src/gromacs/gmxlib/nonbonded/nb_kernel_avx_256_double/nb_kernel_ElecRFCut_VdwLJSw_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 "gmxpre.h"
  39
  40 #include "config.h"
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "gromacs/legacyheaders/types/simple.h"
  46 #include "gromacs/math/vec.h"
  47 #include "gromacs/legacyheaders/nrnb.h"
  48
  49 #include "gromacs/simd/math_x86_avx_256_double.h"
  50 #include "kernelutil_x86_avx_256_double.h"
  51
  52 /*
  53  * Gromacs nonbonded kernel:   nb_kernel_ElecRFCut_VdwLJSw_GeomW3W3_VF_avx_256_double
  54  * Electrostatics interaction: ReactionField
  55  * VdW interaction:            LennardJones
  56  * Geometry:                   Water3-Water3
  57  * Calculate force/pot:        PotentialAndForce
  58  */
  59 void
  60 nb_kernel_ElecRFCut_VdwLJSw_GeomW3W3_VF_avx_256_double
  61                     (t_nblist                    * gmx_restrict       nlist,
  62                      rvec                        * gmx_restrict          xx,
  63                      rvec                        * gmx_restrict          ff,
  64                      t_forcerec                  * gmx_restrict          fr,
  65                      t_mdatoms                   * gmx_restrict     mdatoms,
  66                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  67                      t_nrnb                      * gmx_restrict        nrnb)
  68 {
  69     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  70      * just 0 for non-waters.
  71      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
  72      * jnr indices corresponding to data put in the four positions in the SIMD register.
  73      */
  74     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  75     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  76     int              jnrA,jnrB,jnrC,jnrD;
  77     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  78     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  79     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  80     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  81     real             rcutoff_scalar;
  82     real             *shiftvec,*fshift,*x,*f;
  83     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  84     real             scratch[4*DIM];
  85     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  86     real *           vdwioffsetptr0;
  87     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  88     real *           vdwioffsetptr1;
  89     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  90     real *           vdwioffsetptr2;
  91     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  92     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  93     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  94     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  95     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  96     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  97     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  98     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  99     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
 100     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
 101     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 102     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 103     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 104     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 105     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 106     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 107     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
 108     real             *charge;
 109     int              nvdwtype;
 110     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 111     int              *vdwtype;
 112     real             *vdwparam;
 113     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
 114     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
 115     __m256d          rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
 116     real             rswitch_scalar,d_scalar;
 117     __m256d          dummy_mask,cutoff_mask;
 118     __m128           tmpmask0,tmpmask1;
 119     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
 120     __m256d          one     = _mm256_set1_pd(1.0);
 121     __m256d          two     = _mm256_set1_pd(2.0);
 122     x                = xx[0];
 123     f                = ff[0];
 124
 125     nri              = nlist->nri;
 126     iinr             = nlist->iinr;
 127     jindex           = nlist->jindex;
 128     jjnr             = nlist->jjnr;
 129     shiftidx         = nlist->shift;
 130     gid              = nlist->gid;
 131     shiftvec         = fr->shift_vec[0];
 132     fshift           = fr->fshift[0];
 133     facel            = _mm256_set1_pd(fr->epsfac);
 134     charge           = mdatoms->chargeA;
 135     krf              = _mm256_set1_pd(fr->ic->k_rf);
 136     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
 137     crf              = _mm256_set1_pd(fr->ic->c_rf);
 138     nvdwtype         = fr->ntype;
 139     vdwparam         = fr->nbfp;
 140     vdwtype          = mdatoms->typeA;
 141
 142     /* Setup water-specific parameters */
 143     inr              = nlist->iinr[0];
 144     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
 145     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
 146     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
 147     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 148
 149     jq0              = _mm256_set1_pd(charge[inr+0]);
 150     jq1              = _mm256_set1_pd(charge[inr+1]);
 151     jq2              = _mm256_set1_pd(charge[inr+2]);
 152     vdwjidx0A        = 2*vdwtype[inr+0];
 153     qq00             = _mm256_mul_pd(iq0,jq0);
 154     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
 155     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
 156     qq01             = _mm256_mul_pd(iq0,jq1);
 157     qq02             = _mm256_mul_pd(iq0,jq2);
 158     qq10             = _mm256_mul_pd(iq1,jq0);
 159     qq11             = _mm256_mul_pd(iq1,jq1);
 160     qq12             = _mm256_mul_pd(iq1,jq2);
 161     qq20             = _mm256_mul_pd(iq2,jq0);
 162     qq21             = _mm256_mul_pd(iq2,jq1);
 163     qq22             = _mm256_mul_pd(iq2,jq2);
 164
 165     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 166     rcutoff_scalar   = fr->rcoulomb;
 167     rcutoff          = _mm256_set1_pd(rcutoff_scalar);
 168     rcutoff2         = _mm256_mul_pd(rcutoff,rcutoff);
 169
 170     rswitch_scalar   = fr->rvdw_switch;
 171     rswitch          = _mm256_set1_pd(rswitch_scalar);
 172     /* Setup switch parameters */
 173     d_scalar         = rcutoff_scalar-rswitch_scalar;
 174     d                = _mm256_set1_pd(d_scalar);
 175     swV3             = _mm256_set1_pd(-10.0/(d_scalar*d_scalar*d_scalar));
 176     swV4             = _mm256_set1_pd( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 177     swV5             = _mm256_set1_pd( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 178     swF2             = _mm256_set1_pd(-30.0/(d_scalar*d_scalar*d_scalar));
 179     swF3             = _mm256_set1_pd( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 180     swF4             = _mm256_set1_pd(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 181
 182     /* Avoid stupid compiler warnings */
 183     jnrA = jnrB = jnrC = jnrD = 0;
 184     j_coord_offsetA = 0;
 185     j_coord_offsetB = 0;
 186     j_coord_offsetC = 0;
 187     j_coord_offsetD = 0;
 188
 189     outeriter        = 0;
 190     inneriter        = 0;
 191
 192     for(iidx=0;iidx<4*DIM;iidx++)
 193     {
 194         scratch[iidx] = 0.0;
 195     }
 196
 197     /* Start outer loop over neighborlists */
 198     for(iidx=0; iidx<nri; iidx++)
 199     {
 200         /* Load shift vector for this list */
 201         i_shift_offset   = DIM*shiftidx[iidx];
 202
 203         /* Load limits for loop over neighbors */
 204         j_index_start    = jindex[iidx];
 205         j_index_end      = jindex[iidx+1];
 206
 207         /* Get outer coordinate index */
 208         inr              = iinr[iidx];
 209         i_coord_offset   = DIM*inr;
 210
 211         /* Load i particle coords and add shift vector */
 212         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 213                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 214
 215         fix0             = _mm256_setzero_pd();
 216         fiy0             = _mm256_setzero_pd();
 217         fiz0             = _mm256_setzero_pd();
 218         fix1             = _mm256_setzero_pd();
 219         fiy1             = _mm256_setzero_pd();
 220         fiz1             = _mm256_setzero_pd();
 221         fix2             = _mm256_setzero_pd();
 222         fiy2             = _mm256_setzero_pd();
 223         fiz2             = _mm256_setzero_pd();
 224
 225         /* Reset potential sums */
 226         velecsum         = _mm256_setzero_pd();
 227         vvdwsum          = _mm256_setzero_pd();
 228
 229         /* Start inner kernel loop */
 230         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 231         {
 232
 233             /* Get j neighbor index, and coordinate index */
 234             jnrA             = jjnr[jidx];
 235             jnrB             = jjnr[jidx+1];
 236             jnrC             = jjnr[jidx+2];
 237             jnrD             = jjnr[jidx+3];
 238             j_coord_offsetA  = DIM*jnrA;
 239             j_coord_offsetB  = DIM*jnrB;
 240             j_coord_offsetC  = DIM*jnrC;
 241             j_coord_offsetD  = DIM*jnrD;
 242
 243             /* load j atom coordinates */
 244             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 245                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 246                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 247
 248             /* Calculate displacement vector */
 249             dx00             = _mm256_sub_pd(ix0,jx0);
 250             dy00             = _mm256_sub_pd(iy0,jy0);
 251             dz00             = _mm256_sub_pd(iz0,jz0);
 252             dx01             = _mm256_sub_pd(ix0,jx1);
 253             dy01             = _mm256_sub_pd(iy0,jy1);
 254             dz01             = _mm256_sub_pd(iz0,jz1);
 255             dx02             = _mm256_sub_pd(ix0,jx2);
 256             dy02             = _mm256_sub_pd(iy0,jy2);
 257             dz02             = _mm256_sub_pd(iz0,jz2);
 258             dx10             = _mm256_sub_pd(ix1,jx0);
 259             dy10             = _mm256_sub_pd(iy1,jy0);
 260             dz10             = _mm256_sub_pd(iz1,jz0);
 261             dx11             = _mm256_sub_pd(ix1,jx1);
 262             dy11             = _mm256_sub_pd(iy1,jy1);
 263             dz11             = _mm256_sub_pd(iz1,jz1);
 264             dx12             = _mm256_sub_pd(ix1,jx2);
 265             dy12             = _mm256_sub_pd(iy1,jy2);
 266             dz12             = _mm256_sub_pd(iz1,jz2);
 267             dx20             = _mm256_sub_pd(ix2,jx0);
 268             dy20             = _mm256_sub_pd(iy2,jy0);
 269             dz20             = _mm256_sub_pd(iz2,jz0);
 270             dx21             = _mm256_sub_pd(ix2,jx1);
 271             dy21             = _mm256_sub_pd(iy2,jy1);
 272             dz21             = _mm256_sub_pd(iz2,jz1);
 273             dx22             = _mm256_sub_pd(ix2,jx2);
 274             dy22             = _mm256_sub_pd(iy2,jy2);
 275             dz22             = _mm256_sub_pd(iz2,jz2);
 276
 277             /* Calculate squared distance and things based on it */
 278             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 279             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 280             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 281             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 282             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 283             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 284             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 285             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 286             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 287
 288             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 289             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
 290             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
 291             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
