src/gmxlib/nonbonded/nb_kernel_avx_256_double/nb_kernel_ElecRF_VdwNone_GeomW4W4_avx_256_double.c

   1 /*
   2  * Note: this file was generated by the Gromacs avx_256_double kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_avx_256_double.h"
  34 #include "kernelutil_x86_avx_256_double.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwNone_GeomW4W4_VF_avx_256_double
  38  * Electrostatics interaction: ReactionField
  39  * VdW interaction:            None
  40  * Geometry:                   Water4-Water4
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecRF_VdwNone_GeomW4W4_VF_avx_256_double
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB,jnrC,jnrD;
  61     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  62     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  63     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  64     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  65     real             rcutoff_scalar;
  66     real             *shiftvec,*fshift,*x,*f;
  67     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  68     real             scratch[4*DIM];
  69     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  70     real *           vdwioffsetptr1;
  71     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  72     real *           vdwioffsetptr2;
  73     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  74     real *           vdwioffsetptr3;
  75     __m256d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  76     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  77     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  78     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  79     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  80     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
  81     __m256d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
  82     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
  83     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
  84     __m256d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
  85     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
  86     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
  87     __m256d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
  88     __m256d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
  89     __m256d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
  90     __m256d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
  91     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
  92     real             *charge;
  93     __m256d          dummy_mask,cutoff_mask;
  94     __m128           tmpmask0,tmpmask1;
  95     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
  96     __m256d          one     = _mm256_set1_pd(1.0);
  97     __m256d          two     = _mm256_set1_pd(2.0);
  98     x                = xx[0];
  99     f                = ff[0];
 100
 101     nri              = nlist->nri;
 102     iinr             = nlist->iinr;
 103     jindex           = nlist->jindex;
 104     jjnr             = nlist->jjnr;
 105     shiftidx         = nlist->shift;
 106     gid              = nlist->gid;
 107     shiftvec         = fr->shift_vec[0];
 108     fshift           = fr->fshift[0];
 109     facel            = _mm256_set1_pd(fr->epsfac);
 110     charge           = mdatoms->chargeA;
 111     krf              = _mm256_set1_pd(fr->ic->k_rf);
 112     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
 113     crf              = _mm256_set1_pd(fr->ic->c_rf);
 114
 115     /* Setup water-specific parameters */
 116     inr              = nlist->iinr[0];
 117     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
 118     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
 119     iq3              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
 120
 121     jq1              = _mm256_set1_pd(charge[inr+1]);
 122     jq2              = _mm256_set1_pd(charge[inr+2]);
 123     jq3              = _mm256_set1_pd(charge[inr+3]);
 124     qq11             = _mm256_mul_pd(iq1,jq1);
 125     qq12             = _mm256_mul_pd(iq1,jq2);
 126     qq13             = _mm256_mul_pd(iq1,jq3);
 127     qq21             = _mm256_mul_pd(iq2,jq1);
 128     qq22             = _mm256_mul_pd(iq2,jq2);
 129     qq23             = _mm256_mul_pd(iq2,jq3);
 130     qq31             = _mm256_mul_pd(iq3,jq1);
 131     qq32             = _mm256_mul_pd(iq3,jq2);
 132     qq33             = _mm256_mul_pd(iq3,jq3);
 133
 134     /* Avoid stupid compiler warnings */
 135     jnrA = jnrB = jnrC = jnrD = 0;
 136     j_coord_offsetA = 0;
 137     j_coord_offsetB = 0;
 138     j_coord_offsetC = 0;
 139     j_coord_offsetD = 0;
 140
 141     outeriter        = 0;
 142     inneriter        = 0;
 143
 144     for(iidx=0;iidx<4*DIM;iidx++)
 145     {
 146         scratch[iidx] = 0.0;
 147     }
 148
 149     /* Start outer loop over neighborlists */
 150     for(iidx=0; iidx<nri; iidx++)
 151     {
 152         /* Load shift vector for this list */
 153         i_shift_offset   = DIM*shiftidx[iidx];
 154
 155         /* Load limits for loop over neighbors */
 156         j_index_start    = jindex[iidx];
