src/gromacs/gmxlib/nonbonded/nb_kernel_avx_256_double/nb_kernel_ElecRFCut_VdwNone_GeomW3W3_avx_256_double.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2017,2018, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS 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/gmxlib/nrnb.h"
  46
  47 #include "kernelutil_x86_avx_256_double.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecRFCut_VdwNone_GeomW3W3_VF_avx_256_double
  51  * Electrostatics interaction: ReactionField
  52  * VdW interaction:            None
  53  * Geometry:                   Water3-Water3
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecRFCut_VdwNone_GeomW3W3_VF_avx_256_double
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      struct t_forcerec           * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  67      * just 0 for non-waters.
  68      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
  69      * jnr indices corresponding to data put in the four positions in the SIMD register.
  70      */
  71     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  72     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  73     int              jnrA,jnrB,jnrC,jnrD;
  74     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  75     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  76     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  77     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  78     real             rcutoff_scalar;
  79     real             *shiftvec,*fshift,*x,*f;
  80     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  81     real             scratch[4*DIM];
  82     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  83     real *           vdwioffsetptr0;
  84     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  85     real *           vdwioffsetptr1;
  86     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  87     real *           vdwioffsetptr2;
  88     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  89     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  90     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  91     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  92     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  93     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  94     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  95     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  96     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
  97     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
  98     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  99     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 100     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 101     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 102     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 103     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 104     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
 105     real             *charge;
 106     __m256d          dummy_mask,cutoff_mask;
 107     __m128           tmpmask0,tmpmask1;
 108     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
 109     __m256d          one     = _mm256_set1_pd(1.0);
 110     __m256d          two     = _mm256_set1_pd(2.0);
 111     x                = xx[0];
 112     f                = ff[0];
 113
 114     nri              = nlist->nri;
 115     iinr             = nlist->iinr;
 116     jindex           = nlist->jindex;
 117     jjnr             = nlist->jjnr;
 118     shiftidx         = nlist->shift;
 119     gid              = nlist->gid;
 120     shiftvec         = fr->shift_vec[0];
 121     fshift           = fr->fshift[0];
 122     facel            = _mm256_set1_pd(fr->ic->epsfac);
 123     charge           = mdatoms->chargeA;
 124     krf              = _mm256_set1_pd(fr->ic->k_rf);
 125     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
 126     crf              = _mm256_set1_pd(fr->ic->c_rf);
 127
 128     /* Setup water-specific parameters */
 129     inr              = nlist->iinr[0];
 130     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
 131     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
 132     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
 133
 134     jq0              = _mm256_set1_pd(charge[inr+0]);
 135     jq1              = _mm256_set1_pd(charge[inr+1]);
 136     jq2              = _mm256_set1_pd(charge[inr+2]);
 137     qq00             = _mm256_mul_pd(iq0,jq0);
 138     qq01             = _mm256_mul_pd(iq0,jq1);
 139     qq02             = _mm256_mul_pd(iq0,jq2);
 140     qq10             = _mm256_mul_pd(iq1,jq0);
 141     qq11             = _mm256_mul_pd(iq1,jq1);
 142     qq12             = _mm256_mul_pd(iq1,jq2);
 143     qq20             = _mm256_mul_pd(iq2,jq0);
 144     qq21             = _mm256_mul_pd(iq2,jq1);
 145     qq22             = _mm256_mul_pd(iq2,jq2);
 146
 147     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 148     rcutoff_scalar   = fr->ic->rcoulomb;
 149     rcutoff          = _mm256_set1_pd(rcutoff_scalar);
 150     rcutoff2         = _mm256_mul_pd(rcutoff,rcutoff);
 151
 152     /* Avoid stupid compiler warnings */
 153     jnrA = jnrB = jnrC = jnrD = 0;
 154     j_coord_offsetA = 0;
 155     j_coord_offsetB = 0;
 156     j_coord_offsetC = 0;
 157     j_coord_offsetD = 0;
 158
 159     outeriter        = 0;
 160     inneriter        = 0;
 161
 162     for(iidx=0;iidx<4*DIM;iidx++)
 163     {
 164         scratch[iidx] = 0.0;
 165     }
 166
 167     /* Start outer loop over neighborlists */
 168     for(iidx=0; iidx<nri; iidx++)
 169     {
 170         /* Load shift vector for this list */
 171         i_shift_offset   = DIM*shiftidx[iidx];
 172
 173         /* Load limits for loop over neighbors */
 174         j_index_start    = jindex[iidx];
 175         j_index_end      = jindex[iidx+1];
 176
 177         /* Get outer coordinate index */
 178         inr              = iinr[iidx];
 179         i_coord_offset   = DIM*inr;
 180
 181         /* Load i particle coords and add shift vector */
 182         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 183                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 184
 185         fix0             = _mm256_setzero_pd();
 186         fiy0             = _mm256_setzero_pd();
 187         fiz0             = _mm256_setzero_pd();
 188         fix1             = _mm256_setzero_pd();
 189         fiy1             = _mm256_setzero_pd();
 190         fiz1             = _mm256_setzero_pd();
 191         fix2             = _mm256_setzero_pd();
 192         fiy2             = _mm256_setzero_pd();
 193         fiz2             = _mm256_setzero_pd();
 194
 195         /* Reset potential sums */
 196         velecsum         = _mm256_setzero_pd();
 197
 198         /* Start inner kernel loop */
 199         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 200         {
 201
 202             /* Get j neighbor index, and coordinate index */
 203             jnrA             = jjnr[jidx];
 204             jnrB             = jjnr[jidx+1];
 205             jnrC             = jjnr[jidx+2];
 206             jnrD             = jjnr[jidx+3];
 207             j_coord_offsetA  = DIM*jnrA;
 208             j_coord_offsetB  = DIM*jnrB;
 209             j_coord_offsetC  = DIM*jnrC;
 210             j_coord_offsetD  = DIM*jnrD;
 211
 212             /* load j atom coordinates */
 213             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 214                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 215                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 216
 217             /* Calculate displacement vector */
 218             dx00             = _mm256_sub_pd(ix0,jx0);
 219             dy00             = _mm256_sub_pd(iy0,jy0);
 220             dz00             = _mm256_sub_pd(iz0,jz0);
 221             dx01             = _mm256_sub_pd(ix0,jx1);
 222             dy01             = _mm256_sub_pd(iy0,jy1);
 223             dz01             = _mm256_sub_pd(iz0,jz1);
 224             dx02             = _mm256_sub_pd(ix0,jx2);
 225             dy02             = _mm256_sub_pd(iy0,jy2);
 226             dz02             = _mm256_sub_pd(iz0,jz2);
 227             dx10             = _mm256_sub_pd(ix1,jx0);
 228             dy10             = _mm256_sub_pd(iy1,jy0);
 229             dz10             = _mm256_sub_pd(iz1,jz0);
 230             dx11             = _mm256_sub_pd(ix1,jx1);
 231             dy11             = _mm256_sub_pd(iy1,jy1);
 232             dz11             = _mm256_sub_pd(iz1,jz1);
 233             dx12             = _mm256_sub_pd(ix1,jx2);
 234             dy12             = _mm256_sub_pd(iy1,jy2);
 235             dz12             = _mm256_sub_pd(iz1,jz2);
 236             dx20             = _mm256_sub_pd(ix2,jx0);
 237             dy20             = _mm256_sub_pd(iy2,jy0);
 238             dz20             = _mm256_sub_pd(iz2,jz0);
 239             dx21             = _mm256_sub_pd(ix2,jx1);
 240             dy21             = _mm256_sub_pd(iy2,jy1);
 241             dz21             = _mm256_sub_pd(iz2,jz1);
 242             dx22             = _mm256_sub_pd(ix2,jx2);
 243             dy22             = _mm256_sub_pd(iy2,jy2);
 244             dz22             = _mm256_sub_pd(iz2,jz2);
 245
 246             /* Calculate squared distance and things based on it */
 247             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 248             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 249             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 250             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 251             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 252             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 253             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 254             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 255             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 256
 257             rinv00           = avx256_invsqrt_d(rsq00);
 258             rinv01           = avx256_invsqrt_d(rsq01);
 259             rinv02           = avx256_invsqrt_d(rsq02);
 260             rinv10           = avx256_invsqrt_d(rsq10);
 261             rinv11           = avx256_invsqrt_d(rsq11);
 262             rinv12           = avx256_invsqrt_d(rsq12);
 263             rinv20           = avx256_invsqrt_d(rsq20);
 264             rinv21           = avx256_invsqrt_d(rsq21);
 265             rinv22           = avx256_invsqrt_d(rsq22);
 266
 267             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 268             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 269             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 270             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 271             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 272             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 273             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 274             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 275             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 276
 277             fjx0             = _mm256_setzero_pd();
 278             fjy0             = _mm256_setzero_pd();
 279             fjz0             = _mm256_setzero_pd();
 280             fjx1             = _mm256_setzero_pd();
 281             fjy1             = _mm256_setzero_pd();
 282             fjz1             = _mm256_setzero_pd();
 283             fjx2             = _mm256_setzero_pd();
 284             fjy2             = _mm256_setzero_pd();
 285             fjz2             = _mm256_setzero_pd();
 286
 287             /**************************
 288              * CALCULATE INTERACTIONS *
 289              **************************/
 290
 291             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 292             {
 293
 294             /* REACTION-FIELD ELECTROSTATICS */
 295             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_add_pd(rinv00,_mm256_mul_pd(krf,rsq00)),crf));
 296             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
 297
 298             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
 299
 300             /* Update potential sum for this i atom from the interaction with this j atom. */
 301             velec            = _mm256_and_pd(velec,cutoff_mask);