 292             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 293             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 294             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
 295             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 296             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 297
 298             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 299             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 300             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 301             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 302             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 303             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 304             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 305             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 306             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 307
 308             fjx0             = _mm256_setzero_pd();
 309             fjy0             = _mm256_setzero_pd();
 310             fjz0             = _mm256_setzero_pd();
 311             fjx1             = _mm256_setzero_pd();
 312             fjy1             = _mm256_setzero_pd();
 313             fjz1             = _mm256_setzero_pd();
 314             fjx2             = _mm256_setzero_pd();
 315             fjy2             = _mm256_setzero_pd();
 316             fjz2             = _mm256_setzero_pd();
 317
 318             /**************************
 319              * CALCULATE INTERACTIONS *
 320              **************************/
 321
 322             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 323             {
 324
 325             r00              = _mm256_mul_pd(rsq00,rinv00);
 326
 327             /* REACTION-FIELD ELECTROSTATICS */
 328             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_add_pd(rinv00,_mm256_mul_pd(krf,rsq00)),crf));
 329             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
 330
 331             /* LENNARD-JONES DISPERSION/REPULSION */
 332
 333             rinvsix          = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 334             vvdw6            = _mm256_mul_pd(c6_00,rinvsix);
 335             vvdw12           = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
 336             vvdw             = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
 337             fvdw             = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
 338
 339             d                = _mm256_sub_pd(r00,rswitch);
 340             d                = _mm256_max_pd(d,_mm256_setzero_pd());
 341             d2               = _mm256_mul_pd(d,d);
 342             sw               = _mm256_add_pd(one,_mm256_mul_pd(d2,_mm256_mul_pd(d,_mm256_add_pd(swV3,_mm256_mul_pd(d,_mm256_add_pd(swV4,_mm256_mul_pd(d,swV5)))))));
 343
 344             dsw              = _mm256_mul_pd(d2,_mm256_add_pd(swF2,_mm256_mul_pd(d,_mm256_add_pd(swF3,_mm256_mul_pd(d,swF4)))));
 345
 346             /* Evaluate switch function */
 347             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 348             fvdw             = _mm256_sub_pd( _mm256_mul_pd(fvdw,sw) , _mm256_mul_pd(rinv00,_mm256_mul_pd(vvdw,dsw)) );
 349             vvdw             = _mm256_mul_pd(vvdw,sw);
 350             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
 351
 352             /* Update potential sum for this i atom from the interaction with this j atom. */
 353             velec            = _mm256_and_pd(velec,cutoff_mask);
 354             velecsum         = _mm256_add_pd(velecsum,velec);
 355             vvdw             = _mm256_and_pd(vvdw,cutoff_mask);
 356             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 357
 358             fscal            = _mm256_add_pd(felec,fvdw);
 359
 360             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 361
 362             /* Calculate temporary vectorial force */
 363             tx               = _mm256_mul_pd(fscal,dx00);
 364             ty               = _mm256_mul_pd(fscal,dy00);
 365             tz               = _mm256_mul_pd(fscal,dz00);
 366
 367             /* Update vectorial force */
 368             fix0             = _mm256_add_pd(fix0,tx);
 369             fiy0             = _mm256_add_pd(fiy0,ty);
 370             fiz0             = _mm256_add_pd(fiz0,tz);
 371
 372             fjx0             = _mm256_add_pd(fjx0,tx);
 373             fjy0             = _mm256_add_pd(fjy0,ty);
 374             fjz0             = _mm256_add_pd(fjz0,tz);
 375
 376             }
 377
 378             /**************************
 379              * CALCULATE INTERACTIONS *
 380              **************************/
 381
 382             if (gmx_mm256_any_lt(rsq01,rcutoff2))
 383             {
 384
 385             /* REACTION-FIELD ELECTROSTATICS */
 386             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_add_pd(rinv01,_mm256_mul_pd(krf,rsq01)),crf));
 387             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
 388
 389             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
 390
 391             /* Update potential sum for this i atom from the interaction with this j atom. */
 392             velec            = _mm256_and_pd(velec,cutoff_mask);
 393             velecsum         = _mm256_add_pd(velecsum,velec);
 394
 395             fscal            = felec;
 396
 397             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 398
 399             /* Calculate temporary vectorial force */
 400             tx               = _mm256_mul_pd(fscal,dx01);
 401             ty               = _mm256_mul_pd(fscal,dy01);
 402             tz               = _mm256_mul_pd(fscal,dz01);
 403
 404             /* Update vectorial force */
 405             fix0             = _mm256_add_pd(fix0,tx);
 406             fiy0             = _mm256_add_pd(fiy0,ty);
 407             fiz0             = _mm256_add_pd(fiz0,tz);
 408
 409             fjx1             = _mm256_add_pd(fjx1,tx);
 410             fjy1             = _mm256_add_pd(fjy1,ty);
 411             fjz1             = _mm256_add_pd(fjz1,tz);
 412
 413             }
 414
 415             /**************************
 416              * CALCULATE INTERACTIONS *
 417              **************************/
 418
 419             if (gmx_mm256_any_lt(rsq02,rcutoff2))
 420             {
 421
 422             /* REACTION-FIELD ELECTROSTATICS */
 423             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_add_pd(rinv02,_mm256_mul_pd(krf,rsq02)),crf));
 424             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
 425
 426             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
 427
 428             /* Update potential sum for this i atom from the interaction with this j atom. */
 429             velec            = _mm256_and_pd(velec,cutoff_mask);
 430             velecsum         = _mm256_add_pd(velecsum,velec);
 431
 432             fscal            = felec;
 433
 434             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 435
 436             /* Calculate temporary vectorial force */
 437             tx               = _mm256_mul_pd(fscal,dx02);
 438             ty               = _mm256_mul_pd(fscal,dy02);
 439             tz               = _mm256_mul_pd(fscal,dz02);
 440
 441             /* Update vectorial force */
 442             fix0             = _mm256_add_pd(fix0,tx);
 443             fiy0             = _mm256_add_pd(fiy0,ty);
 444             fiz0             = _mm256_add_pd(fiz0,tz);
 445
 446             fjx2             = _mm256_add_pd(fjx2,tx);
 447             fjy2             = _mm256_add_pd(fjy2,ty);
 448             fjz2             = _mm256_add_pd(fjz2,tz);
 449
 450             }
 451
 452             /**************************
 453              * CALCULATE INTERACTIONS *
 454              **************************/
 455
 456             if (gmx_mm256_any_lt(rsq10,rcutoff2))
 457             {
 458
 459             /* REACTION-FIELD ELECTROSTATICS */
 460             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_add_pd(rinv10,_mm256_mul_pd(krf,rsq10)),crf));
 461             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
 462
 463             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
 464
 465             /* Update potential sum for this i atom from the interaction with this j atom. */
 466             velec            = _mm256_and_pd(velec,cutoff_mask);
 467             velecsum         = _mm256_add_pd(velecsum,velec);
 468
 469             fscal            = felec;
 470
 471             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 472
 473             /* Calculate temporary vectorial force */
 474             tx               = _mm256_mul_pd(fscal,dx10);
 475             ty               = _mm256_mul_pd(fscal,dy10);
 476             tz               = _mm256_mul_pd(fscal,dz10);
 477
 478             /* Update vectorial force */
 479             fix1             = _mm256_add_pd(fix1,tx);
 480             fiy1             = _mm256_add_pd(fiy1,ty);
 481             fiz1             = _mm256_add_pd(fiz1,tz);
 482
 483             fjx0             = _mm256_add_pd(fjx0,tx);
 484             fjy0             = _mm256_add_pd(fjy0,ty);
 485             fjz0             = _mm256_add_pd(fjz0,tz);
 486
 487             }
 488
 489             /**************************
 490              * CALCULATE INTERACTIONS *
 491              **************************/
 492
 493             if (gmx_mm256_any_lt(rsq11,rcutoff2))
 494             {
 495
 496             /* REACTION-FIELD ELECTROSTATICS */
 497             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 498             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 499
 500             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
 501
 502             /* Update potential sum for this i atom from the interaction with this j atom. */
 503             velec            = _mm256_and_pd(velec,cutoff_mask);
 504             velecsum         = _mm256_add_pd(velecsum,velec);
 505
 506             fscal            = felec;
 507
 508             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 509
 510             /* Calculate temporary vectorial force */
 511             tx               = _mm256_mul_pd(fscal,dx11);
 512             ty               = _mm256_mul_pd(fscal,dy11);
 513             tz               = _mm256_mul_pd(fscal,dz11);
 514
 515             /* Update vectorial force */
 516             fix1             = _mm256_add_pd(fix1,tx);
 517             fiy1             = _mm256_add_pd(fiy1,ty);
 518             fiz1             = _mm256_add_pd(fiz1,tz);
 519
 520             fjx1             = _mm256_add_pd(fjx1,tx);
 521             fjy1             = _mm256_add_pd(fjy1,ty);
 522             fjz1             = _mm256_add_pd(fjz1,tz);
 523
 524             }
 525
 526             /**************************
 527              * CALCULATE INTERACTIONS *
 528              **************************/
 529
 530             if (gmx_mm256_any_lt(rsq12,rcutoff2))
 531             {
 532
 533             /* REACTION-FIELD ELECTROSTATICS */
 534             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
 535             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
 536
 537             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
 538
 539             /* Update potential sum for this i atom from the interaction with this j atom. */
 540             velec            = _mm256_and_pd(velec,cutoff_mask);
 541             velecsum         = _mm256_add_pd(velecsum,velec);
 542
 543             fscal            = felec;
 544
 545             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 546
 547             /* Calculate temporary vectorial force */
 548             tx               = _mm256_mul_pd(fscal,dx12);
 549             ty               = _mm256_mul_pd(fscal,dy12);
 550             tz               = _mm256_mul_pd(fscal,dz12);
 551
 552             /* Update vectorial force */
 553             fix1             = _mm256_add_pd(fix1,tx);
 554             fiy1             = _mm256_add_pd(fiy1,ty);
 555             fiz1             = _mm256_add_pd(fiz1,tz);
 556
 557             fjx2             = _mm256_add_pd(fjx2,tx);
 558             fjy2             = _mm256_add_pd(fjy2,ty);
 559             fjz2             = _mm256_add_pd(fjz2,tz);
 560
 561             }
 562
 563             /**************************
 564              * CALCULATE INTERACTIONS *
 565              **************************/
 566
