src/gromacs/gmxlib/nonbonded/nb_kernel_avx_256_double/nb_kernel_ElecEw_VdwCSTab_GeomW3W3_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_GeomW3W3_VF_avx_256_double
  52  * Electrostatics interaction: Ewald
  53  * VdW interaction:            CubicSplineTable
  54  * Geometry:                   Water3-Water3
  55  * Calculate force/pot:        PotentialAndForce
  56  */
  57 void
  58 nb_kernel_ElecEw_VdwCSTab_GeomW3W3_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     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  91     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  92     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  93     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  94     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  95     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  96     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  97     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
  98     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
  99     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 100     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 101     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 102     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 103     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 104     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 105     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
 106     real             *charge;
 107     int              nvdwtype;
 108     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 109     int              *vdwtype;
 110     real             *vdwparam;
 111     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
 112     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
 113     __m128i          vfitab;
 114     __m128i          ifour       = _mm_set1_epi32(4);
 115     __m256d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 116     real             *vftab;
 117     __m128i          ewitab;
 118     __m256d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 119     __m256d          beta,beta2,beta3,zeta2,pmecorrF,pmecorrV,rinv3;
 120     real             *ewtab;
 121     __m256d          dummy_mask,cutoff_mask;
 122     __m128           tmpmask0,tmpmask1;
 123     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
 124     __m256d          one     = _mm256_set1_pd(1.0);
 125     __m256d          two     = _mm256_set1_pd(2.0);
 126     x                = xx[0];
 127     f                = ff[0];
 128
 129     nri              = nlist->nri;
 130     iinr             = nlist->iinr;
 131     jindex           = nlist->jindex;
 132     jjnr             = nlist->jjnr;
 133     shiftidx         = nlist->shift;
 134     gid              = nlist->gid;
 135     shiftvec         = fr->shift_vec[0];
 136     fshift           = fr->fshift[0];
 137     facel            = _mm256_set1_pd(fr->epsfac);
 138     charge           = mdatoms->chargeA;
 139     nvdwtype         = fr->ntype;
 140     vdwparam         = fr->nbfp;
 141     vdwtype          = mdatoms->typeA;
 142
 143     vftab            = kernel_data->table_vdw->data;
 144     vftabscale       = _mm256_set1_pd(kernel_data->table_vdw->scale);
 145
 146     sh_ewald         = _mm256_set1_pd(fr->ic->sh_ewald);
 147     beta             = _mm256_set1_pd(fr->ic->ewaldcoeff_q);
 148     beta2            = _mm256_mul_pd(beta,beta);
 149     beta3            = _mm256_mul_pd(beta,beta2);
 150
 151     ewtab            = fr->ic->tabq_coul_FDV0;
 152     ewtabscale       = _mm256_set1_pd(fr->ic->tabq_scale);
 153     ewtabhalfspace   = _mm256_set1_pd(0.5/fr->ic->tabq_scale);
 154
 155     /* Setup water-specific parameters */
 156     inr              = nlist->iinr[0];
 157     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
 158     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
 159     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
 160     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 161
 162     jq0              = _mm256_set1_pd(charge[inr+0]);
 163     jq1              = _mm256_set1_pd(charge[inr+1]);
 164     jq2              = _mm256_set1_pd(charge[inr+2]);
 165     vdwjidx0A        = 2*vdwtype[inr+0];
 166     qq00             = _mm256_mul_pd(iq0,jq0);
 167     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
 168     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
 169     qq01             = _mm256_mul_pd(iq0,jq1);
 170     qq02             = _mm256_mul_pd(iq0,jq2);
 171     qq10             = _mm256_mul_pd(iq1,jq0);
 172     qq11             = _mm256_mul_pd(iq1,jq1);
 173     qq12             = _mm256_mul_pd(iq1,jq2);
 174     qq20             = _mm256_mul_pd(iq2,jq0);
 175     qq21             = _mm256_mul_pd(iq2,jq1);
 176     qq22             = _mm256_mul_pd(iq2,jq2);
 177
 178     /* Avoid stupid compiler warnings */
 179     jnrA = jnrB = jnrC = jnrD = 0;
 180     j_coord_offsetA = 0;
 181     j_coord_offsetB = 0;
 182     j_coord_offsetC = 0;
 183     j_coord_offsetD = 0;
 184
 185     outeriter        = 0;
 186     inneriter        = 0;
 187
 188     for(iidx=0;iidx<4*DIM;iidx++)
 189     {
 190         scratch[iidx] = 0.0;
 191     }
 192
 193     /* Start outer loop over neighborlists */
 194     for(iidx=0; iidx<nri; iidx++)
 195     {
 196         /* Load shift vector for this list */
 197         i_shift_offset   = DIM*shiftidx[iidx];
 198
 199         /* Load limits for loop over neighbors */
 200         j_index_start    = jindex[iidx];
 201         j_index_end      = jindex[iidx+1];
 202
 203         /* Get outer coordinate index */
 204         inr              = iinr[iidx];
 205         i_coord_offset   = DIM*inr;
 206
 207         /* Load i particle coords and add shift vector */
 208         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 209                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 210
 211         fix0             = _mm256_setzero_pd();
 212         fiy0             = _mm256_setzero_pd();
 213         fiz0             = _mm256_setzero_pd();
 214         fix1             = _mm256_setzero_pd();
 215         fiy1             = _mm256_setzero_pd();
 216         fiz1             = _mm256_setzero_pd();
 217         fix2             = _mm256_setzero_pd();
 218         fiy2             = _mm256_setzero_pd();
 219         fiz2             = _mm256_setzero_pd();
 220
 221         /* Reset potential sums */
 222         velecsum         = _mm256_setzero_pd();
 223         vvdwsum          = _mm256_setzero_pd();
 224
 225         /* Start inner kernel loop */
 226         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 227         {
 228
 229             /* Get j neighbor index, and coordinate index */
 230             jnrA             = jjnr[jidx];
 231             jnrB             = jjnr[jidx+1];
 232             jnrC             = jjnr[jidx+2];
 233             jnrD             = jjnr[jidx+3];
 234             j_coord_offsetA  = DIM*jnrA;
 235             j_coord_offsetB  = DIM*jnrB;
 236             j_coord_offsetC  = DIM*jnrC;
 237             j_coord_offsetD  = DIM*jnrD;
 238
 239             /* load j atom coordinates */
 240             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 241                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 242                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 243
 244             /* Calculate displacement vector */
 245             dx00             = _mm256_sub_pd(ix0,jx0);
 246             dy00             = _mm256_sub_pd(iy0,jy0);
 247             dz00             = _mm256_sub_pd(iz0,jz0);
 248             dx01             = _mm256_sub_pd(ix0,jx1);
 249             dy01             = _mm256_sub_pd(iy0,jy1);
 250             dz01             = _mm256_sub_pd(iz0,jz1);
 251             dx02             = _mm256_sub_pd(ix0,jx2);
 252             dy02             = _mm256_sub_pd(iy0,jy2);
 253             dz02             = _mm256_sub_pd(iz0,jz2);
 254             dx10             = _mm256_sub_pd(ix1,jx0);
 255             dy10             = _mm256_sub_pd(iy1,jy0);
 256             dz10             = _mm256_sub_pd(iz1,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             dx20             = _mm256_sub_pd(ix2,jx0);
 264             dy20             = _mm256_sub_pd(iy2,jy0);
 265             dz20             = _mm256_sub_pd(iz2,jz0);
 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
 273             /* Calculate squared distance and things based on it */
 274             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 275             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 276             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 277             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 278             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 279             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 280             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 281             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 282             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 283
 284             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 285             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
 286             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
 287             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
 288             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 289             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 290             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
 291             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 292             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 293
 294             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 295             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 296             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 297             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 298             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 299             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 300             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 301             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 302             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 303
 304             fjx0             = _mm256_setzero_pd();
 305             fjy0             = _mm256_setzero_pd();
 306             fjz0             = _mm256_setzero_pd();
 307             fjx1             = _mm256_setzero_pd();
 308             fjy1             = _mm256_setzero_pd();
 309             fjz1             = _mm256_setzero_pd();
 310             fjx2             = _mm256_setzero_pd();
 311             fjy2             = _mm256_setzero_pd();
 312             fjz2             = _mm256_setzero_pd();
 313
 314             /**************************
 315              * CALCULATE INTERACTIONS *
 316              **************************/
 317
 318             r00              = _mm256_mul_pd(rsq00,rinv00);
 319
 320             /* Calculate table index by multiplying r with table scale and truncate to integer */
 321             rt               = _mm256_mul_pd(r00,vftabscale);
 322             vfitab           = _mm256_cvttpd_epi32(rt);
 323             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
 324             vfitab           = _mm_slli_epi32(vfitab,3);
 325
 326             /* EWALD ELECTROSTATICS */
 327
 328             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 329             ewrt             = _mm256_mul_pd(r00,ewtabscale);
 330             ewitab           = _mm256_cvttpd_epi32(ewrt);
 331             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 332             ewitab           = _mm_slli_epi32(ewitab,2);
 333             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 334             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 335             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 336             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 337             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 338             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 339             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 340             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(rinv00,velec));
 341             felec            = _mm256_mul_pd(_mm256_mul_pd(qq00,rinv00),_mm256_sub_pd(rinvsq00,felec));
 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             velecsum         = _mm256_add_pd(velecsum,velec);
 374             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 375
 376             fscal            = _mm256_add_pd(felec,fvdw);
 377
 378             /* Calculate temporary vectorial force */
 379             tx               = _mm256_mul_pd(fscal,dx00);
 380             ty               = _mm256_mul_pd(fscal,dy00);
 381             tz               = _mm256_mul_pd(fscal,dz00);
 382
 383             /* Update vectorial force */
 384             fix0             = _mm256_add_pd(fix0,tx);
 385             fiy0             = _mm256_add_pd(fiy0,ty);
 386             fiz0             = _mm256_add_pd(fiz0,tz);
 387
 388             fjx0             = _mm256_add_pd(fjx0,tx);
 389             fjy0             = _mm256_add_pd(fjy0,ty);
 390             fjz0             = _mm256_add_pd(fjz0,tz);
 391
 392             /**************************
 393              * CALCULATE INTERACTIONS *
 394              **************************/
 395
 396             r01              = _mm256_mul_pd(rsq01,rinv01);
 397
 398             /* EWALD ELECTROSTATICS */
 399
 400             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 401             ewrt             = _mm256_mul_pd(r01,ewtabscale);
 402             ewitab           = _mm256_cvttpd_epi32(ewrt);
 403             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 404             ewitab           = _mm_slli_epi32(ewitab,2);
 405             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 406             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 407             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 408             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 409             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 410             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 411             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 412             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(rinv01,velec));
 413             felec            = _mm256_mul_pd(_mm256_mul_pd(qq01,rinv01),_mm256_sub_pd(rinvsq01,felec));
 414
 415             /* Update potential sum for this i atom from the interaction with this j atom. */
 416             velecsum         = _mm256_add_pd(velecsum,velec);
 417
 418             fscal            = felec;
 419
 420             /* Calculate temporary vectorial force */
 421             tx               = _mm256_mul_pd(fscal,dx01);
 422             ty               = _mm256_mul_pd(fscal,dy01);
 423             tz               = _mm256_mul_pd(fscal,dz01);
 424
 425             /* Update vectorial force */
 426             fix0             = _mm256_add_pd(fix0,tx);
 427             fiy0             = _mm256_add_pd(fiy0,ty);
 428             fiz0             = _mm256_add_pd(fiz0,tz);
 429
 430             fjx1             = _mm256_add_pd(fjx1,tx);
 431             fjy1             = _mm256_add_pd(fjy1,ty);
 432             fjz1             = _mm256_add_pd(fjz1,tz);
 433
 434             /**************************
 435              * CALCULATE INTERACTIONS *
