src/gromacs/gmxlib/nonbonded/nb_kernel_avx_256_double/nb_kernel_ElecEw_VdwCSTab_GeomW4W4_avx_256_double.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   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 avx_256_double kernel generator.
  37  */
  38 #include "config.h"
  39
  40 #include <math.h>
  41
  42 #include "../nb_kernel.h"
  43 #include "types/simple.h"
  44 #include "gromacs/math/vec.h"
  45 #include "nrnb.h"
  46
  47 #include "gromacs/simd/math_x86_avx_256_double.h"
  48 #include "kernelutil_x86_avx_256_double.h"
  49
  50 /*
  51  * Gromacs nonbonded kernel:   nb_kernel_ElecEw_VdwCSTab_GeomW4W4_VF_avx_256_double
  52  * Electrostatics interaction: Ewald
  53  * VdW interaction:            CubicSplineTable
  54  * Geometry:                   Water4-Water4
  55  * Calculate force/pot:        PotentialAndForce
  56  */
  57 void
  58 nb_kernel_ElecEw_VdwCSTab_GeomW4W4_VF_avx_256_double
  59                     (t_nblist                    * gmx_restrict       nlist,
  60                      rvec                        * gmx_restrict          xx,
  61                      rvec                        * gmx_restrict          ff,
  62                      t_forcerec                  * gmx_restrict          fr,
  63                      t_mdatoms                   * gmx_restrict     mdatoms,
  64                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  65                      t_nrnb                      * gmx_restrict        nrnb)
  66 {
  67     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  68      * just 0 for non-waters.
  69      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
  70      * jnr indices corresponding to data put in the four positions in the SIMD register.
  71      */
  72     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  73     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  74     int              jnrA,jnrB,jnrC,jnrD;
  75     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  76     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  77     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  78     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  79     real             rcutoff_scalar;
  80     real             *shiftvec,*fshift,*x,*f;
  81     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  82     real             scratch[4*DIM];
  83     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  84     real *           vdwioffsetptr0;
  85     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  86     real *           vdwioffsetptr1;
  87     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  88     real *           vdwioffsetptr2;
  89     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  90     real *           vdwioffsetptr3;
  91     __m256d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  92     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  93     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  94     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  95     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  96     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  97     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  98     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
  99     __m256d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
 100     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 101     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 102     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 103     __m256d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
 104     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 105     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 106     __m256d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
 107     __m256d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
 108     __m256d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
 109     __m256d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
 110     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
 111     real             *charge;
 112     int              nvdwtype;
 113     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 114     int              *vdwtype;
 115     real             *vdwparam;
 116     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
 117     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
 118     __m128i          vfitab;
 119     __m128i          ifour       = _mm_set1_epi32(4);
 120     __m256d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 121     real             *vftab;
 122     __m128i          ewitab;
 123     __m256d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 124     __m256d          beta,beta2,beta3,zeta2,pmecorrF,pmecorrV,rinv3;
 125     real             *ewtab;
 126     __m256d          dummy_mask,cutoff_mask;
 127     __m128           tmpmask0,tmpmask1;
 128     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
 129     __m256d          one     = _mm256_set1_pd(1.0);
 130     __m256d          two     = _mm256_set1_pd(2.0);
 131     x                = xx[0];
 132     f                = ff[0];
 133
 134     nri              = nlist->nri;
 135     iinr             = nlist->iinr;
 136     jindex           = nlist->jindex;
 137     jjnr             = nlist->jjnr;
 138     shiftidx         = nlist->shift;
 139     gid              = nlist->gid;
 140     shiftvec         = fr->shift_vec[0];
 141     fshift           = fr->fshift[0];
 142     facel            = _mm256_set1_pd(fr->epsfac);
 143     charge           = mdatoms->chargeA;
 144     nvdwtype         = fr->ntype;
 145     vdwparam         = fr->nbfp;
 146     vdwtype          = mdatoms->typeA;
 147
 148     vftab            = kernel_data->table_vdw->data;
 149     vftabscale       = _mm256_set1_pd(kernel_data->table_vdw->scale);
 150
 151     sh_ewald         = _mm256_set1_pd(fr->ic->sh_ewald);
 152     beta             = _mm256_set1_pd(fr->ic->ewaldcoeff_q);
 153     beta2            = _mm256_mul_pd(beta,beta);
 154     beta3            = _mm256_mul_pd(beta,beta2);
 155
 156     ewtab            = fr->ic->tabq_coul_FDV0;
 157     ewtabscale       = _mm256_set1_pd(fr->ic->tabq_scale);
 158     ewtabhalfspace   = _mm256_set1_pd(0.5/fr->ic->tabq_scale);
 159
 160     /* Setup water-specific parameters */
 161     inr              = nlist->iinr[0];
 162     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
 163     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
 164     iq3              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
 165     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 166
 167     jq1              = _mm256_set1_pd(charge[inr+1]);
 168     jq2              = _mm256_set1_pd(charge[inr+2]);
 169     jq3              = _mm256_set1_pd(charge[inr+3]);
 170     vdwjidx0A        = 2*vdwtype[inr+0];
 171     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
 172     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
 173     qq11             = _mm256_mul_pd(iq1,jq1);
 174     qq12             = _mm256_mul_pd(iq1,jq2);
 175     qq13             = _mm256_mul_pd(iq1,jq3);
 176     qq21             = _mm256_mul_pd(iq2,jq1);
 177     qq22             = _mm256_mul_pd(iq2,jq2);
 178     qq23             = _mm256_mul_pd(iq2,jq3);
 179     qq31             = _mm256_mul_pd(iq3,jq1);
 180     qq32             = _mm256_mul_pd(iq3,jq2);
 181     qq33             = _mm256_mul_pd(iq3,jq3);
 182
 183     /* Avoid stupid compiler warnings */
 184     jnrA = jnrB = jnrC = jnrD = 0;
 185     j_coord_offsetA = 0;
 186     j_coord_offsetB = 0;
 187     j_coord_offsetC = 0;
 188     j_coord_offsetD = 0;
 189
 190     outeriter        = 0;
 191     inneriter        = 0;
 192
 193     for(iidx=0;iidx<4*DIM;iidx++)
 194     {
 195         scratch[iidx] = 0.0;
 196     }
 197
 198     /* Start outer loop over neighborlists */
 199     for(iidx=0; iidx<nri; iidx++)
 200     {
 201         /* Load shift vector for this list */
 202         i_shift_offset   = DIM*shiftidx[iidx];
 203
 204         /* Load limits for loop over neighbors */
 205         j_index_start    = jindex[iidx];
 206         j_index_end      = jindex[iidx+1];
 207
 208         /* Get outer coordinate index */
 209         inr              = iinr[iidx];
 210         i_coord_offset   = DIM*inr;
 211
 212         /* Load i particle coords and add shift vector */
 213         gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 214                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 215
 216         fix0             = _mm256_setzero_pd();
 217         fiy0             = _mm256_setzero_pd();
 218         fiz0             = _mm256_setzero_pd();
 219         fix1             = _mm256_setzero_pd();
 220         fiy1             = _mm256_setzero_pd();
 221         fiz1             = _mm256_setzero_pd();
 222         fix2             = _mm256_setzero_pd();
 223         fiy2             = _mm256_setzero_pd();
 224         fiz2             = _mm256_setzero_pd();
 225         fix3             = _mm256_setzero_pd();
 226         fiy3             = _mm256_setzero_pd();
 227         fiz3             = _mm256_setzero_pd();
 228
 229         /* Reset potential sums */
 230         velecsum         = _mm256_setzero_pd();
 231         vvdwsum          = _mm256_setzero_pd();
 232
 233         /* Start inner kernel loop */
 234         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 235         {
 236
 237             /* Get j neighbor index, and coordinate index */
 238             jnrA             = jjnr[jidx];
 239             jnrB             = jjnr[jidx+1];
 240             jnrC             = jjnr[jidx+2];
 241             jnrD             = jjnr[jidx+3];
 242             j_coord_offsetA  = DIM*jnrA;
 243             j_coord_offsetB  = DIM*jnrB;
 244             j_coord_offsetC  = DIM*jnrC;
 245             j_coord_offsetD  = DIM*jnrD;
 246
 247             /* load j atom coordinates */
 248             gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 249                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 250                                                  &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 251                                                  &jy2,&jz2,&jx3,&jy3,&jz3);
 252
 253             /* Calculate displacement vector */
 254             dx00             = _mm256_sub_pd(ix0,jx0);
 255             dy00             = _mm256_sub_pd(iy0,jy0);
 256             dz00             = _mm256_sub_pd(iz0,jz0);
 257             dx11             = _mm256_sub_pd(ix1,jx1);
 258             dy11             = _mm256_sub_pd(iy1,jy1);
 259             dz11             = _mm256_sub_pd(iz1,jz1);
 260             dx12             = _mm256_sub_pd(ix1,jx2);
 261             dy12             = _mm256_sub_pd(iy1,jy2);
 262             dz12             = _mm256_sub_pd(iz1,jz2);
 263             dx13             = _mm256_sub_pd(ix1,jx3);
 264             dy13             = _mm256_sub_pd(iy1,jy3);
 265             dz13             = _mm256_sub_pd(iz1,jz3);
 266             dx21             = _mm256_sub_pd(ix2,jx1);
 267             dy21             = _mm256_sub_pd(iy2,jy1);
 268             dz21             = _mm256_sub_pd(iz2,jz1);
 269             dx22             = _mm256_sub_pd(ix2,jx2);
 270             dy22             = _mm256_sub_pd(iy2,jy2);
 271             dz22             = _mm256_sub_pd(iz2,jz2);
 272             dx23             = _mm256_sub_pd(ix2,jx3);
 273             dy23             = _mm256_sub_pd(iy2,jy3);
 274             dz23             = _mm256_sub_pd(iz2,jz3);
 275             dx31             = _mm256_sub_pd(ix3,jx1);
 276             dy31             = _mm256_sub_pd(iy3,jy1);
 277             dz31             = _mm256_sub_pd(iz3,jz1);
 278             dx32             = _mm256_sub_pd(ix3,jx2);
 279             dy32             = _mm256_sub_pd(iy3,jy2);
 280             dz32             = _mm256_sub_pd(iz3,jz2);
 281             dx33             = _mm256_sub_pd(ix3,jx3);
 282             dy33             = _mm256_sub_pd(iy3,jy3);
 283             dz33             = _mm256_sub_pd(iz3,jz3);
 284
 285             /* Calculate squared distance and things based on it */
 286             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 287             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 288             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 289             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
 290             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 291             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 292             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
 293             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
 294             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
 295             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
 296
 297             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 298             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 299             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 300             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
 301             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 302             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 303             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
 304             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
 305             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
 306             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
 307
 308             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 309             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 310             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
 311             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 312             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 313             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
 314             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
 315             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
 316             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
 317
 318             fjx0             = _mm256_setzero_pd();
 319             fjy0             = _mm256_setzero_pd();
 320             fjz0             = _mm256_setzero_pd();
 321             fjx1             = _mm256_setzero_pd();
 322             fjy1             = _mm256_setzero_pd();
 323             fjz1             = _mm256_setzero_pd();
 324             fjx2             = _mm256_setzero_pd();
 325             fjy2             = _mm256_setzero_pd();
 326             fjz2             = _mm256_setzero_pd();
 327             fjx3             = _mm256_setzero_pd();
 328             fjy3             = _mm256_setzero_pd();
 329             fjz3             = _mm256_setzero_pd();
 330
 331             /**************************
 332              * CALCULATE INTERACTIONS *
 333              **************************/
 334
 335             r00              = _mm256_mul_pd(rsq00,rinv00);
 336
 337             /* Calculate table index by multiplying r with table scale and truncate to integer */
 338             rt               = _mm256_mul_pd(r00,vftabscale);
 339             vfitab           = _mm256_cvttpd_epi32(rt);
 340             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
 341             vfitab           = _mm_slli_epi32(vfitab,3);
 342
 343             /* CUBIC SPLINE TABLE DISPERSION */
 344             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 345             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 346             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 347             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 348             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 349             Heps             = _mm256_mul_pd(vfeps,H);
 350             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 351             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 352             vvdw6            = _mm256_mul_pd(c6_00,VV);