 567             if (gmx_mm256_any_lt(rsq20,rcutoff2))
 568             {
 569
 570             /* REACTION-FIELD ELECTROSTATICS */
 571             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_add_pd(rinv20,_mm256_mul_pd(krf,rsq20)),crf));
 572             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
 573
 574             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
 575
 576             /* Update potential sum for this i atom from the interaction with this j atom. */
 577             velec            = _mm256_and_pd(velec,cutoff_mask);
 578             velecsum         = _mm256_add_pd(velecsum,velec);
 579
 580             fscal            = felec;
 581
 582             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 583
 584             /* Calculate temporary vectorial force */
 585             tx               = _mm256_mul_pd(fscal,dx20);
 586             ty               = _mm256_mul_pd(fscal,dy20);
 587             tz               = _mm256_mul_pd(fscal,dz20);
 588
 589             /* Update vectorial force */
 590             fix2             = _mm256_add_pd(fix2,tx);
 591             fiy2             = _mm256_add_pd(fiy2,ty);
 592             fiz2             = _mm256_add_pd(fiz2,tz);
 593
 594             fjx0             = _mm256_add_pd(fjx0,tx);
 595             fjy0             = _mm256_add_pd(fjy0,ty);
 596             fjz0             = _mm256_add_pd(fjz0,tz);
 597
 598             }
 599
 600             /**************************
 601              * CALCULATE INTERACTIONS *
 602              **************************/
 603
 604             if (gmx_mm256_any_lt(rsq21,rcutoff2))
 605             {
 606
 607             /* REACTION-FIELD ELECTROSTATICS */
 608             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
 609             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
 610
 611             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
 612
 613             /* Update potential sum for this i atom from the interaction with this j atom. */
 614             velec            = _mm256_and_pd(velec,cutoff_mask);
 615             velecsum         = _mm256_add_pd(velecsum,velec);
 616
 617             fscal            = felec;
 618
 619             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 620
 621             /* Calculate temporary vectorial force */
 622             tx               = _mm256_mul_pd(fscal,dx21);
 623             ty               = _mm256_mul_pd(fscal,dy21);
 624             tz               = _mm256_mul_pd(fscal,dz21);
 625
 626             /* Update vectorial force */
 627             fix2             = _mm256_add_pd(fix2,tx);
 628             fiy2             = _mm256_add_pd(fiy2,ty);
 629             fiz2             = _mm256_add_pd(fiz2,tz);
 630
 631             fjx1             = _mm256_add_pd(fjx1,tx);
 632             fjy1             = _mm256_add_pd(fjy1,ty);
 633             fjz1             = _mm256_add_pd(fjz1,tz);
 634
 635             }
 636
 637             /**************************
 638              * CALCULATE INTERACTIONS *
 639              **************************/
 640
 641             if (gmx_mm256_any_lt(rsq22,rcutoff2))
 642             {
 643
 644             /* REACTION-FIELD ELECTROSTATICS */
 645             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
 646             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
 647
 648             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
 649
 650             /* Update potential sum for this i atom from the interaction with this j atom. */
 651             velec            = _mm256_and_pd(velec,cutoff_mask);
 652             velecsum         = _mm256_add_pd(velecsum,velec);
 653
 654             fscal            = felec;
 655
 656             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 657
 658             /* Calculate temporary vectorial force */
 659             tx               = _mm256_mul_pd(fscal,dx22);
 660             ty               = _mm256_mul_pd(fscal,dy22);
 661             tz               = _mm256_mul_pd(fscal,dz22);
 662
 663             /* Update vectorial force */
 664             fix2             = _mm256_add_pd(fix2,tx);
 665             fiy2             = _mm256_add_pd(fiy2,ty);
 666             fiz2             = _mm256_add_pd(fiz2,tz);
 667
 668             fjx2             = _mm256_add_pd(fjx2,tx);
 669             fjy2             = _mm256_add_pd(fjy2,ty);
 670             fjz2             = _mm256_add_pd(fjz2,tz);
 671
 672             }
 673
 674             fjptrA             = f+j_coord_offsetA;
 675             fjptrB             = f+j_coord_offsetB;
 676             fjptrC             = f+j_coord_offsetC;
 677             fjptrD             = f+j_coord_offsetD;
 678
 679             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
 680                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
 681
 682             /* Inner loop uses 358 flops */
 683         }
 684
 685         if(jidx<j_index_end)
 686         {
 687
 688             /* Get j neighbor index, and coordinate index */
 689             jnrlistA         = jjnr[jidx];
 690             jnrlistB         = jjnr[jidx+1];
 691             jnrlistC         = jjnr[jidx+2];
 692             jnrlistD         = jjnr[jidx+3];
 693             /* Sign of each element will be negative for non-real atoms.
 694              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 695              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
 696              */
 697             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 698
 699             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
 700             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
 701             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
 702
 703             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 704             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 705             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 706             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 707             j_coord_offsetA  = DIM*jnrA;
 708             j_coord_offsetB  = DIM*jnrB;
 709             j_coord_offsetC  = DIM*jnrC;
 710             j_coord_offsetD  = DIM*jnrD;
 711
 712             /* load j atom coordinates */
 713             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 714                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 715                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 716
 717             /* Calculate displacement vector */
 718             dx00             = _mm256_sub_pd(ix0,jx0);
 719             dy00             = _mm256_sub_pd(iy0,jy0);
 720             dz00             = _mm256_sub_pd(iz0,jz0);
 721             dx01             = _mm256_sub_pd(ix0,jx1);
 722             dy01             = _mm256_sub_pd(iy0,jy1);
 723             dz01             = _mm256_sub_pd(iz0,jz1);
 724             dx02             = _mm256_sub_pd(ix0,jx2);
 725             dy02             = _mm256_sub_pd(iy0,jy2);
 726             dz02             = _mm256_sub_pd(iz0,jz2);
 727             dx10             = _mm256_sub_pd(ix1,jx0);
 728             dy10             = _mm256_sub_pd(iy1,jy0);
 729             dz10             = _mm256_sub_pd(iz1,jz0);
 730             dx11             = _mm256_sub_pd(ix1,jx1);
 731             dy11             = _mm256_sub_pd(iy1,jy1);
 732             dz11             = _mm256_sub_pd(iz1,jz1);
 733             dx12             = _mm256_sub_pd(ix1,jx2);
 734             dy12             = _mm256_sub_pd(iy1,jy2);
 735             dz12             = _mm256_sub_pd(iz1,jz2);
 736             dx20             = _mm256_sub_pd(ix2,jx0);
 737             dy20             = _mm256_sub_pd(iy2,jy0);
 738             dz20             = _mm256_sub_pd(iz2,jz0);
 739             dx21             = _mm256_sub_pd(ix2,jx1);
 740             dy21             = _mm256_sub_pd(iy2,jy1);
 741             dz21             = _mm256_sub_pd(iz2,jz1);
 742             dx22             = _mm256_sub_pd(ix2,jx2);
 743             dy22             = _mm256_sub_pd(iy2,jy2);
 744             dz22             = _mm256_sub_pd(iz2,jz2);
 745
 746             /* Calculate squared distance and things based on it */
 747             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 748             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 749             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 750             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 751             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 752             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 753             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 754             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 755             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 756
 757             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
 758             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
 759             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
 760             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
 761             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 762             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 763             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
 764             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 765             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 766
 767             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 768             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 769             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 770             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 771             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 772             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 773             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 774             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 775             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 776
 777             fjx0             = _mm256_setzero_pd();
 778             fjy0             = _mm256_setzero_pd();
 779             fjz0             = _mm256_setzero_pd();
 780             fjx1             = _mm256_setzero_pd();
 781             fjy1             = _mm256_setzero_pd();
 782             fjz1             = _mm256_setzero_pd();
 783             fjx2             = _mm256_setzero_pd();
 784             fjy2             = _mm256_setzero_pd();
 785             fjz2             = _mm256_setzero_pd();
 786
 787             /**************************
 788              * CALCULATE INTERACTIONS *
 789              **************************/
 790
 791             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 792             {
 793
 794             r00              = _mm256_mul_pd(rsq00,rinv00);
 795             r00              = _mm256_andnot_pd(dummy_mask,r00);
 796
 797             /* REACTION-FIELD ELECTROSTATICS */
 798             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_add_pd(rinv00,_mm256_mul_pd(krf,rsq00)),crf));
 799             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
 800
 801             /* LENNARD-JONES DISPERSION/REPULSION */
 802
 803             rinvsix          = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 804             vvdw6            = _mm256_mul_pd(c6_00,rinvsix);
 805             vvdw12           = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
 806             vvdw             = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
 807             fvdw             = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
 808
 809             d                = _mm256_sub_pd(r00,rswitch);
 810             d                = _mm256_max_pd(d,_mm256_setzero_pd());
 811             d2               = _mm256_mul_pd(d,d);
 812             sw               = _mm256_add_pd(one,_mm256_mul_pd(d2,_mm256_mul_pd(d,_mm256_add_pd(swV3,_mm256_mul_pd(d,_mm256_add_pd(swV4,_mm256_mul_pd(d,swV5)))))));
 813
 814             dsw              = _mm256_mul_pd(d2,_mm256_add_pd(swF2,_mm256_mul_pd(d,_mm256_add_pd(swF3,_mm256_mul_pd(d,swF4)))));
 815
 816             /* Evaluate switch function */
 817             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 818             fvdw             = _mm256_sub_pd( _mm256_mul_pd(fvdw,sw) , _mm256_mul_pd(rinv00,_mm256_mul_pd(vvdw,dsw)) );
 819             vvdw             = _mm256_mul_pd(vvdw,sw);