 436              **************************/
 437
 438             r02              = _mm256_mul_pd(rsq02,rinv02);
 439
 440             /* EWALD ELECTROSTATICS */
 441
 442             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 443             ewrt             = _mm256_mul_pd(r02,ewtabscale);
 444             ewitab           = _mm256_cvttpd_epi32(ewrt);
 445             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 446             ewitab           = _mm_slli_epi32(ewitab,2);
 447             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 448             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 449             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 450             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 451             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 452             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 453             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 454             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(rinv02,velec));
 455             felec            = _mm256_mul_pd(_mm256_mul_pd(qq02,rinv02),_mm256_sub_pd(rinvsq02,felec));
 456
 457             /* Update potential sum for this i atom from the interaction with this j atom. */
 458             velecsum         = _mm256_add_pd(velecsum,velec);
 459
 460             fscal            = felec;
 461
 462             /* Calculate temporary vectorial force */
 463             tx               = _mm256_mul_pd(fscal,dx02);
 464             ty               = _mm256_mul_pd(fscal,dy02);
 465             tz               = _mm256_mul_pd(fscal,dz02);
 466
 467             /* Update vectorial force */
 468             fix0             = _mm256_add_pd(fix0,tx);
 469             fiy0             = _mm256_add_pd(fiy0,ty);
 470             fiz0             = _mm256_add_pd(fiz0,tz);
 471
 472             fjx2             = _mm256_add_pd(fjx2,tx);
 473             fjy2             = _mm256_add_pd(fjy2,ty);
 474             fjz2             = _mm256_add_pd(fjz2,tz);
 475
 476             /**************************
 477              * CALCULATE INTERACTIONS *
 478              **************************/
 479
 480             r10              = _mm256_mul_pd(rsq10,rinv10);
 481
 482             /* EWALD ELECTROSTATICS */
 483
 484             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 485             ewrt             = _mm256_mul_pd(r10,ewtabscale);
 486             ewitab           = _mm256_cvttpd_epi32(ewrt);
 487             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 488             ewitab           = _mm_slli_epi32(ewitab,2);
 489             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 490             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 491             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 492             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 493             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 494             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 495             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 496             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(rinv10,velec));
 497             felec            = _mm256_mul_pd(_mm256_mul_pd(qq10,rinv10),_mm256_sub_pd(rinvsq10,felec));
 498
 499             /* Update potential sum for this i atom from the interaction with this j atom. */
 500             velecsum         = _mm256_add_pd(velecsum,velec);
 501
 502             fscal            = felec;
 503
 504             /* Calculate temporary vectorial force */
 505             tx               = _mm256_mul_pd(fscal,dx10);
 506             ty               = _mm256_mul_pd(fscal,dy10);
 507             tz               = _mm256_mul_pd(fscal,dz10);
 508
 509             /* Update vectorial force */
 510             fix1             = _mm256_add_pd(fix1,tx);
 511             fiy1             = _mm256_add_pd(fiy1,ty);
 512             fiz1             = _mm256_add_pd(fiz1,tz);
 513
 514             fjx0             = _mm256_add_pd(fjx0,tx);
 515             fjy0             = _mm256_add_pd(fjy0,ty);
 516             fjz0             = _mm256_add_pd(fjz0,tz);
 517
 518             /**************************
 519              * CALCULATE INTERACTIONS *
 520              **************************/
 521
 522             r11              = _mm256_mul_pd(rsq11,rinv11);
 523
 524             /* EWALD ELECTROSTATICS */
 525
 526             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 527             ewrt             = _mm256_mul_pd(r11,ewtabscale);
 528             ewitab           = _mm256_cvttpd_epi32(ewrt);
 529             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 530             ewitab           = _mm_slli_epi32(ewitab,2);
 531             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 532             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 533             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 534             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 535             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 536             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 537             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 538             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(rinv11,velec));
 539             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
 540
 541             /* Update potential sum for this i atom from the interaction with this j atom. */
 542             velecsum         = _mm256_add_pd(velecsum,velec);
 543
 544             fscal            = felec;
 545
 546             /* Calculate temporary vectorial force */
 547             tx               = _mm256_mul_pd(fscal,dx11);
 548             ty               = _mm256_mul_pd(fscal,dy11);
 549             tz               = _mm256_mul_pd(fscal,dz11);
 550
 551             /* Update vectorial force */
 552             fix1             = _mm256_add_pd(fix1,tx);
 553             fiy1             = _mm256_add_pd(fiy1,ty);
 554             fiz1             = _mm256_add_pd(fiz1,tz);
 555
 556             fjx1             = _mm256_add_pd(fjx1,tx);
 557             fjy1             = _mm256_add_pd(fjy1,ty);
 558             fjz1             = _mm256_add_pd(fjz1,tz);
 559
 560             /**************************
 561              * CALCULATE INTERACTIONS *
 562              **************************/
 563
 564             r12              = _mm256_mul_pd(rsq12,rinv12);
 565
 566             /* EWALD ELECTROSTATICS */
 567
 568             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 569             ewrt             = _mm256_mul_pd(r12,ewtabscale);
 570             ewitab           = _mm256_cvttpd_epi32(ewrt);
 571             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 572             ewitab           = _mm_slli_epi32(ewitab,2);
 573             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 574             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 575             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 576             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 577             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 578             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 579             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 580             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(rinv12,velec));
 581             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
 582
 583             /* Update potential sum for this i atom from the interaction with this j atom. */
 584             velecsum         = _mm256_add_pd(velecsum,velec);
 585
 586             fscal            = felec;
 587
 588             /* Calculate temporary vectorial force */
 589             tx               = _mm256_mul_pd(fscal,dx12);
 590             ty               = _mm256_mul_pd(fscal,dy12);
 591             tz               = _mm256_mul_pd(fscal,dz12);
 592
 593             /* Update vectorial force */
 594             fix1             = _mm256_add_pd(fix1,tx);
 595             fiy1             = _mm256_add_pd(fiy1,ty);
 596             fiz1             = _mm256_add_pd(fiz1,tz);
 597
 598             fjx2             = _mm256_add_pd(fjx2,tx);
 599             fjy2             = _mm256_add_pd(fjy2,ty);
 600             fjz2             = _mm256_add_pd(fjz2,tz);
 601
 602             /**************************
 603              * CALCULATE INTERACTIONS *
 604              **************************/
 605
 606             r20              = _mm256_mul_pd(rsq20,rinv20);
 607
 608             /* EWALD ELECTROSTATICS */
 609
 610             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 611             ewrt             = _mm256_mul_pd(r20,ewtabscale);
 612             ewitab           = _mm256_cvttpd_epi32(ewrt);
 613             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 614             ewitab           = _mm_slli_epi32(ewitab,2);
 615             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 616             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 617             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 618             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 619             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 620             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 621             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 622             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(rinv20,velec));
 623             felec            = _mm256_mul_pd(_mm256_mul_pd(qq20,rinv20),_mm256_sub_pd(rinvsq20,felec));
 624
 625             /* Update potential sum for this i atom from the interaction with this j atom. */
 626             velecsum         = _mm256_add_pd(velecsum,velec);
 627
 628             fscal            = felec;
 629
 630             /* Calculate temporary vectorial force */
 631             tx               = _mm256_mul_pd(fscal,dx20);
 632             ty               = _mm256_mul_pd(fscal,dy20);
 633             tz               = _mm256_mul_pd(fscal,dz20);
 634
 635             /* Update vectorial force */
 636             fix2             = _mm256_add_pd(fix2,tx);
 637             fiy2             = _mm256_add_pd(fiy2,ty);
 638             fiz2             = _mm256_add_pd(fiz2,tz);
 639
 640             fjx0             = _mm256_add_pd(fjx0,tx);
 641             fjy0             = _mm256_add_pd(fjy0,ty);
 642             fjz0             = _mm256_add_pd(fjz0,tz);
 643
 644             /**************************
 645              * CALCULATE INTERACTIONS *
 646              **************************/
 647
 648             r21              = _mm256_mul_pd(rsq21,rinv21);
 649
 650             /* EWALD ELECTROSTATICS */
 651
 652             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 653             ewrt             = _mm256_mul_pd(r21,ewtabscale);
 654             ewitab           = _mm256_cvttpd_epi32(ewrt);
 655             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 656             ewitab           = _mm_slli_epi32(ewitab,2);
 657             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 658             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 659             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 660             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 661             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 662             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 663             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 664             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(rinv21,velec));
 665             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
 666
 667             /* Update potential sum for this i atom from the interaction with this j atom. */
 668             velecsum         = _mm256_add_pd(velecsum,velec);
 669
 670             fscal            = felec;
 671
 672             /* Calculate temporary vectorial force */
 673             tx               = _mm256_mul_pd(fscal,dx21);
 674             ty               = _mm256_mul_pd(fscal,dy21);
 675             tz               = _mm256_mul_pd(fscal,dz21);
 676
 677             /* Update vectorial force */
 678             fix2             = _mm256_add_pd(fix2,tx);
 679             fiy2             = _mm256_add_pd(fiy2,ty);
 680             fiz2             = _mm256_add_pd(fiz2,tz);
 681
 682             fjx1             = _mm256_add_pd(fjx1,tx);
 683             fjy1             = _mm256_add_pd(fjy1,ty);
 684             fjz1             = _mm256_add_pd(fjz1,tz);
 685
 686             /**************************
 687              * CALCULATE INTERACTIONS *
 688              **************************/
 689
 690             r22              = _mm256_mul_pd(rsq22,rinv22);
 691
 692             /* EWALD ELECTROSTATICS */
 693
 694             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 695             ewrt             = _mm256_mul_pd(r22,ewtabscale);
 696             ewitab           = _mm256_cvttpd_epi32(ewrt);
 697             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 698             ewitab           = _mm_slli_epi32(ewitab,2);
 699             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 700             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 701             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 702             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 703             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 704             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 705             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 706             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(rinv22,velec));
 707             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
 708
 709             /* Update potential sum for this i atom from the interaction with this j atom. */
 710             velecsum         = _mm256_add_pd(velecsum,velec);
 711
 712             fscal            = felec;
 713
 714             /* Calculate temporary vectorial force */
 715             tx               = _mm256_mul_pd(fscal,dx22);
 716             ty               = _mm256_mul_pd(fscal,dy22);
 717             tz               = _mm256_mul_pd(fscal,dz22);
 718
 719             /* Update vectorial force */
 720             fix2             = _mm256_add_pd(fix2,tx);
 721             fiy2             = _mm256_add_pd(fiy2,ty);
 722             fiz2             = _mm256_add_pd(fiz2,tz);
 723
 724             fjx2             = _mm256_add_pd(fjx2,tx);
 725             fjy2             = _mm256_add_pd(fjy2,ty);
 726             fjz2             = _mm256_add_pd(fjz2,tz);
 727
 728             fjptrA             = f+j_coord_offsetA;
 729             fjptrB             = f+j_coord_offsetB;
 730             fjptrC             = f+j_coord_offsetC;
 731             fjptrD             = f+j_coord_offsetD;
 732
 733             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
 734                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
 735
 736             /* Inner loop uses 403 flops */
 737         }
 738
 739         if(jidx<j_index_end)
 740         {
 741
 742             /* Get j neighbor index, and coordinate index */
 743             jnrlistA         = jjnr[jidx];
 744             jnrlistB         = jjnr[jidx+1];
 745             jnrlistC         = jjnr[jidx+2];
 746             jnrlistD         = jjnr[jidx+3];
 747             /* Sign of each element will be negative for non-real atoms.