 302             velecsum         = _mm256_add_pd(velecsum,velec);
 303
 304             fscal            = felec;
 305
 306             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 307
 308             /* Calculate temporary vectorial force */
 309             tx               = _mm256_mul_pd(fscal,dx00);
 310             ty               = _mm256_mul_pd(fscal,dy00);
 311             tz               = _mm256_mul_pd(fscal,dz00);
 312
 313             /* Update vectorial force */
 314             fix0             = _mm256_add_pd(fix0,tx);
 315             fiy0             = _mm256_add_pd(fiy0,ty);
 316             fiz0             = _mm256_add_pd(fiz0,tz);
 317
 318             fjx0             = _mm256_add_pd(fjx0,tx);
 319             fjy0             = _mm256_add_pd(fjy0,ty);
 320             fjz0             = _mm256_add_pd(fjz0,tz);
 321
 322             }
 323
 324             /**************************
 325              * CALCULATE INTERACTIONS *
 326              **************************/
 327
 328             if (gmx_mm256_any_lt(rsq01,rcutoff2))
 329             {
 330
 331             /* REACTION-FIELD ELECTROSTATICS */
 332             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_add_pd(rinv01,_mm256_mul_pd(krf,rsq01)),crf));
 333             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
 334
 335             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
 336
 337             /* Update potential sum for this i atom from the interaction with this j atom. */
 338             velec            = _mm256_and_pd(velec,cutoff_mask);
 339             velecsum         = _mm256_add_pd(velecsum,velec);
 340
 341             fscal            = felec;
 342
 343             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 344
 345             /* Calculate temporary vectorial force */
 346             tx               = _mm256_mul_pd(fscal,dx01);
 347             ty               = _mm256_mul_pd(fscal,dy01);
 348             tz               = _mm256_mul_pd(fscal,dz01);
 349
 350             /* Update vectorial force */
 351             fix0             = _mm256_add_pd(fix0,tx);
 352             fiy0             = _mm256_add_pd(fiy0,ty);
 353             fiz0             = _mm256_add_pd(fiz0,tz);
 354
 355             fjx1             = _mm256_add_pd(fjx1,tx);
 356             fjy1             = _mm256_add_pd(fjy1,ty);
 357             fjz1             = _mm256_add_pd(fjz1,tz);
 358
 359             }
 360
 361             /**************************
 362              * CALCULATE INTERACTIONS *
 363              **************************/
 364
 365             if (gmx_mm256_any_lt(rsq02,rcutoff2))
 366             {
 367
 368             /* REACTION-FIELD ELECTROSTATICS */
 369             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_add_pd(rinv02,_mm256_mul_pd(krf,rsq02)),crf));
 370             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
 371
 372             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
 373
 374             /* Update potential sum for this i atom from the interaction with this j atom. */
 375             velec            = _mm256_and_pd(velec,cutoff_mask);
 376             velecsum         = _mm256_add_pd(velecsum,velec);
 377
 378             fscal            = felec;
 379
 380             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 381
 382             /* Calculate temporary vectorial force */
 383             tx               = _mm256_mul_pd(fscal,dx02);
 384             ty               = _mm256_mul_pd(fscal,dy02);
 385             tz               = _mm256_mul_pd(fscal,dz02);
 386
 387             /* Update vectorial force */
 388             fix0             = _mm256_add_pd(fix0,tx);
 389             fiy0             = _mm256_add_pd(fiy0,ty);
 390             fiz0             = _mm256_add_pd(fiz0,tz);
 391
 392             fjx2             = _mm256_add_pd(fjx2,tx);
 393             fjy2             = _mm256_add_pd(fjy2,ty);
 394             fjz2             = _mm256_add_pd(fjz2,tz);
 395
 396             }
 397
 398             /**************************
 399              * CALCULATE INTERACTIONS *
 400              **************************/
 401
 402             if (gmx_mm256_any_lt(rsq10,rcutoff2))
 403             {
 404
 405             /* REACTION-FIELD ELECTROSTATICS */
 406             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_add_pd(rinv10,_mm256_mul_pd(krf,rsq10)),crf));
 407             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
 408
 409             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
 410
 411             /* Update potential sum for this i atom from the interaction with this j atom. */
 412             velec            = _mm256_and_pd(velec,cutoff_mask);
 413             velecsum         = _mm256_add_pd(velecsum,velec);
 414
 415             fscal            = felec;
 416
 417             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 418
 419             /* Calculate temporary vectorial force */
 420             tx               = _mm256_mul_pd(fscal,dx10);
 421             ty               = _mm256_mul_pd(fscal,dy10);
 422             tz               = _mm256_mul_pd(fscal,dz10);
 423
 424             /* Update vectorial force */
 425             fix1             = _mm256_add_pd(fix1,tx);
 426             fiy1             = _mm256_add_pd(fiy1,ty);
 427             fiz1             = _mm256_add_pd(fiz1,tz);
 428
 429             fjx0             = _mm256_add_pd(fjx0,tx);
 430             fjy0             = _mm256_add_pd(fjy0,ty);
 431             fjz0             = _mm256_add_pd(fjz0,tz);
 432
 433             }
 434
 435             /**************************
 436              * CALCULATE INTERACTIONS *
 437              **************************/
 438
 439             if (gmx_mm256_any_lt(rsq11,rcutoff2))
 440             {
 441
 442             /* REACTION-FIELD ELECTROSTATICS */
 443             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 444             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 445
 446             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
 447
 448             /* Update potential sum for this i atom from the interaction with this j atom. */
 449             velec            = _mm256_and_pd(velec,cutoff_mask);
 450             velecsum         = _mm256_add_pd(velecsum,velec);
 451
 452             fscal            = felec;
 453
 454             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 455
 456             /* Calculate temporary vectorial force */
 457             tx               = _mm256_mul_pd(fscal,dx11);
 458             ty               = _mm256_mul_pd(fscal,dy11);
 459             tz               = _mm256_mul_pd(fscal,dz11);
 460
 461             /* Update vectorial force */
 462             fix1             = _mm256_add_pd(fix1,tx);
 463             fiy1             = _mm256_add_pd(fiy1,ty);
 464             fiz1             = _mm256_add_pd(fiz1,tz);
 465
 466             fjx1             = _mm256_add_pd(fjx1,tx);
 467             fjy1             = _mm256_add_pd(fjy1,ty);
 468             fjz1             = _mm256_add_pd(fjz1,tz);
 469
 470             }
 471
 472             /**************************
 473              * CALCULATE INTERACTIONS *
 474              **************************/
 475
 476             if (gmx_mm256_any_lt(rsq12,rcutoff2))
 477             {
 478
 479             /* REACTION-FIELD ELECTROSTATICS */
 480             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
 481             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
 482
 483             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
 484
 485             /* Update potential sum for this i atom from the interaction with this j atom. */
 486             velec            = _mm256_and_pd(velec,cutoff_mask);
 487             velecsum         = _mm256_add_pd(velecsum,velec);
 488
 489             fscal            = felec;
 490
 491             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 492
 493             /* Calculate temporary vectorial force */
 494             tx               = _mm256_mul_pd(fscal,dx12);
 495             ty               = _mm256_mul_pd(fscal,dy12);
 496             tz               = _mm256_mul_pd(fscal,dz12);
 497
 498             /* Update vectorial force */
 499             fix1             = _mm256_add_pd(fix1,tx);
 500             fiy1             = _mm256_add_pd(fiy1,ty);
 501             fiz1             = _mm256_add_pd(fiz1,tz);
 502
 503             fjx2             = _mm256_add_pd(fjx2,tx);
 504             fjy2             = _mm256_add_pd(fjy2,ty);
 505             fjz2             = _mm256_add_pd(fjz2,tz);
 506
 507             }
 508
 509             /**************************
 510              * CALCULATE INTERACTIONS *
 511              **************************/
 512
 513             if (gmx_mm256_any_lt(rsq20,rcutoff2))
 514             {
 515
 516             /* REACTION-FIELD ELECTROSTATICS */
 517             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_add_pd(rinv20,_mm256_mul_pd(krf,rsq20)),crf));
 518             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
 519
 520             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
 521
 522             /* Update potential sum for this i atom from the interaction with this j atom. */
 523             velec            = _mm256_and_pd(velec,cutoff_mask);
 524             velecsum         = _mm256_add_pd(velecsum,velec);
 525
 526             fscal            = felec;
 527
 528             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 529
 530             /* Calculate temporary vectorial force */
 531             tx               = _mm256_mul_pd(fscal,dx20);
 532             ty               = _mm256_mul_pd(fscal,dy20);
 533             tz               = _mm256_mul_pd(fscal,dz20);
 534
 535             /* Update vectorial force */
 536             fix2             = _mm256_add_pd(fix2,tx);
 537             fiy2             = _mm256_add_pd(fiy2,ty);
 538             fiz2             = _mm256_add_pd(fiz2,tz);
 539
 540             fjx0             = _mm256_add_pd(fjx0,tx);
 541             fjy0             = _mm256_add_pd(fjy0,ty);
 542             fjz0             = _mm256_add_pd(fjz0,tz);
 543
 544             }
 545
 546             /**************************
 547              * CALCULATE INTERACTIONS *
 548              **************************/
 549
 550             if (gmx_mm256_any_lt(rsq21,rcutoff2))
 551             {
 552
 553             /* REACTION-FIELD ELECTROSTATICS */
 554             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
 555             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
 556
 557             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
 558
 559             /* Update potential sum for this i atom from the interaction with this j atom. */
 560             velec            = _mm256_and_pd(velec,cutoff_mask);
 561             velecsum         = _mm256_add_pd(velecsum,velec);
 562
 563             fscal            = felec;
 564
 565             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 566
 567             /* Calculate temporary vectorial force */
 568             tx               = _mm256_mul_pd(fscal,dx21);
 569             ty               = _mm256_mul_pd(fscal,dy21);
 570             tz               = _mm256_mul_pd(fscal,dz21);
 571
 572             /* Update vectorial force */
 573             fix2             = _mm256_add_pd(fix2,tx);
 574             fiy2             = _mm256_add_pd(fiy2,ty);
 575             fiz2             = _mm256_add_pd(fiz2,tz);
 576
 577             fjx1             = _mm256_add_pd(fjx1,tx);
 578             fjy1             = _mm256_add_pd(fjy1,ty);
 579             fjz1             = _mm256_add_pd(fjz1,tz);
 580
 581             }
 582
 583             /**************************
 584              * CALCULATE INTERACTIONS *
 585              **************************/
 586
 587             if (gmx_mm256_any_lt(rsq22,rcutoff2))
 588             {
 589
 590             /* REACTION-FIELD ELECTROSTATICS */
 591             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
 592             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
 593
 594             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
 595
 596             /* Update potential sum for this i atom from the interaction with this j atom. */