 820             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
 821
 822             /* Update potential sum for this i atom from the interaction with this j atom. */
 823             velec            = _mm256_and_pd(velec,cutoff_mask);
 824             velec            = _mm256_andnot_pd(dummy_mask,velec);
 825             velecsum         = _mm256_add_pd(velecsum,velec);
 826             vvdw             = _mm256_and_pd(vvdw,cutoff_mask);
 827             vvdw             = _mm256_andnot_pd(dummy_mask,vvdw);
 828             vvdwsum          = _mm256_add_pd(vvdwsum,vvdw);
 829
 830             fscal            = _mm256_add_pd(felec,fvdw);
 831
 832             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 833
 834             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 835
 836             /* Calculate temporary vectorial force */
 837             tx               = _mm256_mul_pd(fscal,dx00);
 838             ty               = _mm256_mul_pd(fscal,dy00);
 839             tz               = _mm256_mul_pd(fscal,dz00);
 840
 841             /* Update vectorial force */
 842             fix0             = _mm256_add_pd(fix0,tx);
 843             fiy0             = _mm256_add_pd(fiy0,ty);
 844             fiz0             = _mm256_add_pd(fiz0,tz);
 845
 846             fjx0             = _mm256_add_pd(fjx0,tx);
 847             fjy0             = _mm256_add_pd(fjy0,ty);
 848             fjz0             = _mm256_add_pd(fjz0,tz);
 849
 850             }
 851
 852             /**************************
 853              * CALCULATE INTERACTIONS *
 854              **************************/
 855
 856             if (gmx_mm256_any_lt(rsq01,rcutoff2))
 857             {
 858
 859             /* REACTION-FIELD ELECTROSTATICS */
 860             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_add_pd(rinv01,_mm256_mul_pd(krf,rsq01)),crf));
 861             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
 862
 863             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
 864
 865             /* Update potential sum for this i atom from the interaction with this j atom. */
 866             velec            = _mm256_and_pd(velec,cutoff_mask);
 867             velec            = _mm256_andnot_pd(dummy_mask,velec);
 868             velecsum         = _mm256_add_pd(velecsum,velec);
 869
 870             fscal            = felec;
 871
 872             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 873
 874             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 875
 876             /* Calculate temporary vectorial force */
 877             tx               = _mm256_mul_pd(fscal,dx01);
 878             ty               = _mm256_mul_pd(fscal,dy01);
 879             tz               = _mm256_mul_pd(fscal,dz01);
 880
 881             /* Update vectorial force */
 882             fix0             = _mm256_add_pd(fix0,tx);
 883             fiy0             = _mm256_add_pd(fiy0,ty);
 884             fiz0             = _mm256_add_pd(fiz0,tz);
 885
 886             fjx1             = _mm256_add_pd(fjx1,tx);
 887             fjy1             = _mm256_add_pd(fjy1,ty);
 888             fjz1             = _mm256_add_pd(fjz1,tz);
 889
 890             }
 891
 892             /**************************
 893              * CALCULATE INTERACTIONS *
 894              **************************/
 895
 896             if (gmx_mm256_any_lt(rsq02,rcutoff2))
 897             {
 898
 899             /* REACTION-FIELD ELECTROSTATICS */
 900             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_add_pd(rinv02,_mm256_mul_pd(krf,rsq02)),crf));
 901             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
 902
 903             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
 904
 905             /* Update potential sum for this i atom from the interaction with this j atom. */
 906             velec            = _mm256_and_pd(velec,cutoff_mask);
 907             velec            = _mm256_andnot_pd(dummy_mask,velec);
 908             velecsum         = _mm256_add_pd(velecsum,velec);
 909
 910             fscal            = felec;
 911
 912             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 913
 914             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 915
 916             /* Calculate temporary vectorial force */
 917             tx               = _mm256_mul_pd(fscal,dx02);
 918             ty               = _mm256_mul_pd(fscal,dy02);
 919             tz               = _mm256_mul_pd(fscal,dz02);
 920
 921             /* Update vectorial force */
 922             fix0             = _mm256_add_pd(fix0,tx);
 923             fiy0             = _mm256_add_pd(fiy0,ty);
 924             fiz0             = _mm256_add_pd(fiz0,tz);
 925
 926             fjx2             = _mm256_add_pd(fjx2,tx);
 927             fjy2             = _mm256_add_pd(fjy2,ty);
 928             fjz2             = _mm256_add_pd(fjz2,tz);
 929
 930             }
 931
 932             /**************************
 933              * CALCULATE INTERACTIONS *
 934              **************************/
 935
 936             if (gmx_mm256_any_lt(rsq10,rcutoff2))
 937             {
 938
 939             /* REACTION-FIELD ELECTROSTATICS */
 940             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_add_pd(rinv10,_mm256_mul_pd(krf,rsq10)),crf));
 941             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
 942
 943             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
 944
 945             /* Update potential sum for this i atom from the interaction with this j atom. */
 946             velec            = _mm256_and_pd(velec,cutoff_mask);
 947             velec            = _mm256_andnot_pd(dummy_mask,velec);
 948             velecsum         = _mm256_add_pd(velecsum,velec);
 949
 950             fscal            = felec;
 951
 952             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 953
 954             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 955
 956             /* Calculate temporary vectorial force */
 957             tx               = _mm256_mul_pd(fscal,dx10);
 958             ty               = _mm256_mul_pd(fscal,dy10);
 959             tz               = _mm256_mul_pd(fscal,dz10);
 960
 961             /* Update vectorial force */
 962             fix1             = _mm256_add_pd(fix1,tx);
 963             fiy1             = _mm256_add_pd(fiy1,ty);
 964             fiz1             = _mm256_add_pd(fiz1,tz);
 965
 966             fjx0             = _mm256_add_pd(fjx0,tx);
 967             fjy0             = _mm256_add_pd(fjy0,ty);
 968             fjz0             = _mm256_add_pd(fjz0,tz);
 969
 970             }
 971
 972             /**************************
 973              * CALCULATE INTERACTIONS *
 974              **************************/
 975
 976             if (gmx_mm256_any_lt(rsq11,rcutoff2))
 977             {
 978
 979             /* REACTION-FIELD ELECTROSTATICS */
 980             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 981             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 982
 983             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
 984
 985             /* Update potential sum for this i atom from the interaction with this j atom. */
 986             velec            = _mm256_and_pd(velec,cutoff_mask);
 987             velec            = _mm256_andnot_pd(dummy_mask,velec);
 988             velecsum         = _mm256_add_pd(velecsum,velec);
 989
 990             fscal            = felec;
 991
 992             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 993
 994             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 995
 996             /* Calculate temporary vectorial force */
 997             tx               = _mm256_mul_pd(fscal,dx11);
 998             ty               = _mm256_mul_pd(fscal,dy11);
 999             tz               = _mm256_mul_pd(fscal,dz11);
1000
1001             /* Update vectorial force */
1002             fix1             = _mm256_add_pd(fix1,tx);
1003             fiy1             = _mm256_add_pd(fiy1,ty);
1004             fiz1             = _mm256_add_pd(fiz1,tz);
1005
1006             fjx1             = _mm256_add_pd(fjx1,tx);
1007             fjy1             = _mm256_add_pd(fjy1,ty);
1008             fjz1             = _mm256_add_pd(fjz1,tz);
1009
1010             }
1011
1012             /**************************
1013              * CALCULATE INTERACTIONS *
1014              **************************/
1015
1016             if (gmx_mm256_any_lt(rsq12,rcutoff2))
1017             {
1018
1019             /* REACTION-FIELD ELECTROSTATICS */
1020             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
1021             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1022
1023             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
1024
1025             /* Update potential sum for this i atom from the interaction with this j atom. */
1026             velec            = _mm256_and_pd(velec,cutoff_mask);
1027             velec            = _mm256_andnot_pd(dummy_mask,velec);
1028             velecsum         = _mm256_add_pd(velecsum,velec);
1029
1030             fscal            = felec;
1031
1032             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1033
1034             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1035
1036             /* Calculate temporary vectorial force */
1037             tx               = _mm256_mul_pd(fscal,dx12);
1038             ty               = _mm256_mul_pd(fscal,dy12);
1039             tz               = _mm256_mul_pd(fscal,dz12);
1040
1041             /* Update vectorial force */
1042             fix1             = _mm256_add_pd(fix1,tx);
1043             fiy1             = _mm256_add_pd(fiy1,ty);
1044             fiz1             = _mm256_add_pd(fiz1,tz);
1045
1046             fjx2             = _mm256_add_pd(fjx2,tx);
1047             fjy2             = _mm256_add_pd(fjy2,ty);
1048             fjz2             = _mm256_add_pd(fjz2,tz);
1049
1050             }
1051
1052             /**************************
1053              * CALCULATE INTERACTIONS *
1054              **************************/
1055
1056             if (gmx_mm256_any_lt(rsq20,rcutoff2))
1057             {
1058
1059             /* REACTION-FIELD ELECTROSTATICS */
1060             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_add_pd(rinv20,_mm256_mul_pd(krf,rsq20)),crf));
1061             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
1062
1063             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
1064
1065             /* Update potential sum for this i atom from the interaction with this j atom. */
1066             velec            = _mm256_and_pd(velec,cutoff_mask);
1067             velec            = _mm256_andnot_pd(dummy_mask,velec);
1068             velecsum         = _mm256_add_pd(velecsum,velec);
1069
1070             fscal            = felec;
1071
1072             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1073
1074             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1075
1076             /* Calculate temporary vectorial force */
1077             tx               = _mm256_mul_pd(fscal,dx20);
1078             ty               = _mm256_mul_pd(fscal,dy20);
1079             tz               = _mm256_mul_pd(fscal,dz20);
1080
1081             /* Update vectorial force */
1082             fix2             = _mm256_add_pd(fix2,tx);
1083             fiy2             = _mm256_add_pd(fiy2,ty);
1084             fiz2             = _mm256_add_pd(fiz2,tz);
1085
1086             fjx0             = _mm256_add_pd(fjx0,tx);
1087             fjy0             = _mm256_add_pd(fjy0,ty);
1088             fjz0             = _mm256_add_pd(fjz0,tz);
1089
1090             }
1091
1092             /**************************
1093              * CALCULATE INTERACTIONS *
1094              **************************/
1095
1096             if (gmx_mm256_any_lt(rsq21,rcutoff2))
1097             {
1098
1099             /* REACTION-FIELD ELECTROSTATICS */
1100             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
1101             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1102
1103             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
1104
1105             /* Update potential sum for this i atom from the interaction with this j atom. */
1106             velec            = _mm256_and_pd(velec,cutoff_mask);
1107             velec            = _mm256_andnot_pd(dummy_mask,velec);
1108             velecsum         = _mm256_add_pd(velecsum,velec);