 748              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 749              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
 750              */
 751             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 752
 753             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
 754             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
 755             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
 756
 757             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 758             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 759             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 760             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 761             j_coord_offsetA  = DIM*jnrA;
 762             j_coord_offsetB  = DIM*jnrB;
 763             j_coord_offsetC  = DIM*jnrC;
 764             j_coord_offsetD  = DIM*jnrD;
 765
 766             /* load j atom coordinates */
 767             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 768                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 769                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 770
 771             /* Calculate displacement vector */
 772             dx00             = _mm256_sub_pd(ix0,jx0);
 773             dy00             = _mm256_sub_pd(iy0,jy0);
 774             dz00             = _mm256_sub_pd(iz0,jz0);
 775             dx01             = _mm256_sub_pd(ix0,jx1);
 776             dy01             = _mm256_sub_pd(iy0,jy1);
 777             dz01             = _mm256_sub_pd(iz0,jz1);
 778             dx02             = _mm256_sub_pd(ix0,jx2);
 779             dy02             = _mm256_sub_pd(iy0,jy2);
 780             dz02             = _mm256_sub_pd(iz0,jz2);
 781             dx10             = _mm256_sub_pd(ix1,jx0);
 782             dy10             = _mm256_sub_pd(iy1,jy0);
 783             dz10             = _mm256_sub_pd(iz1,jz0);
 784             dx11             = _mm256_sub_pd(ix1,jx1);
 785             dy11             = _mm256_sub_pd(iy1,jy1);
 786             dz11             = _mm256_sub_pd(iz1,jz1);
 787             dx12             = _mm256_sub_pd(ix1,jx2);
 788             dy12             = _mm256_sub_pd(iy1,jy2);
 789             dz12             = _mm256_sub_pd(iz1,jz2);
 790             dx20             = _mm256_sub_pd(ix2,jx0);
 791             dy20             = _mm256_sub_pd(iy2,jy0);
 792             dz20             = _mm256_sub_pd(iz2,jz0);
 793             dx21             = _mm256_sub_pd(ix2,jx1);
 794             dy21             = _mm256_sub_pd(iy2,jy1);
 795             dz21             = _mm256_sub_pd(iz2,jz1);
 796             dx22             = _mm256_sub_pd(ix2,jx2);
 797             dy22             = _mm256_sub_pd(iy2,jy2);
 798             dz22             = _mm256_sub_pd(iz2,jz2);
 799
 800             /* Calculate squared distance and things based on it */
 801             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 802             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 803             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 804             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 805             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 806             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 807             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 808             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 809             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 810
 811             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 812             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
 813             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
 814             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
 815             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 816             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 817             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
 818             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 819             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 820
 821             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 822             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 823             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 824             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 825             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 826             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 827             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 828             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 829             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 830
 831             fjx0             = _mm256_setzero_pd();
 832             fjy0             = _mm256_setzero_pd();
 833             fjz0             = _mm256_setzero_pd();
 834             fjx1             = _mm256_setzero_pd();
 835             fjy1             = _mm256_setzero_pd();
 836             fjz1             = _mm256_setzero_pd();
 837             fjx2             = _mm256_setzero_pd();
 838             fjy2             = _mm256_setzero_pd();
 839             fjz2             = _mm256_setzero_pd();
 840
 841             /**************************
 842              * CALCULATE INTERACTIONS *
 843              **************************/
 844
 845             r00              = _mm256_mul_pd(rsq00,rinv00);
 846             r00              = _mm256_andnot_pd(dummy_mask,r00);
 847
 848             /* Calculate table index by multiplying r with table scale and truncate to integer */
 849             rt               = _mm256_mul_pd(r00,vftabscale);
 850             vfitab           = _mm256_cvttpd_epi32(rt);
 851             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
 852             vfitab           = _mm_slli_epi32(vfitab,3);
 853
 854             /* EWALD ELECTROSTATICS */
 855
 856             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 857             ewrt             = _mm256_mul_pd(r00,ewtabscale);
 858             ewitab           = _mm256_cvttpd_epi32(ewrt);
 859             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 860             ewitab           = _mm_slli_epi32(ewitab,2);
 861             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 862             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 863             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 864             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 865             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 866             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 867             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 868             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(rinv00,velec));
 869             felec            = _mm256_mul_pd(_mm256_mul_pd(qq00,rinv00),_mm256_sub_pd(rinvsq00,felec));
 870
 871             /* CUBIC SPLINE TABLE DISPERSION */
 872             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 873             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 874             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 875             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 876             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 877             Heps             = _mm256_mul_pd(vfeps,H);
 878             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 879             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 880             vvdw6            = _mm256_mul_pd(c6_00,VV);
 881             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 882             fvdw6            = _mm256_mul_pd(c6_00,FF);
 883
 884             /* CUBIC SPLINE TABLE REPULSION */
 885             vfitab           = _mm_add_epi32(vfitab,ifour);
 886             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 887             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 888             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
 889             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
 890             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
 891             Heps             = _mm256_mul_pd(vfeps,H);
 892             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
 893             VV               = _mm256_add_pd(Y,_mm256_mul_pd(vfeps,Fp));
 894             vvdw12           = _mm256_mul_pd(c12_00,VV);
 895             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
 896             fvdw12           = _mm256_mul_pd(c12_00,FF);
 897             vvdw             = _mm256_add_pd(vvdw12,vvdw6);
 898             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
 899
 900             /* Update potential sum for this i atom from the interaction with this j atom. */
 901             velec            = _mm256_andnot_pd(dummy_mask,velec);
 902             velecsum         = _mm256_add_pd(velecsum,velec);
 903             vvdw             = _mm256_andnot_pd(dummy_mask,vvdw);
 904             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 905
 906             fscal            = _mm256_add_pd(felec,fvdw);
 907
 908             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 909
 910             /* Calculate temporary vectorial force */
 911             tx               = _mm256_mul_pd(fscal,dx00);
 912             ty               = _mm256_mul_pd(fscal,dy00);
 913             tz               = _mm256_mul_pd(fscal,dz00);
 914
 915             /* Update vectorial force */
 916             fix0             = _mm256_add_pd(fix0,tx);
 917             fiy0             = _mm256_add_pd(fiy0,ty);
 918             fiz0             = _mm256_add_pd(fiz0,tz);
 919
 920             fjx0             = _mm256_add_pd(fjx0,tx);
 921             fjy0             = _mm256_add_pd(fjy0,ty);
 922             fjz0             = _mm256_add_pd(fjz0,tz);
 923
 924             /**************************
 925              * CALCULATE INTERACTIONS *
 926              **************************/
 927
 928             r01              = _mm256_mul_pd(rsq01,rinv01);
 929             r01              = _mm256_andnot_pd(dummy_mask,r01);
 930
 931             /* EWALD ELECTROSTATICS */
 932
 933             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 934             ewrt             = _mm256_mul_pd(r01,ewtabscale);
 935             ewitab           = _mm256_cvttpd_epi32(ewrt);
 936             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 937             ewitab           = _mm_slli_epi32(ewitab,2);
 938             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 939             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 940             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 941             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 942             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 943             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 944             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 945             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(rinv01,velec));
 946             felec            = _mm256_mul_pd(_mm256_mul_pd(qq01,rinv01),_mm256_sub_pd(rinvsq01,felec));
 947
 948             /* Update potential sum for this i atom from the interaction with this j atom. */
 949             velec            = _mm256_andnot_pd(dummy_mask,velec);
 950             velecsum         = _mm256_add_pd(velecsum,velec);
 951
 952             fscal            = felec;
 953
 954             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 955
 956             /* Calculate temporary vectorial force */
 957             tx               = _mm256_mul_pd(fscal,dx01);
 958             ty               = _mm256_mul_pd(fscal,dy01);
 959             tz               = _mm256_mul_pd(fscal,dz01);
 960
 961             /* Update vectorial force */
 962             fix0             = _mm256_add_pd(fix0,tx);
 963             fiy0             = _mm256_add_pd(fiy0,ty);
 964             fiz0             = _mm256_add_pd(fiz0,tz);
 965
 966             fjx1             = _mm256_add_pd(fjx1,tx);
 967             fjy1             = _mm256_add_pd(fjy1,ty);
 968             fjz1             = _mm256_add_pd(fjz1,tz);
 969
 970             /**************************
 971              * CALCULATE INTERACTIONS *
 972              **************************/
 973
 974             r02              = _mm256_mul_pd(rsq02,rinv02);
 975             r02              = _mm256_andnot_pd(dummy_mask,r02);
 976
 977             /* EWALD ELECTROSTATICS */
 978
 979             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 980             ewrt             = _mm256_mul_pd(r02,ewtabscale);
 981             ewitab           = _mm256_cvttpd_epi32(ewrt);
 982             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
 983             ewitab           = _mm_slli_epi32(ewitab,2);
 984             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 985             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 986             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
 987             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
 988             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
 989             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
 990             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
 991             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(rinv02,velec));
 992             felec            = _mm256_mul_pd(_mm256_mul_pd(qq02,rinv02),_mm256_sub_pd(rinvsq02,felec));
 993
 994             /* Update potential sum for this i atom from the interaction with this j atom. */
 995             velec            = _mm256_andnot_pd(dummy_mask,velec);
 996             velecsum         = _mm256_add_pd(velecsum,velec);
 997
 998             fscal            = felec;
 999
1000             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1001
1002             /* Calculate temporary vectorial force */
1003             tx               = _mm256_mul_pd(fscal,dx02);
1004             ty               = _mm256_mul_pd(fscal,dy02);
1005             tz               = _mm256_mul_pd(fscal,dz02);
1006
1007             /* Update vectorial force */
1008             fix0             = _mm256_add_pd(fix0,tx);
1009             fiy0             = _mm256_add_pd(fiy0,ty);
1010             fiz0             = _mm256_add_pd(fiz0,tz);
1011
1012             fjx2             = _mm256_add_pd(fjx2,tx);
1013             fjy2             = _mm256_add_pd(fjy2,ty);
1014             fjz2             = _mm256_add_pd(fjz2,tz);
1015
1016             /**************************
1017              * CALCULATE INTERACTIONS *
1018              **************************/
1019
1020             r10              = _mm256_mul_pd(rsq10,rinv10);
1021             r10              = _mm256_andnot_pd(dummy_mask,r10);
1022
1023             /* EWALD ELECTROSTATICS */
1024
1025             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1026             ewrt             = _mm256_mul_pd(r10,ewtabscale);
1027             ewitab           = _mm256_cvttpd_epi32(ewrt);
1028             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1029             ewitab           = _mm_slli_epi32(ewitab,2);