 597             velec            = _mm256_and_pd(velec,cutoff_mask);
 598             velecsum         = _mm256_add_pd(velecsum,velec);
 599
 600             fscal            = felec;
 601
 602             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 603
 604             /* Calculate temporary vectorial force */
 605             tx               = _mm256_mul_pd(fscal,dx22);
 606             ty               = _mm256_mul_pd(fscal,dy22);
 607             tz               = _mm256_mul_pd(fscal,dz22);
 608
 609             /* Update vectorial force */
 610             fix2             = _mm256_add_pd(fix2,tx);
 611             fiy2             = _mm256_add_pd(fiy2,ty);
 612             fiz2             = _mm256_add_pd(fiz2,tz);
 613
 614             fjx2             = _mm256_add_pd(fjx2,tx);
 615             fjy2             = _mm256_add_pd(fjy2,ty);
 616             fjz2             = _mm256_add_pd(fjz2,tz);
 617
 618             }
 619
 620             fjptrA             = f+j_coord_offsetA;
 621             fjptrB             = f+j_coord_offsetB;
 622             fjptrC             = f+j_coord_offsetC;
 623             fjptrD             = f+j_coord_offsetD;
 624
 625             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
 626                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
 627
 628             /* Inner loop uses 324 flops */
 629         }
 630
 631         if(jidx<j_index_end)
 632         {
 633
 634             /* Get j neighbor index, and coordinate index */
 635             jnrlistA         = jjnr[jidx];
 636             jnrlistB         = jjnr[jidx+1];
 637             jnrlistC         = jjnr[jidx+2];
 638             jnrlistD         = jjnr[jidx+3];
 639             /* Sign of each element will be negative for non-real atoms.
 640              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 641              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
 642              */
 643             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 644
 645             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
 646             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
 647             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
 648
 649             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 650             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 651             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 652             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 653             j_coord_offsetA  = DIM*jnrA;
 654             j_coord_offsetB  = DIM*jnrB;
 655             j_coord_offsetC  = DIM*jnrC;
 656             j_coord_offsetD  = DIM*jnrD;
 657
 658             /* load j atom coordinates */
 659             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 660                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 661                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 662
 663             /* Calculate displacement vector */
 664             dx00             = _mm256_sub_pd(ix0,jx0);
 665             dy00             = _mm256_sub_pd(iy0,jy0);
 666             dz00             = _mm256_sub_pd(iz0,jz0);
 667             dx01             = _mm256_sub_pd(ix0,jx1);
 668             dy01             = _mm256_sub_pd(iy0,jy1);
 669             dz01             = _mm256_sub_pd(iz0,jz1);
 670             dx02             = _mm256_sub_pd(ix0,jx2);
 671             dy02             = _mm256_sub_pd(iy0,jy2);
 672             dz02             = _mm256_sub_pd(iz0,jz2);
 673             dx10             = _mm256_sub_pd(ix1,jx0);
 674             dy10             = _mm256_sub_pd(iy1,jy0);
 675             dz10             = _mm256_sub_pd(iz1,jz0);
 676             dx11             = _mm256_sub_pd(ix1,jx1);
 677             dy11             = _mm256_sub_pd(iy1,jy1);
 678             dz11             = _mm256_sub_pd(iz1,jz1);
 679             dx12             = _mm256_sub_pd(ix1,jx2);
 680             dy12             = _mm256_sub_pd(iy1,jy2);
 681             dz12             = _mm256_sub_pd(iz1,jz2);
 682             dx20             = _mm256_sub_pd(ix2,jx0);
 683             dy20             = _mm256_sub_pd(iy2,jy0);
 684             dz20             = _mm256_sub_pd(iz2,jz0);
 685             dx21             = _mm256_sub_pd(ix2,jx1);
 686             dy21             = _mm256_sub_pd(iy2,jy1);
 687             dz21             = _mm256_sub_pd(iz2,jz1);
 688             dx22             = _mm256_sub_pd(ix2,jx2);
 689             dy22             = _mm256_sub_pd(iy2,jy2);
 690             dz22             = _mm256_sub_pd(iz2,jz2);
 691
 692             /* Calculate squared distance and things based on it */
 693             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
 694             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
 695             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
 696             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
 697             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 698             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 699             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
 700             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 701             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 702
 703             rinv00           = avx256_invsqrt_d(rsq00);
 704             rinv01           = avx256_invsqrt_d(rsq01);
 705             rinv02           = avx256_invsqrt_d(rsq02);
 706             rinv10           = avx256_invsqrt_d(rsq10);
 707             rinv11           = avx256_invsqrt_d(rsq11);
 708             rinv12           = avx256_invsqrt_d(rsq12);
 709             rinv20           = avx256_invsqrt_d(rsq20);
 710             rinv21           = avx256_invsqrt_d(rsq21);
 711             rinv22           = avx256_invsqrt_d(rsq22);
 712
 713             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
 714             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
 715             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
 716             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
 717             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 718             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 719             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
 720             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 721             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 722
 723             fjx0             = _mm256_setzero_pd();
 724             fjy0             = _mm256_setzero_pd();
 725             fjz0             = _mm256_setzero_pd();
 726             fjx1             = _mm256_setzero_pd();
 727             fjy1             = _mm256_setzero_pd();
 728             fjz1             = _mm256_setzero_pd();
 729             fjx2             = _mm256_setzero_pd();
 730             fjy2             = _mm256_setzero_pd();
 731             fjz2             = _mm256_setzero_pd();
 732
 733             /**************************
 734              * CALCULATE INTERACTIONS *
 735              **************************/
 736
 737             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 738             {
 739
 740             /* REACTION-FIELD ELECTROSTATICS */
 741             velec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_add_pd(rinv00,_mm256_mul_pd(krf,rsq00)),crf));
 742             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
 743
 744             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
 745
 746             /* Update potential sum for this i atom from the interaction with this j atom. */
 747             velec            = _mm256_and_pd(velec,cutoff_mask);
 748             velec            = _mm256_andnot_pd(dummy_mask,velec);
 749             velecsum         = _mm256_add_pd(velecsum,velec);
 750
 751             fscal            = felec;
 752
 753             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 754
 755             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 756
 757             /* Calculate temporary vectorial force */
 758             tx               = _mm256_mul_pd(fscal,dx00);
 759             ty               = _mm256_mul_pd(fscal,dy00);
 760             tz               = _mm256_mul_pd(fscal,dz00);
 761
 762             /* Update vectorial force */
 763             fix0             = _mm256_add_pd(fix0,tx);
 764             fiy0             = _mm256_add_pd(fiy0,ty);
 765             fiz0             = _mm256_add_pd(fiz0,tz);
 766
 767             fjx0             = _mm256_add_pd(fjx0,tx);
 768             fjy0             = _mm256_add_pd(fjy0,ty);
 769             fjz0             = _mm256_add_pd(fjz0,tz);
 770
 771             }
 772
 773             /**************************
 774              * CALCULATE INTERACTIONS *
 775              **************************/
 776
 777             if (gmx_mm256_any_lt(rsq01,rcutoff2))
 778             {
 779
 780             /* REACTION-FIELD ELECTROSTATICS */
 781             velec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_add_pd(rinv01,_mm256_mul_pd(krf,rsq01)),crf));
 782             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
 783
 784             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
 785
 786             /* Update potential sum for this i atom from the interaction with this j atom. */
 787             velec            = _mm256_and_pd(velec,cutoff_mask);
 788             velec            = _mm256_andnot_pd(dummy_mask,velec);
 789             velecsum         = _mm256_add_pd(velecsum,velec);
 790
 791             fscal            = felec;
 792
 793             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 794
 795             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 796
 797             /* Calculate temporary vectorial force */
 798             tx               = _mm256_mul_pd(fscal,dx01);
 799             ty               = _mm256_mul_pd(fscal,dy01);
 800             tz               = _mm256_mul_pd(fscal,dz01);
 801
 802             /* Update vectorial force */
 803             fix0             = _mm256_add_pd(fix0,tx);
 804             fiy0             = _mm256_add_pd(fiy0,ty);
 805             fiz0             = _mm256_add_pd(fiz0,tz);
 806
 807             fjx1             = _mm256_add_pd(fjx1,tx);
 808             fjy1             = _mm256_add_pd(fjy1,ty);
 809             fjz1             = _mm256_add_pd(fjz1,tz);
 810
 811             }
 812
 813             /**************************
 814              * CALCULATE INTERACTIONS *
 815              **************************/
 816
 817             if (gmx_mm256_any_lt(rsq02,rcutoff2))
 818             {
 819
 820             /* REACTION-FIELD ELECTROSTATICS */
 821             velec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_add_pd(rinv02,_mm256_mul_pd(krf,rsq02)),crf));
 822             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
 823
 824             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
 825
 826             /* Update potential sum for this i atom from the interaction with this j atom. */
 827             velec            = _mm256_and_pd(velec,cutoff_mask);
 828             velec            = _mm256_andnot_pd(dummy_mask,velec);
 829             velecsum         = _mm256_add_pd(velecsum,velec);
 830
 831             fscal            = felec;
 832
 833             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 834
 835             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 836
 837             /* Calculate temporary vectorial force */
 838             tx               = _mm256_mul_pd(fscal,dx02);
 839             ty               = _mm256_mul_pd(fscal,dy02);
 840             tz               = _mm256_mul_pd(fscal,dz02);
 841
 842             /* Update vectorial force */
 843             fix0             = _mm256_add_pd(fix0,tx);
 844             fiy0             = _mm256_add_pd(fiy0,ty);
 845             fiz0             = _mm256_add_pd(fiz0,tz);
 846
 847             fjx2             = _mm256_add_pd(fjx2,tx);
 848             fjy2             = _mm256_add_pd(fjy2,ty);
 849             fjz2             = _mm256_add_pd(fjz2,tz);
 850
 851             }
 852
 853             /**************************
 854              * CALCULATE INTERACTIONS *
 855              **************************/