 353             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 354             fvdw6            = _mm256_mul_pd(c6_00,FF);
 355
 356             /* CUBIC SPLINE TABLE REPULSION */
 357             vfitab           = _mm_add_epi32(vfitab,ifour);
 358             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 359             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 360             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 361             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 362             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 363             Heps             = _mm256_mul_pd(vfeps,H);
 364             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 365             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 366             vvdw12           = _mm256_mul_pd(c12_00,VV);
 367             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 368             fvdw12           = _mm256_mul_pd(c12_00,FF);
 369             vvdw             = _mm256_add_pd(vvdw12,vvdw6);
 370             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
 371
 372             /* Update potential sum for this i atom from the interaction with this j atom. */
 373             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 374
 375             fscal            = fvdw;
 376
 377             /* Calculate temporary vectorial force */
 378             tx               = _mm256_mul_pd(fscal,dx00);
 379             ty               = _mm256_mul_pd(fscal,dy00);
 380             tz               = _mm256_mul_pd(fscal,dz00);
 381
 382             /* Update vectorial force */
 383             fix0             = _mm256_add_pd(fix0,tx);
 384             fiy0             = _mm256_add_pd(fiy0,ty);
 385             fiz0             = _mm256_add_pd(fiz0,tz);
 386
 387             fjx0             = _mm256_add_pd(fjx0,tx);
 388             fjy0             = _mm256_add_pd(fjy0,ty);
 389             fjz0             = _mm256_add_pd(fjz0,tz);
 390
 391             /**************************
 392              * CALCULATE INTERACTIONS *
 393              **************************/
 394
 395             r11              = _mm256_mul_pd(rsq11,rinv11);
 396
 397             /* EWALD ELECTROSTATICS */
 398
 399             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 400             ewrt             = _mm256_mul_pd(r11,ewtabscale);
 401             ewitab           = _mm256_cvttpd_epi32(ewrt);
 402             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 403             ewitab           = _mm_slli_epi32(ewitab,2);
 404             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 405             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 406             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 407             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 408             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 409             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 410             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 411             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(rinv11,velec));
 412             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
 413
 414             /* Update potential sum for this i atom from the interaction with this j atom. */
 415             velecsum         = _mm256_add_pd(velecsum,velec);
 416
 417             fscal            = felec;
 418
 419             /* Calculate temporary vectorial force */
 420             tx               = _mm256_mul_pd(fscal,dx11);
 421             ty               = _mm256_mul_pd(fscal,dy11);
 422             tz               = _mm256_mul_pd(fscal,dz11);
 423
 424             /* Update vectorial force */
 425             fix1             = _mm256_add_pd(fix1,tx);
 426             fiy1             = _mm256_add_pd(fiy1,ty);
 427             fiz1             = _mm256_add_pd(fiz1,tz);
 428
 429             fjx1             = _mm256_add_pd(fjx1,tx);
 430             fjy1             = _mm256_add_pd(fjy1,ty);
 431             fjz1             = _mm256_add_pd(fjz1,tz);
 432
 433             /**************************
 434              * CALCULATE INTERACTIONS *
 435              **************************/
 436
 437             r12              = _mm256_mul_pd(rsq12,rinv12);
 438
 439             /* EWALD ELECTROSTATICS */
 440
 441             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 442             ewrt             = _mm256_mul_pd(r12,ewtabscale);
 443             ewitab           = _mm256_cvttpd_epi32(ewrt);
 444             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 445             ewitab           = _mm_slli_epi32(ewitab,2);
 446             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 447             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 448             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 449             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 450             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 451             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 452             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 453             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(rinv12,velec));
 454             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
 455
 456             /* Update potential sum for this i atom from the interaction with this j atom. */
 457             velecsum         = _mm256_add_pd(velecsum,velec);
 458
 459             fscal            = felec;
 460
 461             /* Calculate temporary vectorial force */
 462             tx               = _mm256_mul_pd(fscal,dx12);
 463             ty               = _mm256_mul_pd(fscal,dy12);
 464             tz               = _mm256_mul_pd(fscal,dz12);
 465
 466             /* Update vectorial force */
 467             fix1             = _mm256_add_pd(fix1,tx);
 468             fiy1             = _mm256_add_pd(fiy1,ty);
 469             fiz1             = _mm256_add_pd(fiz1,tz);
 470
 471             fjx2             = _mm256_add_pd(fjx2,tx);
 472             fjy2             = _mm256_add_pd(fjy2,ty);
 473             fjz2             = _mm256_add_pd(fjz2,tz);
 474
 475             /**************************
 476              * CALCULATE INTERACTIONS *
 477              **************************/
 478
 479             r13              = _mm256_mul_pd(rsq13,rinv13);
 480
 481             /* EWALD ELECTROSTATICS */
 482
 483             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 484             ewrt             = _mm256_mul_pd(r13,ewtabscale);
 485             ewitab           = _mm256_cvttpd_epi32(ewrt);
 486             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 487             ewitab           = _mm_slli_epi32(ewitab,2);
 488             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 489             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 490             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 491             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 492             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 493             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 494             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 495             velec            = _mm256_mul_pd(qq13,_mm256_sub_pd(rinv13,velec));
 496             felec            = _mm256_mul_pd(_mm256_mul_pd(qq13,rinv13),_mm256_sub_pd(rinvsq13,felec));
 497
 498             /* Update potential sum for this i atom from the interaction with this j atom. */
 499             velecsum         = _mm256_add_pd(velecsum,velec);
 500
 501             fscal            = felec;
 502
 503             /* Calculate temporary vectorial force */
 504             tx               = _mm256_mul_pd(fscal,dx13);
 505             ty               = _mm256_mul_pd(fscal,dy13);
 506             tz               = _mm256_mul_pd(fscal,dz13);
 507
 508             /* Update vectorial force */
 509             fix1             = _mm256_add_pd(fix1,tx);
 510             fiy1             = _mm256_add_pd(fiy1,ty);
 511             fiz1             = _mm256_add_pd(fiz1,tz);
 512
 513             fjx3             = _mm256_add_pd(fjx3,tx);
 514             fjy3             = _mm256_add_pd(fjy3,ty);
 515             fjz3             = _mm256_add_pd(fjz3,tz);
 516
 517             /**************************
 518              * CALCULATE INTERACTIONS *
 519              **************************/
 520
 521             r21              = _mm256_mul_pd(rsq21,rinv21);
 522
 523             /* EWALD ELECTROSTATICS */
 524
 525             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 526             ewrt             = _mm256_mul_pd(r21,ewtabscale);
 527             ewitab           = _mm256_cvttpd_epi32(ewrt);
 528             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 529             ewitab           = _mm_slli_epi32(ewitab,2);
 530             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 531             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 532             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 533             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 534             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 535             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 536             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 537             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(rinv21,velec));
 538             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
 539
 540             /* Update potential sum for this i atom from the interaction with this j atom. */
 541             velecsum         = _mm256_add_pd(velecsum,velec);
 542
 543             fscal            = felec;
 544
 545             /* Calculate temporary vectorial force */
 546             tx               = _mm256_mul_pd(fscal,dx21);
 547             ty               = _mm256_mul_pd(fscal,dy21);
 548             tz               = _mm256_mul_pd(fscal,dz21);
 549
 550             /* Update vectorial force */
 551             fix2             = _mm256_add_pd(fix2,tx);
 552             fiy2             = _mm256_add_pd(fiy2,ty);
 553             fiz2             = _mm256_add_pd(fiz2,tz);
 554
 555             fjx1             = _mm256_add_pd(fjx1,tx);
 556             fjy1             = _mm256_add_pd(fjy1,ty);
 557             fjz1             = _mm256_add_pd(fjz1,tz);
 558
 559             /**************************
 560              * CALCULATE INTERACTIONS *
 561              **************************/
 562
 563             r22              = _mm256_mul_pd(rsq22,rinv22);
 564
 565             /* EWALD ELECTROSTATICS */
 566
 567             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 568             ewrt             = _mm256_mul_pd(r22,ewtabscale);
 569             ewitab           = _mm256_cvttpd_epi32(ewrt);
 570             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 571             ewitab           = _mm_slli_epi32(ewitab,2);
 572             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 573             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 574             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 575             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 576             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 577             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 578             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 579             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(rinv22,velec));
 580             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
 581
 582             /* Update potential sum for this i atom from the interaction with this j atom. */
 583             velecsum         = _mm256_add_pd(velecsum,velec);
 584
 585             fscal            = felec;
 586
 587             /* Calculate temporary vectorial force */
 588             tx               = _mm256_mul_pd(fscal,dx22);
 589             ty               = _mm256_mul_pd(fscal,dy22);
 590             tz               = _mm256_mul_pd(fscal,dz22);
 591
 592             /* Update vectorial force */
 593             fix2             = _mm256_add_pd(fix2,tx);
 594             fiy2             = _mm256_add_pd(fiy2,ty);
 595             fiz2             = _mm256_add_pd(fiz2,tz);
 596
 597             fjx2             = _mm256_add_pd(fjx2,tx);
 598             fjy2             = _mm256_add_pd(fjy2,ty);
 599             fjz2             = _mm256_add_pd(fjz2,tz);
 600
 601             /**************************
 602              * CALCULATE INTERACTIONS *
 603              **************************/
 604
 605             r23              = _mm256_mul_pd(rsq23,rinv23);
 606
 607             /* EWALD ELECTROSTATICS */
 608
 609             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 610             ewrt             = _mm256_mul_pd(r23,ewtabscale);
 611             ewitab           = _mm256_cvttpd_epi32(ewrt);
 612             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 613             ewitab           = _mm_slli_epi32(ewitab,2);
 614             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 615             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 616             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 617             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 618             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 619             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 620             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 621             velec            = _mm256_mul_pd(qq23,_mm256_sub_pd(rinv23,velec));
 622             felec            = _mm256_mul_pd(_mm256_mul_pd(qq23,rinv23),_mm256_sub_pd(rinvsq23,felec));
 623
 624             /* Update potential sum for this i atom from the interaction with this j atom. */
 625             velecsum         = _mm256_add_pd(velecsum,velec);
 626
 627             fscal            = felec;
 628
 629             /* Calculate temporary vectorial force */
 630             tx               = _mm256_mul_pd(fscal,dx23);
 631             ty               = _mm256_mul_pd(fscal,dy23);
 632             tz               = _mm256_mul_pd(fscal,dz23);
 633
 634             /* Update vectorial force */
 635             fix2             = _mm256_add_pd(fix2,tx);
 636             fiy2             = _mm256_add_pd(fiy2,ty);
 637             fiz2             = _mm256_add_pd(fiz2,tz);
 638
 639             fjx3             = _mm256_add_pd(fjx3,tx);
 640             fjy3             = _mm256_add_pd(fjy3,ty);
 641             fjz3             = _mm256_add_pd(fjz3,tz);
 642
 643             /**************************
 644              * CALCULATE INTERACTIONS *
 645              **************************/
 646
 647             r31              = _mm256_mul_pd(rsq31,rinv31);
 648
 649             /* EWALD ELECTROSTATICS */
 650
 651             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 652             ewrt             = _mm256_mul_pd(r31,ewtabscale);
 653             ewitab           = _mm256_cvttpd_epi32(ewrt);
 654             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 655             ewitab           = _mm_slli_epi32(ewitab,2);
 656             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 657             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 658             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 659             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 660             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 661             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 662             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 663             velec            = _mm256_mul_pd(qq31,_mm256_sub_pd(rinv31,velec));
 664             felec            = _mm256_mul_pd(_mm256_mul_pd(qq31,rinv31),_mm256_sub_pd(rinvsq31,felec));