1030             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1031             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1032             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1033             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1034             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1035             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1036             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1037             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(rinv10,velec));
1038             felec            = _mm256_mul_pd(_mm256_mul_pd(qq10,rinv10),_mm256_sub_pd(rinvsq10,felec));
1039
1040             /* Update potential sum for this i atom from the interaction with this j atom. */
1041             velec            = _mm256_andnot_pd(dummy_mask,velec);
1042             velecsum         = _mm256_add_pd(velecsum,velec);
1043
1044             fscal            = felec;
1045
1046             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1047
1048             /* Calculate temporary vectorial force */
1049             tx               = _mm256_mul_pd(fscal,dx10);
1050             ty               = _mm256_mul_pd(fscal,dy10);
1051             tz               = _mm256_mul_pd(fscal,dz10);
1052
1053             /* Update vectorial force */
1054             fix1             = _mm256_add_pd(fix1,tx);
1055             fiy1             = _mm256_add_pd(fiy1,ty);
1056             fiz1             = _mm256_add_pd(fiz1,tz);
1057
1058             fjx0             = _mm256_add_pd(fjx0,tx);
1059             fjy0             = _mm256_add_pd(fjy0,ty);
1060             fjz0             = _mm256_add_pd(fjz0,tz);
1061
1062             /**************************
1063              * CALCULATE INTERACTIONS *
1064              **************************/
1065
1066             r11              = _mm256_mul_pd(rsq11,rinv11);
1067             r11              = _mm256_andnot_pd(dummy_mask,r11);
1068
1069             /* EWALD ELECTROSTATICS */
1070
1071             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1072             ewrt             = _mm256_mul_pd(r11,ewtabscale);
1073             ewitab           = _mm256_cvttpd_epi32(ewrt);
1074             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1075             ewitab           = _mm_slli_epi32(ewitab,2);
1076             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1077             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1078             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1079             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1080             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1081             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1082             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1083             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(rinv11,velec));
1084             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
1085
1086             /* Update potential sum for this i atom from the interaction with this j atom. */
1087             velec            = _mm256_andnot_pd(dummy_mask,velec);
1088             velecsum         = _mm256_add_pd(velecsum,velec);
1089
1090             fscal            = felec;
1091
1092             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1093
1094             /* Calculate temporary vectorial force */
1095             tx               = _mm256_mul_pd(fscal,dx11);
1096             ty               = _mm256_mul_pd(fscal,dy11);
1097             tz               = _mm256_mul_pd(fscal,dz11);
1098
1099             /* Update vectorial force */
1100             fix1             = _mm256_add_pd(fix1,tx);
1101             fiy1             = _mm256_add_pd(fiy1,ty);
1102             fiz1             = _mm256_add_pd(fiz1,tz);
1103
1104             fjx1             = _mm256_add_pd(fjx1,tx);
1105             fjy1             = _mm256_add_pd(fjy1,ty);
1106             fjz1             = _mm256_add_pd(fjz1,tz);
1107
1108             /**************************
1109              * CALCULATE INTERACTIONS *
1110              **************************/
1111
1112             r12              = _mm256_mul_pd(rsq12,rinv12);
1113             r12              = _mm256_andnot_pd(dummy_mask,r12);
1114
1115             /* EWALD ELECTROSTATICS */
1116
1117             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1118             ewrt             = _mm256_mul_pd(r12,ewtabscale);
1119             ewitab           = _mm256_cvttpd_epi32(ewrt);
1120             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1121             ewitab           = _mm_slli_epi32(ewitab,2);
1122             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1123             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1124             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1125             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1126             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1127             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1128             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1129             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(rinv12,velec));
1130             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
1131
1132             /* Update potential sum for this i atom from the interaction with this j atom. */
1133             velec            = _mm256_andnot_pd(dummy_mask,velec);
1134             velecsum         = _mm256_add_pd(velecsum,velec);
1135
1136             fscal            = felec;
1137
1138             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1139
1140             /* Calculate temporary vectorial force */
1141             tx               = _mm256_mul_pd(fscal,dx12);
1142             ty               = _mm256_mul_pd(fscal,dy12);
1143             tz               = _mm256_mul_pd(fscal,dz12);
1144
1145             /* Update vectorial force */
1146             fix1             = _mm256_add_pd(fix1,tx);
1147             fiy1             = _mm256_add_pd(fiy1,ty);
1148             fiz1             = _mm256_add_pd(fiz1,tz);
1149
1150             fjx2             = _mm256_add_pd(fjx2,tx);
1151             fjy2             = _mm256_add_pd(fjy2,ty);
1152             fjz2             = _mm256_add_pd(fjz2,tz);
1153
1154             /**************************
1155              * CALCULATE INTERACTIONS *
1156              **************************/
1157
1158             r20              = _mm256_mul_pd(rsq20,rinv20);
1159             r20              = _mm256_andnot_pd(dummy_mask,r20);
1160
1161             /* EWALD ELECTROSTATICS */
1162
1163             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1164             ewrt             = _mm256_mul_pd(r20,ewtabscale);
1165             ewitab           = _mm256_cvttpd_epi32(ewrt);
1166             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1167             ewitab           = _mm_slli_epi32(ewitab,2);
1168             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1169             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1170             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1171             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1172             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1173             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1174             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1175             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(rinv20,velec));
1176             felec            = _mm256_mul_pd(_mm256_mul_pd(qq20,rinv20),_mm256_sub_pd(rinvsq20,felec));
1177
1178             /* Update potential sum for this i atom from the interaction with this j atom. */
1179             velec            = _mm256_andnot_pd(dummy_mask,velec);
1180             velecsum         = _mm256_add_pd(velecsum,velec);
1181
1182             fscal            = felec;
1183
1184             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1185
1186             /* Calculate temporary vectorial force */
1187             tx               = _mm256_mul_pd(fscal,dx20);
1188             ty               = _mm256_mul_pd(fscal,dy20);
1189             tz               = _mm256_mul_pd(fscal,dz20);
1190
1191             /* Update vectorial force */
1192             fix2             = _mm256_add_pd(fix2,tx);
1193             fiy2             = _mm256_add_pd(fiy2,ty);
1194             fiz2             = _mm256_add_pd(fiz2,tz);
1195
1196             fjx0             = _mm256_add_pd(fjx0,tx);
1197             fjy0             = _mm256_add_pd(fjy0,ty);
1198             fjz0             = _mm256_add_pd(fjz0,tz);
1199
1200             /**************************
1201              * CALCULATE INTERACTIONS *
1202              **************************/
1203
1204             r21              = _mm256_mul_pd(rsq21,rinv21);
1205             r21              = _mm256_andnot_pd(dummy_mask,r21);
1206
1207             /* EWALD ELECTROSTATICS */
1208
1209             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1210             ewrt             = _mm256_mul_pd(r21,ewtabscale);
1211             ewitab           = _mm256_cvttpd_epi32(ewrt);
1212             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1213             ewitab           = _mm_slli_epi32(ewitab,2);
1214             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1215             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1216             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1217             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1218             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1219             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1220             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1221             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(rinv21,velec));
1222             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
1223
1224             /* Update potential sum for this i atom from the interaction with this j atom. */
1225             velec            = _mm256_andnot_pd(dummy_mask,velec);
1226             velecsum         = _mm256_add_pd(velecsum,velec);
1227
1228             fscal            = felec;
1229
1230             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1231
1232             /* Calculate temporary vectorial force */
1233             tx               = _mm256_mul_pd(fscal,dx21);
1234             ty               = _mm256_mul_pd(fscal,dy21);
1235             tz               = _mm256_mul_pd(fscal,dz21);
1236
1237             /* Update vectorial force */
1238             fix2             = _mm256_add_pd(fix2,tx);
1239             fiy2             = _mm256_add_pd(fiy2,ty);
1240             fiz2             = _mm256_add_pd(fiz2,tz);
1241
1242             fjx1             = _mm256_add_pd(fjx1,tx);
1243             fjy1             = _mm256_add_pd(fjy1,ty);
1244             fjz1             = _mm256_add_pd(fjz1,tz);
1245
1246             /**************************
1247              * CALCULATE INTERACTIONS *
1248              **************************/
1249
1250             r22              = _mm256_mul_pd(rsq22,rinv22);
1251             r22              = _mm256_andnot_pd(dummy_mask,r22);
1252
1253             /* EWALD ELECTROSTATICS */
1254
1255             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1256             ewrt             = _mm256_mul_pd(r22,ewtabscale);
1257             ewitab           = _mm256_cvttpd_epi32(ewrt);
1258             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1259             ewitab           = _mm_slli_epi32(ewitab,2);
1260             ewtabF           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1261             ewtabD           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
1262             ewtabV           = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,2) );
1263             ewtabFn          = _mm256_load_pd( ewtab + _mm_extract_epi32(ewitab,3) );
1264             GMX_MM256_FULLTRANSPOSE4_PD(ewtabF,ewtabD,ewtabV,ewtabFn);
1265             felec            = _mm256_add_pd(ewtabF,_mm256_mul_pd(eweps,ewtabD));
1266             velec            = _mm256_sub_pd(ewtabV,_mm256_mul_pd(_mm256_mul_pd(ewtabhalfspace,eweps),_mm256_add_pd(ewtabF,felec)));
1267             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(rinv22,velec));
1268             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
1269
1270             /* Update potential sum for this i atom from the interaction with this j atom. */
1271             velec            = _mm256_andnot_pd(dummy_mask,velec);
1272             velecsum         = _mm256_add_pd(velecsum,velec);
1273
1274             fscal            = felec;
1275
1276             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1277
1278             /* Calculate temporary vectorial force */
1279             tx               = _mm256_mul_pd(fscal,dx22);
1280             ty               = _mm256_mul_pd(fscal,dy22);
1281             tz               = _mm256_mul_pd(fscal,dz22);
1282
1283             /* Update vectorial force */
1284             fix2             = _mm256_add_pd(fix2,tx);
1285             fiy2             = _mm256_add_pd(fiy2,ty);
1286             fiz2             = _mm256_add_pd(fiz2,tz);
1287
1288             fjx2             = _mm256_add_pd(fjx2,tx);
1289             fjy2             = _mm256_add_pd(fjy2,ty);
1290             fjz2             = _mm256_add_pd(fjz2,tz);
1291
1292             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1293             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1294             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1295             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1296
1297             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1298                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1299
1300             /* Inner loop uses 412 flops */
1301         }
1302
1303         /* End of innermost loop */
1304
1305         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1306                                                  f+i_coord_offset,fshift+i_shift_offset);
1307
1308         ggid                        = gid[iidx];
1309         /* Update potential energies */
1310         gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1311         gmx_mm256_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1312
1313         /* Increment number of inner iterations */
1314         inneriter                  += j_index_end - j_index_start;
1315
1316         /* Outer loop uses 20 flops */
1317     }
1318
1319     /* Increment number of outer iterations */
1320     outeriter        += nri;
1321
1322     /* Update outer/inner flops */
1323
1324     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_VF,outeriter*20 + inneriter*412);
1325 }
1326 /*
1327  * Gromacs nonbonded kernel:   nb_kernel_ElecEw_VdwCSTab_GeomW3W3_F_avx_256_double
1328  * Electrostatics interaction: Ewald
1329  * VdW interaction:            CubicSplineTable
1330  * Geometry:                   Water3-Water3
1331  * Calculate force/pot:        Force
1332  */
1333 void
1334 nb_kernel_ElecEw_VdwCSTab_GeomW3W3_F_avx_256_double
1335                     (t_nblist                    * gmx_restrict       nlist,
1336                      rvec                        * gmx_restrict          xx,
1337                      rvec                        * gmx_restrict          ff,
1338                      t_forcerec                  * gmx_restrict          fr,
1339                      t_mdatoms                   * gmx_restrict     mdatoms,
1340                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1341                      t_nrnb                      * gmx_restrict        nrnb)
1342 {
1343     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1344      * just 0 for non-waters.