 157         j_index_end      = jindex[iidx+1];
 158
 159         /* Get outer coordinate index */
 160         inr              = iinr[iidx];
 161         i_coord_offset   = DIM*inr;
 162
 163         /* Load i particle coords and add shift vector */
 164         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
 165                                                     &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 166
 167         fix1             = _mm256_setzero_pd();
 168         fiy1             = _mm256_setzero_pd();
 169         fiz1             = _mm256_setzero_pd();
 170         fix2             = _mm256_setzero_pd();
 171         fiy2             = _mm256_setzero_pd();
 172         fiz2             = _mm256_setzero_pd();
 173         fix3             = _mm256_setzero_pd();
 174         fiy3             = _mm256_setzero_pd();
 175         fiz3             = _mm256_setzero_pd();
 176
 177         /* Reset potential sums */
 178         velecsum         = _mm256_setzero_pd();
 179
 180         /* Start inner kernel loop */
 181         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 182         {
 183
 184             /* Get j neighbor index, and coordinate index */
 185             jnrA             = jjnr[jidx];
 186             jnrB             = jjnr[jidx+1];
 187             jnrC             = jjnr[jidx+2];
 188             jnrD             = jjnr[jidx+3];
 189             j_coord_offsetA  = DIM*jnrA;
 190             j_coord_offsetB  = DIM*jnrB;
 191             j_coord_offsetC  = DIM*jnrC;
 192             j_coord_offsetD  = DIM*jnrD;
 193
 194             /* load j atom coordinates */
 195             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
 196                                                  x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
 197                                                  &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
 198
 199             /* Calculate displacement vector */
 200             dx11             = _mm256_sub_pd(ix1,jx1);
 201             dy11             = _mm256_sub_pd(iy1,jy1);
 202             dz11             = _mm256_sub_pd(iz1,jz1);
 203             dx12             = _mm256_sub_pd(ix1,jx2);
 204             dy12             = _mm256_sub_pd(iy1,jy2);
 205             dz12             = _mm256_sub_pd(iz1,jz2);
 206             dx13             = _mm256_sub_pd(ix1,jx3);
 207             dy13             = _mm256_sub_pd(iy1,jy3);
 208             dz13             = _mm256_sub_pd(iz1,jz3);
 209             dx21             = _mm256_sub_pd(ix2,jx1);
 210             dy21             = _mm256_sub_pd(iy2,jy1);
 211             dz21             = _mm256_sub_pd(iz2,jz1);
 212             dx22             = _mm256_sub_pd(ix2,jx2);
 213             dy22             = _mm256_sub_pd(iy2,jy2);
 214             dz22             = _mm256_sub_pd(iz2,jz2);
 215             dx23             = _mm256_sub_pd(ix2,jx3);
 216             dy23             = _mm256_sub_pd(iy2,jy3);
 217             dz23             = _mm256_sub_pd(iz2,jz3);
 218             dx31             = _mm256_sub_pd(ix3,jx1);
 219             dy31             = _mm256_sub_pd(iy3,jy1);
 220             dz31             = _mm256_sub_pd(iz3,jz1);
 221             dx32             = _mm256_sub_pd(ix3,jx2);
 222             dy32             = _mm256_sub_pd(iy3,jy2);
 223             dz32             = _mm256_sub_pd(iz3,jz2);
 224             dx33             = _mm256_sub_pd(ix3,jx3);
 225             dy33             = _mm256_sub_pd(iy3,jy3);
 226             dz33             = _mm256_sub_pd(iz3,jz3);
 227
 228             /* Calculate squared distance and things based on it */
 229             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 230             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 231             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
 232             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 233             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 234             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
 235             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
 236             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
 237             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
 238
 239             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 240             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 241             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
 242             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 243             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 244             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
 245             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
 246             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
 247             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
 248
 249             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 250             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 251             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
