1248             ewcljrsq         = _mm_mul_pd(ewclj2,rsq00);
1249             ewclj6           = _mm_mul_pd(ewclj2,_mm_mul_pd(ewclj2,ewclj2));
1250             exponent         = sse41_exp_d(ewcljrsq);
1251             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
1252             poly             = _mm_mul_pd(exponent,_mm_add_pd(_mm_sub_pd(one,ewcljrsq),_mm_mul_pd(_mm_mul_pd(ewcljrsq,ewcljrsq),one_half)));
1253             /* f6A = 6 * C6grid * (1 - poly) */
1254             f6A              = _mm_mul_pd(c6grid_00,_mm_sub_pd(one,poly));
1255             /* f6B = C6grid * exponent * beta^6 */
1256             f6B              = _mm_mul_pd(_mm_mul_pd(c6grid_00,one_sixth),_mm_mul_pd(exponent,ewclj6));
1257             /* fvdw = 12*C12/r13 - ((6*C6 - f6A)/r6 + f6B)/r */
1258             fvdw              = _mm_mul_pd(_mm_add_pd(_mm_mul_pd(_mm_sub_pd(_mm_mul_pd(c12_00,rinvsix),_mm_sub_pd(c6_00,f6A)),rinvsix),f6B),rinvsq00);
1259
1260             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
1261
1262             fscal            = fvdw;
1263
1264             fscal            = _mm_and_pd(fscal,cutoff_mask);
1265
1266             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1267
1268             /* Calculate temporary vectorial force */
1269             tx               = _mm_mul_pd(fscal,dx00);
1270             ty               = _mm_mul_pd(fscal,dy00);
1271             tz               = _mm_mul_pd(fscal,dz00);
1272
1273             /* Update vectorial force */
1274             fix0             = _mm_add_pd(fix0,tx);
1275             fiy0             = _mm_add_pd(fiy0,ty);
1276             fiz0             = _mm_add_pd(fiz0,tz);
1277
1278             fjx0             = _mm_add_pd(fjx0,tx);
1279             fjy0             = _mm_add_pd(fjy0,ty);
1280             fjz0             = _mm_add_pd(fjz0,tz);
1281
1282             }
1283
1284             /**************************
1285              * CALCULATE INTERACTIONS *
1286              **************************/
1287
1288             if (gmx_mm_any_lt(rsq10,rcutoff2))
1289             {
1290
1291             r10              = _mm_mul_pd(rsq10,rinv10);
1292
1293             /* Compute parameters for interactions between i and j atoms */
1294             qq10             = _mm_mul_pd(iq1,jq0);
1295
1296             /* EWALD ELECTROSTATICS */
1297
1298             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1299             ewrt             = _mm_mul_pd(r10,ewtabscale);
1300             ewitab           = _mm_cvttpd_epi32(ewrt);
1301             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1302             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1303             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1304             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1305
1306             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
1307
1308             fscal            = felec;
1309
1310             fscal            = _mm_and_pd(fscal,cutoff_mask);
1311
1312             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1313
1314             /* Calculate temporary vectorial force */
1315             tx               = _mm_mul_pd(fscal,dx10);
1316             ty               = _mm_mul_pd(fscal,dy10);
1317             tz               = _mm_mul_pd(fscal,dz10);
1318
1319             /* Update vectorial force */
1320             fix1             = _mm_add_pd(fix1,tx);
1321             fiy1             = _mm_add_pd(fiy1,ty);
1322             fiz1             = _mm_add_pd(fiz1,tz);
1323
1324             fjx0             = _mm_add_pd(fjx0,tx);
1325             fjy0             = _mm_add_pd(fjy0,ty);
1326             fjz0             = _mm_add_pd(fjz0,tz);
1327
1328             }
1329
1330             /**************************
1331              * CALCULATE INTERACTIONS *
1332              **************************/
1333
1334             if (gmx_mm_any_lt(rsq20,rcutoff2))
1335             {
1336
1337             r20              = _mm_mul_pd(rsq20,rinv20);
1338
1339             /* Compute parameters for interactions between i and j atoms */
1340             qq20             = _mm_mul_pd(iq2,jq0);
1341
1342             /* EWALD ELECTROSTATICS */
1343
1344             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1345             ewrt             = _mm_mul_pd(r20,ewtabscale);
1346             ewitab           = _mm_cvttpd_epi32(ewrt);
1347             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1348             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1349             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1350             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1351
1352             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
1353
1354             fscal            = felec;
1355
1356             fscal            = _mm_and_pd(fscal,cutoff_mask);
1357
1358             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1359
1360             /* Calculate temporary vectorial force */
1361             tx               = _mm_mul_pd(fscal,dx20);
1362             ty               = _mm_mul_pd(fscal,dy20);
1363             tz               = _mm_mul_pd(fscal,dz20);
1364
1365             /* Update vectorial force */
1366             fix2             = _mm_add_pd(fix2,tx);
1367             fiy2             = _mm_add_pd(fiy2,ty);
1368             fiz2             = _mm_add_pd(fiz2,tz);
1369
1370             fjx0             = _mm_add_pd(fjx0,tx);
1371             fjy0             = _mm_add_pd(fjy0,ty);
1372             fjz0             = _mm_add_pd(fjz0,tz);
1373
1374             }
1375
1376             /**************************
1377              * CALCULATE INTERACTIONS *
1378              **************************/
1379
1380             if (gmx_mm_any_lt(rsq30,rcutoff2))
1381             {
1382
1383             r30              = _mm_mul_pd(rsq30,rinv30);
1384
1385             /* Compute parameters for interactions between i and j atoms */
1386             qq30             = _mm_mul_pd(iq3,jq0);
1387
1388             /* EWALD ELECTROSTATICS */
1389
1390             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1391             ewrt             = _mm_mul_pd(r30,ewtabscale);
1392             ewitab           = _mm_cvttpd_epi32(ewrt);
1393             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1394             gmx_mm_load_1pair_swizzle_pd(ewtab+gmx_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1395             felec            = _mm_add_pd(_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF),_mm_mul_pd(eweps,ewtabFn));
1396             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
1397
1398             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
1399
1400             fscal            = felec;
1401
1402             fscal            = _mm_and_pd(fscal,cutoff_mask);
1403
1404             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1405
1406             /* Calculate temporary vectorial force */
1407             tx               = _mm_mul_pd(fscal,dx30);
1408             ty               = _mm_mul_pd(fscal,dy30);
1409             tz               = _mm_mul_pd(fscal,dz30);
1410
1411             /* Update vectorial force */
1412             fix3             = _mm_add_pd(fix3,tx);
1413             fiy3             = _mm_add_pd(fiy3,ty);
1414             fiz3             = _mm_add_pd(fiz3,tz);
1415
1416             fjx0             = _mm_add_pd(fjx0,tx);
1417             fjy0             = _mm_add_pd(fjy0,ty);
1418             fjz0             = _mm_add_pd(fjz0,tz);
1419
1420             }
1421
1422             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1423
1424             /* Inner loop uses 169 flops */
1425         }
1426
1427         /* End of innermost loop */
1428
1429         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1430                                               f+i_coord_offset,fshift+i_shift_offset);
1431
1432         /* Increment number of inner iterations */
1433         inneriter                  += j_index_end - j_index_start;
1434
1435         /* Outer loop uses 24 flops */
1436     }
1437
1438     /* Increment number of outer iterations */
1439     outeriter        += nri;
1440
1441     /* Update outer/inner flops */
1442
1443     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*24 + inneriter*169);
1444 }