 665
 666             /* Update potential sum for this i atom from the interaction with this j atom. */
 667             velecsum         = _mm256_add_pd(velecsum,velec);
 668
 669             fscal            = felec;
 670
 671             /* Calculate temporary vectorial force */
 672             tx               = _mm256_mul_pd(fscal,dx31);
 673             ty               = _mm256_mul_pd(fscal,dy31);
 674             tz               = _mm256_mul_pd(fscal,dz31);
 675
 676             /* Update vectorial force */
 677             fix3             = _mm256_add_pd(fix3,tx);
 678             fiy3             = _mm256_add_pd(fiy3,ty);
 679             fiz3             = _mm256_add_pd(fiz3,tz);
 680
 681             fjx1             = _mm256_add_pd(fjx1,tx);
 682             fjy1             = _mm256_add_pd(fjy1,ty);
 683             fjz1             = _mm256_add_pd(fjz1,tz);
 684
 685             /**************************
 686              * CALCULATE INTERACTIONS *
 687              **************************/
 688
 689             r32              = _mm256_mul_pd(rsq32,rinv32);
 690
 691             /* EWALD ELECTROSTATICS */
 692
 693             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 694             ewrt             = _mm256_mul_pd(r32,ewtabscale);
 695             ewitab           = _mm256_cvttpd_epi32(ewrt);
 696             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 697             ewitab           = _mm_slli_epi32(ewitab,2);
 698             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 699             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 700             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 701             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 702             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 703             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 704             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 705             velec            = _mm256_mul_pd(qq32,_mm256_sub_pd(rinv32,velec));
 706             felec            = _mm256_mul_pd(_mm256_mul_pd(qq32,rinv32),_mm256_sub_pd(rinvsq32,felec));
 707
 708             /* Update potential sum for this i atom from the interaction with this j atom. */
 709             velecsum         = _mm256_add_pd(velecsum,velec);
 710
 711             fscal            = felec;
 712
 713             /* Calculate temporary vectorial force */
 714             tx               = _mm256_mul_pd(fscal,dx32);
 715             ty               = _mm256_mul_pd(fscal,dy32);
 716             tz               = _mm256_mul_pd(fscal,dz32);
 717
 718             /* Update vectorial force */
 719             fix3             = _mm256_add_pd(fix3,tx);
 720             fiy3             = _mm256_add_pd(fiy3,ty);
 721             fiz3             = _mm256_add_pd(fiz3,tz);
 722
 723             fjx2             = _mm256_add_pd(fjx2,tx);
 724             fjy2             = _mm256_add_pd(fjy2,ty);
 725             fjz2             = _mm256_add_pd(fjz2,tz);
 726
 727             /**************************
 728              * CALCULATE INTERACTIONS *
 729              **************************/
 730
 731             r33              = _mm256_mul_pd(rsq33,rinv33);
 732
 733             /* EWALD ELECTROSTATICS */
 734
 735             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 736             ewrt             = _mm256_mul_pd(r33,ewtabscale);
 737             ewitab           = _mm256_cvttpd_epi32(ewrt);
 738             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 739             ewitab           = _mm_slli_epi32(ewitab,2);
 740             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 741             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 742             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 743             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 744             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 745             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 746             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 747             velec            = _mm256_mul_pd(qq33,_mm256_sub_pd(rinv33,velec));
 748             felec            = _mm256_mul_pd(_mm256_mul_pd(qq33,rinv33),_mm256_sub_pd(rinvsq33,felec));
 749
 750             /* Update potential sum for this i atom from the interaction with this j atom. */
 751             velecsum         = _mm256_add_pd(velecsum,velec);
 752
 753             fscal            = felec;
 754
 755             /* Calculate temporary vectorial force */
 756             tx               = _mm256_mul_pd(fscal,dx33);
 757             ty               = _mm256_mul_pd(fscal,dy33);
 758             tz               = _mm256_mul_pd(fscal,dz33);
 759
 760             /* Update vectorial force */
 761             fix3             = _mm256_add_pd(fix3,tx);
 762             fiy3             = _mm256_add_pd(fiy3,ty);
 763             fiz3             = _mm256_add_pd(fiz3,tz);
 764
 765             fjx3             = _mm256_add_pd(fjx3,tx);
 766             fjy3             = _mm256_add_pd(fjy3,ty);
 767             fjz3             = _mm256_add_pd(fjz3,tz);
 768
 769             fjptrA             = f+j_coord_offsetA;
 770             fjptrB             = f+j_coord_offsetB;
 771             fjptrC             = f+j_coord_offsetC;
 772             fjptrD             = f+j_coord_offsetD;
 773
 774             gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
 775                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
 776                                                       fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 777
 778             /* Inner loop uses 428 flops */
 779         }
 780
 781         if(jidx<j_index_end)
 782         {
 783
 784             /* Get j neighbor index, and coordinate index */
 785             jnrlistA         = jjnr[jidx];
 786             jnrlistB         = jjnr[jidx+1];
 787             jnrlistC         = jjnr[jidx+2];
 788             jnrlistD         = jjnr[jidx+3];
 789             /* Sign of each element will be negative for non-real atoms.
 790              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 791              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
 792              */
 793             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 794
 795             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
 796             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
 797             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
 798
 799             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 800             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 801             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 802             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 803             j_coord_offsetA  = DIM*jnrA;
 804             j_coord_offsetB  = DIM*jnrB;
 805             j_coord_offsetC  = DIM*jnrC;
 806             j_coord_offsetD  = DIM*jnrD;
 807
 808             /* load j atom coordinates */
 809             gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 810                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 811                                                  &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 812                                                  &jy2,&jz2,&jx3,&jy3,&jz3);
 813
 814             /* Calculate displacement vector */
 815             dx00             = _mm256_sub_pd(ix0,jx0);
 816             dy00             = _mm256_sub_pd(iy0,jy0);
 817             dz00             = _mm256_sub_pd(iz0,jz0);
 818             dx11             = _mm256_sub_pd(ix1,jx1);
 819             dy11             = _mm256_sub_pd(iy1,jy1);
 820             dz11             = _mm256_sub_pd(iz1,jz1);
 821             dx12             = _mm256_sub_pd(ix1,jx2);
 822             dy12             = _mm256_sub_pd(iy1,jy2);
 823             dz12             = _mm256_sub_pd(iz1,jz2);
 824             dx13             = _mm256_sub_pd(ix1,jx3);
 825             dy13             = _mm256_sub_pd(iy1,jy3);
 826             dz13             = _mm256_sub_pd(iz1,jz3);
 827             dx21             = _mm256_sub_pd(ix2,jx1);
 828             dy21             = _mm256_sub_pd(iy2,jy1);
 829             dz21             = _mm256_sub_pd(iz2,jz1);
 830             dx22             = _mm256_sub_pd(ix2,jx2);
 831             dy22             = _mm256_sub_pd(iy2,jy2);
 832             dz22             = _mm256_sub_pd(iz2,jz2);
 833             dx23             = _mm256_sub_pd(ix2,jx3);
 834             dy23             = _mm256_sub_pd(iy2,jy3);
 835             dz23             = _mm256_sub_pd(iz2,jz3);
 836             dx31             = _mm256_sub_pd(ix3,jx1);
 837             dy31             = _mm256_sub_pd(iy3,jy1);
 838             dz31             = _mm256_sub_pd(iz3,jz1);
 839             dx32             = _mm256_sub_pd(ix3,jx2);
 840             dy32             = _mm256_sub_pd(iy3,jy2);
 841             dz32             = _mm256_sub_pd(iz3,jz2);
 842             dx33             = _mm256_sub_pd(ix3,jx3);
 843             dy33             = _mm256_sub_pd(iy3,jy3);
 844             dz33             = _mm256_sub_pd(iz3,jz3);
 845
 846             /* Calculate squared distance and things based on it */
 847             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 848             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 849             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 850             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
 851             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 852             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 853             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
 854             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
 855             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
 856             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
 857
 858             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 859             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 860             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 861             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
 862             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 863             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 864             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
 865             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
 866             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
 867             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
 868
 869             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 870             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 871             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
 872             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 873             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 874             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
 875             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
 876             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
 877             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
 878
 879             fjx0             = _mm256_setzero_pd();
 880             fjy0             = _mm256_setzero_pd();
 881             fjz0             = _mm256_setzero_pd();
 882             fjx1             = _mm256_setzero_pd();
 883             fjy1             = _mm256_setzero_pd();
 884             fjz1             = _mm256_setzero_pd();
 885             fjx2             = _mm256_setzero_pd();
 886             fjy2             = _mm256_setzero_pd();
 887             fjz2             = _mm256_setzero_pd();
 888             fjx3             = _mm256_setzero_pd();
 889             fjy3             = _mm256_setzero_pd();
 890             fjz3             = _mm256_setzero_pd();
 891
 892             /**************************
 893              * CALCULATE INTERACTIONS *
 894              **************************/
 895
 896             r00              = _mm256_mul_pd(rsq00,rinv00);
 897             r00              = _mm256_andnot_pd(dummy_mask,r00);
 898
 899             /* Calculate table index by multiplying r with table scale and truncate to integer */
 900             rt               = _mm256_mul_pd(r00,vftabscale);
 901             vfitab           = _mm256_cvttpd_epi32(rt);
 902             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
 903             vfitab           = _mm_slli_epi32(vfitab,3);
 904
 905             /* CUBIC SPLINE TABLE DISPERSION */
 906             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 907             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 908             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 909             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 910             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 911             Heps             = _mm256_mul_pd(vfeps,H);
 912             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 913             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 914             vvdw6            = _mm256_mul_pd(c6_00,VV);
 915             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 916             fvdw6            = _mm256_mul_pd(c6_00,FF);
 917
 918             /* CUBIC SPLINE TABLE REPULSION */
 919             vfitab           = _mm_add_epi32(vfitab,ifour);
 920             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 921             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 922             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 923             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 924             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 925             Heps             = _mm256_mul_pd(vfeps,H);
 926             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 927             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 928             vvdw12           = _mm256_mul_pd(c12_00,VV);
 929             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 930             fvdw12           = _mm256_mul_pd(c12_00,FF);
 931             vvdw             = _mm256_add_pd(vvdw12,vvdw6);
 932             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
 933
 934             /* Update potential sum for this i atom from the interaction with this j atom. */
 935             vvdw             = _mm256_andnot_pd(dummy_mask,vvdw);
 936             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 937
 938             fscal            = fvdw;
 939
 940             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 941
 942             /* Calculate temporary vectorial force */
 943             tx               = _mm256_mul_pd(fscal,dx00);
 944             ty               = _mm256_mul_pd(fscal,dy00);
 945             tz               = _mm256_mul_pd(fscal,dz00);
 946
 947             /* Update vectorial force */
 948             fix0             = _mm256_add_pd(fix0,tx);
 949             fiy0             = _mm256_add_pd(fiy0,ty);
 950             fiz0             = _mm256_add_pd(fiz0,tz);
 951
 952             fjx0             = _mm256_add_pd(fjx0,tx);
 953             fjy0             = _mm256_add_pd(fjy0,ty);
 954             fjz0             = _mm256_add_pd(fjz0,tz);
 955
 956             /**************************
 957              * CALCULATE INTERACTIONS *
 958              **************************/
 959
 960             r11              = _mm256_mul_pd(rsq11,rinv11);
 961             r11              = _mm256_andnot_pd(dummy_mask,r11);
 962
 963             /* EWALD ELECTROSTATICS */
 964
 965             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 966             ewrt             = _mm256_mul_pd(r11,ewtabscale);
 967             ewitab           = _mm256_cvttpd_epi32(ewrt);