 856
 857             if (gmx_mm256_any_lt(rsq10,rcutoff2))
 858             {
 859
 860             /* REACTION-FIELD ELECTROSTATICS */
 861             velec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_add_pd(rinv10,_mm256_mul_pd(krf,rsq10)),crf));
 862             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
 863
 864             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
 865
 866             /* Update potential sum for this i atom from the interaction with this j atom. */
 867             velec            = _mm256_and_pd(velec,cutoff_mask);
 868             velec            = _mm256_andnot_pd(dummy_mask,velec);
 869             velecsum         = _mm256_add_pd(velecsum,velec);
 870
 871             fscal            = felec;
 872
 873             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 874
 875             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 876
 877             /* Calculate temporary vectorial force */
 878             tx               = _mm256_mul_pd(fscal,dx10);
 879             ty               = _mm256_mul_pd(fscal,dy10);
 880             tz               = _mm256_mul_pd(fscal,dz10);
 881
 882             /* Update vectorial force */
 883             fix1             = _mm256_add_pd(fix1,tx);
 884             fiy1             = _mm256_add_pd(fiy1,ty);
 885             fiz1             = _mm256_add_pd(fiz1,tz);
 886
 887             fjx0             = _mm256_add_pd(fjx0,tx);
 888             fjy0             = _mm256_add_pd(fjy0,ty);
 889             fjz0             = _mm256_add_pd(fjz0,tz);
 890
 891             }
 892
 893             /**************************
 894              * CALCULATE INTERACTIONS *
 895              **************************/
 896
 897             if (gmx_mm256_any_lt(rsq11,rcutoff2))
 898             {
 899
 900             /* REACTION-FIELD ELECTROSTATICS */
 901             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 902             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 903
 904             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
 905
 906             /* Update potential sum for this i atom from the interaction with this j atom. */
 907             velec            = _mm256_and_pd(velec,cutoff_mask);
 908             velec            = _mm256_andnot_pd(dummy_mask,velec);
 909             velecsum         = _mm256_add_pd(velecsum,velec);
 910
 911             fscal            = felec;
 912
 913             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 914
 915             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 916
 917             /* Calculate temporary vectorial force */
 918             tx               = _mm256_mul_pd(fscal,dx11);
 919             ty               = _mm256_mul_pd(fscal,dy11);
 920             tz               = _mm256_mul_pd(fscal,dz11);
 921
 922             /* Update vectorial force */
 923             fix1             = _mm256_add_pd(fix1,tx);
 924             fiy1             = _mm256_add_pd(fiy1,ty);
 925             fiz1             = _mm256_add_pd(fiz1,tz);
 926
 927             fjx1             = _mm256_add_pd(fjx1,tx);
 928             fjy1             = _mm256_add_pd(fjy1,ty);
 929             fjz1             = _mm256_add_pd(fjz1,tz);
 930
 931             }
 932
 933             /**************************
 934              * CALCULATE INTERACTIONS *
 935              **************************/
 936
 937             if (gmx_mm256_any_lt(rsq12,rcutoff2))
 938             {
 939
 940             /* REACTION-FIELD ELECTROSTATICS */
 941             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
 942             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
 943
 944             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
 945
 946             /* Update potential sum for this i atom from the interaction with this j atom. */
 947             velec            = _mm256_and_pd(velec,cutoff_mask);
 948             velec            = _mm256_andnot_pd(dummy_mask,velec);
 949             velecsum         = _mm256_add_pd(velecsum,velec);
 950
 951             fscal            = felec;
 952
 953             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 954
 955             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 956
 957             /* Calculate temporary vectorial force */
 958             tx               = _mm256_mul_pd(fscal,dx12);
 959             ty               = _mm256_mul_pd(fscal,dy12);
 960             tz               = _mm256_mul_pd(fscal,dz12);
 961
 962             /* Update vectorial force */
 963             fix1             = _mm256_add_pd(fix1,tx);
 964             fiy1             = _mm256_add_pd(fiy1,ty);
 965             fiz1             = _mm256_add_pd(fiz1,tz);
 966
 967             fjx2             = _mm256_add_pd(fjx2,tx);
 968             fjy2             = _mm256_add_pd(fjy2,ty);
 969             fjz2             = _mm256_add_pd(fjz2,tz);
 970
 971             }
 972
 973             /**************************
 974              * CALCULATE INTERACTIONS *
 975              **************************/
 976
 977             if (gmx_mm256_any_lt(rsq20,rcutoff2))
 978             {
 979
 980             /* REACTION-FIELD ELECTROSTATICS */
 981             velec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_add_pd(rinv20,_mm256_mul_pd(krf,rsq20)),crf));
 982             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
 983
 984             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
 985
 986             /* Update potential sum for this i atom from the interaction with this j atom. */
 987             velec            = _mm256_and_pd(velec,cutoff_mask);
 988             velec            = _mm256_andnot_pd(dummy_mask,velec);
 989             velecsum         = _mm256_add_pd(velecsum,velec);
 990
 991             fscal            = felec;
 992
 993             fscal            = _mm256_and_pd(fscal,cutoff_mask);
 994
 995             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 996
 997             /* Calculate temporary vectorial force */
 998             tx               = _mm256_mul_pd(fscal,dx20);
 999             ty               = _mm256_mul_pd(fscal,dy20);
1000             tz               = _mm256_mul_pd(fscal,dz20);
1001
1002             /* Update vectorial force */
1003             fix2             = _mm256_add_pd(fix2,tx);
1004             fiy2             = _mm256_add_pd(fiy2,ty);
1005             fiz2             = _mm256_add_pd(fiz2,tz);
1006
1007             fjx0             = _mm256_add_pd(fjx0,tx);
1008             fjy0             = _mm256_add_pd(fjy0,ty);
1009             fjz0             = _mm256_add_pd(fjz0,tz);
1010
1011             }
1012
1013             /**************************
1014              * CALCULATE INTERACTIONS *
1015              **************************/
1016
1017             if (gmx_mm256_any_lt(rsq21,rcutoff2))
1018             {
1019
1020             /* REACTION-FIELD ELECTROSTATICS */
1021             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
1022             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1023
1024             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
1025
1026             /* Update potential sum for this i atom from the interaction with this j atom. */
1027             velec            = _mm256_and_pd(velec,cutoff_mask);
1028             velec            = _mm256_andnot_pd(dummy_mask,velec);
1029             velecsum         = _mm256_add_pd(velecsum,velec);
1030
1031             fscal            = felec;
1032
1033             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1034
1035             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1036
1037             /* Calculate temporary vectorial force */
1038             tx               = _mm256_mul_pd(fscal,dx21);
1039             ty               = _mm256_mul_pd(fscal,dy21);
1040             tz               = _mm256_mul_pd(fscal,dz21);
1041
1042             /* Update vectorial force */
1043             fix2             = _mm256_add_pd(fix2,tx);
1044             fiy2             = _mm256_add_pd(fiy2,ty);
1045             fiz2             = _mm256_add_pd(fiz2,tz);
1046
1047             fjx1             = _mm256_add_pd(fjx1,tx);
1048             fjy1             = _mm256_add_pd(fjy1,ty);
1049             fjz1             = _mm256_add_pd(fjz1,tz);
1050
1051             }
1052
1053             /**************************
1054              * CALCULATE INTERACTIONS *
1055              **************************/
1056
1057             if (gmx_mm256_any_lt(rsq22,rcutoff2))
1058             {
1059
1060             /* REACTION-FIELD ELECTROSTATICS */
1061             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
1062             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1063
1064             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
1065
1066             /* Update potential sum for this i atom from the interaction with this j atom. */
1067             velec            = _mm256_and_pd(velec,cutoff_mask);
1068             velec            = _mm256_andnot_pd(dummy_mask,velec);
1069             velecsum         = _mm256_add_pd(velecsum,velec);
1070
1071             fscal            = felec;
1072
1073             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1074
1075             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1076
1077             /* Calculate temporary vectorial force */
1078             tx               = _mm256_mul_pd(fscal,dx22);
1079             ty               = _mm256_mul_pd(fscal,dy22);
1080             tz               = _mm256_mul_pd(fscal,dz22);
1081
1082             /* Update vectorial force */
1083             fix2             = _mm256_add_pd(fix2,tx);
1084             fiy2             = _mm256_add_pd(fiy2,ty);
1085             fiz2             = _mm256_add_pd(fiz2,tz);
1086
1087             fjx2             = _mm256_add_pd(fjx2,tx);
1088             fjy2             = _mm256_add_pd(fjy2,ty);
1089             fjz2             = _mm256_add_pd(fjz2,tz);
1090
1091             }
1092
1093             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1094             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1095             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1096             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1097
1098             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1099                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1100
1101             /* Inner loop uses 324 flops */
1102         }
1103
1104         /* End of innermost loop */
1105
1106         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1107                                                  f+i_coord_offset,fshift+i_shift_offset);
1108
1109         ggid                        = gid[iidx];
1110         /* Update potential energies */
1111         gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1112
1113         /* Increment number of inner iterations */
1114         inneriter                  += j_index_end - j_index_start;
1115
1116         /* Outer loop uses 19 flops */
1117     }
1118
1119     /* Increment number of outer iterations */
1120     outeriter        += nri;
1121
1122     /* Update outer/inner flops */
1123
1124     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3W3_VF,outeriter*19 + inneriter*324);
1125 }
1126 /*
1127  * Gromacs nonbonded kernel:   nb_kernel_ElecRFCut_VdwNone_GeomW3W3_F_avx_256_double
1128  * Electrostatics interaction: ReactionField
1129  * VdW interaction:            None
1130  * Geometry:                   Water3-Water3
1131  * Calculate force/pot:        Force
1132  */
1133 void
1134 nb_kernel_ElecRFCut_VdwNone_GeomW3W3_F_avx_256_double
1135                     (t_nblist                    * gmx_restrict       nlist,
1136                      rvec                        * gmx_restrict          xx,
1137                      rvec                        * gmx_restrict          ff,
1138                      struct t_forcerec           * gmx_restrict          fr,
1139                      t_mdatoms                   * gmx_restrict     mdatoms,
1140                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1141                      t_nrnb                      * gmx_restrict        nrnb)
1142 {
1143     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1144      * just 0 for non-waters.