1345      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
1346      * jnr indices corresponding to data put in the four positions in the SIMD register.
1347      */
1348     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1349     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1350     int              jnrA,jnrB,jnrC,jnrD;
1351     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1352     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
1353     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1354     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1355     real             rcutoff_scalar;
1356     real             *shiftvec,*fshift,*x,*f;
1357     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1358     real             scratch[4*DIM];
1359     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1360     real *           vdwioffsetptr0;
1361     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1362     real *           vdwioffsetptr1;
1363     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1364     real *           vdwioffsetptr2;
1365     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1366     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1367     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1368     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1369     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1370     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1371     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1372     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1373     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1374     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1375     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1376     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1377     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1378     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1379     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1380     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1381     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
1382     real             *charge;
1383     int              nvdwtype;
1384     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1385     int              *vdwtype;
1386     real             *vdwparam;
1387     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
1388     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
1389     __m128i          vfitab;
1390     __m128i          ifour       = _mm_set1_epi32(4);
1391     __m256d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1392     real             *vftab;
1393     __m128i          ewitab;
1394     __m256d          ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1395     __m256d          beta,beta2,beta3,zeta2,pmecorrF,pmecorrV,rinv3;
1396     real             *ewtab;
1397     __m256d          dummy_mask,cutoff_mask;
1398     __m128           tmpmask0,tmpmask1;
1399     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
1400     __m256d          one     = _mm256_set1_pd(1.0);
1401     __m256d          two     = _mm256_set1_pd(2.0);
1402     x                = xx[0];
1403     f                = ff[0];
1404
1405     nri              = nlist->nri;
1406     iinr             = nlist->iinr;
1407     jindex           = nlist->jindex;
1408     jjnr             = nlist->jjnr;
1409     shiftidx         = nlist->shift;
1410     gid              = nlist->gid;
1411     shiftvec         = fr->shift_vec[0];
1412     fshift           = fr->fshift[0];
1413     facel            = _mm256_set1_pd(fr->epsfac);
1414     charge           = mdatoms->chargeA;
1415     nvdwtype         = fr->ntype;
1416     vdwparam         = fr->nbfp;
1417     vdwtype          = mdatoms->typeA;
1418
1419     vftab            = kernel_data->table_vdw->data;
1420     vftabscale       = _mm256_set1_pd(kernel_data->table_vdw->scale);
1421
1422     sh_ewald         = _mm256_set1_pd(fr->ic->sh_ewald);
1423     beta             = _mm256_set1_pd(fr->ic->ewaldcoeff_q);
1424     beta2            = _mm256_mul_pd(beta,beta);
1425     beta3            = _mm256_mul_pd(beta,beta2);
1426
1427     ewtab            = fr->ic->tabq_coul_F;
1428     ewtabscale       = _mm256_set1_pd(fr->ic->tabq_scale);
1429     ewtabhalfspace   = _mm256_set1_pd(0.5/fr->ic->tabq_scale);
1430
1431     /* Setup water-specific parameters */
1432     inr              = nlist->iinr[0];
1433     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
1434     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1435     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1436     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
1437
1438     jq0              = _mm256_set1_pd(charge[inr+0]);
1439     jq1              = _mm256_set1_pd(charge[inr+1]);
1440     jq2              = _mm256_set1_pd(charge[inr+2]);
1441     vdwjidx0A        = 2*vdwtype[inr+0];
1442     qq00             = _mm256_mul_pd(iq0,jq0);
1443     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
1444     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
1445     qq01             = _mm256_mul_pd(iq0,jq1);
1446     qq02             = _mm256_mul_pd(iq0,jq2);
1447     qq10             = _mm256_mul_pd(iq1,jq0);
1448     qq11             = _mm256_mul_pd(iq1,jq1);
1449     qq12             = _mm256_mul_pd(iq1,jq2);
1450     qq20             = _mm256_mul_pd(iq2,jq0);
1451     qq21             = _mm256_mul_pd(iq2,jq1);
1452     qq22             = _mm256_mul_pd(iq2,jq2);
1453
1454     /* Avoid stupid compiler warnings */
1455     jnrA = jnrB = jnrC = jnrD = 0;
1456     j_coord_offsetA = 0;
1457     j_coord_offsetB = 0;
1458     j_coord_offsetC = 0;
1459     j_coord_offsetD = 0;
1460
1461     outeriter        = 0;
1462     inneriter        = 0;
1463
1464     for(iidx=0;iidx<4*DIM;iidx++)
1465     {
1466         scratch[iidx] = 0.0;
1467     }
1468
1469     /* Start outer loop over neighborlists */
1470     for(iidx=0; iidx<nri; iidx++)
1471     {
1472         /* Load shift vector for this list */
1473         i_shift_offset   = DIM*shiftidx[iidx];
1474
1475         /* Load limits for loop over neighbors */
1476         j_index_start    = jindex[iidx];
1477         j_index_end      = jindex[iidx+1];
1478
1479         /* Get outer coordinate index */
1480         inr              = iinr[iidx];
1481         i_coord_offset   = DIM*inr;
1482
1483         /* Load i particle coords and add shift vector */
1484         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1485                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1486
1487         fix0             = _mm256_setzero_pd();
1488         fiy0             = _mm256_setzero_pd();
1489         fiz0             = _mm256_setzero_pd();
1490         fix1             = _mm256_setzero_pd();
1491         fiy1             = _mm256_setzero_pd();
1492         fiz1             = _mm256_setzero_pd();
1493         fix2             = _mm256_setzero_pd();
1494         fiy2             = _mm256_setzero_pd();
1495         fiz2             = _mm256_setzero_pd();
1496
1497         /* Start inner kernel loop */
1498         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1499         {
1500
1501             /* Get j neighbor index, and coordinate index */
1502             jnrA             = jjnr[jidx];
1503             jnrB             = jjnr[jidx+1];
1504             jnrC             = jjnr[jidx+2];
1505             jnrD             = jjnr[jidx+3];
1506             j_coord_offsetA  = DIM*jnrA;
1507             j_coord_offsetB  = DIM*jnrB;
1508             j_coord_offsetC  = DIM*jnrC;
1509             j_coord_offsetD  = DIM*jnrD;
1510
1511             /* load j atom coordinates */
1512             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1513                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1514                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1515
1516             /* Calculate displacement vector */
1517             dx00             = _mm256_sub_pd(ix0,jx0);
1518             dy00             = _mm256_sub_pd(iy0,jy0);
1519             dz00             = _mm256_sub_pd(iz0,jz0);
1520             dx01             = _mm256_sub_pd(ix0,jx1);
1521             dy01             = _mm256_sub_pd(iy0,jy1);
1522             dz01             = _mm256_sub_pd(iz0,jz1);
1523             dx02             = _mm256_sub_pd(ix0,jx2);
1524             dy02             = _mm256_sub_pd(iy0,jy2);
1525             dz02             = _mm256_sub_pd(iz0,jz2);
1526             dx10             = _mm256_sub_pd(ix1,jx0);
1527             dy10             = _mm256_sub_pd(iy1,jy0);
1528             dz10             = _mm256_sub_pd(iz1,jz0);
1529             dx11             = _mm256_sub_pd(ix1,jx1);
1530             dy11             = _mm256_sub_pd(iy1,jy1);
1531             dz11             = _mm256_sub_pd(iz1,jz1);
1532             dx12             = _mm256_sub_pd(ix1,jx2);
1533             dy12             = _mm256_sub_pd(iy1,jy2);
1534             dz12             = _mm256_sub_pd(iz1,jz2);
1535             dx20             = _mm256_sub_pd(ix2,jx0);
1536             dy20             = _mm256_sub_pd(iy2,jy0);
1537             dz20             = _mm256_sub_pd(iz2,jz0);
1538             dx21             = _mm256_sub_pd(ix2,jx1);
1539             dy21             = _mm256_sub_pd(iy2,jy1);
1540             dz21             = _mm256_sub_pd(iz2,jz1);
1541             dx22             = _mm256_sub_pd(ix2,jx2);
1542             dy22             = _mm256_sub_pd(iy2,jy2);
1543             dz22             = _mm256_sub_pd(iz2,jz2);
1544
1545             /* Calculate squared distance and things based on it */
1546             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1547             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1548             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1549             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1550             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1551             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1552             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
1553             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1554             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1555
1556             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
1557             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
1558             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
1559             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
1560             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1561             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1562             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
1563             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1564             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1565
1566             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
1567             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
1568             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
1569             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
1570             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1571             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1572             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
1573             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1574             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1575
1576             fjx0             = _mm256_setzero_pd();
1577             fjy0             = _mm256_setzero_pd();
1578             fjz0             = _mm256_setzero_pd();
1579             fjx1             = _mm256_setzero_pd();
1580             fjy1             = _mm256_setzero_pd();
1581             fjz1             = _mm256_setzero_pd();
1582             fjx2             = _mm256_setzero_pd();
1583             fjy2             = _mm256_setzero_pd();
1584             fjz2             = _mm256_setzero_pd();
1585
1586             /**************************
1587              * CALCULATE INTERACTIONS *
1588              **************************/
1589
1590             r00              = _mm256_mul_pd(rsq00,rinv00);
1591
1592             /* Calculate table index by multiplying r with table scale and truncate to integer */
1593             rt               = _mm256_mul_pd(r00,vftabscale);
1594             vfitab           = _mm256_cvttpd_epi32(rt);
1595             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
1596             vfitab           = _mm_slli_epi32(vfitab,3);
1597
1598             /* EWALD ELECTROSTATICS */
1599
1600             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1601             ewrt             = _mm256_mul_pd(r00,ewtabscale);
1602             ewitab           = _mm256_cvttpd_epi32(ewrt);
1603             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1604             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1605                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1606                                             &ewtabF,&ewtabFn);
1607             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1608             felec            = _mm256_mul_pd(_mm256_mul_pd(qq00,rinv00),_mm256_sub_pd(rinvsq00,felec));
1609
1610             /* CUBIC SPLINE TABLE DISPERSION */
1611             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1612             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1613             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1614             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1615             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1616             Heps             = _mm256_mul_pd(vfeps,H);
1617             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1618             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1619             fvdw6            = _mm256_mul_pd(c6_00,FF);
1620
1621             /* CUBIC SPLINE TABLE REPULSION */
1622             vfitab           = _mm_add_epi32(vfitab,ifour);
1623             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
1624             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
1625             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
1626             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
1627             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
1628             Heps             = _mm256_mul_pd(vfeps,H);
1629             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