 968             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 969             ewitab           = _mm_slli_epi32(ewitab,2);
 970             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 971             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 972             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 973             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 974             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 975             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 976             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 977             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(rinv11,velec));
 978             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
 979
 980             /* Update potential sum for this i atom from the interaction with this j atom. */
 981             velec            = _mm256_andnot_pd(dummy_mask,velec);
 982             velecsum         = _mm256_add_pd(velecsum,velec);
 983
 984             fscal            = felec;
 985
 986             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 987
 988             /* Calculate temporary vectorial force */
 989             tx               = _mm256_mul_pd(fscal,dx11);
 990             ty               = _mm256_mul_pd(fscal,dy11);
 991             tz               = _mm256_mul_pd(fscal,dz11);
 992
 993             /* Update vectorial force */
 994             fix1             = _mm256_add_pd(fix1,tx);
 995             fiy1             = _mm256_add_pd(fiy1,ty);
 996             fiz1             = _mm256_add_pd(fiz1,tz);
 997
 998             fjx1             = _mm256_add_pd(fjx1,tx);
 999             fjy1             = _mm256_add_pd(fjy1,ty);
1000             fjz1             = _mm256_add_pd(fjz1,tz);
1001
1002             /**************************
1003              * CALCULATE INTERACTIONS *
1004              **************************/
1005
1006             r12              = _mm256_mul_pd(rsq12,rinv12);
1007             r12              = _mm256_andnot_pd(dummy_mask,r12);
1008
1009             /* EWALD ELECTROSTATICS */
1010
1011             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1012             ewrt             = _mm256_mul_pd(r12,ewtabscale);
1013             ewitab           = _mm256_cvttpd_epi32(ewrt);
1014             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1015             ewitab           = _mm_slli_epi32(ewitab,2);
1016             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1017             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1018             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1019             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1020             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1021             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1022             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1023             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(rinv12,velec));
1024             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
1025
1026             /* Update potential sum for this i atom from the interaction with this j atom. */
1027             velec            = _mm256_andnot_pd(dummy_mask,velec);
1028             velecsum         = _mm256_add_pd(velecsum,velec);
1029
1030             fscal            = felec;
1031
1032             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1033
1034             /* Calculate temporary vectorial force */
1035             tx               = _mm256_mul_pd(fscal,dx12);
1036             ty               = _mm256_mul_pd(fscal,dy12);
1037             tz               = _mm256_mul_pd(fscal,dz12);
1038
1039             /* Update vectorial force */
1040             fix1             = _mm256_add_pd(fix1,tx);
1041             fiy1             = _mm256_add_pd(fiy1,ty);
1042             fiz1             = _mm256_add_pd(fiz1,tz);
1043
1044             fjx2             = _mm256_add_pd(fjx2,tx);
1045             fjy2             = _mm256_add_pd(fjy2,ty);
1046             fjz2             = _mm256_add_pd(fjz2,tz);
1047
1048             /**************************
1049              * CALCULATE INTERACTIONS *
1050              **************************/
1051
1052             r13              = _mm256_mul_pd(rsq13,rinv13);
1053             r13              = _mm256_andnot_pd(dummy_mask,r13);
1054
1055             /* EWALD ELECTROSTATICS */
1056
1057             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1058             ewrt             = _mm256_mul_pd(r13,ewtabscale);
1059             ewitab           = _mm256_cvttpd_epi32(ewrt);
1060             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1061             ewitab           = _mm_slli_epi32(ewitab,2);
1062             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1063             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1064             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1065             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1066             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1067             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1068             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1069             velec            = _mm256_mul_pd(qq13,_mm256_sub_pd(rinv13,velec));
1070             felec            = _mm256_mul_pd(_mm256_mul_pd(qq13,rinv13),_mm256_sub_pd(rinvsq13,felec));
1071
1072             /* Update potential sum for this i atom from the interaction with this j atom. */
1073             velec            = _mm256_andnot_pd(dummy_mask,velec);
1074             velecsum         = _mm256_add_pd(velecsum,velec);
1075
1076             fscal            = felec;
1077
1078             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1079
1080             /* Calculate temporary vectorial force */
1081             tx               = _mm256_mul_pd(fscal,dx13);
1082             ty               = _mm256_mul_pd(fscal,dy13);
1083             tz               = _mm256_mul_pd(fscal,dz13);
1084
1085             /* Update vectorial force */
1086             fix1             = _mm256_add_pd(fix1,tx);
1087             fiy1             = _mm256_add_pd(fiy1,ty);
1088             fiz1             = _mm256_add_pd(fiz1,tz);
1089
1090             fjx3             = _mm256_add_pd(fjx3,tx);
1091             fjy3             = _mm256_add_pd(fjy3,ty);
1092             fjz3             = _mm256_add_pd(fjz3,tz);
1093
1094             /**************************
1095              * CALCULATE INTERACTIONS *
1096              **************************/
1097
1098             r21              = _mm256_mul_pd(rsq21,rinv21);
1099             r21              = _mm256_andnot_pd(dummy_mask,r21);
1100
1101             /* EWALD ELECTROSTATICS */
1102
1103             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1104             ewrt             = _mm256_mul_pd(r21,ewtabscale);
1105             ewitab           = _mm256_cvttpd_epi32(ewrt);
1106             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1107             ewitab           = _mm_slli_epi32(ewitab,2);
1108             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1109             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1110             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1111             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1112             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1113             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1114             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1115             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(rinv21,velec));
1116             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
1117
1118             /* Update potential sum for this i atom from the interaction with this j atom. */
1119             velec            = _mm256_andnot_pd(dummy_mask,velec);
1120             velecsum         = _mm256_add_pd(velecsum,velec);
1121
1122             fscal            = felec;
1123
1124             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1125
1126             /* Calculate temporary vectorial force */
1127             tx               = _mm256_mul_pd(fscal,dx21);
1128             ty               = _mm256_mul_pd(fscal,dy21);
1129             tz               = _mm256_mul_pd(fscal,dz21);
1130
1131             /* Update vectorial force */
1132             fix2             = _mm256_add_pd(fix2,tx);
1133             fiy2             = _mm256_add_pd(fiy2,ty);
1134             fiz2             = _mm256_add_pd(fiz2,tz);
1135
1136             fjx1             = _mm256_add_pd(fjx1,tx);
1137             fjy1             = _mm256_add_pd(fjy1,ty);
1138             fjz1             = _mm256_add_pd(fjz1,tz);
1139
1140             /**************************
1141              * CALCULATE INTERACTIONS *
1142              **************************/
1143
1144             r22              = _mm256_mul_pd(rsq22,rinv22);
1145             r22              = _mm256_andnot_pd(dummy_mask,r22);
1146
1147             /* EWALD ELECTROSTATICS */
1148
1149             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1150             ewrt             = _mm256_mul_pd(r22,ewtabscale);
1151             ewitab           = _mm256_cvttpd_epi32(ewrt);
1152             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1153             ewitab           = _mm_slli_epi32(ewitab,2);
1154             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1155             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1156             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1157             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1158             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1159             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1160             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1161             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(rinv22,velec));
1162             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
1163
1164             /* Update potential sum for this i atom from the interaction with this j atom. */
1165             velec            = _mm256_andnot_pd(dummy_mask,velec);
1166             velecsum         = _mm256_add_pd(velecsum,velec);
1167
1168             fscal            = felec;
1169
1170             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1171
1172             /* Calculate temporary vectorial force */
1173             tx               = _mm256_mul_pd(fscal,dx22);
1174             ty               = _mm256_mul_pd(fscal,dy22);
1175             tz               = _mm256_mul_pd(fscal,dz22);
1176
1177             /* Update vectorial force */
1178             fix2             = _mm256_add_pd(fix2,tx);
1179             fiy2             = _mm256_add_pd(fiy2,ty);
1180             fiz2             = _mm256_add_pd(fiz2,tz);
1181
1182             fjx2             = _mm256_add_pd(fjx2,tx);
1183             fjy2             = _mm256_add_pd(fjy2,ty);
1184             fjz2             = _mm256_add_pd(fjz2,tz);
1185
1186             /**************************
1187              * CALCULATE INTERACTIONS *
1188              **************************/
1189
1190             r23              = _mm256_mul_pd(rsq23,rinv23);
1191             r23              = _mm256_andnot_pd(dummy_mask,r23);
1192
1193             /* EWALD ELECTROSTATICS */
1194
1195             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1196             ewrt             = _mm256_mul_pd(r23,ewtabscale);
1197             ewitab           = _mm256_cvttpd_epi32(ewrt);
1198             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1199             ewitab           = _mm_slli_epi32(ewitab,2);
1200             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1201             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1202             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1203             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1204             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1205             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1206             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1207             velec            = _mm256_mul_pd(qq23,_mm256_sub_pd(rinv23,velec));
1208             felec            = _mm256_mul_pd(_mm256_mul_pd(qq23,rinv23),_mm256_sub_pd(rinvsq23,felec));
1209
1210             /* Update potential sum for this i atom from the interaction with this j atom. */
1211             velec            = _mm256_andnot_pd(dummy_mask,velec);
1212             velecsum         = _mm256_add_pd(velecsum,velec);
1213
1214             fscal            = felec;
1215
1216             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1217
1218             /* Calculate temporary vectorial force */
1219             tx               = _mm256_mul_pd(fscal,dx23);
1220             ty               = _mm256_mul_pd(fscal,dy23);
1221             tz               = _mm256_mul_pd(fscal,dz23);
1222
1223             /* Update vectorial force */
1224             fix2             = _mm256_add_pd(fix2,tx);
1225             fiy2             = _mm256_add_pd(fiy2,ty);
1226             fiz2             = _mm256_add_pd(fiz2,tz);
1227
1228             fjx3             = _mm256_add_pd(fjx3,tx);
1229             fjy3             = _mm256_add_pd(fjy3,ty);
1230             fjz3             = _mm256_add_pd(fjz3,tz);
1231
1232             /**************************
1233              * CALCULATE INTERACTIONS *
1234              **************************/
1235
1236             r31              = _mm256_mul_pd(rsq31,rinv31);
1237             r31              = _mm256_andnot_pd(dummy_mask,r31);
1238
1239             /* EWALD ELECTROSTATICS */
1240
1241             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1242             ewrt             = _mm256_mul_pd(r31,ewtabscale);
1243             ewitab           = _mm256_cvttpd_epi32(ewrt);
1244             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1245             ewitab           = _mm_slli_epi32(ewitab,2);
1246             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1247             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1248             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1249             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1250             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1251             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1252             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1253             velec            = _mm256_mul_pd(qq31,_mm256_sub_pd(rinv31,velec));
1254             felec            = _mm256_mul_pd(_mm256_mul_pd(qq31,rinv31),_mm256_sub_pd(rinvsq31,felec));
1255
1256             /* Update potential sum for this i atom from the interaction with this j atom. */
1257             velec            = _mm256_andnot_pd(dummy_mask,velec);
1258             velecsum         = _mm256_add_pd(velecsum,velec);
1259
1260             fscal            = felec;
1261
1262             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1263
1264             /* Calculate temporary vectorial force */
1265             tx               = _mm256_mul_pd(fscal,dx31);
1266             ty               = _mm256_mul_pd(fscal,dy31);
1267             tz               = _mm256_mul_pd(fscal,dz31);
1268
1269             /* Update vectorial force */
1270             fix3             = _mm256_add_pd(fix3,tx);
1271             fiy3             = _mm256_add_pd(fiy3,ty);
1272             fiz3             = _mm256_add_pd(fiz3,tz);
1273
1274             fjx1             = _mm256_add_pd(fjx1,tx);
1275             fjy1             = _mm256_add_pd(fjy1,ty);
1276             fjz1             = _mm256_add_pd(fjz1,tz);
1277
1278             /**************************
1279              * CALCULATE INTERACTIONS *
1280              **************************/
1281
1282             r32              = _mm256_mul_pd(rsq32,rinv32);