1145      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
1146      * jnr indices corresponding to data put in the four positions in the SIMD register.
1147      */
1148     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1149     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1150     int              jnrA,jnrB,jnrC,jnrD;
1151     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1152     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
1153     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1154     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1155     real             rcutoff_scalar;
1156     real             *shiftvec,*fshift,*x,*f;
1157     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1158     real             scratch[4*DIM];
1159     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1160     real *           vdwioffsetptr0;
1161     __m256d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1162     real *           vdwioffsetptr1;
1163     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1164     real *           vdwioffsetptr2;
1165     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1166     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1167     __m256d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1168     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1169     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1170     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1171     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1172     __m256d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1173     __m256d          dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1174     __m256d          dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1175     __m256d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1176     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1177     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1178     __m256d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1179     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1180     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1181     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
1182     real             *charge;
1183     __m256d          dummy_mask,cutoff_mask;
1184     __m128           tmpmask0,tmpmask1;
1185     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
1186     __m256d          one     = _mm256_set1_pd(1.0);
1187     __m256d          two     = _mm256_set1_pd(2.0);
1188     x                = xx[0];
1189     f                = ff[0];
1190
1191     nri              = nlist->nri;
1192     iinr             = nlist->iinr;
1193     jindex           = nlist->jindex;
1194     jjnr             = nlist->jjnr;
1195     shiftidx         = nlist->shift;
1196     gid              = nlist->gid;
1197     shiftvec         = fr->shift_vec[0];
1198     fshift           = fr->fshift[0];
1199     facel            = _mm256_set1_pd(fr->ic->epsfac);
1200     charge           = mdatoms->chargeA;
1201     krf              = _mm256_set1_pd(fr->ic->k_rf);
1202     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
1203     crf              = _mm256_set1_pd(fr->ic->c_rf);
1204
1205     /* Setup water-specific parameters */
1206     inr              = nlist->iinr[0];
1207     iq0              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+0]));
1208     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1209     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1210
1211     jq0              = _mm256_set1_pd(charge[inr+0]);
1212     jq1              = _mm256_set1_pd(charge[inr+1]);
1213     jq2              = _mm256_set1_pd(charge[inr+2]);
1214     qq00             = _mm256_mul_pd(iq0,jq0);
1215     qq01             = _mm256_mul_pd(iq0,jq1);
1216     qq02             = _mm256_mul_pd(iq0,jq2);
1217     qq10             = _mm256_mul_pd(iq1,jq0);
1218     qq11             = _mm256_mul_pd(iq1,jq1);
1219     qq12             = _mm256_mul_pd(iq1,jq2);
1220     qq20             = _mm256_mul_pd(iq2,jq0);
1221     qq21             = _mm256_mul_pd(iq2,jq1);
1222     qq22             = _mm256_mul_pd(iq2,jq2);
1223
1224     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1225     rcutoff_scalar   = fr->ic->rcoulomb;
1226     rcutoff          = _mm256_set1_pd(rcutoff_scalar);
1227     rcutoff2         = _mm256_mul_pd(rcutoff,rcutoff);
1228
1229     /* Avoid stupid compiler warnings */
1230     jnrA = jnrB = jnrC = jnrD = 0;
1231     j_coord_offsetA = 0;
1232     j_coord_offsetB = 0;
1233     j_coord_offsetC = 0;
1234     j_coord_offsetD = 0;
1235
1236     outeriter        = 0;
1237     inneriter        = 0;
1238
1239     for(iidx=0;iidx<4*DIM;iidx++)
1240     {
1241         scratch[iidx] = 0.0;
1242     }
1243
1244     /* Start outer loop over neighborlists */
1245     for(iidx=0; iidx<nri; iidx++)
1246     {
1247         /* Load shift vector for this list */
1248         i_shift_offset   = DIM*shiftidx[iidx];
1249
1250         /* Load limits for loop over neighbors */
1251         j_index_start    = jindex[iidx];
1252         j_index_end      = jindex[iidx+1];
1253
1254         /* Get outer coordinate index */
1255         inr              = iinr[iidx];
1256         i_coord_offset   = DIM*inr;
1257
1258         /* Load i particle coords and add shift vector */
1259         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1260                                                     &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1261
1262         fix0             = _mm256_setzero_pd();
1263         fiy0             = _mm256_setzero_pd();
1264         fiz0             = _mm256_setzero_pd();
1265         fix1             = _mm256_setzero_pd();
1266         fiy1             = _mm256_setzero_pd();
1267         fiz1             = _mm256_setzero_pd();
1268         fix2             = _mm256_setzero_pd();
1269         fiy2             = _mm256_setzero_pd();
1270         fiz2             = _mm256_setzero_pd();
1271
1272         /* Start inner kernel loop */
1273         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1274         {
1275
1276             /* Get j neighbor index, and coordinate index */
1277             jnrA             = jjnr[jidx];
1278             jnrB             = jjnr[jidx+1];
1279             jnrC             = jjnr[jidx+2];
1280             jnrD             = jjnr[jidx+3];
1281             j_coord_offsetA  = DIM*jnrA;
1282             j_coord_offsetB  = DIM*jnrB;
1283             j_coord_offsetC  = DIM*jnrC;
1284             j_coord_offsetD  = DIM*jnrD;
1285
1286             /* load j atom coordinates */
1287             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1288                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1289                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1290
1291             /* Calculate displacement vector */
1292             dx00             = _mm256_sub_pd(ix0,jx0);
1293             dy00             = _mm256_sub_pd(iy0,jy0);
1294             dz00             = _mm256_sub_pd(iz0,jz0);
1295             dx01             = _mm256_sub_pd(ix0,jx1);
1296             dy01             = _mm256_sub_pd(iy0,jy1);
1297             dz01             = _mm256_sub_pd(iz0,jz1);
1298             dx02             = _mm256_sub_pd(ix0,jx2);
1299             dy02             = _mm256_sub_pd(iy0,jy2);
1300             dz02             = _mm256_sub_pd(iz0,jz2);
1301             dx10             = _mm256_sub_pd(ix1,jx0);
1302             dy10             = _mm256_sub_pd(iy1,jy0);
1303             dz10             = _mm256_sub_pd(iz1,jz0);
1304             dx11             = _mm256_sub_pd(ix1,jx1);
1305             dy11             = _mm256_sub_pd(iy1,jy1);
1306             dz11             = _mm256_sub_pd(iz1,jz1);
1307             dx12             = _mm256_sub_pd(ix1,jx2);
1308             dy12             = _mm256_sub_pd(iy1,jy2);
1309             dz12             = _mm256_sub_pd(iz1,jz2);
1310             dx20             = _mm256_sub_pd(ix2,jx0);
1311             dy20             = _mm256_sub_pd(iy2,jy0);
1312             dz20             = _mm256_sub_pd(iz2,jz0);
1313             dx21             = _mm256_sub_pd(ix2,jx1);
1314             dy21             = _mm256_sub_pd(iy2,jy1);
1315             dz21             = _mm256_sub_pd(iz2,jz1);
1316             dx22             = _mm256_sub_pd(ix2,jx2);
1317             dy22             = _mm256_sub_pd(iy2,jy2);
1318             dz22             = _mm256_sub_pd(iz2,jz2);
1319
1320             /* Calculate squared distance and things based on it */
1321             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1322             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1323             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1324             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1325             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1326             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1327             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
1328             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1329             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1330
1331             rinv00           = avx256_invsqrt_d(rsq00);
1332             rinv01           = avx256_invsqrt_d(rsq01);
1333             rinv02           = avx256_invsqrt_d(rsq02);
1334             rinv10           = avx256_invsqrt_d(rsq10);
1335             rinv11           = avx256_invsqrt_d(rsq11);
1336             rinv12           = avx256_invsqrt_d(rsq12);
1337             rinv20           = avx256_invsqrt_d(rsq20);
1338             rinv21           = avx256_invsqrt_d(rsq21);
1339             rinv22           = avx256_invsqrt_d(rsq22);
1340
1341             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
1342             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
1343             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
1344             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
1345             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1346             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1347             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
1348             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1349             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1350
1351             fjx0             = _mm256_setzero_pd();
1352             fjy0             = _mm256_setzero_pd();
1353             fjz0             = _mm256_setzero_pd();
1354             fjx1             = _mm256_setzero_pd();
1355             fjy1             = _mm256_setzero_pd();
1356             fjz1             = _mm256_setzero_pd();
1357             fjx2             = _mm256_setzero_pd();
1358             fjy2             = _mm256_setzero_pd();
1359             fjz2             = _mm256_setzero_pd();
1360
1361             /**************************
1362              * CALCULATE INTERACTIONS *
1363              **************************/
1364
1365             if (gmx_mm256_any_lt(rsq00,rcutoff2))
1366             {
1367
1368             /* REACTION-FIELD ELECTROSTATICS */
1369             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
1370
1371             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
1372
1373             fscal            = felec;
1374
1375             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1376