1109
1110             fscal            = felec;
1111
1112             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1113
1114             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1115
1116             /* Calculate temporary vectorial force */
1117             tx               = _mm256_mul_pd(fscal,dx21);
1118             ty               = _mm256_mul_pd(fscal,dy21);
1119             tz               = _mm256_mul_pd(fscal,dz21);
1120
1121             /* Update vectorial force */
1122             fix2             = _mm256_add_pd(fix2,tx);
1123             fiy2             = _mm256_add_pd(fiy2,ty);
1124             fiz2             = _mm256_add_pd(fiz2,tz);
1125
1126             fjx1             = _mm256_add_pd(fjx1,tx);
1127             fjy1             = _mm256_add_pd(fjy1,ty);
1128             fjz1             = _mm256_add_pd(fjz1,tz);
1129
1130             }
1131
1132             /**************************
1133              * CALCULATE INTERACTIONS *
1134              **************************/
1135
1136             if (gmx_mm256_any_lt(rsq22,rcutoff2))
1137             {
1138
1139             /* REACTION-FIELD ELECTROSTATICS */
1140             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
1141             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1142
1143             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
1144
1145             /* Update potential sum for this i atom from the interaction with this j atom. */
1146             velec            = _mm256_and_pd(velec,cutoff_mask);
1147             velec            = _mm256_andnot_pd(dummy_mask,velec);
1148             velecsum         = _mm256_add_pd(velecsum,velec);
1149
1150             fscal            = felec;
1151
1152             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1153
1154             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1155
1156             /* Calculate temporary vectorial force */
1157             tx               = _mm256_mul_pd(fscal,dx22);
1158             ty               = _mm256_mul_pd(fscal,dy22);
1159             tz               = _mm256_mul_pd(fscal,dz22);
1160
1161             /* Update vectorial force */
1162             fix2             = _mm256_add_pd(fix2,tx);
1163             fiy2             = _mm256_add_pd(fiy2,ty);
1164             fiz2             = _mm256_add_pd(fiz2,tz);
1165
1166             fjx2             = _mm256_add_pd(fjx2,tx);
1167             fjy2             = _mm256_add_pd(fjy2,ty);
1168             fjz2             = _mm256_add_pd(fjz2,tz);
1169
1170             }
1171
1172             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1173             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1174             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1175             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1176
1177             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1178                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1179
1180             /* Inner loop uses 359 flops */
1181         }
1182
1183         /* End of innermost loop */
1184
1185         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1186                                                  f+i_coord_offset,fshift+i_shift_offset);
1187
1188         ggid                        = gid[iidx];
1189         /* Update potential energies */
1190         gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1191         gmx_mm256_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1192
1193         /* Increment number of inner iterations */
1194         inneriter                  += j_index_end - j_index_start;
1195
1196         /* Outer loop uses 20 flops */
1197     }
1198
1199     /* Increment number of outer iterations */
1200     outeriter        += nri;
1201
1202     /* Update outer/inner flops */
1203
1204     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_VF,outeriter*20 + inneriter*359);
1205 }
1206 /*
1207  * Gromacs nonbonded kernel:   nb_kernel_ElecRFCut_VdwLJSw_GeomW3W3_F_avx_256_double
1208  * Electrostatics interaction: ReactionField
1209  * VdW interaction:            LennardJones
1210  * Geometry:                   Water3-Water3
1211  * Calculate force/pot:        Force
1212  */
1213 void
1214 nb_kernel_ElecRFCut_VdwLJSw_GeomW3W3_F_avx_256_double
1215                     (t_nblist                    * gmx_restrict       nlist,
1216                      rvec                        * gmx_restrict          xx,
1217                      rvec                        * gmx_restrict          ff,
1218                      t_forcerec                  * gmx_restrict          fr,
1219                      t_mdatoms                   * gmx_restrict     mdatoms,
1220                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1221                      t_nrnb                      * gmx_restrict        nrnb)
1222 {
1223     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1224      * just 0 for non-waters.
1225      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
1226      * jnr indices corresponding to data put in the four positions in the SIMD register.
1227      */
1228     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1229     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1230     int              jnrA,jnrB,jnrC,jnrD;
1231     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1232     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
1233     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1234     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1235     real             rcutoff_scalar;
1236     real             *shiftvec,*fshift,*x,*f;
1237     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1238     real             scratch[4*DIM];
1239     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1240     real *           vdwioffsetptr0;
1241     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1242     real *           vdwioffsetptr1;
1243     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1244     real *           vdwioffsetptr2;
1245     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1246     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1247     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1248     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1249     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1250     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1251     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1252     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1253     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1254     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1255     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1256     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1257     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1258     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1259     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1260     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1261     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
1262     real             *charge;
1263     int              nvdwtype;
1264     __m256d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1265     int              *vdwtype;
1266     real             *vdwparam;
1267     __m256d          one_sixth   = _mm256_set1_pd(1.0/6.0);
1268     __m256d          one_twelfth = _mm256_set1_pd(1.0/12.0);
1269     __m256d          rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
1270     real             rswitch_scalar,d_scalar;
1271     __m256d          dummy_mask,cutoff_mask;
1272     __m128           tmpmask0,tmpmask1;
1273     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
1274     __m256d          one     = _mm256_set1_pd(1.0);
1275     __m256d          two     = _mm256_set1_pd(2.0);
1276     x                = xx[0];
1277     f                = ff[0];
1278
1279     nri              = nlist->nri;
1280     iinr             = nlist->iinr;
1281     jindex           = nlist->jindex;
1282     jjnr             = nlist->jjnr;
1283     shiftidx         = nlist->shift;
1284     gid              = nlist->gid;
1285     shiftvec         = fr->shift_vec[0];
1286     fshift           = fr->fshift[0];
1287     facel            = _mm256_set1_pd(fr->epsfac);
1288     charge           = mdatoms->chargeA;
1289     krf              = _mm256_set1_pd(fr->ic->k_rf);
1290     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
1291     crf              = _mm256_set1_pd(fr->ic->c_rf);
1292     nvdwtype         = fr->ntype;
1293     vdwparam         = fr->nbfp;
1294     vdwtype          = mdatoms->typeA;
1295
1296     /* Setup water-specific parameters */
1297     inr              = nlist->iinr[0];
1298     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
1299     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1300     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1301     vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
1302
1303     jq0              = _mm256_set1_pd(charge[inr+0]);
1304     jq1              = _mm256_set1_pd(charge[inr+1]);
1305     jq2              = _mm256_set1_pd(charge[inr+2]);
1306     vdwjidx0A        = 2*vdwtype[inr+0];
1307     qq00             = _mm256_mul_pd(iq0,jq0);
1308     c6_00            = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A]);
1309     c12_00           = _mm256_set1_pd(vdwioffsetptr0[vdwjidx0A+1]);
1310     qq01             = _mm256_mul_pd(iq0,jq1);
1311     qq02             = _mm256_mul_pd(iq0,jq2);
1312     qq10             = _mm256_mul_pd(iq1,jq0);
1313     qq11             = _mm256_mul_pd(iq1,jq1);
1314     qq12             = _mm256_mul_pd(iq1,jq2);
1315     qq20             = _mm256_mul_pd(iq2,jq0);
1316     qq21             = _mm256_mul_pd(iq2,jq1);
1317     qq22             = _mm256_mul_pd(iq2,jq2);
1318
1319     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1320     rcutoff_scalar   = fr->rcoulomb;
1321     rcutoff          = _mm256_set1_pd(rcutoff_scalar);
1322     rcutoff2         = _mm256_mul_pd(rcutoff,rcutoff);
1323
1324     rswitch_scalar   = fr->rvdw_switch;
1325     rswitch          = _mm256_set1_pd(rswitch_scalar);
1326     /* Setup switch parameters */
1327     d_scalar         = rcutoff_scalar-rswitch_scalar;
1328     d                = _mm256_set1_pd(d_scalar);
1329     swV3             = _mm256_set1_pd(-10.0/(d_scalar*d_scalar*d_scalar));
1330     swV4             = _mm256_set1_pd( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
1331     swV5             = _mm256_set1_pd( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
1332     swF2             = _mm256_set1_pd(-30.0/(d_scalar*d_scalar*d_scalar));
1333     swF3             = _mm256_set1_pd( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
1334     swF4             = _mm256_set1_pd(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
1335
1336     /* Avoid stupid compiler warnings */
1337     jnrA = jnrB = jnrC = jnrD = 0;
1338     j_coord_offsetA = 0;
1339     j_coord_offsetB = 0;
1340     j_coord_offsetC = 0;
1341     j_coord_offsetD = 0;
1342
1343     outeriter        = 0;
1344     inneriter        = 0;
1345
1346     for(iidx=0;iidx<4*DIM;iidx++)
1347     {
1348         scratch[iidx] = 0.0;
1349     }
1350
1351     /* Start outer loop over neighborlists */
1352     for(iidx=0; iidx<nri; iidx++)
1353     {
1354         /* Load shift vector for this list */
1355         i_shift_offset   = DIM*shiftidx[iidx];
1356
1357         /* Load limits for loop over neighbors */
1358         j_index_start    = jindex[iidx];
1359         j_index_end      = jindex[iidx+1];
1360
1361         /* Get outer coordinate index */
1362         inr              = iinr[iidx];
1363         i_coord_offset   = DIM*inr;
1364
1365         /* Load i particle coords and add shift vector */
1366         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1367                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1368
1369         fix0             = _mm256_setzero_pd();
1370         fiy0             = _mm256_setzero_pd();
1371         fiz0             = _mm256_setzero_pd();
1372         fix1             = _mm256_setzero_pd();
1373         fiy1             = _mm256_setzero_pd();
1374         fiz1             = _mm256_setzero_pd();
1375         fix2             = _mm256_setzero_pd();
1376         fiy2             = _mm256_setzero_pd();
1377         fiz2             = _mm256_setzero_pd();
1378
1379         /* Start inner kernel loop */
1380         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1381         {