1283             r32              = _mm256_andnot_pd(dummy_mask,r32);
1284
1285             /* EWALD ELECTROSTATICS */
1286
1287             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1288             ewrt             = _mm256_mul_pd(r32,ewtabscale);
1289             ewitab           = _mm256_cvttpd_epi32(ewrt);
1290             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1291             ewitab           = _mm_slli_epi32(ewitab,2);
1292             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1293             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1294             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1295             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1296             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1297             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1298             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1299             velec            = _mm256_mul_pd(qq32,_mm256_sub_pd(rinv32,velec));
1300             felec            = _mm256_mul_pd(_mm256_mul_pd(qq32,rinv32),_mm256_sub_pd(rinvsq32,felec));
1301
1302             /* Update potential sum for this i atom from the interaction with this j atom. */
1303             velec            = _mm256_andnot_pd(dummy_mask,velec);
1304             velecsum         = _mm256_add_pd(velecsum,velec);
1305
1306             fscal            = felec;
1307
1308             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1309
1310             /* Calculate temporary vectorial force */
1311             tx               = _mm256_mul_pd(fscal,dx32);
1312             ty               = _mm256_mul_pd(fscal,dy32);
1313             tz               = _mm256_mul_pd(fscal,dz32);
1314
1315             /* Update vectorial force */
1316             fix3             = _mm256_add_pd(fix3,tx);
1317             fiy3             = _mm256_add_pd(fiy3,ty);
1318             fiz3             = _mm256_add_pd(fiz3,tz);
1319
1320             fjx2             = _mm256_add_pd(fjx2,tx);
1321             fjy2             = _mm256_add_pd(fjy2,ty);
1322             fjz2             = _mm256_add_pd(fjz2,tz);
1323
1324             /**************************
1325              * CALCULATE INTERACTIONS *
1326              **************************/
1327
1328             r33              = _mm256_mul_pd(rsq33,rinv33);
1329             r33              = _mm256_andnot_pd(dummy_mask,r33);
1330
1331             /* EWALD ELECTROSTATICS */
1332
1333             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1334             ewrt             = _mm256_mul_pd(r33,ewtabscale);
1335             ewitab           = _mm256_cvttpd_epi32(ewrt);
1336             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1337             ewitab           = _mm_slli_epi32(ewitab,2);
1338             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1339             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1340             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1341             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1342             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1343             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1344             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1345             velec            = _mm256_mul_pd(qq33,_mm256_sub_pd(rinv33,velec));
1346             felec            = _mm256_mul_pd(_mm256_mul_pd(qq33,rinv33),_mm256_sub_pd(rinvsq33,felec));
1347
1348             /* Update potential sum for this i atom from the interaction with this j atom. */
1349             velec            = _mm256_andnot_pd(dummy_mask,velec);
1350             velecsum         = _mm256_add_pd(velecsum,velec);
1351
1352             fscal            = felec;
1353
1354             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1355
1356             /* Calculate temporary vectorial force */
1357             tx               = _mm256_mul_pd(fscal,dx33);
1358             ty               = _mm256_mul_pd(fscal,dy33);
1359             tz               = _mm256_mul_pd(fscal,dz33);
1360
1361             /* Update vectorial force */
1362             fix3             = _mm256_add_pd(fix3,tx);
1363             fiy3             = _mm256_add_pd(fiy3,ty);
1364             fiz3             = _mm256_add_pd(fiz3,tz);
1365
1366             fjx3             = _mm256_add_pd(fjx3,tx);
1367             fjy3             = _mm256_add_pd(fjy3,ty);
1368             fjz3             = _mm256_add_pd(fjz3,tz);
1369
1370             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1371             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1372             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1373             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1374
1375             gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1376                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1377                                                       fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1378
1379             /* Inner loop uses 438 flops */
1380         }
1381
1382         /* End of innermost loop */
1383
1384         gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1385                                                  f+i_coord_offset,fshift+i_shift_offset);
1386
1387         ggid                        = gid[iidx];
1388         /* Update potential energies */
1389         gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1390         gmx_mm256_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1391
1392         /* Increment number of inner iterations */
1393         inneriter                  += j_index_end - j_index_start;
1394
1395         /* Outer loop uses 26 flops */
1396     }
1397
1398     /* Increment number of outer iterations */
1399     outeriter        += nri;
1400
1401     /* Update outer/inner flops */
1402
1403     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*438);
1404 }
1405 /*
1406  * Gromacs nonbonded kernel:   nb_kernel_ElecEw_VdwCSTab_GeomW4W4_F_avx_256_double
1407  * Electrostatics interaction: Ewald
1408  * VdW interaction:            CubicSplineTable
1409  * Geometry:                   Water4-Water4
1410  * Calculate force/pot:        Force
1411  */
1412 void
1413 nb_kernel_ElecEw_VdwCSTab_GeomW4W4_F_avx_256_double
1414                     (t_nblist                    * gmx_restrict       nlist,
1415                      rvec                        * gmx_restrict          xx,
1416                      rvec                        * gmx_restrict          ff,
1417                      t_forcerec                  * gmx_restrict          fr,
1418                      t_mdatoms                   * gmx_restrict     mdatoms,
1419                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1420                      t_nrnb                      * gmx_restrict        nrnb)
1421 {
1422     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1423      * just 0 for non-waters.
1424      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
1425      * jnr indices corresponding to data put in the four positions in the SIMD register.
1426      */
1427     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1428     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1429     int              jnrA,jnrB,jnrC,jnrD;
1430     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1431     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
1432     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1433     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1434     real             rcutoff_scalar;
1435     real             *shiftvec,*fshift,*x,*f;
1436     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1437     real             scratch[4*DIM];
1438     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1439     real *           vdwioffsetptr0;
1440     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1441     real *           vdwioffsetptr1;
1442     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1443     real *           vdwioffsetptr2;
1444     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1445     real *           vdwioffsetptr3;
1446     __m256d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1447     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1448     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1449     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1450     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1451     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1452     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1453     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1454     __m256d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1455     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1456     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1457     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1458     __m256d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1459     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1460     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1461     __m256d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1462     __m256d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1463     __m256d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1464     __m256d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1465     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
1466     real             *charge;
1467     int              nvdwtype;
1468     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1469     int              *vdwtype;
1470     real             *vdwparam;
1471     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
1472     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
1473     __m128i          vfitab;
1474     __m128i          ifour       = _mm_set1_epi32(4);
1475     __m256d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1476     real             *vftab;
1477     __m128i          ewitab;
1478     __m256d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1479     __m256d          beta,beta2,beta3,zeta2,pmecorrF,pmecorrV,rinv3;
1480     real             *ewtab;
1481     __m256d          dummy_mask,cutoff_mask;
1482     __m128           tmpmask0,tmpmask1;
1483     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
1484     __m256d          one     = _mm256_set1_pd(1.0);
1485     __m256d          two     = _mm256_set1_pd(2.0);
1486     x                = xx[0];
1487     f                = ff[0];
1488
1489     nri              = nlist->nri;
1490     iinr             = nlist->iinr;
1491     jindex           = nlist->jindex;
1492     jjnr             = nlist->jjnr;
1493     shiftidx         = nlist->shift;
1494     gid              = nlist->gid;
1495     shiftvec         = fr->shift_vec[0];
1496     fshift           = fr->fshift[0];
1497     facel            = _mm256_set1_pd(fr->epsfac);
1498     charge           = mdatoms->chargeA;
1499     nvdwtype         = fr->ntype;
1500     vdwparam         = fr->nbfp;
1501     vdwtype          = mdatoms->typeA;
1502
1503     vftab            = kernel_data->table_vdw->data;
1504     vftabscale       = _mm256_set1_pd(kernel_data->table_vdw->scale);
1505
1506     sh_ewald         = _mm256_set1_pd(fr->ic->sh_ewald);
1507     beta             = _mm256_set1_pd(fr->ic->ewaldcoeff_q);
1508     beta2            = _mm256_mul_pd(beta,beta);
1509     beta3            = _mm256_mul_pd(beta,beta2);
1510
1511     ewtab            = fr->ic->tabq_coul_F;
1512     ewtabscale       = _mm256_set1_pd(fr->ic->tabq_scale);
1513     ewtabhalfspace   = _mm256_set1_pd(0.5/fr->ic->tabq_scale);
1514
1515     /* Setup water-specific parameters */
1516     inr              = nlist->iinr[0];
1517     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1518     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1519     iq3              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
1520     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
1521
1522     jq1              = _mm256_set1_pd(charge[inr+1]);
1523     jq2              = _mm256_set1_pd(charge[inr+2]);
1524     jq3              = _mm256_set1_pd(charge[inr+3]);
1525     vdwjidx0A        = 2*vdwtype[inr+0];
1526     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
1527     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
1528     qq11             = _mm256_mul_pd(iq1,jq1);
1529     qq12             = _mm256_mul_pd(iq1,jq2);
1530     qq13             = _mm256_mul_pd(iq1,jq3);
1531     qq21             = _mm256_mul_pd(iq2,jq1);
1532     qq22             = _mm256_mul_pd(iq2,jq2);
1533     qq23             = _mm256_mul_pd(iq2,jq3);
1534     qq31             = _mm256_mul_pd(iq3,jq1);
1535     qq32             = _mm256_mul_pd(iq3,jq2);
1536     qq33             = _mm256_mul_pd(iq3,jq3);
1537
1538     /* Avoid stupid compiler warnings */
1539     jnrA = jnrB = jnrC = jnrD = 0;
1540     j_coord_offsetA = 0;
1541     j_coord_offsetB = 0;
1542     j_coord_offsetC = 0;
1543     j_coord_offsetD = 0;
1544
1545     outeriter        = 0;
1546     inneriter        = 0;
1547
1548     for(iidx=0;iidx<4*DIM;iidx++)
1549     {
1550         scratch[iidx] = 0.0;
1551     }
1552
1553     /* Start outer loop over neighborlists */
1554     for(iidx=0; iidx<nri; iidx++)
1555     {
1556         /* Load shift vector for this list */
1557         i_shift_offset   = DIM*shiftidx[iidx];
1558
1559         /* Load limits for loop over neighbors */
1560         j_index_start    = jindex[iidx];
1561         j_index_end      = jindex[iidx+1];
1562
1563         /* Get outer coordinate index */
1564         inr              = iinr[iidx];
1565         i_coord_offset   = DIM*inr;
1566
1567         /* Load i particle coords and add shift vector */
1568         gmx_mm256_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1569                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1570
1571         fix0             = _mm256_setzero_pd();
1572         fiy0             = _mm256_setzero_pd();
1573         fiz0             = _mm256_setzero_pd();
1574         fix1             = _mm256_setzero_pd();
1575         fiy1             = _mm256_setzero_pd();
1576         fiz1             = _mm256_setzero_pd();
1577         fix2             = _mm256_setzero_pd();
1578         fiy2             = _mm256_setzero_pd();
1579         fiz2             = _mm256_setzero_pd();
1580         fix3             = _mm256_setzero_pd();
1581         fiy3             = _mm256_setzero_pd();
1582         fiz3             = _mm256_setzero_pd();
1583
1584         /* Start inner kernel loop */
1585         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1586         {
1587
1588             /* Get j neighbor index, and coordinate index */
1589             jnrA             = jjnr[jidx];
1590             jnrB             = jjnr[jidx+1];
1591             jnrC             = jjnr[jidx+2];
1592             jnrD             = jjnr[jidx+3];
1593             j_coord_offsetA  = DIM*jnrA;
1594             j_coord_offsetB  = DIM*jnrB;
1595             j_coord_offsetC  = DIM*jnrC;
1596             j_coord_offsetD  = DIM*jnrD;
1597
1598             /* load j atom coordinates */
1599             gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1600                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1601                                                  &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1602                                                  &jy2,&jz2,&jx3,&jy3,&jz3);
1603
1604             /* Calculate displacement vector */
1605             dx00             = _mm256_sub_pd(ix0,jx0);
1606             dy00             = _mm256_sub_pd(iy0,jy0);
1607             dz00             = _mm256_sub_pd(iz0,jz0);
1608             dx11             = _mm256_sub_pd(ix1,jx1);
1609             dy11             = _mm256_sub_pd(iy1,jy1);
1610             dz11             = _mm256_sub_pd(iz1,jz1);
1611             dx12             = _mm256_sub_pd(ix1,jx2);
1612             dy12             = _mm256_sub_pd(iy1,jy2);