1377             /* Calculate temporary vectorial force */
1378             tx               = _mm256_mul_pd(fscal,dx00);
1379             ty               = _mm256_mul_pd(fscal,dy00);
1380             tz               = _mm256_mul_pd(fscal,dz00);
1381
1382             /* Update vectorial force */
1383             fix0             = _mm256_add_pd(fix0,tx);
1384             fiy0             = _mm256_add_pd(fiy0,ty);
1385             fiz0             = _mm256_add_pd(fiz0,tz);
1386
1387             fjx0             = _mm256_add_pd(fjx0,tx);
1388             fjy0             = _mm256_add_pd(fjy0,ty);
1389             fjz0             = _mm256_add_pd(fjz0,tz);
1390
1391             }
1392
1393             /**************************
1394              * CALCULATE INTERACTIONS *
1395              **************************/
1396
1397             if (gmx_mm256_any_lt(rsq01,rcutoff2))
1398             {
1399
1400             /* REACTION-FIELD ELECTROSTATICS */
1401             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
1402
1403             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
1404
1405             fscal            = felec;
1406
1407             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1408
1409             /* Calculate temporary vectorial force */
1410             tx               = _mm256_mul_pd(fscal,dx01);
1411             ty               = _mm256_mul_pd(fscal,dy01);
1412             tz               = _mm256_mul_pd(fscal,dz01);
1413
1414             /* Update vectorial force */
1415             fix0             = _mm256_add_pd(fix0,tx);
1416             fiy0             = _mm256_add_pd(fiy0,ty);
1417             fiz0             = _mm256_add_pd(fiz0,tz);
1418
1419             fjx1             = _mm256_add_pd(fjx1,tx);
1420             fjy1             = _mm256_add_pd(fjy1,ty);
1421             fjz1             = _mm256_add_pd(fjz1,tz);
1422
1423             }
1424
1425             /**************************
1426              * CALCULATE INTERACTIONS *
1427              **************************/
1428
1429             if (gmx_mm256_any_lt(rsq02,rcutoff2))
1430             {
1431
1432             /* REACTION-FIELD ELECTROSTATICS */
1433             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
1434
1435             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
1436
1437             fscal            = felec;
1438
1439             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1440
1441             /* Calculate temporary vectorial force */
1442             tx               = _mm256_mul_pd(fscal,dx02);
1443             ty               = _mm256_mul_pd(fscal,dy02);
1444             tz               = _mm256_mul_pd(fscal,dz02);
1445
1446             /* Update vectorial force */
1447             fix0             = _mm256_add_pd(fix0,tx);
1448             fiy0             = _mm256_add_pd(fiy0,ty);
1449             fiz0             = _mm256_add_pd(fiz0,tz);
1450
1451             fjx2             = _mm256_add_pd(fjx2,tx);
1452             fjy2             = _mm256_add_pd(fjy2,ty);
1453             fjz2             = _mm256_add_pd(fjz2,tz);
1454
1455             }
1456
1457             /**************************
1458              * CALCULATE INTERACTIONS *
1459              **************************/
1460
1461             if (gmx_mm256_any_lt(rsq10,rcutoff2))
1462             {
1463
1464             /* REACTION-FIELD ELECTROSTATICS */
1465             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
1466
1467             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
1468
1469             fscal            = felec;
1470
1471             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1472
1473             /* Calculate temporary vectorial force */
1474             tx               = _mm256_mul_pd(fscal,dx10);
1475             ty               = _mm256_mul_pd(fscal,dy10);
1476             tz               = _mm256_mul_pd(fscal,dz10);
1477
1478             /* Update vectorial force */
1479             fix1             = _mm256_add_pd(fix1,tx);
1480             fiy1             = _mm256_add_pd(fiy1,ty);
1481             fiz1             = _mm256_add_pd(fiz1,tz);
1482
1483             fjx0             = _mm256_add_pd(fjx0,tx);
1484             fjy0             = _mm256_add_pd(fjy0,ty);
1485             fjz0             = _mm256_add_pd(fjz0,tz);
1486
1487             }
1488
1489             /**************************
1490              * CALCULATE INTERACTIONS *
1491              **************************/
1492
1493             if (gmx_mm256_any_lt(rsq11,rcutoff2))
1494             {
1495
1496             /* REACTION-FIELD ELECTROSTATICS */
1497             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1498
1499             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
1500
1501             fscal            = felec;
1502
1503             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1504
1505             /* Calculate temporary vectorial force */
1506             tx               = _mm256_mul_pd(fscal,dx11);
1507             ty               = _mm256_mul_pd(fscal,dy11);
1508             tz               = _mm256_mul_pd(fscal,dz11);
1509
1510             /* Update vectorial force */
1511             fix1             = _mm256_add_pd(fix1,tx);
1512             fiy1             = _mm256_add_pd(fiy1,ty);
1513             fiz1             = _mm256_add_pd(fiz1,tz);
1514
1515             fjx1             = _mm256_add_pd(fjx1,tx);
1516             fjy1             = _mm256_add_pd(fjy1,ty);
1517             fjz1             = _mm256_add_pd(fjz1,tz);
1518
1519             }
1520
1521             /**************************
1522              * CALCULATE INTERACTIONS *
1523              **************************/
1524
1525             if (gmx_mm256_any_lt(rsq12,rcutoff2))
1526             {
1527
1528             /* REACTION-FIELD ELECTROSTATICS */
1529             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1530
1531             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
1532
1533             fscal            = felec;
1534
1535             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1536
1537             /* Calculate temporary vectorial force */
1538             tx               = _mm256_mul_pd(fscal,dx12);
1539             ty               = _mm256_mul_pd(fscal,dy12);
1540             tz               = _mm256_mul_pd(fscal,dz12);
1541
1542             /* Update vectorial force */
1543             fix1             = _mm256_add_pd(fix1,tx);
1544             fiy1             = _mm256_add_pd(fiy1,ty);
1545             fiz1             = _mm256_add_pd(fiz1,tz);
1546
1547             fjx2             = _mm256_add_pd(fjx2,tx);
1548             fjy2             = _mm256_add_pd(fjy2,ty);
1549             fjz2             = _mm256_add_pd(fjz2,tz);
1550
1551             }
1552
1553             /**************************
1554              * CALCULATE INTERACTIONS *
1555              **************************/
1556
1557             if (gmx_mm256_any_lt(rsq20,rcutoff2))
1558             {
1559
1560             /* REACTION-FIELD ELECTROSTATICS */
1561             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
1562
1563             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
1564
1565             fscal            = felec;
1566
1567             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1568
1569             /* Calculate temporary vectorial force */
1570             tx               = _mm256_mul_pd(fscal,dx20);
1571             ty               = _mm256_mul_pd(fscal,dy20);
1572             tz               = _mm256_mul_pd(fscal,dz20);
1573
1574             /* Update vectorial force */
1575             fix2             = _mm256_add_pd(fix2,tx);
1576             fiy2             = _mm256_add_pd(fiy2,ty);
1577             fiz2             = _mm256_add_pd(fiz2,tz);
1578
1579             fjx0             = _mm256_add_pd(fjx0,tx);
1580             fjy0             = _mm256_add_pd(fjy0,ty);
1581             fjz0             = _mm256_add_pd(fjz0,tz);
1582
1583             }
1584
1585             /**************************
1586              * CALCULATE INTERACTIONS *
1587              **************************/
1588
1589             if (gmx_mm256_any_lt(rsq21,rcutoff2))
1590             {
1591
1592             /* REACTION-FIELD ELECTROSTATICS */
1593             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1594
1595             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
1596
1597             fscal            = felec;
1598
1599             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1600
1601             /* Calculate temporary vectorial force */
1602             tx               = _mm256_mul_pd(fscal,dx21);
1603             ty               = _mm256_mul_pd(fscal,dy21);
1604             tz               = _mm256_mul_pd(fscal,dz21);
1605
1606             /* Update vectorial force */
1607             fix2             = _mm256_add_pd(fix2,tx);
1608             fiy2             = _mm256_add_pd(fiy2,ty);
1609             fiz2             = _mm256_add_pd(fiz2,tz);
1610
1611             fjx1             = _mm256_add_pd(fjx1,tx);
1612             fjy1             = _mm256_add_pd(fjy1,ty);
1613             fjz1             = _mm256_add_pd(fjz1,tz);
1614
1615             }
1616
1617             /**************************
1618              * CALCULATE INTERACTIONS *
1619              **************************/
1620
1621             if (gmx_mm256_any_lt(rsq22,rcutoff2))
1622             {
1623
1624             /* REACTION-FIELD ELECTROSTATICS */
1625             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1626
1627             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
1628
1629             fscal            = felec;
1630
1631             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1632
1633             /* Calculate temporary vectorial force */
1634             tx               = _mm256_mul_pd(fscal,dx22);
1635             ty               = _mm256_mul_pd(fscal,dy22);
1636             tz               = _mm256_mul_pd(fscal,dz22);
1637
1638             /* Update vectorial force */
1639             fix2             = _mm256_add_pd(fix2,tx);
1640             fiy2             = _mm256_add_pd(fiy2,ty);
1641             fiz2             = _mm256_add_pd(fiz2,tz);
1642
1643             fjx2             = _mm256_add_pd(fjx2,tx);
1644             fjy2             = _mm256_add_pd(fjy2,ty);
1645             fjz2             = _mm256_add_pd(fjz2,tz);
1646
1647             }
1648
1649             fjptrA             = f+j_coord_offsetA;
1650             fjptrB             = f+j_coord_offsetB;
1651             fjptrC             = f+j_coord_offsetC;
1652             fjptrD             = f+j_coord_offsetD;
1653
1654             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
1655                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1656
1657             /* Inner loop uses 270 flops */
1658         }
1659
1660         if(jidx<j_index_end)
1661         {
1662
1663             /* Get j neighbor index, and coordinate index */
1664             jnrlistA         = jjnr[jidx];
1665             jnrlistB         = jjnr[jidx+1];
1666             jnrlistC         = jjnr[jidx+2];
1667             jnrlistD         = jjnr[jidx+3];
1668             /* Sign of each element will be negative for non-real atoms.