1630             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
1631             fvdw12           = _mm256_mul_pd(c12_00,FF);
1632             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
1633
1634             fscal            = _mm256_add_pd(felec,fvdw);
1635
1636             /* Calculate temporary vectorial force */
1637             tx               = _mm256_mul_pd(fscal,dx00);
1638             ty               = _mm256_mul_pd(fscal,dy00);
1639             tz               = _mm256_mul_pd(fscal,dz00);
1640
1641             /* Update vectorial force */
1642             fix0             = _mm256_add_pd(fix0,tx);
1643             fiy0             = _mm256_add_pd(fiy0,ty);
1644             fiz0             = _mm256_add_pd(fiz0,tz);
1645
1646             fjx0             = _mm256_add_pd(fjx0,tx);
1647             fjy0             = _mm256_add_pd(fjy0,ty);
1648             fjz0             = _mm256_add_pd(fjz0,tz);
1649
1650             /**************************
1651              * CALCULATE INTERACTIONS *
1652              **************************/
1653
1654             r01              = _mm256_mul_pd(rsq01,rinv01);
1655
1656             /* EWALD ELECTROSTATICS */
1657
1658             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1659             ewrt             = _mm256_mul_pd(r01,ewtabscale);
1660             ewitab           = _mm256_cvttpd_epi32(ewrt);
1661             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1662             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1663                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1664                                             &ewtabF,&ewtabFn);
1665             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1666             felec            = _mm256_mul_pd(_mm256_mul_pd(qq01,rinv01),_mm256_sub_pd(rinvsq01,felec));
1667
1668             fscal            = felec;
1669
1670             /* Calculate temporary vectorial force */
1671             tx               = _mm256_mul_pd(fscal,dx01);
1672             ty               = _mm256_mul_pd(fscal,dy01);
1673             tz               = _mm256_mul_pd(fscal,dz01);
1674
1675             /* Update vectorial force */
1676             fix0             = _mm256_add_pd(fix0,tx);
1677             fiy0             = _mm256_add_pd(fiy0,ty);
1678             fiz0             = _mm256_add_pd(fiz0,tz);
1679
1680             fjx1             = _mm256_add_pd(fjx1,tx);
1681             fjy1             = _mm256_add_pd(fjy1,ty);
1682             fjz1             = _mm256_add_pd(fjz1,tz);
1683
1684             /**************************
1685              * CALCULATE INTERACTIONS *
1686              **************************/
1687
1688             r02              = _mm256_mul_pd(rsq02,rinv02);
1689
1690             /* EWALD ELECTROSTATICS */
1691
1692             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1693             ewrt             = _mm256_mul_pd(r02,ewtabscale);
1694             ewitab           = _mm256_cvttpd_epi32(ewrt);
1695             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1696             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1697                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1698                                             &ewtabF,&ewtabFn);
1699             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1700             felec            = _mm256_mul_pd(_mm256_mul_pd(qq02,rinv02),_mm256_sub_pd(rinvsq02,felec));
1701
1702             fscal            = felec;
1703
1704             /* Calculate temporary vectorial force */
1705             tx               = _mm256_mul_pd(fscal,dx02);
1706             ty               = _mm256_mul_pd(fscal,dy02);
1707             tz               = _mm256_mul_pd(fscal,dz02);
1708
1709             /* Update vectorial force */
1710             fix0             = _mm256_add_pd(fix0,tx);
1711             fiy0             = _mm256_add_pd(fiy0,ty);
1712             fiz0             = _mm256_add_pd(fiz0,tz);
1713
1714             fjx2             = _mm256_add_pd(fjx2,tx);
1715             fjy2             = _mm256_add_pd(fjy2,ty);
1716             fjz2             = _mm256_add_pd(fjz2,tz);
1717
1718             /**************************
1719              * CALCULATE INTERACTIONS *
1720              **************************/
1721
1722             r10              = _mm256_mul_pd(rsq10,rinv10);
1723
1724             /* EWALD ELECTROSTATICS */
1725
1726             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1727             ewrt             = _mm256_mul_pd(r10,ewtabscale);
1728             ewitab           = _mm256_cvttpd_epi32(ewrt);
1729             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1730             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1731                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1732                                             &ewtabF,&ewtabFn);
1733             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1734             felec            = _mm256_mul_pd(_mm256_mul_pd(qq10,rinv10),_mm256_sub_pd(rinvsq10,felec));
1735
1736             fscal            = felec;
1737
1738             /* Calculate temporary vectorial force */
1739             tx               = _mm256_mul_pd(fscal,dx10);
1740             ty               = _mm256_mul_pd(fscal,dy10);
1741             tz               = _mm256_mul_pd(fscal,dz10);
1742
1743             /* Update vectorial force */
1744             fix1             = _mm256_add_pd(fix1,tx);
1745             fiy1             = _mm256_add_pd(fiy1,ty);
1746             fiz1             = _mm256_add_pd(fiz1,tz);
1747
1748             fjx0             = _mm256_add_pd(fjx0,tx);
1749             fjy0             = _mm256_add_pd(fjy0,ty);
1750             fjz0             = _mm256_add_pd(fjz0,tz);
1751
1752             /**************************
1753              * CALCULATE INTERACTIONS *
1754              **************************/
1755
1756             r11              = _mm256_mul_pd(rsq11,rinv11);
1757
1758             /* EWALD ELECTROSTATICS */
1759
1760             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1761             ewrt             = _mm256_mul_pd(r11,ewtabscale);
1762             ewitab           = _mm256_cvttpd_epi32(ewrt);
1763             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1764             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1765                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1766                                             &ewtabF,&ewtabFn);
1767             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1768             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
1769
1770             fscal            = felec;
1771
1772             /* Calculate temporary vectorial force */
1773             tx               = _mm256_mul_pd(fscal,dx11);
1774             ty               = _mm256_mul_pd(fscal,dy11);
1775             tz               = _mm256_mul_pd(fscal,dz11);
1776
1777             /* Update vectorial force */
1778             fix1             = _mm256_add_pd(fix1,tx);
1779             fiy1             = _mm256_add_pd(fiy1,ty);
1780             fiz1             = _mm256_add_pd(fiz1,tz);
1781
1782             fjx1             = _mm256_add_pd(fjx1,tx);
1783             fjy1             = _mm256_add_pd(fjy1,ty);
1784             fjz1             = _mm256_add_pd(fjz1,tz);
1785
1786             /**************************
1787              * CALCULATE INTERACTIONS *
1788              **************************/
1789
1790             r12              = _mm256_mul_pd(rsq12,rinv12);
1791
1792             /* EWALD ELECTROSTATICS */
1793
1794             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1795             ewrt             = _mm256_mul_pd(r12,ewtabscale);
1796             ewitab           = _mm256_cvttpd_epi32(ewrt);
1797             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1798             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1799                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1800                                             &ewtabF,&ewtabFn);
1801             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1802             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
1803
1804             fscal            = felec;
1805
1806             /* Calculate temporary vectorial force */
1807             tx               = _mm256_mul_pd(fscal,dx12);
1808             ty               = _mm256_mul_pd(fscal,dy12);
1809             tz               = _mm256_mul_pd(fscal,dz12);
1810
1811             /* Update vectorial force */
1812             fix1             = _mm256_add_pd(fix1,tx);
1813             fiy1             = _mm256_add_pd(fiy1,ty);
1814             fiz1             = _mm256_add_pd(fiz1,tz);
1815
1816             fjx2             = _mm256_add_pd(fjx2,tx);
1817             fjy2             = _mm256_add_pd(fjy2,ty);
1818             fjz2             = _mm256_add_pd(fjz2,tz);
1819
1820             /**************************
1821              * CALCULATE INTERACTIONS *
1822              **************************/
1823
1824             r20              = _mm256_mul_pd(rsq20,rinv20);
1825
1826             /* EWALD ELECTROSTATICS */
1827
1828             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1829             ewrt             = _mm256_mul_pd(r20,ewtabscale);
1830             ewitab           = _mm256_cvttpd_epi32(ewrt);
1831             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1832             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1833                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1834                                             &ewtabF,&ewtabFn);
1835             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1836             felec            = _mm256_mul_pd(_mm256_mul_pd(qq20,rinv20),_mm256_sub_pd(rinvsq20,felec));
1837
1838             fscal            = felec;
1839
1840             /* Calculate temporary vectorial force */
1841             tx               = _mm256_mul_pd(fscal,dx20);
1842             ty               = _mm256_mul_pd(fscal,dy20);
1843             tz               = _mm256_mul_pd(fscal,dz20);
1844
1845             /* Update vectorial force */
1846             fix2             = _mm256_add_pd(fix2,tx);
1847             fiy2             = _mm256_add_pd(fiy2,ty);
1848             fiz2             = _mm256_add_pd(fiz2,tz);
1849
1850             fjx0             = _mm256_add_pd(fjx0,tx);
1851             fjy0             = _mm256_add_pd(fjy0,ty);
1852             fjz0             = _mm256_add_pd(fjz0,tz);
1853
1854             /**************************
1855              * CALCULATE INTERACTIONS *
1856              **************************/
1857
1858             r21              = _mm256_mul_pd(rsq21,rinv21);
1859
1860             /* EWALD ELECTROSTATICS */
1861
1862             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1863             ewrt             = _mm256_mul_pd(r21,ewtabscale);
1864             ewitab           = _mm256_cvttpd_epi32(ewrt);
1865             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1866             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1867                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1868                                             &ewtabF,&ewtabFn);
1869             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1870             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
1871
1872             fscal            = felec;
1873
1874             /* Calculate temporary vectorial force */
1875             tx               = _mm256_mul_pd(fscal,dx21);
1876             ty               = _mm256_mul_pd(fscal,dy21);
1877             tz               = _mm256_mul_pd(fscal,dz21);
1878
1879             /* Update vectorial force */
1880             fix2             = _mm256_add_pd(fix2,tx);
1881             fiy2             = _mm256_add_pd(fiy2,ty);
1882             fiz2             = _mm256_add_pd(fiz2,tz);
1883
1884             fjx1             = _mm256_add_pd(fjx1,tx);
1885             fjy1             = _mm256_add_pd(fjy1,ty);
1886             fjz1             = _mm256_add_pd(fjz1,tz);
1887
1888             /**************************
1889              * CALCULATE INTERACTIONS *
1890              **************************/
1891
1892             r22              = _mm256_mul_pd(rsq22,rinv22);
1893
1894             /* EWALD ELECTROSTATICS */
1895
1896             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1897             ewrt             = _mm256_mul_pd(r22,ewtabscale);
1898             ewitab           = _mm256_cvttpd_epi32(ewrt);
1899             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
1900             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
1901                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
1902                                             &ewtabF,&ewtabFn);
1903             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
1904             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
1905
1906             fscal            = felec;
1907
1908             /* Calculate temporary vectorial force */
1909             tx               = _mm256_mul_pd(fscal,dx22);
1910             ty               = _mm256_mul_pd(fscal,dy22);
1911             tz               = _mm256_mul_pd(fscal,dz22);
1912
1913             /* Update vectorial force */
1914             fix2             = _mm256_add_pd(fix2,tx);
1915             fiy2             = _mm256_add_pd(fiy2,ty);
1916             fiz2             = _mm256_add_pd(fiz2,tz);
1917
1918             fjx2             = _mm256_add_pd(fjx2,tx);
1919             fjy2             = _mm256_add_pd(fjy2,ty);
1920             fjz2             = _mm256_add_pd(fjz2,tz);
1921
1922             fjptrA             = f+j_coord_offsetA;
1923             fjptrB             = f+j_coord_offsetB;
1924             fjptrC             = f+j_coord_offsetC;
1925             fjptrD             = f+j_coord_offsetD;
1926
1927             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1928                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1929
1930             /* Inner loop uses 350 flops */
1931         }
1932
1933         if(jidx<j_index_end)
1934         {
1935
1936             /* Get j neighbor index, and coordinate index */
1937             jnrlistA         = jjnr[jidx];
1938             jnrlistB         = jjnr[jidx+1];
1939             jnrlistC         = jjnr[jidx+2];
1940             jnrlistD         = jjnr[jidx+3];
1941             /* Sign of each element will be negative for non-real atoms.