1613             dz12             = _mm256_sub_pd(iz1,jz2);
1614             dx13             = _mm256_sub_pd(ix1,jx3);
1615             dy13             = _mm256_sub_pd(iy1,jy3);
1616             dz13             = _mm256_sub_pd(iz1,jz3);
1617             dx21             = _mm256_sub_pd(ix2,jx1);
1618             dy21             = _mm256_sub_pd(iy2,jy1);
1619             dz21             = _mm256_sub_pd(iz2,jz1);
1620             dx22             = _mm256_sub_pd(ix2,jx2);
1621             dy22             = _mm256_sub_pd(iy2,jy2);
1622             dz22             = _mm256_sub_pd(iz2,jz2);
1623             dx23             = _mm256_sub_pd(ix2,jx3);
1624             dy23             = _mm256_sub_pd(iy2,jy3);
1625             dz23             = _mm256_sub_pd(iz2,jz3);
1626             dx31             = _mm256_sub_pd(ix3,jx1);
1627             dy31             = _mm256_sub_pd(iy3,jy1);
1628             dz31             = _mm256_sub_pd(iz3,jz1);
1629             dx32             = _mm256_sub_pd(ix3,jx2);
1630             dy32             = _mm256_sub_pd(iy3,jy2);
1631             dz32             = _mm256_sub_pd(iz3,jz2);
1632             dx33             = _mm256_sub_pd(ix3,jx3);
1633             dy33             = _mm256_sub_pd(iy3,jy3);
1634             dz33             = _mm256_sub_pd(iz3,jz3);
1635
1636             /* Calculate squared distance and things based on it */
1637             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1638             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1639             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1640             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
1641             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1642             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1643             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
1644             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
1645             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
1646             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
1647
1648             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
1649             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1650             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1651             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
1652             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1653             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1654             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
1655             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
1656             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
1657             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
1658
1659             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1660             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1661             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
1662             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1663             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1664             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
1665             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
1666             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
1667             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
1668
1669             fjx0             = _mm256_setzero_pd();
1670             fjy0             = _mm256_setzero_pd();
1671             fjz0             = _mm256_setzero_pd();
1672             fjx1             = _mm256_setzero_pd();
1673             fjy1             = _mm256_setzero_pd();
1674             fjz1             = _mm256_setzero_pd();
1675             fjx2             = _mm256_setzero_pd();
1676             fjy2             = _mm256_setzero_pd();
1677             fjz2             = _mm256_setzero_pd();
1678             fjx3             = _mm256_setzero_pd();
1679             fjy3             = _mm256_setzero_pd();
1680             fjz3             = _mm256_setzero_pd();
1681
1682             /**************************
1683              * CALCULATE INTERACTIONS *
1684              **************************/
1685
1686             r00              = _mm256_mul_pd(rsq00,rinv00);
1687
1688             /* Calculate table index by multiplying r with table scale and truncate to integer */
1689             rt               = _mm256_mul_pd(r00,vftabscale);
1690             vfitab           = _mm256_cvttpd_epi32(rt);
1691             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
1692             vfitab           = _mm_slli_epi32(vfitab,3);
1693
1694             /* CUBIC SPLINE TABLE DISPERSION */
1695             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1696             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1697             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1698             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1699             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1700             Heps             = _mm256_mul_pd(vfeps,H);
1701             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1702             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1703             fvdw6            = _mm256_mul_pd(c6_00,FF);
1704
1705             /* CUBIC SPLINE TABLE REPULSION */
1706             vfitab           = _mm_add_epi32(vfitab,ifour);
1707             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1708             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1709             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1710             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1711             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1712             Heps             = _mm256_mul_pd(vfeps,H);
1713             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1714             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1715             fvdw12           = _mm256_mul_pd(c12_00,FF);
1716             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
1717
1718             fscal            = fvdw;
1719
1720             /* Calculate temporary vectorial force */
1721             tx               = _mm256_mul_pd(fscal,dx00);
1722             ty               = _mm256_mul_pd(fscal,dy00);
1723             tz               = _mm256_mul_pd(fscal,dz00);
1724
1725             /* Update vectorial force */
1726             fix0             = _mm256_add_pd(fix0,tx);
1727             fiy0             = _mm256_add_pd(fiy0,ty);
1728             fiz0             = _mm256_add_pd(fiz0,tz);
1729
1730             fjx0             = _mm256_add_pd(fjx0,tx);
1731             fjy0             = _mm256_add_pd(fjy0,ty);
1732             fjz0             = _mm256_add_pd(fjz0,tz);
1733
1734             /**************************
1735              * CALCULATE INTERACTIONS *
1736              **************************/
1737
1738             r11              = _mm256_mul_pd(rsq11,rinv11);
1739
1740             /* EWALD ELECTROSTATICS */
1741
1742             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1743             ewrt             = _mm256_mul_pd(r11,ewtabscale);
1744             ewitab           = _mm256_cvttpd_epi32(ewrt);
1745             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1746             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1747                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1748                                             &ewtabF,&ewtabFn);
1749             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1750             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
1751
1752             fscal            = felec;
1753
1754             /* Calculate temporary vectorial force */
1755             tx               = _mm256_mul_pd(fscal,dx11);
1756             ty               = _mm256_mul_pd(fscal,dy11);
1757             tz               = _mm256_mul_pd(fscal,dz11);
1758
1759             /* Update vectorial force */
1760             fix1             = _mm256_add_pd(fix1,tx);
1761             fiy1             = _mm256_add_pd(fiy1,ty);
1762             fiz1             = _mm256_add_pd(fiz1,tz);
1763
1764             fjx1             = _mm256_add_pd(fjx1,tx);
1765             fjy1             = _mm256_add_pd(fjy1,ty);
1766             fjz1             = _mm256_add_pd(fjz1,tz);
1767
1768             /**************************
1769              * CALCULATE INTERACTIONS *
1770              **************************/
1771
1772             r12              = _mm256_mul_pd(rsq12,rinv12);
1773
1774             /* EWALD ELECTROSTATICS */
1775
1776             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1777             ewrt             = _mm256_mul_pd(r12,ewtabscale);
1778             ewitab           = _mm256_cvttpd_epi32(ewrt);
1779             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1780             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1781                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1782                                             &ewtabF,&ewtabFn);
1783             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1784             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
1785
1786             fscal            = felec;
1787
1788             /* Calculate temporary vectorial force */
1789             tx               = _mm256_mul_pd(fscal,dx12);
1790             ty               = _mm256_mul_pd(fscal,dy12);
1791             tz               = _mm256_mul_pd(fscal,dz12);
1792
1793             /* Update vectorial force */
1794             fix1             = _mm256_add_pd(fix1,tx);
1795             fiy1             = _mm256_add_pd(fiy1,ty);
1796             fiz1             = _mm256_add_pd(fiz1,tz);
1797
1798             fjx2             = _mm256_add_pd(fjx2,tx);
1799             fjy2             = _mm256_add_pd(fjy2,ty);
1800             fjz2             = _mm256_add_pd(fjz2,tz);
1801
1802             /**************************
1803              * CALCULATE INTERACTIONS *
1804              **************************/
1805
1806             r13              = _mm256_mul_pd(rsq13,rinv13);
1807
1808             /* EWALD ELECTROSTATICS */
1809
1810             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1811             ewrt             = _mm256_mul_pd(r13,ewtabscale);
1812             ewitab           = _mm256_cvttpd_epi32(ewrt);
1813             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1814             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1815                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1816                                             &ewtabF,&ewtabFn);
1817             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1818             felec            = _mm256_mul_pd(_mm256_mul_pd(qq13,rinv13),_mm256_sub_pd(rinvsq13,felec));
1819
1820             fscal            = felec;
1821
1822             /* Calculate temporary vectorial force */
1823             tx               = _mm256_mul_pd(fscal,dx13);
1824             ty               = _mm256_mul_pd(fscal,dy13);
1825             tz               = _mm256_mul_pd(fscal,dz13);
1826
1827             /* Update vectorial force */
1828             fix1             = _mm256_add_pd(fix1,tx);
1829             fiy1             = _mm256_add_pd(fiy1,ty);
1830             fiz1             = _mm256_add_pd(fiz1,tz);
1831
1832             fjx3             = _mm256_add_pd(fjx3,tx);
1833             fjy3             = _mm256_add_pd(fjy3,ty);
1834             fjz3             = _mm256_add_pd(fjz3,tz);
1835
1836             /**************************
1837              * CALCULATE INTERACTIONS *
1838              **************************/
1839
1840             r21              = _mm256_mul_pd(rsq21,rinv21);
1841
1842             /* EWALD ELECTROSTATICS */
1843
1844             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1845             ewrt             = _mm256_mul_pd(r21,ewtabscale);
1846             ewitab           = _mm256_cvttpd_epi32(ewrt);
1847             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1848             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1849                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1850                                             &ewtabF,&ewtabFn);
1851             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1852             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
1853
1854             fscal            = felec;
1855
1856             /* Calculate temporary vectorial force */
1857             tx               = _mm256_mul_pd(fscal,dx21);
1858             ty               = _mm256_mul_pd(fscal,dy21);
1859             tz               = _mm256_mul_pd(fscal,dz21);
1860
1861             /* Update vectorial force */
1862             fix2             = _mm256_add_pd(fix2,tx);
1863             fiy2             = _mm256_add_pd(fiy2,ty);
1864             fiz2             = _mm256_add_pd(fiz2,tz);
1865
1866             fjx1             = _mm256_add_pd(fjx1,tx);
1867             fjy1             = _mm256_add_pd(fjy1,ty);
1868             fjz1             = _mm256_add_pd(fjz1,tz);
1869
1870             /**************************
1871              * CALCULATE INTERACTIONS *
1872              **************************/
1873
1874             r22              = _mm256_mul_pd(rsq22,rinv22);
1875
1876             /* EWALD ELECTROSTATICS */
1877
1878             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1879             ewrt             = _mm256_mul_pd(r22,ewtabscale);
1880             ewitab           = _mm256_cvttpd_epi32(ewrt);
1881             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1882             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1883                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1884                                             &ewtabF,&ewtabFn);
1885             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1886             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
1887
1888             fscal            = felec;
1889
1890             /* Calculate temporary vectorial force */
1891             tx               = _mm256_mul_pd(fscal,dx22);
1892             ty               = _mm256_mul_pd(fscal,dy22);
1893             tz               = _mm256_mul_pd(fscal,dz22);
1894
1895             /* Update vectorial force */
1896             fix2             = _mm256_add_pd(fix2,tx);
1897             fiy2             = _mm256_add_pd(fiy2,ty);
1898             fiz2             = _mm256_add_pd(fiz2,tz);
1899
1900             fjx2             = _mm256_add_pd(fjx2,tx);
1901             fjy2             = _mm256_add_pd(fjy2,ty);
1902             fjz2             = _mm256_add_pd(fjz2,tz);
1903
1904             /**************************
1905              * CALCULATE INTERACTIONS *
1906              **************************/
1907
1908             r23              = _mm256_mul_pd(rsq23,rinv23);
1909
1910             /* EWALD ELECTROSTATICS */
1911
1912             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1913             ewrt             = _mm256_mul_pd(r23,ewtabscale);
1914             ewitab           = _mm256_cvttpd_epi32(ewrt);
1915             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1916             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1917                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1918                                             &ewtabF,&ewtabFn);
1919             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1920             felec            = _mm256_mul_pd(_mm256_mul_pd(qq23,rinv23),_mm256_sub_pd(rinvsq23,felec));
1921
1922             fscal            = felec;
1923
1924             /* Calculate temporary vectorial force */
1925             tx               = _mm256_mul_pd(fscal,dx23);
1926             ty               = _mm256_mul_pd(fscal,dy23);
1927             tz               = _mm256_mul_pd(fscal,dz23);
1928
1929             /* Update vectorial force */