1669              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1670              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
1671              */
1672             tmpmask0 = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1673
1674             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
1675             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
1676             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
1677
1678             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1679             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1680             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1681             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1682             j_coord_offsetA  = DIM*jnrA;
1683             j_coord_offsetB  = DIM*jnrB;
1684             j_coord_offsetC  = DIM*jnrC;
1685             j_coord_offsetD  = DIM*jnrD;
1686
1687             /* load j atom coordinates */
1688             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1689                                                  x+j_coord_offsetC,x+j_coord_offsetD,
1690                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1691
1692             /* Calculate displacement vector */
1693             dx00             = _mm256_sub_pd(ix0,jx0);
1694             dy00             = _mm256_sub_pd(iy0,jy0);
1695             dz00             = _mm256_sub_pd(iz0,jz0);
1696             dx01             = _mm256_sub_pd(ix0,jx1);
1697             dy01             = _mm256_sub_pd(iy0,jy1);
1698             dz01             = _mm256_sub_pd(iz0,jz1);
1699             dx02             = _mm256_sub_pd(ix0,jx2);
1700             dy02             = _mm256_sub_pd(iy0,jy2);
1701             dz02             = _mm256_sub_pd(iz0,jz2);
1702             dx10             = _mm256_sub_pd(ix1,jx0);
1703             dy10             = _mm256_sub_pd(iy1,jy0);
1704             dz10             = _mm256_sub_pd(iz1,jz0);
1705             dx11             = _mm256_sub_pd(ix1,jx1);
1706             dy11             = _mm256_sub_pd(iy1,jy1);
1707             dz11             = _mm256_sub_pd(iz1,jz1);
1708             dx12             = _mm256_sub_pd(ix1,jx2);
1709             dy12             = _mm256_sub_pd(iy1,jy2);
1710             dz12             = _mm256_sub_pd(iz1,jz2);
1711             dx20             = _mm256_sub_pd(ix2,jx0);
1712             dy20             = _mm256_sub_pd(iy2,jy0);
1713             dz20             = _mm256_sub_pd(iz2,jz0);
1714             dx21             = _mm256_sub_pd(ix2,jx1);
1715             dy21             = _mm256_sub_pd(iy2,jy1);
1716             dz21             = _mm256_sub_pd(iz2,jz1);
1717             dx22             = _mm256_sub_pd(ix2,jx2);
1718             dy22             = _mm256_sub_pd(iy2,jy2);
1719             dz22             = _mm256_sub_pd(iz2,jz2);
1720
1721             /* Calculate squared distance and things based on it */
1722             rsq00            = gmx_mm256_calc_rsq_pd(dx00,dy00,dz00);
1723             rsq01            = gmx_mm256_calc_rsq_pd(dx01,dy01,dz01);
1724             rsq02            = gmx_mm256_calc_rsq_pd(dx02,dy02,dz02);
1725             rsq10            = gmx_mm256_calc_rsq_pd(dx10,dy10,dz10);
1726             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1727             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1728             rsq20            = gmx_mm256_calc_rsq_pd(dx20,dy20,dz20);
1729             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1730             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1731
1732             rinv00           = avx256_invsqrt_d(rsq00);
1733             rinv01           = avx256_invsqrt_d(rsq01);
1734             rinv02           = avx256_invsqrt_d(rsq02);
1735             rinv10           = avx256_invsqrt_d(rsq10);
1736             rinv11           = avx256_invsqrt_d(rsq11);
1737             rinv12           = avx256_invsqrt_d(rsq12);
1738             rinv20           = avx256_invsqrt_d(rsq20);
1739             rinv21           = avx256_invsqrt_d(rsq21);
1740             rinv22           = avx256_invsqrt_d(rsq22);
1741
1742             rinvsq00         = _mm256_mul_pd(rinv00,rinv00);
1743             rinvsq01         = _mm256_mul_pd(rinv01,rinv01);
1744             rinvsq02         = _mm256_mul_pd(rinv02,rinv02);
1745             rinvsq10         = _mm256_mul_pd(rinv10,rinv10);
1746             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1747             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1748             rinvsq20         = _mm256_mul_pd(rinv20,rinv20);
1749             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1750             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1751
1752             fjx0             = _mm256_setzero_pd();
1753             fjy0             = _mm256_setzero_pd();
1754             fjz0             = _mm256_setzero_pd();
1755             fjx1             = _mm256_setzero_pd();
1756             fjy1             = _mm256_setzero_pd();
1757             fjz1             = _mm256_setzero_pd();
1758             fjx2             = _mm256_setzero_pd();
1759             fjy2             = _mm256_setzero_pd();
1760             fjz2             = _mm256_setzero_pd();
1761
1762             /**************************
1763              * CALCULATE INTERACTIONS *
1764              **************************/
1765
1766             if (gmx_mm256_any_lt(rsq00,rcutoff2))
1767             {
1768
1769             /* REACTION-FIELD ELECTROSTATICS */
1770             felec            = _mm256_mul_pd(qq00,_mm256_sub_pd(_mm256_mul_pd(rinv00,rinvsq00),krf2));
1771
1772             cutoff_mask      = _mm256_cmp_pd(rsq00,rcutoff2,_CMP_LT_OQ);
1773
1774             fscal            = felec;
1775
1776             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1777
1778             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1779
1780             /* Calculate temporary vectorial force */
1781             tx               = _mm256_mul_pd(fscal,dx00);
1782             ty               = _mm256_mul_pd(fscal,dy00);
1783             tz               = _mm256_mul_pd(fscal,dz00);
1784
1785             /* Update vectorial force */
1786             fix0             = _mm256_add_pd(fix0,tx);
1787             fiy0             = _mm256_add_pd(fiy0,ty);
1788             fiz0             = _mm256_add_pd(fiz0,tz);
1789
1790             fjx0             = _mm256_add_pd(fjx0,tx);
1791             fjy0             = _mm256_add_pd(fjy0,ty);
1792             fjz0             = _mm256_add_pd(fjz0,tz);
1793
1794             }
1795
1796             /**************************
1797              * CALCULATE INTERACTIONS *
1798              **************************/
1799
1800             if (gmx_mm256_any_lt(rsq01,rcutoff2))
1801             {
1802
1803             /* REACTION-FIELD ELECTROSTATICS */
1804             felec            = _mm256_mul_pd(qq01,_mm256_sub_pd(_mm256_mul_pd(rinv01,rinvsq01),krf2));
1805
1806             cutoff_mask      = _mm256_cmp_pd(rsq01,rcutoff2,_CMP_LT_OQ);
1807
1808             fscal            = felec;
1809
1810             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1811
1812             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1813
1814             /* Calculate temporary vectorial force */
1815             tx               = _mm256_mul_pd(fscal,dx01);
1816             ty               = _mm256_mul_pd(fscal,dy01);
1817             tz               = _mm256_mul_pd(fscal,dz01);
1818
1819             /* Update vectorial force */
1820             fix0             = _mm256_add_pd(fix0,tx);
1821             fiy0             = _mm256_add_pd(fiy0,ty);
1822             fiz0             = _mm256_add_pd(fiz0,tz);
1823
1824             fjx1             = _mm256_add_pd(fjx1,tx);
1825             fjy1             = _mm256_add_pd(fjy1,ty);
1826             fjz1             = _mm256_add_pd(fjz1,tz);
1827
1828             }
1829
1830             /**************************
1831              * CALCULATE INTERACTIONS *
1832              **************************/
1833
1834             if (gmx_mm256_any_lt(rsq02,rcutoff2))
1835             {
1836
1837             /* REACTION-FIELD ELECTROSTATICS */
1838             felec            = _mm256_mul_pd(qq02,_mm256_sub_pd(_mm256_mul_pd(rinv02,rinvsq02),krf2));
1839
1840             cutoff_mask      = _mm256_cmp_pd(rsq02,rcutoff2,_CMP_LT_OQ);
1841
1842             fscal            = felec;
1843
1844             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1845
1846             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1847
1848             /* Calculate temporary vectorial force */
1849             tx               = _mm256_mul_pd(fscal,dx02);
1850             ty               = _mm256_mul_pd(fscal,dy02);
1851             tz               = _mm256_mul_pd(fscal,dz02);
1852
1853             /* Update vectorial force */
1854             fix0             = _mm256_add_pd(fix0,tx);
1855             fiy0             = _mm256_add_pd(fiy0,ty);
1856             fiz0             = _mm256_add_pd(fiz0,tz);
1857
1858             fjx2             = _mm256_add_pd(fjx2,tx);
1859             fjy2             = _mm256_add_pd(fjy2,ty);