 252             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 253             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 254             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
 255             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
 256             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
 257             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
 258
 259             fjx1             = _mm256_setzero_pd();
 260             fjy1             = _mm256_setzero_pd();
 261             fjz1             = _mm256_setzero_pd();
 262             fjx2             = _mm256_setzero_pd();
 263             fjy2             = _mm256_setzero_pd();
 264             fjz2             = _mm256_setzero_pd();
 265             fjx3             = _mm256_setzero_pd();
 266             fjy3             = _mm256_setzero_pd();
 267             fjz3             = _mm256_setzero_pd();
 268
 269             /**************************
 270              * CALCULATE INTERACTIONS *
 271              **************************/
 272
 273             /* REACTION-FIELD ELECTROSTATICS */
 274             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 275             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 276
 277             /* Update potential sum for this i atom from the interaction with this j atom. */
 278             velecsum         = _mm256_add_pd(velecsum,velec);
 279
 280             fscal            = felec;
 281
 282             /* Calculate temporary vectorial force */
 283             tx               = _mm256_mul_pd(fscal,dx11);
 284             ty               = _mm256_mul_pd(fscal,dy11);
 285             tz               = _mm256_mul_pd(fscal,dz11);
 286
 287             /* Update vectorial force */
 288             fix1             = _mm256_add_pd(fix1,tx);
 289             fiy1             = _mm256_add_pd(fiy1,ty);
 290             fiz1             = _mm256_add_pd(fiz1,tz);
 291
 292             fjx1             = _mm256_add_pd(fjx1,tx);
 293             fjy1             = _mm256_add_pd(fjy1,ty);
 294             fjz1             = _mm256_add_pd(fjz1,tz);
 295
 296             /**************************
 297              * CALCULATE INTERACTIONS *
 298              **************************/
 299
 300             /* REACTION-FIELD ELECTROSTATICS */
 301             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
 302             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
 303
 304             /* Update potential sum for this i atom from the interaction with this j atom. */
 305             velecsum         = _mm256_add_pd(velecsum,velec);
 306
 307             fscal            = felec;
 308
 309             /* Calculate temporary vectorial force */
 310             tx               = _mm256_mul_pd(fscal,dx12);
 311             ty               = _mm256_mul_pd(fscal,dy12);
 312             tz               = _mm256_mul_pd(fscal,dz12);
 313
 314             /* Update vectorial force */
 315             fix1             = _mm256_add_pd(fix1,tx);
 316             fiy1             = _mm256_add_pd(fiy1,ty);
 317             fiz1             = _mm256_add_pd(fiz1,tz);
 318
 319             fjx2             = _mm256_add_pd(fjx2,tx);
 320             fjy2             = _mm256_add_pd(fjy2,ty);
 321             fjz2             = _mm256_add_pd(fjz2,tz);
 322
 323             /**************************
 324              * CALCULATE INTERACTIONS *
 325              **************************/
 326
 327             /* REACTION-FIELD ELECTROSTATICS */
 328             velec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_add_pd(rinv13,_mm256_mul_pd(krf,rsq13)),crf));
 329             felec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
 330
 331             /* Update potential sum for this i atom from the interaction with this j atom. */
 332             velecsum         = _mm256_add_pd(velecsum,velec);
 333
 334             fscal            = felec;
 335
 336             /* Calculate temporary vectorial force */
 337             tx               = _mm256_mul_pd(fscal,dx13);
 338             ty               = _mm256_mul_pd(fscal,dy13);
 339             tz               = _mm256_mul_pd(fscal,dz13);
 340
 341             /* Update vectorial force */
 342             fix1             = _mm256_add_pd(fix1,tx);
 343             fiy1             = _mm256_add_pd(fiy1,ty);
 344             fiz1             = _mm256_add_pd(fiz1,tz);
 345
 346             fjx3             = _mm256_add_pd(fjx3,tx);
 347             fjy3             = _mm256_add_pd(fjy3,ty);
 348             fjz3             = _mm256_add_pd(fjz3,tz);
 349
 350             /**************************
 351              * CALCULATE INTERACTIONS *
 352              **************************/
 353
 354             /* REACTION-FIELD ELECTROSTATICS */
 355             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
 356             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
 357
 358             /* Update potential sum for this i atom from the interaction with this j atom. */
 359             velecsum         = _mm256_add_pd(velecsum,velec);
 360
 361             fscal            = felec;