1930             fix2             = _mm256_add_pd(fix2,tx);
1931             fiy2             = _mm256_add_pd(fiy2,ty);
1932             fiz2             = _mm256_add_pd(fiz2,tz);
1933
1934             fjx3             = _mm256_add_pd(fjx3,tx);
1935             fjy3             = _mm256_add_pd(fjy3,ty);
1936             fjz3             = _mm256_add_pd(fjz3,tz);
1937
1938             /**************************
1939              * CALCULATE INTERACTIONS *
1940              **************************/
1941
1942             r31              = _mm256_mul_pd(rsq31,rinv31);
1943
1944             /* EWALD ELECTROSTATICS */
1945
1946             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1947             ewrt             = _mm256_mul_pd(r31,ewtabscale);
1948             ewitab           = _mm256_cvttpd_epi32(ewrt);
1949             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1950             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1951                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1952                                             &ewtabF,&ewtabFn);
1953             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1954             felec            = _mm256_mul_pd(_mm256_mul_pd(qq31,rinv31),_mm256_sub_pd(rinvsq31,felec));
1955
1956             fscal            = felec;
1957
1958             /* Calculate temporary vectorial force */
1959             tx               = _mm256_mul_pd(fscal,dx31);
1960             ty               = _mm256_mul_pd(fscal,dy31);
1961             tz               = _mm256_mul_pd(fscal,dz31);
1962
1963             /* Update vectorial force */
1964             fix3             = _mm256_add_pd(fix3,tx);
1965             fiy3             = _mm256_add_pd(fiy3,ty);
1966             fiz3             = _mm256_add_pd(fiz3,tz);
1967
1968             fjx1             = _mm256_add_pd(fjx1,tx);
1969             fjy1             = _mm256_add_pd(fjy1,ty);
1970             fjz1             = _mm256_add_pd(fjz1,tz);
1971
1972             /**************************
1973              * CALCULATE INTERACTIONS *
1974              **************************/
1975
1976             r32              = _mm256_mul_pd(rsq32,rinv32);
1977
1978             /* EWALD ELECTROSTATICS */
1979
1980             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1981             ewrt             = _mm256_mul_pd(r32,ewtabscale);
1982             ewitab           = _mm256_cvttpd_epi32(ewrt);
1983             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1984             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1985                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1986                                             &ewtabF,&ewtabFn);
1987             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1988             felec            = _mm256_mul_pd(_mm256_mul_pd(qq32,rinv32),_mm256_sub_pd(rinvsq32,felec));
1989
1990             fscal            = felec;
1991
1992             /* Calculate temporary vectorial force */
1993             tx               = _mm256_mul_pd(fscal,dx32);
1994             ty               = _mm256_mul_pd(fscal,dy32);
1995             tz               = _mm256_mul_pd(fscal,dz32);
1996
1997             /* Update vectorial force */
1998             fix3             = _mm256_add_pd(fix3,tx);
1999             fiy3             = _mm256_add_pd(fiy3,ty);
2000             fiz3             = _mm256_add_pd(fiz3,tz);
2001
2002             fjx2             = _mm256_add_pd(fjx2,tx);
2003             fjy2             = _mm256_add_pd(fjy2,ty);
2004             fjz2             = _mm256_add_pd(fjz2,tz);
2005
2006             /**************************
2007              * CALCULATE INTERACTIONS *
2008              **************************/
2009
2010             r33              = _mm256_mul_pd(rsq33,rinv33);
2011
2012             /* EWALD ELECTROSTATICS */
2013
2014             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2015             ewrt             = _mm256_mul_pd(r33,ewtabscale);
2016             ewitab           = _mm256_cvttpd_epi32(ewrt);
2017             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2018             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2019                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2020                                             &ewtabF,&ewtabFn);
2021             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2022             felec            = _mm256_mul_pd(_mm256_mul_pd(qq33,rinv33),_mm256_sub_pd(rinvsq33,felec));
2023
2024             fscal            = felec;
2025
2026             /* Calculate temporary vectorial force */
2027             tx               = _mm256_mul_pd(fscal,dx33);
2028             ty               = _mm256_mul_pd(fscal,dy33);
2029             tz               = _mm256_mul_pd(fscal,dz33);
2030
2031             /* Update vectorial force */
2032             fix3             = _mm256_add_pd(fix3,tx);
2033             fiy3             = _mm256_add_pd(fiy3,ty);
2034             fiz3             = _mm256_add_pd(fiz3,tz);
2035
2036             fjx3             = _mm256_add_pd(fjx3,tx);
2037             fjy3             = _mm256_add_pd(fjy3,ty);
2038             fjz3             = _mm256_add_pd(fjz3,tz);
2039
2040             fjptrA             = f+j_coord_offsetA;
2041             fjptrB             = f+j_coord_offsetB;
2042             fjptrC             = f+j_coord_offsetC;
2043             fjptrD             = f+j_coord_offsetD;
2044
2045             gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2046                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2047                                                       fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2048
2049             /* Inner loop uses 375 flops */
2050         }
2051
2052         if(jidx<j_index_end)
2053         {
2054
2055             /* Get j neighbor index, and coordinate index */
2056             jnrlistA         = jjnr[jidx];
2057             jnrlistB         = jjnr[jidx+1];
2058             jnrlistC         = jjnr[jidx+2];
2059             jnrlistD         = jjnr[jidx+3];
2060             /* Sign of each element will be negative for non-real atoms.
2061              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
2062              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
2063              */
2064             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
2065
2066             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
2067             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
2068             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
2069
2070             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
2071             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
2072             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
2073             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
2074             j_coord_offsetA  = DIM*jnrA;
2075             j_coord_offsetB  = DIM*jnrB;
2076             j_coord_offsetC  = DIM*jnrC;
2077             j_coord_offsetD  = DIM*jnrD;
2078
2079             /* load j atom coordinates */
2080             gmx_mm256_load_4rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
2081                                                  x+j_coord_offsetC,x+j_coord_offsetD,
2082                                                  &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
2083                                                  &jy2,&jz2,&jx3,&jy3,&jz3);
2084
2085             /* Calculate displacement vector */
2086             dx00             = _mm256_sub_pd(ix0,jx0);
2087             dy00             = _mm256_sub_pd(iy0,jy0);
2088             dz00             = _mm256_sub_pd(iz0,jz0);
2089             dx11             = _mm256_sub_pd(ix1,jx1);
2090             dy11             = _mm256_sub_pd(iy1,jy1);
2091             dz11             = _mm256_sub_pd(iz1,jz1);
2092             dx12             = _mm256_sub_pd(ix1,jx2);
2093             dy12             = _mm256_sub_pd(iy1,jy2);
2094             dz12             = _mm256_sub_pd(iz1,jz2);
2095             dx13             = _mm256_sub_pd(ix1,jx3);
2096             dy13             = _mm256_sub_pd(iy1,jy3);
2097             dz13             = _mm256_sub_pd(iz1,jz3);
2098             dx21             = _mm256_sub_pd(ix2,jx1);
2099             dy21             = _mm256_sub_pd(iy2,jy1);
2100             dz21             = _mm256_sub_pd(iz2,jz1);
2101             dx22             = _mm256_sub_pd(ix2,jx2);
2102             dy22             = _mm256_sub_pd(iy2,jy2);
2103             dz22             = _mm256_sub_pd(iz2,jz2);
2104             dx23             = _mm256_sub_pd(ix2,jx3);
2105             dy23             = _mm256_sub_pd(iy2,jy3);
2106             dz23             = _mm256_sub_pd(iz2,jz3);
2107             dx31             = _mm256_sub_pd(ix3,jx1);
2108             dy31             = _mm256_sub_pd(iy3,jy1);
2109             dz31             = _mm256_sub_pd(iz3,jz1);
2110             dx32             = _mm256_sub_pd(ix3,jx2);
2111             dy32             = _mm256_sub_pd(iy3,jy2);
2112             dz32             = _mm256_sub_pd(iz3,jz2);
2113             dx33             = _mm256_sub_pd(ix3,jx3);
2114             dy33             = _mm256_sub_pd(iy3,jy3);
2115             dz33             = _mm256_sub_pd(iz3,jz3);
2116
2117             /* Calculate squared distance and things based on it */
2118             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
2119             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
2120             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
2121             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
2122             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
2123             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
2124             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
2125             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
2126             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
2127             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
2128
2129             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
2130             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
2131             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
2132             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
2133             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
2134             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
2135             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
2136             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
2137             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
2138             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
2139
2140             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
2141             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
2142             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
2143             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
2144             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
2145             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
2146             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
2147             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
2148             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
2149
2150             fjx0             = _mm256_setzero_pd();
2151             fjy0             = _mm256_setzero_pd();
2152             fjz0             = _mm256_setzero_pd();
2153             fjx1             = _mm256_setzero_pd();
2154             fjy1             = _mm256_setzero_pd();
2155             fjz1             = _mm256_setzero_pd();
2156             fjx2             = _mm256_setzero_pd();
2157             fjy2             = _mm256_setzero_pd();
2158             fjz2             = _mm256_setzero_pd();
2159             fjx3             = _mm256_setzero_pd();
2160             fjy3             = _mm256_setzero_pd();
2161             fjz3             = _mm256_setzero_pd();
2162
2163             /**************************
2164              * CALCULATE INTERACTIONS *
2165              **************************/
2166
2167             r00              = _mm256_mul_pd(rsq00,rinv00);
2168             r00              = _mm256_andnot_pd(dummy_mask,r00);
2169
2170             /* Calculate table index by multiplying r with table scale and truncate to integer */
2171             rt               = _mm256_mul_pd(r00,vftabscale);
2172             vfitab           = _mm256_cvttpd_epi32(rt);
2173             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
2174             vfitab           = _mm_slli_epi32(vfitab,3);
2175
2176             /* CUBIC SPLINE TABLE DISPERSION */
2177             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
2178             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
2179             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
2180             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
2181             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
2182             Heps             = _mm256_mul_pd(vfeps,H);
2183             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
2184             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
2185             fvdw6            = _mm256_mul_pd(c6_00,FF);
2186
2187             /* CUBIC SPLINE TABLE REPULSION */
2188             vfitab           = _mm_add_epi32(vfitab,ifour);
2189             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
2190             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
2191             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
2192             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
2193             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
2194             Heps             = _mm256_mul_pd(vfeps,H);
2195             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
2196             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
2197             fvdw12           = _mm256_mul_pd(c12_00,FF);
2198             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
2199
2200             fscal            = fvdw;
2201
2202             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2203
2204             /* Calculate temporary vectorial force */
2205             tx               = _mm256_mul_pd(fscal,dx00);
2206             ty               = _mm256_mul_pd(fscal,dy00);
2207             tz               = _mm256_mul_pd(fscal,dz00);
2208
2209             /* Update vectorial force */
2210             fix0             = _mm256_add_pd(fix0,tx);
2211             fiy0             = _mm256_add_pd(fiy0,ty);
2212             fiz0             = _mm256_add_pd(fiz0,tz);
2213
2214             fjx0             = _mm256_add_pd(fjx0,tx);
2215             fjy0             = _mm256_add_pd(fjy0,ty);
2216             fjz0             = _mm256_add_pd(fjz0,tz);
2217
2218             /**************************
2219              * CALCULATE INTERACTIONS *
2220              **************************/
2221
2222             r11              = _mm256_mul_pd(rsq11,rinv11);
2223             r11              = _mm256_andnot_pd(dummy_mask,r11);
2224
2225             /* EWALD ELECTROSTATICS */
2226
2227             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2228             ewrt             = _mm256_mul_pd(r11,ewtabscale);
2229             ewitab           = _mm256_cvttpd_epi32(ewrt);
2230             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2231             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2232                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2233                                             &ewtabF,&ewtabFn);
2234             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2235             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
2236