1860             fjz2             = _mm256_add_pd(fjz2,tz);
1861
1862             }
1863
1864             /**************************
1865              * CALCULATE INTERACTIONS *
1866              **************************/
1867
1868             if (gmx_mm256_any_lt(rsq10,rcutoff2))
1869             {
1870
1871             /* REACTION-FIELD ELECTROSTATICS */
1872             felec            = _mm256_mul_pd(qq10,_mm256_sub_pd(_mm256_mul_pd(rinv10,rinvsq10),krf2));
1873
1874             cutoff_mask      = _mm256_cmp_pd(rsq10,rcutoff2,_CMP_LT_OQ);
1875
1876             fscal            = felec;
1877
1878             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1879
1880             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1881
1882             /* Calculate temporary vectorial force */
1883             tx               = _mm256_mul_pd(fscal,dx10);
1884             ty               = _mm256_mul_pd(fscal,dy10);
1885             tz               = _mm256_mul_pd(fscal,dz10);
1886
1887             /* Update vectorial force */
1888             fix1             = _mm256_add_pd(fix1,tx);
1889             fiy1             = _mm256_add_pd(fiy1,ty);
1890             fiz1             = _mm256_add_pd(fiz1,tz);
1891
1892             fjx0             = _mm256_add_pd(fjx0,tx);
1893             fjy0             = _mm256_add_pd(fjy0,ty);
1894             fjz0             = _mm256_add_pd(fjz0,tz);
1895
1896             }
1897
1898             /**************************
1899              * CALCULATE INTERACTIONS *
1900              **************************/
1901
1902             if (gmx_mm256_any_lt(rsq11,rcutoff2))
1903             {
1904
1905             /* REACTION-FIELD ELECTROSTATICS */
1906             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1907
1908             cutoff_mask      = _mm256_cmp_pd(rsq11,rcutoff2,_CMP_LT_OQ);
1909
1910             fscal            = felec;
1911
1912             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1913
1914             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1915
1916             /* Calculate temporary vectorial force */
1917             tx               = _mm256_mul_pd(fscal,dx11);
1918             ty               = _mm256_mul_pd(fscal,dy11);
1919             tz               = _mm256_mul_pd(fscal,dz11);
1920
1921             /* Update vectorial force */
1922             fix1             = _mm256_add_pd(fix1,tx);
1923             fiy1             = _mm256_add_pd(fiy1,ty);
1924             fiz1             = _mm256_add_pd(fiz1,tz);
1925
1926             fjx1             = _mm256_add_pd(fjx1,tx);
1927             fjy1             = _mm256_add_pd(fjy1,ty);
1928             fjz1             = _mm256_add_pd(fjz1,tz);
1929
1930             }
1931
1932             /**************************
1933              * CALCULATE INTERACTIONS *
1934              **************************/
1935
1936             if (gmx_mm256_any_lt(rsq12,rcutoff2))
1937             {
1938
1939             /* REACTION-FIELD ELECTROSTATICS */
1940             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1941
1942             cutoff_mask      = _mm256_cmp_pd(rsq12,rcutoff2,_CMP_LT_OQ);
1943
1944             fscal            = felec;
1945
1946             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1947
1948             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1949
1950             /* Calculate temporary vectorial force */
1951             tx               = _mm256_mul_pd(fscal,dx12);
1952             ty               = _mm256_mul_pd(fscal,dy12);
1953             tz               = _mm256_mul_pd(fscal,dz12);
1954
1955             /* Update vectorial force */
1956             fix1             = _mm256_add_pd(fix1,tx);
1957             fiy1             = _mm256_add_pd(fiy1,ty);
1958             fiz1             = _mm256_add_pd(fiz1,tz);
1959
1960             fjx2             = _mm256_add_pd(fjx2,tx);
1961             fjy2             = _mm256_add_pd(fjy2,ty);
1962             fjz2             = _mm256_add_pd(fjz2,tz);
1963
1964             }
1965
1966             /**************************
1967              * CALCULATE INTERACTIONS *
1968              **************************/
1969
1970             if (gmx_mm256_any_lt(rsq20,rcutoff2))
1971             {
1972
1973             /* REACTION-FIELD ELECTROSTATICS */
1974             felec            = _mm256_mul_pd(qq20,_mm256_sub_pd(_mm256_mul_pd(rinv20,rinvsq20),krf2));
1975
1976             cutoff_mask      = _mm256_cmp_pd(rsq20,rcutoff2,_CMP_LT_OQ);
1977
1978             fscal            = felec;
1979
1980             fscal            = _mm256_and_pd(fscal,cutoff_mask);
1981
1982             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1983
1984             /* Calculate temporary vectorial force */
1985             tx               = _mm256_mul_pd(fscal,dx20);
1986             ty               = _mm256_mul_pd(fscal,dy20);
1987             tz               = _mm256_mul_pd(fscal,dz20);
1988
1989             /* Update vectorial force */
1990             fix2             = _mm256_add_pd(fix2,tx);
1991             fiy2             = _mm256_add_pd(fiy2,ty);
1992             fiz2             = _mm256_add_pd(fiz2,tz);
1993
1994             fjx0             = _mm256_add_pd(fjx0,tx);
1995             fjy0             = _mm256_add_pd(fjy0,ty);
1996             fjz0             = _mm256_add_pd(fjz0,tz);
1997
1998             }
1999
2000             /**************************
2001              * CALCULATE INTERACTIONS *
2002              **************************/
2003
2004             if (gmx_mm256_any_lt(rsq21,rcutoff2))
2005             {
2006
2007             /* REACTION-FIELD ELECTROSTATICS */
2008             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
2009
2010             cutoff_mask      = _mm256_cmp_pd(rsq21,rcutoff2,_CMP_LT_OQ);
2011
2012             fscal            = felec;
2013
2014             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2015
2016             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2017
2018             /* Calculate temporary vectorial force */
2019             tx               = _mm256_mul_pd(fscal,dx21);
2020             ty               = _mm256_mul_pd(fscal,dy21);
2021             tz               = _mm256_mul_pd(fscal,dz21);
2022
2023             /* Update vectorial force */
2024             fix2             = _mm256_add_pd(fix2,tx);
2025             fiy2             = _mm256_add_pd(fiy2,ty);
2026             fiz2             = _mm256_add_pd(fiz2,tz);
2027
2028             fjx1             = _mm256_add_pd(fjx1,tx);
2029             fjy1             = _mm256_add_pd(fjy1,ty);
2030             fjz1             = _mm256_add_pd(fjz1,tz);
2031
2032             }
2033
2034             /**************************
2035              * CALCULATE INTERACTIONS *
2036              **************************/
2037
2038             if (gmx_mm256_any_lt(rsq22,rcutoff2))
2039             {
2040
2041             /* REACTION-FIELD ELECTROSTATICS */
2042             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
2043
2044             cutoff_mask      = _mm256_cmp_pd(rsq22,rcutoff2,_CMP_LT_OQ);
2045
2046             fscal            = felec;
2047
2048             fscal            = _mm256_and_pd(fscal,cutoff_mask);
2049
2050             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
2051
2052             /* Calculate temporary vectorial force */
2053             tx               = _mm256_mul_pd(fscal,dx22);
2054             ty               = _mm256_mul_pd(fscal,dy22);
2055             tz               = _mm256_mul_pd(fscal,dz22);
2056
2057             /* Update vectorial force */
2058             fix2             = _mm256_add_pd(fix2,tx);
2059             fiy2             = _mm256_add_pd(fiy2,ty);
2060             fiz2             = _mm256_add_pd(fiz2,tz);
2061
2062             fjx2             = _mm256_add_pd(fjx2,tx);
2063             fjy2             = _mm256_add_pd(fjy2,ty);
2064             fjz2             = _mm256_add_pd(fjz2,tz);
2065
2066             }
2067
2068             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2069             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2070             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2071             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2072
2073             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA,fjptrB,fjptrC,fjptrD,
2074                                                       fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2075
2076             /* Inner loop uses 270 flops */
2077         }
2078
2079         /* End of innermost loop */
2080
2081         gmx_mm256_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
2082                                                  f+i_coord_offset,fshift+i_shift_offset);
2083
2084         /* Increment number of inner iterations */
2085         inneriter                  += j_index_end - j_index_start;
2086
2087         /* Outer loop uses 18 flops */
2088     }
2089
2090     /* Increment number of outer iterations */
2091     outeriter        += nri;
2092
2093     /* Update outer/inner flops */
2094
2095     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W3W3_F,outeriter*18 + inneriter*270);
2096 }