 362
 363             /* Calculate temporary vectorial force */
 364             tx               = _mm256_mul_pd(fscal,dx21);
 365             ty               = _mm256_mul_pd(fscal,dy21);
 366             tz               = _mm256_mul_pd(fscal,dz21);
 367
 368             /* Update vectorial force */
 369             fix2             = _mm256_add_pd(fix2,tx);
 370             fiy2             = _mm256_add_pd(fiy2,ty);
 371             fiz2             = _mm256_add_pd(fiz2,tz);
 372
 373             fjx1             = _mm256_add_pd(fjx1,tx);
 374             fjy1             = _mm256_add_pd(fjy1,ty);
 375             fjz1             = _mm256_add_pd(fjz1,tz);
 376
 377             /**************************
 378              * CALCULATE INTERACTIONS *
 379              **************************/
 380
 381             /* REACTION-FIELD ELECTROSTATICS */
 382             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
 383             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
 384
 385             /* Update potential sum for this i atom from the interaction with this j atom. */
 386             velecsum         = _mm256_add_pd(velecsum,velec);
 387
 388             fscal            = felec;
 389
 390             /* Calculate temporary vectorial force */
 391             tx               = _mm256_mul_pd(fscal,dx22);
 392             ty               = _mm256_mul_pd(fscal,dy22);
 393             tz               = _mm256_mul_pd(fscal,dz22);
 394
 395             /* Update vectorial force */
 396             fix2             = _mm256_add_pd(fix2,tx);
 397             fiy2             = _mm256_add_pd(fiy2,ty);
 398             fiz2             = _mm256_add_pd(fiz2,tz);
 399
 400             fjx2             = _mm256_add_pd(fjx2,tx);
 401             fjy2             = _mm256_add_pd(fjy2,ty);
 402             fjz2             = _mm256_add_pd(fjz2,tz);
 403
 404             /**************************
 405              * CALCULATE INTERACTIONS *
 406              **************************/
 407
 408             /* REACTION-FIELD ELECTROSTATICS */
 409             velec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_add_pd(rinv23,_mm256_mul_pd(krf,rsq23)),crf));
 410             felec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
 411
 412             /* Update potential sum for this i atom from the interaction with this j atom. */
 413             velecsum         = _mm256_add_pd(velecsum,velec);
 414
 415             fscal            = felec;
 416
 417             /* Calculate temporary vectorial force */
 418             tx               = _mm256_mul_pd(fscal,dx23);
 419             ty               = _mm256_mul_pd(fscal,dy23);
 420             tz               = _mm256_mul_pd(fscal,dz23);
 421
 422             /* Update vectorial force */
 423             fix2             = _mm256_add_pd(fix2,tx);
 424             fiy2             = _mm256_add_pd(fiy2,ty);
 425             fiz2             = _mm256_add_pd(fiz2,tz);
 426
 427             fjx3             = _mm256_add_pd(fjx3,tx);
 428             fjy3             = _mm256_add_pd(fjy3,ty);
 429             fjz3             = _mm256_add_pd(fjz3,tz);
 430
 431             /**************************
 432              * CALCULATE INTERACTIONS *
 433              **************************/
 434
 435             /* REACTION-FIELD ELECTROSTATICS */
 436             velec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_add_pd(rinv31,_mm256_mul_pd(krf,rsq31)),crf));
 437             felec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
 438
 439             /* Update potential sum for this i atom from the interaction with this j atom. */
 440             velecsum         = _mm256_add_pd(velecsum,velec);
 441
 442             fscal            = felec;
 443
 444             /* Calculate temporary vectorial force */
 445             tx               = _mm256_mul_pd(fscal,dx31);
 446             ty               = _mm256_mul_pd(fscal,dy31);
 447             tz               = _mm256_mul_pd(fscal,dz31);
 448
 449             /* Update vectorial force */
 450             fix3             = _mm256_add_pd(fix3,tx);
 451             fiy3             = _mm256_add_pd(fiy3,ty);
 452             fiz3             = _mm256_add_pd(fiz3,tz);
 453
 454             fjx1             = _mm256_add_pd(fjx1,tx);
 455             fjy1             = _mm256_add_pd(fjy1,ty);
 456             fjz1             = _mm256_add_pd(fjz1,tz);
 457
 458             /**************************
 459              * CALCULATE INTERACTIONS *
 460              **************************/
 461
 462             /* REACTION-FIELD ELECTROSTATICS */
 463             velec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_add_pd(rinv32,_mm256_mul_pd(krf,rsq32)),crf));
 464             felec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
 465
 466             /* Update potential sum for this i atom from the interaction with this j atom. */
 467             velecsum         = _mm256_add_pd(velecsum,velec);
 468
 469             fscal            = felec;
 470
 471             /* Calculate temporary vectorial force */
 472             tx               = _mm256_mul_pd(fscal,dx32);