1382
1383             /* Get j neighbor index, and coordinate index */
1384             jnrA             = jjnr[jidx];
1385             jnrB             = jjnr[jidx+1];
1386             jnrC             = jjnr[jidx+2];
1387             jnrD             = jjnr[jidx+3];
1388             j_coord_offsetA  = DIM*jnrA;
1389             j_coord_offsetB  = DIM*jnrB;
1390             j_coord_offsetC  = DIM*jnrC;
1391             j_coord_offsetD  = DIM*jnrD;
1392
1393             /* load j atom coordinates */
1394             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1395                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1396                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1397
1398             /* Calculate displacement vector */
1399             dx00             = _mm256_sub_pd(ix0,jx0);
1400             dy00             = _mm256_sub_pd(iy0,jy0);
1401             dz00             = _mm256_sub_pd(iz0,jz0);
1402             dx01             = _mm256_sub_pd(ix0,jx1);
1403             dy01             = _mm256_sub_pd(iy0,jy1);
1404             dz01             = _mm256_sub_pd(iz0,jz1);
1405             dx02             = _mm256_sub_pd(ix0,jx2);
1406             dy02             = _mm256_sub_pd(iy0,jy2);
1407             dz02             = _mm256_sub_pd(iz0,jz2);
1408             dx10             = _mm256_sub_pd(ix1,jx0);
1409             dy10             = _mm256_sub_pd(iy1,jy0);
1410             dz10             = _mm256_sub_pd(iz1,jz0);
1411             dx11             = _mm256_sub_pd(ix1,jx1);
1412             dy11             = _mm256_sub_pd(iy1,jy1);
1413             dz11             = _mm256_sub_pd(iz1,jz1);
1414             dx12             = _mm256_sub_pd(ix1,jx2);
1415             dy12             = _mm256_sub_pd(iy1,jy2);
1416             dz12             = _mm256_sub_pd(iz1,jz2);
1417             dx20             = _mm256_sub_pd(ix2,jx0);
1418             dy20             = _mm256_sub_pd(iy2,jy0);
1419             dz20             = _mm256_sub_pd(iz2,jz0);
1420             dx21             = _mm256_sub_pd(ix2,jx1);
1421             dy21             = _mm256_sub_pd(iy2,jy1);
1422             dz21             = _mm256_sub_pd(iz2,jz1);
1423             dx22             = _mm256_sub_pd(ix2,jx2);
1424             dy22             = _mm256_sub_pd(iy2,jy2);
1425             dz22             = _mm256_sub_pd(iz2,jz2);
1426
1427             /* Calculate squared distance and things based on it */
1428             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1429             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1430             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1431             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1432             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1433             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1434             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
1435             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1436             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1437
1438             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
1439             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
1440             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
1441             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
1442             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1443             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1444             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
1445             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1446             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1447
1448             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
1449             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
1450             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
1451             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
1452             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1453             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1454             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
1455             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1456             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1457
1458             fjx0             = _mm256_setzero_pd();
1459             fjy0             = _mm256_setzero_pd();
1460             fjz0             = _mm256_setzero_pd();
1461             fjx1             = _mm256_setzero_pd();
1462             fjy1             = _mm256_setzero_pd();
1463             fjz1             = _mm256_setzero_pd();
1464             fjx2             = _mm256_setzero_pd();
1465             fjy2             = _mm256_setzero_pd();
1466             fjz2             = _mm256_setzero_pd();
1467
1468             /**************************
1469              * CALCULATE INTERACTIONS *
1470              **************************/
1471
1472             if (gmx_mm256_any_lt(rsq00,rcutoff2))
1473             {
1474
1475             r00              = _mm256_mul_pd(rsq00,rinv00);
1476
1477             /* REACTION-FIELD ELECTROSTATICS */
1478             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
1479
1480             /* LENNARD-JONES DISPERSION/REPULSION */
1481
1482             rinvsix          = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1483             vvdw6            = _mm256_mul_pd(c6_00,rinvsix);
1484             vvdw12           = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
1485             vvdw             = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
1486             fvdw             = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
1487
1488             d                = _mm256_sub_pd(r00,rswitch);
1489             d                = _mm256_max_pd(d,_mm256_setzero_pd());
1490             d2               = _mm256_mul_pd(d,d);
1491             sw               = _mm256_add_pd(one,_mm256_mul_pd(d2,_mm256_mul_pd(d,_mm256_add_pd(swV3,_mm256_mul_pd(d,_mm256_add_pd(swV4,_mm256_mul_pd(d,swV5)))))));
1492
1493             dsw              = _mm256_mul_pd(d2,_mm256_add_pd(swF2,_mm256_mul_pd(d,_mm256_add_pd(swF3,_mm256_mul_pd(d,swF4)))));
1494
1495             /* Evaluate switch function */
1496             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1497             fvdw             = _mm256_sub_pd( _mm256_mul_pd(fvdw,sw) , _mm256_mul_pd(rinv00,_mm256_mul_pd(vvdw,dsw)) );
1498             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
1499
1500             fscal            = _mm256_add_pd(felec,fvdw);
1501
1502             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1503
1504             /* Calculate temporary vectorial force */
1505             tx               = _mm256_mul_pd(fscal,dx00);
1506             ty               = _mm256_mul_pd(fscal,dy00);
1507             tz               = _mm256_mul_pd(fscal,dz00);
1508
1509             /* Update vectorial force */
1510             fix0             = _mm256_add_pd(fix0,tx);
1511             fiy0             = _mm256_add_pd(fiy0,ty);
1512             fiz0             = _mm256_add_pd(fiz0,tz);
1513
1514             fjx0             = _mm256_add_pd(fjx0,tx);
1515             fjy0             = _mm256_add_pd(fjy0,ty);
1516             fjz0             = _mm256_add_pd(fjz0,tz);
1517
1518             }
1519
1520             /**************************
1521              * CALCULATE INTERACTIONS *
1522              **************************/
1523
1524             if (gmx_mm256_any_lt(rsq01,rcutoff2))
1525             {
1526
1527             /* REACTION-FIELD ELECTROSTATICS */
1528             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
1529
1530             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
1531
1532             fscal            = felec;
1533
1534             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1535
1536             /* Calculate temporary vectorial force */
1537             tx               = _mm256_mul_pd(fscal,dx01);
1538             ty               = _mm256_mul_pd(fscal,dy01);
1539             tz               = _mm256_mul_pd(fscal,dz01);
1540
1541             /* Update vectorial force */
1542             fix0             = _mm256_add_pd(fix0,tx);
1543             fiy0             = _mm256_add_pd(fiy0,ty);
1544             fiz0             = _mm256_add_pd(fiz0,tz);
1545
1546             fjx1             = _mm256_add_pd(fjx1,tx);
1547             fjy1             = _mm256_add_pd(fjy1,ty);
1548             fjz1             = _mm256_add_pd(fjz1,tz);
1549
1550             }
1551
1552             /**************************
1553              * CALCULATE INTERACTIONS *
1554              **************************/
1555
1556             if (gmx_mm256_any_lt(rsq02,rcutoff2))
1557             {
1558
1559             /* REACTION-FIELD ELECTROSTATICS */
1560             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
1561
1562             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
1563
1564             fscal            = felec;
1565
1566             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1567
1568             /* Calculate temporary vectorial force */
1569             tx               = _mm256_mul_pd(fscal,dx02);
1570             ty               = _mm256_mul_pd(fscal,dy02);
1571             tz               = _mm256_mul_pd(fscal,dz02);
1572
1573             /* Update vectorial force */
1574             fix0             = _mm256_add_pd(fix0,tx);
1575             fiy0             = _mm256_add_pd(fiy0,ty);
1576             fiz0             = _mm256_add_pd(fiz0,tz);
1577
1578             fjx2             = _mm256_add_pd(fjx2,tx);
1579             fjy2             = _mm256_add_pd(fjy2,ty);
1580             fjz2             = _mm256_add_pd(fjz2,tz);
1581
1582             }
1583
1584             /**************************
1585              * CALCULATE INTERACTIONS *
1586              **************************/
1587
1588             if (gmx_mm256_any_lt(rsq10,rcutoff2))
1589             {
1590
1591             /* REACTION-FIELD ELECTROSTATICS */
1592             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
1593
1594             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
1595
1596             fscal            = felec;
1597
1598             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1599
1600             /* Calculate temporary vectorial force */
1601             tx               = _mm256_mul_pd(fscal,dx10);
1602             ty               = _mm256_mul_pd(fscal,dy10);
1603             tz               = _mm256_mul_pd(fscal,dz10);
1604
1605             /* Update vectorial force */
1606             fix1             = _mm256_add_pd(fix1,tx);
1607             fiy1             = _mm256_add_pd(fiy1,ty);
1608             fiz1             = _mm256_add_pd(fiz1,tz);
1609
1610             fjx0             = _mm256_add_pd(fjx0,tx);
1611             fjy0             = _mm256_add_pd(fjy0,ty);
1612             fjz0             = _mm256_add_pd(fjz0,tz);
1613
1614             }
1615
1616             /**************************
1617              * CALCULATE INTERACTIONS *
1618              **************************/
1619
1620             if (gmx_mm256_any_lt(rsq11,rcutoff2))
1621             {
1622
1623             /* REACTION-FIELD ELECTROSTATICS */
1624             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1625
1626             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
1627
1628             fscal            = felec;
1629
1630             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1631
1632             /* Calculate temporary vectorial force */
1633             tx               = _mm256_mul_pd(fscal,dx11);
1634             ty               = _mm256_mul_pd(fscal,dy11);
1635             tz               = _mm256_mul_pd(fscal,dz11);
1636
1637             /* Update vectorial force */
1638             fix1             = _mm256_add_pd(fix1,tx);
1639             fiy1             = _mm256_add_pd(fiy1,ty);
1640             fiz1             = _mm256_add_pd(fiz1,tz);
1641
1642             fjx1             = _mm256_add_pd(fjx1,tx);
1643             fjy1             = _mm256_add_pd(fjy1,ty);
1644             fjz1             = _mm256_add_pd(fjz1,tz);
1645
1646             }
1647
1648             /**************************
1649              * CALCULATE INTERACTIONS *
1650              **************************/
1651
1652             if (gmx_mm256_any_lt(rsq12,rcutoff2))
1653             {
1654