1942              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1943              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
1944              */
1945             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1946
1947             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
1948             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
1949             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
1950
1951             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1952             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1953             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1954             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1955             j_coord_offsetA  = DIM*jnrA;
1956             j_coord_offsetB  = DIM*jnrB;
1957             j_coord_offsetC  = DIM*jnrC;
1958             j_coord_offsetD  = DIM*jnrD;
1959
1960             /* load j atom coordinates */
1961             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1962                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1963                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1964
1965             /* Calculate displacement vector */
1966             dx00             = _mm256_sub_pd(ix0,jx0);
1967             dy00             = _mm256_sub_pd(iy0,jy0);
1968             dz00             = _mm256_sub_pd(iz0,jz0);
1969             dx01             = _mm256_sub_pd(ix0,jx1);
1970             dy01             = _mm256_sub_pd(iy0,jy1);
1971             dz01             = _mm256_sub_pd(iz0,jz1);
1972             dx02             = _mm256_sub_pd(ix0,jx2);
1973             dy02             = _mm256_sub_pd(iy0,jy2);
1974             dz02             = _mm256_sub_pd(iz0,jz2);
1975             dx10             = _mm256_sub_pd(ix1,jx0);
1976             dy10             = _mm256_sub_pd(iy1,jy0);
1977             dz10             = _mm256_sub_pd(iz1,jz0);
1978             dx11             = _mm256_sub_pd(ix1,jx1);
1979             dy11             = _mm256_sub_pd(iy1,jy1);
1980             dz11             = _mm256_sub_pd(iz1,jz1);
1981             dx12             = _mm256_sub_pd(ix1,jx2);
1982             dy12             = _mm256_sub_pd(iy1,jy2);
1983             dz12             = _mm256_sub_pd(iz1,jz2);
1984             dx20             = _mm256_sub_pd(ix2,jx0);
1985             dy20             = _mm256_sub_pd(iy2,jy0);
1986             dz20             = _mm256_sub_pd(iz2,jz0);
1987             dx21             = _mm256_sub_pd(ix2,jx1);
1988             dy21             = _mm256_sub_pd(iy2,jy1);
1989             dz21             = _mm256_sub_pd(iz2,jz1);
1990             dx22             = _mm256_sub_pd(ix2,jx2);
1991             dy22             = _mm256_sub_pd(iy2,jy2);
1992             dz22             = _mm256_sub_pd(iz2,jz2);
1993
1994             /* Calculate squared distance and things based on it */
1995             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1996             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1997             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1998             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1999             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
2000             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
2001             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
2002             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
2003             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
2004
2005             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
2006             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
2007             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
2008             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
2009             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
2010             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
2011             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
2012             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
2013             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
2014
2015             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
2016             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
2017             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
2018             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
2019             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
2020             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
2021             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
2022             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
2023             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
2024
2025             fjx0             = _mm256_setzero_pd();
2026             fjy0             = _mm256_setzero_pd();
2027             fjz0             = _mm256_setzero_pd();
2028             fjx1             = _mm256_setzero_pd();
2029             fjy1             = _mm256_setzero_pd();
2030             fjz1             = _mm256_setzero_pd();
2031             fjx2             = _mm256_setzero_pd();
2032             fjy2             = _mm256_setzero_pd();
2033             fjz2             = _mm256_setzero_pd();
2034
2035             /**************************
2036              * CALCULATE INTERACTIONS *
2037              **************************/
2038
2039             r00              = _mm256_mul_pd(rsq00,rinv00);
2040             r00              = _mm256_andnot_pd(dummy_mask,r00);
2041
2042             /* Calculate table index by multiplying r with table scale and truncate to integer */
2043             rt               = _mm256_mul_pd(r00,vftabscale);
2044             vfitab           = _mm256_cvttpd_epi32(rt);
2045             vfeps            = _mm256_sub_pd(rt,_mm256_round_pd(rt, _MM_FROUND_FLOOR));
2046             vfitab           = _mm_slli_epi32(vfitab,3);
2047
2048             /* EWALD ELECTROSTATICS */
2049
2050             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2051             ewrt             = _mm256_mul_pd(r00,ewtabscale);
2052             ewitab           = _mm256_cvttpd_epi32(ewrt);
2053             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2054             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2055                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2056                                             &ewtabF,&ewtabFn);
2057             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2058             felec            = _mm256_mul_pd(_mm256_mul_pd(qq00,rinv00),_mm256_sub_pd(rinvsq00,felec));
2059
2060             /* CUBIC SPLINE TABLE DISPERSION */
2061             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
2062             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
2063             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
2064             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
2065             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
2066             Heps             = _mm256_mul_pd(vfeps,H);
2067             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
2068             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
2069             fvdw6            = _mm256_mul_pd(c6_00,FF);
2070
2071             /* CUBIC SPLINE TABLE REPULSION */
2072             vfitab           = _mm_add_epi32(vfitab,ifour);
2073             Y                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
2074             F                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
2075             G                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,2) );
2076             H                = _mm256_load_pd( vftab + _mm_extract_epi32(vfitab,3) );
2077             GMX_MM256_FULLTRANSPOSE4_PD(Y,F,G,H);
2078             Heps             = _mm256_mul_pd(vfeps,H);
2079             Fp               = _mm256_add_pd(F,_mm256_mul_pd(vfeps,_mm256_add_pd(G,Heps)));
2080             FF               = _mm256_add_pd(Fp,_mm256_mul_pd(vfeps,_mm256_add_pd(G,_mm256_add_pd(Heps,Heps))));
2081             fvdw12           = _mm256_mul_pd(c12_00,FF);
2082             fvdw             = _mm256_xor_pd(signbit,_mm256_mul_pd(_mm256_add_pd(fvdw6,fvdw12),_mm256_mul_pd(vftabscale,rinv00)));
2083
2084             fscal            = _mm256_add_pd(felec,fvdw);
2085
2086             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2087
2088             /* Calculate temporary vectorial force */
2089             tx               = _mm256_mul_pd(fscal,dx00);
2090             ty               = _mm256_mul_pd(fscal,dy00);
2091             tz               = _mm256_mul_pd(fscal,dz00);
2092
2093             /* Update vectorial force */
2094             fix0             = _mm256_add_pd(fix0,tx);
2095             fiy0             = _mm256_add_pd(fiy0,ty);
2096             fiz0             = _mm256_add_pd(fiz0,tz);
2097
2098             fjx0             = _mm256_add_pd(fjx0,tx);
2099             fjy0             = _mm256_add_pd(fjy0,ty);
2100             fjz0             = _mm256_add_pd(fjz0,tz);
2101
2102             /**************************
2103              * CALCULATE INTERACTIONS *
2104              **************************/
2105
2106             r01              = _mm256_mul_pd(rsq01,rinv01);
2107             r01              = _mm256_andnot_pd(dummy_mask,r01);
2108
2109             /* EWALD ELECTROSTATICS */
2110
2111             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2112             ewrt             = _mm256_mul_pd(r01,ewtabscale);
2113             ewitab           = _mm256_cvttpd_epi32(ewrt);
2114             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2115             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2116                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2117                                             &ewtabF,&ewtabFn);
2118             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2119             felec            = _mm256_mul_pd(_mm256_mul_pd(qq01,rinv01),_mm256_sub_pd(rinvsq01,felec));
2120
2121             fscal            = felec;
2122
2123             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2124
2125             /* Calculate temporary vectorial force */
2126             tx               = _mm256_mul_pd(fscal,dx01);
2127             ty               = _mm256_mul_pd(fscal,dy01);
2128             tz               = _mm256_mul_pd(fscal,dz01);
2129
2130             /* Update vectorial force */
2131             fix0             = _mm256_add_pd(fix0,tx);
2132             fiy0             = _mm256_add_pd(fiy0,ty);
2133             fiz0             = _mm256_add_pd(fiz0,tz);
2134
2135             fjx1             = _mm256_add_pd(fjx1,tx);
2136             fjy1             = _mm256_add_pd(fjy1,ty);
2137             fjz1             = _mm256_add_pd(fjz1,tz);
2138
2139             /**************************
2140              * CALCULATE INTERACTIONS *
2141              **************************/
2142
2143             r02              = _mm256_mul_pd(rsq02,rinv02);
2144             r02              = _mm256_andnot_pd(dummy_mask,r02);
2145
2146             /* EWALD ELECTROSTATICS */
2147
2148             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2149             ewrt             = _mm256_mul_pd(r02,ewtabscale);
2150             ewitab           = _mm256_cvttpd_epi32(ewrt);
2151             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2152             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2153                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2154                                             &ewtabF,&ewtabFn);
2155             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2156             felec            = _mm256_mul_pd(_mm256_mul_pd(qq02,rinv02),_mm256_sub_pd(rinvsq02,felec));
2157
2158             fscal            = felec;
2159
2160             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2161
2162             /* Calculate temporary vectorial force */
2163             tx               = _mm256_mul_pd(fscal,dx02);
2164             ty               = _mm256_mul_pd(fscal,dy02);
2165             tz               = _mm256_mul_pd(fscal,dz02);
2166