2237             fscal            = felec;
2238
2239             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2240
2241             /* Calculate temporary vectorial force */
2242             tx               = _mm256_mul_pd(fscal,dx11);
2243             ty               = _mm256_mul_pd(fscal,dy11);
2244             tz               = _mm256_mul_pd(fscal,dz11);
2245
2246             /* Update vectorial force */
2247             fix1             = _mm256_add_pd(fix1,tx);
2248             fiy1             = _mm256_add_pd(fiy1,ty);
2249             fiz1             = _mm256_add_pd(fiz1,tz);
2250
2251             fjx1             = _mm256_add_pd(fjx1,tx);
2252             fjy1             = _mm256_add_pd(fjy1,ty);
2253             fjz1             = _mm256_add_pd(fjz1,tz);
2254
2255             /**************************
2256              * CALCULATE INTERACTIONS *
2257              **************************/
2258
2259             r12              = _mm256_mul_pd(rsq12,rinv12);
2260             r12              = _mm256_andnot_pd(dummy_mask,r12);
2261
2262             /* EWALD ELECTROSTATICS */
2263
2264             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2265             ewrt             = _mm256_mul_pd(r12,ewtabscale);
2266             ewitab           = _mm256_cvttpd_epi32(ewrt);
2267             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2268             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2269                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2270                                             &ewtabF,&ewtabFn);
2271             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2272             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
2273
2274             fscal            = felec;
2275
2276             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2277
2278             /* Calculate temporary vectorial force */
2279             tx               = _mm256_mul_pd(fscal,dx12);
2280             ty               = _mm256_mul_pd(fscal,dy12);
2281             tz               = _mm256_mul_pd(fscal,dz12);
2282
2283             /* Update vectorial force */
2284             fix1             = _mm256_add_pd(fix1,tx);
2285             fiy1             = _mm256_add_pd(fiy1,ty);
2286             fiz1             = _mm256_add_pd(fiz1,tz);
2287
2288             fjx2             = _mm256_add_pd(fjx2,tx);
2289             fjy2             = _mm256_add_pd(fjy2,ty);
2290             fjz2             = _mm256_add_pd(fjz2,tz);
2291
2292             /**************************
2293              * CALCULATE INTERACTIONS *
2294              **************************/
2295
2296             r13              = _mm256_mul_pd(rsq13,rinv13);
2297             r13              = _mm256_andnot_pd(dummy_mask,r13);
2298
2299             /* EWALD ELECTROSTATICS */
2300
2301             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2302             ewrt             = _mm256_mul_pd(r13,ewtabscale);
2303             ewitab           = _mm256_cvttpd_epi32(ewrt);
2304             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2305             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2306                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2307                                             &ewtabF,&ewtabFn);
2308             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2309             felec            = _mm256_mul_pd(_mm256_mul_pd(qq13,rinv13),_mm256_sub_pd(rinvsq13,felec));
2310
2311             fscal            = felec;
2312
2313             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2314
2315             /* Calculate temporary vectorial force */
2316             tx               = _mm256_mul_pd(fscal,dx13);
2317             ty               = _mm256_mul_pd(fscal,dy13);
2318             tz               = _mm256_mul_pd(fscal,dz13);
2319
2320             /* Update vectorial force */
2321             fix1             = _mm256_add_pd(fix1,tx);
2322             fiy1             = _mm256_add_pd(fiy1,ty);
2323             fiz1             = _mm256_add_pd(fiz1,tz);
2324
2325             fjx3             = _mm256_add_pd(fjx3,tx);
2326             fjy3             = _mm256_add_pd(fjy3,ty);
2327             fjz3             = _mm256_add_pd(fjz3,tz);
2328
2329             /**************************
2330              * CALCULATE INTERACTIONS *
2331              **************************/
2332
2333             r21              = _mm256_mul_pd(rsq21,rinv21);
2334             r21              = _mm256_andnot_pd(dummy_mask,r21);
2335
2336             /* EWALD ELECTROSTATICS */
2337
2338             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2339             ewrt             = _mm256_mul_pd(r21,ewtabscale);
2340             ewitab           = _mm256_cvttpd_epi32(ewrt);
2341             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2342             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2343                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2344                                             &ewtabF,&ewtabFn);
2345             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2346             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
2347
2348             fscal            = felec;
2349
2350             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2351
2352             /* Calculate temporary vectorial force */
2353             tx               = _mm256_mul_pd(fscal,dx21);
2354             ty               = _mm256_mul_pd(fscal,dy21);
2355             tz               = _mm256_mul_pd(fscal,dz21);
2356
2357             /* Update vectorial force */
2358             fix2             = _mm256_add_pd(fix2,tx);
2359             fiy2             = _mm256_add_pd(fiy2,ty);
2360             fiz2             = _mm256_add_pd(fiz2,tz);
2361
2362             fjx1             = _mm256_add_pd(fjx1,tx);
2363             fjy1             = _mm256_add_pd(fjy1,ty);
2364             fjz1             = _mm256_add_pd(fjz1,tz);
2365
2366             /**************************
2367              * CALCULATE INTERACTIONS *
2368              **************************/
2369
2370             r22              = _mm256_mul_pd(rsq22,rinv22);
2371             r22              = _mm256_andnot_pd(dummy_mask,r22);
2372
2373             /* EWALD ELECTROSTATICS */
2374
2375             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2376             ewrt             = _mm256_mul_pd(r22,ewtabscale);
2377             ewitab           = _mm256_cvttpd_epi32(ewrt);
2378             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2379             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2380                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2381                                             &ewtabF,&ewtabFn);
2382             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2383             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
2384
2385             fscal            = felec;
2386
2387             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2388
2389             /* Calculate temporary vectorial force */
2390             tx               = _mm256_mul_pd(fscal,dx22);
2391             ty               = _mm256_mul_pd(fscal,dy22);
2392             tz               = _mm256_mul_pd(fscal,dz22);
2393
2394             /* Update vectorial force */
2395             fix2             = _mm256_add_pd(fix2,tx);
2396             fiy2             = _mm256_add_pd(fiy2,ty);
2397             fiz2             = _mm256_add_pd(fiz2,tz);
2398
2399             fjx2             = _mm256_add_pd(fjx2,tx);
2400             fjy2             = _mm256_add_pd(fjy2,ty);
2401             fjz2             = _mm256_add_pd(fjz2,tz);
2402
2403             /**************************
2404              * CALCULATE INTERACTIONS *
2405              **************************/
2406
2407             r23              = _mm256_mul_pd(rsq23,rinv23);
2408             r23              = _mm256_andnot_pd(dummy_mask,r23);
2409
2410             /* EWALD ELECTROSTATICS */
2411
2412             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2413             ewrt             = _mm256_mul_pd(r23,ewtabscale);
2414             ewitab           = _mm256_cvttpd_epi32(ewrt);
2415             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2416             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2417                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2418                                             &ewtabF,&ewtabFn);
2419             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2420             felec            = _mm256_mul_pd(_mm256_mul_pd(qq23,rinv23),_mm256_sub_pd(rinvsq23,felec));
2421
2422             fscal            = felec;
2423
2424             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2425
2426             /* Calculate temporary vectorial force */
2427             tx               = _mm256_mul_pd(fscal,dx23);
2428             ty               = _mm256_mul_pd(fscal,dy23);
2429             tz               = _mm256_mul_pd(fscal,dz23);
2430
2431             /* Update vectorial force */
2432             fix2             = _mm256_add_pd(fix2,tx);
2433             fiy2             = _mm256_add_pd(fiy2,ty);
2434             fiz2             = _mm256_add_pd(fiz2,tz);
2435
2436             fjx3             = _mm256_add_pd(fjx3,tx);
2437             fjy3             = _mm256_add_pd(fjy3,ty);
2438             fjz3             = _mm256_add_pd(fjz3,tz);
2439
2440             /**************************
2441              * CALCULATE INTERACTIONS *
2442              **************************/
2443
2444             r31              = _mm256_mul_pd(rsq31,rinv31);
2445             r31              = _mm256_andnot_pd(dummy_mask,r31);
2446
2447             /* EWALD ELECTROSTATICS */
2448
2449             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2450             ewrt             = _mm256_mul_pd(r31,ewtabscale);
2451             ewitab           = _mm256_cvttpd_epi32(ewrt);
2452             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2453             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2454                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2455                                             &ewtabF,&ewtabFn);
2456             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2457             felec            = _mm256_mul_pd(_mm256_mul_pd(qq31,rinv31),_mm256_sub_pd(rinvsq31,felec));
2458
2459             fscal            = felec;
2460
2461             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2462
2463             /* Calculate temporary vectorial force */
2464             tx               = _mm256_mul_pd(fscal,dx31);
2465             ty               = _mm256_mul_pd(fscal,dy31);
2466             tz               = _mm256_mul_pd(fscal,dz31);
2467
2468             /* Update vectorial force */
2469             fix3             = _mm256_add_pd(fix3,tx);
2470             fiy3             = _mm256_add_pd(fiy3,ty);
2471             fiz3             = _mm256_add_pd(fiz3,tz);
2472
2473             fjx1             = _mm256_add_pd(fjx1,tx);
2474             fjy1             = _mm256_add_pd(fjy1,ty);
2475             fjz1             = _mm256_add_pd(fjz1,tz);
2476
2477             /**************************
2478              * CALCULATE INTERACTIONS *
2479              **************************/
2480
2481             r32              = _mm256_mul_pd(rsq32,rinv32);
2482             r32              = _mm256_andnot_pd(dummy_mask,r32);
2483
2484             /* EWALD ELECTROSTATICS */
2485
2486             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2487             ewrt             = _mm256_mul_pd(r32,ewtabscale);
2488             ewitab           = _mm256_cvttpd_epi32(ewrt);
2489             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2490             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2491                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2492                                             &ewtabF,&ewtabFn);
2493             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2494             felec            = _mm256_mul_pd(_mm256_mul_pd(qq32,rinv32),_mm256_sub_pd(rinvsq32,felec));
2495
2496             fscal            = felec;
2497
2498             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2499
2500             /* Calculate temporary vectorial force */
2501             tx               = _mm256_mul_pd(fscal,dx32);
2502             ty               = _mm256_mul_pd(fscal,dy32);
2503             tz               = _mm256_mul_pd(fscal,dz32);
2504
2505             /* Update vectorial force */
2506             fix3             = _mm256_add_pd(fix3,tx);
2507             fiy3             = _mm256_add_pd(fiy3,ty);
2508             fiz3             = _mm256_add_pd(fiz3,tz);
2509
2510             fjx2             = _mm256_add_pd(fjx2,tx);
2511             fjy2             = _mm256_add_pd(fjy2,ty);
2512             fjz2             = _mm256_add_pd(fjz2,tz);
2513
2514             /**************************
2515              * CALCULATE INTERACTIONS *
2516              **************************/
2517
2518             r33              = _mm256_mul_pd(rsq33,rinv33);
2519             r33              = _mm256_andnot_pd(dummy_mask,r33);
2520
2521             /* EWALD ELECTROSTATICS */
2522
2523             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2524             ewrt             = _mm256_mul_pd(r33,ewtabscale);
2525             ewitab           = _mm256_cvttpd_epi32(ewrt);
2526             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2527             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2528                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2529                                             &ewtabF,&ewtabFn);
2530             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2531             felec            = _mm256_mul_pd(_mm256_mul_pd(qq33,rinv33),_mm256_sub_pd(rinvsq33,felec));
2532
2533             fscal            = felec;
2534
2535             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2536
2537             /* Calculate temporary vectorial force */
2538             tx               = _mm256_mul_pd(fscal,dx33);
2539             ty               = _mm256_mul_pd(fscal,dy33);
2540             tz               = _mm256_mul_pd(fscal,dz33);
2541
2542             /* Update vectorial force */
2543             fix3             = _mm256_add_pd(fix3,tx);
2544             fiy3             = _mm256_add_pd(fiy3,ty);
2545             fiz3             = _mm256_add_pd(fiz3,tz);
2546
2547             fjx3             = _mm256_add_pd(fjx3,tx);
2548             fjy3             = _mm256_add_pd(fjy3,ty);
2549             fjz3             = _mm256_add_pd(fjz3,tz);
2550
2551             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2552             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2553             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2554             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2555
2556             gmx_mm256_decrement_4rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2557                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2558                                                       fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2559
2560             /* Inner loop uses 385 flops */
2561         }
2562
2563         /* End of innermost loop */
2564
2565         gmx_mm256_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2566                                                  f+i_coord_offset,fshift+i_shift_offset);
2567
2568         /* Increment number of inner iterations */
2569         inneriter                  += j_index_end - j_index_start;
2570
2571         /* Outer loop uses 24 flops */
2572     }
2573
2574     /* Increment number of outer iterations */
2575     outeriter        += nri;
2576
2577     /* Update outer/inner flops */
2578
2579     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*385);
2580 }