1655             /* REACTION-FIELD ELECTROSTATICS */
1656             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1657
1658             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
1659
1660             fscal            = felec;
1661
1662             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1663
1664             /* Calculate temporary vectorial force */
1665             tx               = _mm256_mul_pd(fscal,dx12);
1666             ty               = _mm256_mul_pd(fscal,dy12);
1667             tz               = _mm256_mul_pd(fscal,dz12);
1668
1669             /* Update vectorial force */
1670             fix1             = _mm256_add_pd(fix1,tx);
1671             fiy1             = _mm256_add_pd(fiy1,ty);
1672             fiz1             = _mm256_add_pd(fiz1,tz);
1673
1674             fjx2             = _mm256_add_pd(fjx2,tx);
1675             fjy2             = _mm256_add_pd(fjy2,ty);
1676             fjz2             = _mm256_add_pd(fjz2,tz);
1677
1678             }
1679
1680             /**************************
1681              * CALCULATE INTERACTIONS *
1682              **************************/
1683
1684             if (gmx_mm256_any_lt(rsq20,rcutoff2))
1685             {
1686
1687             /* REACTION-FIELD ELECTROSTATICS */
1688             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
1689
1690             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
1691
1692             fscal            = felec;
1693
1694             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1695
1696             /* Calculate temporary vectorial force */
1697             tx               = _mm256_mul_pd(fscal,dx20);
1698             ty               = _mm256_mul_pd(fscal,dy20);
1699             tz               = _mm256_mul_pd(fscal,dz20);
1700
1701             /* Update vectorial force */
1702             fix2             = _mm256_add_pd(fix2,tx);
1703             fiy2             = _mm256_add_pd(fiy2,ty);
1704             fiz2             = _mm256_add_pd(fiz2,tz);
1705
1706             fjx0             = _mm256_add_pd(fjx0,tx);
1707             fjy0             = _mm256_add_pd(fjy0,ty);
1708             fjz0             = _mm256_add_pd(fjz0,tz);
1709
1710             }
1711
1712             /**************************
1713              * CALCULATE INTERACTIONS *
1714              **************************/
1715
1716             if (gmx_mm256_any_lt(rsq21,rcutoff2))
1717             {
1718
1719             /* REACTION-FIELD ELECTROSTATICS */
1720             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1721
1722             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
1723
1724             fscal            = felec;
1725
1726             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1727
1728             /* Calculate temporary vectorial force */
1729             tx               = _mm256_mul_pd(fscal,dx21);
1730             ty               = _mm256_mul_pd(fscal,dy21);
1731             tz               = _mm256_mul_pd(fscal,dz21);
1732
1733             /* Update vectorial force */
1734             fix2             = _mm256_add_pd(fix2,tx);
1735             fiy2             = _mm256_add_pd(fiy2,ty);
1736             fiz2             = _mm256_add_pd(fiz2,tz);
1737
1738             fjx1             = _mm256_add_pd(fjx1,tx);
1739             fjy1             = _mm256_add_pd(fjy1,ty);
1740             fjz1             = _mm256_add_pd(fjz1,tz);
1741
1742             }
1743
1744             /**************************
1745              * CALCULATE INTERACTIONS *
1746              **************************/
1747
1748             if (gmx_mm256_any_lt(rsq22,rcutoff2))
1749             {
1750
1751             /* REACTION-FIELD ELECTROSTATICS */
1752             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1753
1754             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
1755
1756             fscal            = felec;
1757
1758             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1759
1760             /* Calculate temporary vectorial force */
1761             tx               = _mm256_mul_pd(fscal,dx22);
1762             ty               = _mm256_mul_pd(fscal,dy22);
1763             tz               = _mm256_mul_pd(fscal,dz22);
1764
1765             /* Update vectorial force */
1766             fix2             = _mm256_add_pd(fix2,tx);
1767             fiy2             = _mm256_add_pd(fiy2,ty);
1768             fiz2             = _mm256_add_pd(fiz2,tz);
1769
1770             fjx2             = _mm256_add_pd(fjx2,tx);
1771             fjy2             = _mm256_add_pd(fjy2,ty);
1772             fjz2             = _mm256_add_pd(fjz2,tz);
1773
1774             }
1775
1776             fjptrA             = f+j_coord_offsetA;
1777             fjptrB             = f+j_coord_offsetB;
1778             fjptrC             = f+j_coord_offsetC;
1779             fjptrD             = f+j_coord_offsetD;
1780
1781             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1782                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1783
1784             /* Inner loop uses 301 flops */
1785         }
1786
1787         if(jidx<j_index_end)
1788         {
1789
1790             /* Get j neighbor index, and coordinate index */
1791             jnrlistA         = jjnr[jidx];
1792             jnrlistB         = jjnr[jidx+1];
1793             jnrlistC         = jjnr[jidx+2];
1794             jnrlistD         = jjnr[jidx+3];
1795             /* Sign of each element will be negative for non-real atoms.
1796              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1797              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
1798              */
1799             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1800
1801             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
1802             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
1803             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
1804
1805             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1806             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1807             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1808             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1809             j_coord_offsetA  = DIM*jnrA;
1810             j_coord_offsetB  = DIM*jnrB;
1811             j_coord_offsetC  = DIM*jnrC;
1812             j_coord_offsetD  = DIM*jnrD;
1813
1814             /* load j atom coordinates */
1815             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1816                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1817                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1818
1819             /* Calculate displacement vector */
1820             dx00             = _mm256_sub_pd(ix0,jx0);
1821             dy00             = _mm256_sub_pd(iy0,jy0);
1822             dz00             = _mm256_sub_pd(iz0,jz0);
1823             dx01             = _mm256_sub_pd(ix0,jx1);
1824             dy01             = _mm256_sub_pd(iy0,jy1);
1825             dz01             = _mm256_sub_pd(iz0,jz1);
1826             dx02             = _mm256_sub_pd(ix0,jx2);
1827             dy02             = _mm256_sub_pd(iy0,jy2);
1828             dz02             = _mm256_sub_pd(iz0,jz2);
1829             dx10             = _mm256_sub_pd(ix1,jx0);
1830             dy10             = _mm256_sub_pd(iy1,jy0);
1831             dz10             = _mm256_sub_pd(iz1,jz0);
1832             dx11             = _mm256_sub_pd(ix1,jx1);
1833             dy11             = _mm256_sub_pd(iy1,jy1);
1834             dz11             = _mm256_sub_pd(iz1,jz1);
1835             dx12             = _mm256_sub_pd(ix1,jx2);
1836             dy12             = _mm256_sub_pd(iy1,jy2);
1837             dz12             = _mm256_sub_pd(iz1,jz2);
1838             dx20             = _mm256_sub_pd(ix2,jx0);
1839             dy20             = _mm256_sub_pd(iy2,jy0);
1840             dz20             = _mm256_sub_pd(iz2,jz0);
1841             dx21             = _mm256_sub_pd(ix2,jx1);
1842             dy21             = _mm256_sub_pd(iy2,jy1);
1843             dz21             = _mm256_sub_pd(iz2,jz1);
1844             dx22             = _mm256_sub_pd(ix2,jx2);
1845             dy22             = _mm256_sub_pd(iy2,jy2);
1846             dz22             = _mm256_sub_pd(iz2,jz2);
1847
1848             /* Calculate squared distance and things based on it */
1849             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1850             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1851             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1852             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1853             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1854             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1855             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
1856             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1857             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1858
1859             rinv00           = gmx_mm256_invsqrt_pd(rsq00);
1860             rinv01           = gmx_mm256_invsqrt_pd(rsq01);
1861             rinv02           = gmx_mm256_invsqrt_pd(rsq02);
1862             rinv10           = gmx_mm256_invsqrt_pd(rsq10);
1863             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1864             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1865             rinv20           = gmx_mm256_invsqrt_pd(rsq20);
1866             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1867             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1868
1869             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
1870             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
1871             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
1872             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
1873             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1874             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1875             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
1876             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1877             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1878
1879             fjx0             = _mm256_setzero_pd();
1880             fjy0             = _mm256_setzero_pd();
1881             fjz0             = _mm256_setzero_pd();
1882             fjx1             = _mm256_setzero_pd();
1883             fjy1             = _mm256_setzero_pd();
1884             fjz1             = _mm256_setzero_pd();
1885             fjx2             = _mm256_setzero_pd();
1886             fjy2             = _mm256_setzero_pd();
1887             fjz2             = _mm256_setzero_pd();
1888
1889             /**************************
1890              * CALCULATE INTERACTIONS *
1891              **************************/
1892
1893             if (gmx_mm256_any_lt(rsq00,rcutoff2))
1894             {
1895
1896             r00              = _mm256_mul_pd(rsq00,rinv00);
1897             r00              = _mm256_andnot_pd(dummy_mask,r00);
1898
1899             /* REACTION-FIELD ELECTROSTATICS */
1900             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
1901
1902             /* LENNARD-JONES DISPERSION/REPULSION */
1903
1904             rinvsix          = _mm256_mul_pd(_mm256_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1905             vvdw6            = _mm256_mul_pd(c6_00,rinvsix);
1906             vvdw12           = _mm256_mul_pd(c12_00,_mm256_mul_pd(rinvsix,rinvsix));
1907             vvdw             = _mm256_sub_pd( _mm256_mul_pd(vvdw12,one_twelfth) , _mm256_mul_pd(vvdw6,one_sixth) );
1908             fvdw             = _mm256_mul_pd(_mm256_sub_pd(vvdw12,vvdw6),rinvsq00);
1909
1910             d                = _mm256_sub_pd(r00,rswitch);
1911             d                = _mm256_max_pd(d,_mm256_setzero_pd());
1912             d2               = _mm256_mul_pd(d,d);
1913             sw               = _mm256_add_pd(one,_mm256_mul_pd(d2,_mm256_mul_pd(d,_mm256_add_pd(swV3,_mm256_mul_pd(d,_mm256_add_pd(swV4,_mm256_mul_pd(d,swV5)))))));
1914
1915             dsw              = _mm256_mul_pd(d2,_mm256_add_pd(swF2,_mm256_mul_pd(d,_mm256_add_pd(swF3,_mm256_mul_pd(d,swF4)))));
1916
1917             /* Evaluate switch function */
1918             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1919             fvdw             = _mm256_sub_pd( _mm256_mul_pd(fvdw,sw) , _mm256_mul_pd(rinv00,_mm256_mul_pd(vvdw,dsw)) );