2167             /* Update vectorial force */
2168             fix0             = _mm256_add_pd(fix0,tx);
2169             fiy0             = _mm256_add_pd(fiy0,ty);
2170             fiz0             = _mm256_add_pd(fiz0,tz);
2171
2172             fjx2             = _mm256_add_pd(fjx2,tx);
2173             fjy2             = _mm256_add_pd(fjy2,ty);
2174             fjz2             = _mm256_add_pd(fjz2,tz);
2175
2176             /**************************
2177              * CALCULATE INTERACTIONS *
2178              **************************/
2179
2180             r10              = _mm256_mul_pd(rsq10,rinv10);
2181             r10              = _mm256_andnot_pd(dummy_mask,r10);
2182
2183             /* EWALD ELECTROSTATICS */
2184
2185             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2186             ewrt             = _mm256_mul_pd(r10,ewtabscale);
2187             ewitab           = _mm256_cvttpd_epi32(ewrt);
2188             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2189             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2190                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2191                                             &ewtabF,&ewtabFn);
2192             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2193             felec            = _mm256_mul_pd(_mm256_mul_pd(qq10,rinv10),_mm256_sub_pd(rinvsq10,felec));
2194
2195             fscal            = felec;
2196
2197             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2198
2199             /* Calculate temporary vectorial force */
2200             tx               = _mm256_mul_pd(fscal,dx10);
2201             ty               = _mm256_mul_pd(fscal,dy10);
2202             tz               = _mm256_mul_pd(fscal,dz10);
2203
2204             /* Update vectorial force */
2205             fix1             = _mm256_add_pd(fix1,tx);
2206             fiy1             = _mm256_add_pd(fiy1,ty);
2207             fiz1             = _mm256_add_pd(fiz1,tz);
2208
2209             fjx0             = _mm256_add_pd(fjx0,tx);
2210             fjy0             = _mm256_add_pd(fjy0,ty);
2211             fjz0             = _mm256_add_pd(fjz0,tz);
2212
2213             /**************************
2214              * CALCULATE INTERACTIONS *
2215              **************************/
2216
2217             r11              = _mm256_mul_pd(rsq11,rinv11);
2218             r11              = _mm256_andnot_pd(dummy_mask,r11);
2219
2220             /* EWALD ELECTROSTATICS */
2221
2222             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2223             ewrt             = _mm256_mul_pd(r11,ewtabscale);
2224             ewitab           = _mm256_cvttpd_epi32(ewrt);
2225             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2226             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2227                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2228                                             &ewtabF,&ewtabFn);
2229             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2230             felec            = _mm256_mul_pd(_mm256_mul_pd(qq11,rinv11),_mm256_sub_pd(rinvsq11,felec));
2231
2232             fscal            = felec;
2233
2234             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2235
2236             /* Calculate temporary vectorial force */
2237             tx               = _mm256_mul_pd(fscal,dx11);
2238             ty               = _mm256_mul_pd(fscal,dy11);
2239             tz               = _mm256_mul_pd(fscal,dz11);
2240
2241             /* Update vectorial force */
2242             fix1             = _mm256_add_pd(fix1,tx);
2243             fiy1             = _mm256_add_pd(fiy1,ty);
2244             fiz1             = _mm256_add_pd(fiz1,tz);
2245
2246             fjx1             = _mm256_add_pd(fjx1,tx);
2247             fjy1             = _mm256_add_pd(fjy1,ty);
2248             fjz1             = _mm256_add_pd(fjz1,tz);
2249
2250             /**************************
2251              * CALCULATE INTERACTIONS *
2252              **************************/
2253
2254             r12              = _mm256_mul_pd(rsq12,rinv12);
2255             r12              = _mm256_andnot_pd(dummy_mask,r12);
2256
2257             /* EWALD ELECTROSTATICS */
2258
2259             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2260             ewrt             = _mm256_mul_pd(r12,ewtabscale);
2261             ewitab           = _mm256_cvttpd_epi32(ewrt);
2262             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2263             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2264                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2265                                             &ewtabF,&ewtabFn);
2266             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2267             felec            = _mm256_mul_pd(_mm256_mul_pd(qq12,rinv12),_mm256_sub_pd(rinvsq12,felec));
2268
2269             fscal            = felec;
2270
2271             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2272
2273             /* Calculate temporary vectorial force */
2274             tx               = _mm256_mul_pd(fscal,dx12);
2275             ty               = _mm256_mul_pd(fscal,dy12);
2276             tz               = _mm256_mul_pd(fscal,dz12);
2277
2278             /* Update vectorial force */
2279             fix1             = _mm256_add_pd(fix1,tx);
2280             fiy1             = _mm256_add_pd(fiy1,ty);
2281             fiz1             = _mm256_add_pd(fiz1,tz);
2282
2283             fjx2             = _mm256_add_pd(fjx2,tx);
2284             fjy2             = _mm256_add_pd(fjy2,ty);
2285             fjz2             = _mm256_add_pd(fjz2,tz);
2286
2287             /**************************
2288              * CALCULATE INTERACTIONS *
2289              **************************/
2290
2291             r20              = _mm256_mul_pd(rsq20,rinv20);
2292             r20              = _mm256_andnot_pd(dummy_mask,r20);
2293
2294             /* EWALD ELECTROSTATICS */
2295
2296             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2297             ewrt             = _mm256_mul_pd(r20,ewtabscale);
2298             ewitab           = _mm256_cvttpd_epi32(ewrt);
2299             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2300             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2301                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2302                                             &ewtabF,&ewtabFn);
2303             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2304             felec            = _mm256_mul_pd(_mm256_mul_pd(qq20,rinv20),_mm256_sub_pd(rinvsq20,felec));
2305
2306             fscal            = felec;
2307
2308             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2309
2310             /* Calculate temporary vectorial force */
2311             tx               = _mm256_mul_pd(fscal,dx20);
2312             ty               = _mm256_mul_pd(fscal,dy20);
2313             tz               = _mm256_mul_pd(fscal,dz20);
2314
2315             /* Update vectorial force */
2316             fix2             = _mm256_add_pd(fix2,tx);
2317             fiy2             = _mm256_add_pd(fiy2,ty);
2318             fiz2             = _mm256_add_pd(fiz2,tz);
2319
2320             fjx0             = _mm256_add_pd(fjx0,tx);
2321             fjy0             = _mm256_add_pd(fjy0,ty);
2322             fjz0             = _mm256_add_pd(fjz0,tz);
2323
2324             /**************************
2325              * CALCULATE INTERACTIONS *
2326              **************************/
2327
2328             r21              = _mm256_mul_pd(rsq21,rinv21);
2329             r21              = _mm256_andnot_pd(dummy_mask,r21);
2330
2331             /* EWALD ELECTROSTATICS */
2332
2333             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2334             ewrt             = _mm256_mul_pd(r21,ewtabscale);
2335             ewitab           = _mm256_cvttpd_epi32(ewrt);
2336             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2337             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2338                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2339                                             &ewtabF,&ewtabFn);
2340             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2341             felec            = _mm256_mul_pd(_mm256_mul_pd(qq21,rinv21),_mm256_sub_pd(rinvsq21,felec));
2342
2343             fscal            = felec;
2344
2345             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2346
2347             /* Calculate temporary vectorial force */
2348             tx               = _mm256_mul_pd(fscal,dx21);
2349             ty               = _mm256_mul_pd(fscal,dy21);
2350             tz               = _mm256_mul_pd(fscal,dz21);
2351
2352             /* Update vectorial force */
2353             fix2             = _mm256_add_pd(fix2,tx);
2354             fiy2             = _mm256_add_pd(fiy2,ty);
2355             fiz2             = _mm256_add_pd(fiz2,tz);
2356
2357             fjx1             = _mm256_add_pd(fjx1,tx);
2358             fjy1             = _mm256_add_pd(fjy1,ty);
2359             fjz1             = _mm256_add_pd(fjz1,tz);
2360
2361             /**************************
2362              * CALCULATE INTERACTIONS *
2363              **************************/
2364
2365             r22              = _mm256_mul_pd(rsq22,rinv22);
2366             r22              = _mm256_andnot_pd(dummy_mask,r22);
2367
2368             /* EWALD ELECTROSTATICS */
2369
2370             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2371             ewrt             = _mm256_mul_pd(r22,ewtabscale);
2372             ewitab           = _mm256_cvttpd_epi32(ewrt);
2373             eweps            = _mm256_sub_pd(ewrt,_mm256_round_pd(ewrt, _MM_FROUND_FLOOR));
2374             gmx_mm256_load_4pair_swizzle_pd(ewtab + _mm_extract_epi32(ewitab,0),ewtab + _mm_extract_epi32(ewitab,1),
2375                                             ewtab + _mm_extract_epi32(ewitab,2),ewtab + _mm_extract_epi32(ewitab,3),
2376                                             &ewtabF,&ewtabFn);
2377             felec            = _mm256_add_pd(_mm256_mul_pd( _mm256_sub_pd(one,eweps),ewtabF),_mm256_mul_pd(eweps,ewtabFn));
2378             felec            = _mm256_mul_pd(_mm256_mul_pd(qq22,rinv22),_mm256_sub_pd(rinvsq22,felec));
2379
2380             fscal            = felec;
2381
2382             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2383
2384             /* Calculate temporary vectorial force */
2385             tx               = _mm256_mul_pd(fscal,dx22);
2386             ty               = _mm256_mul_pd(fscal,dy22);
2387             tz               = _mm256_mul_pd(fscal,dz22);
2388
2389             /* Update vectorial force */
2390             fix2             = _mm256_add_pd(fix2,tx);
2391             fiy2             = _mm256_add_pd(fiy2,ty);
2392             fiz2             = _mm256_add_pd(fiz2,tz);
2393
2394             fjx2             = _mm256_add_pd(fjx2,tx);
2395             fjy2             = _mm256_add_pd(fjy2,ty);
2396             fjz2             = _mm256_add_pd(fjz2,tz);
2397
2398             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2399             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2400             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2401             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2402
2403             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2404                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2405
2406             /* Inner loop uses 359 flops */
2407         }
2408
2409         /* End of innermost loop */
2410
2411         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
2412                                                  f+i_coord_offset,fshift+i_shift_offset);
2413
2414         /* Increment number of inner iterations */
2415         inneriter                  += j_index_end - j_index_start;
2416
2417         /* Outer loop uses 18 flops */
2418     }
2419
2420     /* Increment number of outer iterations */
2421     outeriter        += nri;
2422
2423     /* Update outer/inner flops */
2424
2425     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_F,outeriter*18 + inneriter*359);
2426 }