1920             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
1921
1922             fscal            = _mm256_add_pd(felec,fvdw);
1923
1924             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1925
1926             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1927
1928             /* Calculate temporary vectorial force */
1929             tx               = _mm256_mul_pd(fscal,dx00);
1930             ty               = _mm256_mul_pd(fscal,dy00);
1931             tz               = _mm256_mul_pd(fscal,dz00);
1932
1933             /* Update vectorial force */
1934             fix0             = _mm256_add_pd(fix0,tx);
1935             fiy0             = _mm256_add_pd(fiy0,ty);
1936             fiz0             = _mm256_add_pd(fiz0,tz);
1937
1938             fjx0             = _mm256_add_pd(fjx0,tx);
1939             fjy0             = _mm256_add_pd(fjy0,ty);
1940             fjz0             = _mm256_add_pd(fjz0,tz);
1941
1942             }
1943
1944             /**************************
1945              * CALCULATE INTERACTIONS *
1946              **************************/
1947
1948             if (gmx_mm256_any_lt(rsq01,rcutoff2))
1949             {
1950
1951             /* REACTION-FIELD ELECTROSTATICS */
1952             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
1953
1954             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
1955
1956             fscal            = felec;
1957
1958             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1959
1960             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1961
1962             /* Calculate temporary vectorial force */
1963             tx               = _mm256_mul_pd(fscal,dx01);
1964             ty               = _mm256_mul_pd(fscal,dy01);
1965             tz               = _mm256_mul_pd(fscal,dz01);
1966
1967             /* Update vectorial force */
1968             fix0             = _mm256_add_pd(fix0,tx);
1969             fiy0             = _mm256_add_pd(fiy0,ty);
1970             fiz0             = _mm256_add_pd(fiz0,tz);
1971
1972             fjx1             = _mm256_add_pd(fjx1,tx);
1973             fjy1             = _mm256_add_pd(fjy1,ty);
1974             fjz1             = _mm256_add_pd(fjz1,tz);
1975
1976             }
1977
1978             /**************************
1979              * CALCULATE INTERACTIONS *
1980              **************************/
1981
1982             if (gmx_mm256_any_lt(rsq02,rcutoff2))
1983             {
1984
1985             /* REACTION-FIELD ELECTROSTATICS */
1986             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
1987
1988             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
1989
1990             fscal            = felec;
1991
1992             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1993
1994             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1995
1996             /* Calculate temporary vectorial force */
1997             tx               = _mm256_mul_pd(fscal,dx02);
1998             ty               = _mm256_mul_pd(fscal,dy02);
1999             tz               = _mm256_mul_pd(fscal,dz02);
2000
2001             /* Update vectorial force */
2002             fix0             = _mm256_add_pd(fix0,tx);
2003             fiy0             = _mm256_add_pd(fiy0,ty);
2004             fiz0             = _mm256_add_pd(fiz0,tz);
2005
2006             fjx2             = _mm256_add_pd(fjx2,tx);
2007             fjy2             = _mm256_add_pd(fjy2,ty);
2008             fjz2             = _mm256_add_pd(fjz2,tz);
2009
2010             }
2011
2012             /**************************
2013              * CALCULATE INTERACTIONS *
2014              **************************/
2015
2016             if (gmx_mm256_any_lt(rsq10,rcutoff2))
2017             {
2018
2019             /* REACTION-FIELD ELECTROSTATICS */
2020             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
2021
2022             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
2023
2024             fscal            = felec;
2025
2026             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2027
2028             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2029
2030             /* Calculate temporary vectorial force */
2031             tx               = _mm256_mul_pd(fscal,dx10);
2032             ty               = _mm256_mul_pd(fscal,dy10);
2033             tz               = _mm256_mul_pd(fscal,dz10);
2034
2035             /* Update vectorial force */
2036             fix1             = _mm256_add_pd(fix1,tx);
2037             fiy1             = _mm256_add_pd(fiy1,ty);
2038             fiz1             = _mm256_add_pd(fiz1,tz);
2039
2040             fjx0             = _mm256_add_pd(fjx0,tx);
2041             fjy0             = _mm256_add_pd(fjy0,ty);
2042             fjz0             = _mm256_add_pd(fjz0,tz);
2043
2044             }
2045
2046             /**************************
2047              * CALCULATE INTERACTIONS *
2048              **************************/
2049
2050             if (gmx_mm256_any_lt(rsq11,rcutoff2))
2051             {
2052
2053             /* REACTION-FIELD ELECTROSTATICS */
2054             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
2055
2056             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
2057
2058             fscal            = felec;
2059
2060             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2061
2062             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2063
2064             /* Calculate temporary vectorial force */
2065             tx               = _mm256_mul_pd(fscal,dx11);
2066             ty               = _mm256_mul_pd(fscal,dy11);
2067             tz               = _mm256_mul_pd(fscal,dz11);
2068
2069             /* Update vectorial force */
2070             fix1             = _mm256_add_pd(fix1,tx);
2071             fiy1             = _mm256_add_pd(fiy1,ty);
2072             fiz1             = _mm256_add_pd(fiz1,tz);
2073
2074             fjx1             = _mm256_add_pd(fjx1,tx);
2075             fjy1             = _mm256_add_pd(fjy1,ty);
2076             fjz1             = _mm256_add_pd(fjz1,tz);
2077
2078             }
2079
2080             /**************************
2081              * CALCULATE INTERACTIONS *
2082              **************************/
2083
2084             if (gmx_mm256_any_lt(rsq12,rcutoff2))
2085             {
2086
2087             /* REACTION-FIELD ELECTROSTATICS */
2088             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
2089
2090             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
2091
2092             fscal            = felec;
2093
2094             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2095
2096             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2097
2098             /* Calculate temporary vectorial force */
2099             tx               = _mm256_mul_pd(fscal,dx12);
2100             ty               = _mm256_mul_pd(fscal,dy12);
2101             tz               = _mm256_mul_pd(fscal,dz12);
2102
2103             /* Update vectorial force */
2104             fix1             = _mm256_add_pd(fix1,tx);
2105             fiy1             = _mm256_add_pd(fiy1,ty);
2106             fiz1             = _mm256_add_pd(fiz1,tz);
2107
2108             fjx2             = _mm256_add_pd(fjx2,tx);
2109             fjy2             = _mm256_add_pd(fjy2,ty);
2110             fjz2             = _mm256_add_pd(fjz2,tz);
2111
2112             }
2113
2114             /**************************
2115              * CALCULATE INTERACTIONS *
2116              **************************/
2117
2118             if (gmx_mm256_any_lt(rsq20,rcutoff2))
2119             {
2120
2121             /* REACTION-FIELD ELECTROSTATICS */
2122             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
2123
2124             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
2125
2126             fscal            = felec;
2127
2128             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2129
2130             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2131
2132             /* Calculate temporary vectorial force */
2133             tx               = _mm256_mul_pd(fscal,dx20);
2134             ty               = _mm256_mul_pd(fscal,dy20);
2135             tz               = _mm256_mul_pd(fscal,dz20);
2136
2137             /* Update vectorial force */
2138             fix2             = _mm256_add_pd(fix2,tx);
2139             fiy2             = _mm256_add_pd(fiy2,ty);
2140             fiz2             = _mm256_add_pd(fiz2,tz);
2141
2142             fjx0             = _mm256_add_pd(fjx0,tx);
2143             fjy0             = _mm256_add_pd(fjy0,ty);
2144             fjz0             = _mm256_add_pd(fjz0,tz);
2145
2146             }
2147
2148             /**************************
2149              * CALCULATE INTERACTIONS *
2150              **************************/
2151
2152             if (gmx_mm256_any_lt(rsq21,rcutoff2))
2153             {
2154
2155             /* REACTION-FIELD ELECTROSTATICS */
2156             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
2157
2158             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
2159
2160             fscal            = felec;
2161
2162             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2163
2164             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2165
2166             /* Calculate temporary vectorial force */
2167             tx               = _mm256_mul_pd(fscal,dx21);
2168             ty               = _mm256_mul_pd(fscal,dy21);
2169             tz               = _mm256_mul_pd(fscal,dz21);
2170
2171             /* Update vectorial force */
2172             fix2             = _mm256_add_pd(fix2,tx);
2173             fiy2             = _mm256_add_pd(fiy2,ty);
2174             fiz2             = _mm256_add_pd(fiz2,tz);
2175
2176             fjx1             = _mm256_add_pd(fjx1,tx);
2177             fjy1             = _mm256_add_pd(fjy1,ty);
2178             fjz1             = _mm256_add_pd(fjz1,tz);
2179
2180             }
2181
2182             /**************************
2183              * CALCULATE INTERACTIONS *
2184              **************************/
2185
2186             if (gmx_mm256_any_lt(rsq22,rcutoff2))
2187             {
2188
2189             /* REACTION-FIELD ELECTROSTATICS */
2190             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
2191
2192             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
2193
2194             fscal            = felec;
2195
2196             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2197
2198             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2199
2200             /* Calculate temporary vectorial force */
2201             tx               = _mm256_mul_pd(fscal,dx22);
2202             ty               = _mm256_mul_pd(fscal,dy22);
2203             tz               = _mm256_mul_pd(fscal,dz22);
2204
2205             /* Update vectorial force */
2206             fix2             = _mm256_add_pd(fix2,tx);
2207             fiy2             = _mm256_add_pd(fiy2,ty);
2208             fiz2             = _mm256_add_pd(fiz2,tz);
2209
2210             fjx2             = _mm256_add_pd(fjx2,tx);
2211             fjy2             = _mm256_add_pd(fjy2,ty);
2212             fjz2             = _mm256_add_pd(fjz2,tz);
2213
2214             }
2215
2216             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2217             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2218             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2219             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2220
2221             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2222                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2223
2224             /* Inner loop uses 302 flops */
2225         }
2226
2227         /* End of innermost loop */
2228
2229         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
2230                                                  f+i_coord_offset,fshift+i_shift_offset);
2231
2232         /* Increment number of inner iterations */
2233         inneriter                  += j_index_end - j_index_start;
2234
2235         /* Outer loop uses 18 flops */
2236     }
2237
2238     /* Increment number of outer iterations */
2239     outeriter        += nri;
2240
2241     /* Update outer/inner flops */
2242
2243     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_F,outeriter*18 + inneriter*302);
2244 }