 473             ty               = _mm256_mul_pd(fscal,dy32);
 474             tz               = _mm256_mul_pd(fscal,dz32);
 475
 476             /* Update vectorial force */
 477             fix3             = _mm256_add_pd(fix3,tx);
 478             fiy3             = _mm256_add_pd(fiy3,ty);
 479             fiz3             = _mm256_add_pd(fiz3,tz);
 480
 481             fjx2             = _mm256_add_pd(fjx2,tx);
 482             fjy2             = _mm256_add_pd(fjy2,ty);
 483             fjz2             = _mm256_add_pd(fjz2,tz);
 484
 485             /**************************
 486              * CALCULATE INTERACTIONS *
 487              **************************/
 488
 489             /* REACTION-FIELD ELECTROSTATICS */
 490             velec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_add_pd(rinv33,_mm256_mul_pd(krf,rsq33)),crf));
 491             felec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
 492
 493             /* Update potential sum for this i atom from the interaction with this j atom. */
 494             velecsum         = _mm256_add_pd(velecsum,velec);
 495
 496             fscal            = felec;
 497
 498             /* Calculate temporary vectorial force */
 499             tx               = _mm256_mul_pd(fscal,dx33);
 500             ty               = _mm256_mul_pd(fscal,dy33);
 501             tz               = _mm256_mul_pd(fscal,dz33);
 502
 503             /* Update vectorial force */
 504             fix3             = _mm256_add_pd(fix3,tx);
 505             fiy3             = _mm256_add_pd(fiy3,ty);
 506             fiz3             = _mm256_add_pd(fiz3,tz);
 507
 508             fjx3             = _mm256_add_pd(fjx3,tx);
 509             fjy3             = _mm256_add_pd(fjy3,ty);
 510             fjz3             = _mm256_add_pd(fjz3,tz);
 511
 512             fjptrA             = f+j_coord_offsetA;
 513             fjptrB             = f+j_coord_offsetB;
 514             fjptrC             = f+j_coord_offsetC;
 515             fjptrD             = f+j_coord_offsetD;
 516
 517             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
 518                                                       fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 519
 520             /* Inner loop uses 288 flops */
 521         }
 522
 523         if(jidx<j_index_end)
 524         {
 525
 526             /* Get j neighbor index, and coordinate index */
 527             jnrlistA         = jjnr[jidx];
 528             jnrlistB         = jjnr[jidx+1];
 529             jnrlistC         = jjnr[jidx+2];
 530             jnrlistD         = jjnr[jidx+3];
 531             /* Sign of each element will be negative for non-real atoms.
 532              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 533              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
 534              */
 535             tmpmask0 = gmx_mm_castsi128_pd(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 536
 537             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
 538             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
 539             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
 540
 541             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 542             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 543             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 544             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 545             j_coord_offsetA  = DIM*jnrA;
 546             j_coord_offsetB  = DIM*jnrB;
 547             j_coord_offsetC  = DIM*jnrC;
 548             j_coord_offsetD  = DIM*jnrD;
 549
 550             /* load j atom coordinates */
 551             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
 552                                                  x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
 553                                                  &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
 554
 555             /* Calculate displacement vector */
 556             dx11             = _mm256_sub_pd(ix1,jx1);
 557             dy11             = _mm256_sub_pd(iy1,jy1);
 558             dz11             = _mm256_sub_pd(iz1,jz1);
 559             dx12             = _mm256_sub_pd(ix1,jx2);
 560             dy12             = _mm256_sub_pd(iy1,jy2);
 561             dz12             = _mm256_sub_pd(iz1,jz2);
 562             dx13             = _mm256_sub_pd(ix1,jx3);
 563             dy13             = _mm256_sub_pd(iy1,jy3);
 564             dz13             = _mm256_sub_pd(iz1,jz3);
 565             dx21             = _mm256_sub_pd(ix2,jx1);
 566             dy21             = _mm256_sub_pd(iy2,jy1);
 567             dz21             = _mm256_sub_pd(iz2,jz1);
 568             dx22             = _mm256_sub_pd(ix2,jx2);
 569             dy22             = _mm256_sub_pd(iy2,jy2);
 570             dz22             = _mm256_sub_pd(iz2,jz2);
 571             dx23             = _mm256_sub_pd(ix2,jx3);
 572             dy23             = _mm256_sub_pd(iy2,jy3);
 573             dz23             = _mm256_sub_pd(iz2,jz3);
 574             dx31             = _mm256_sub_pd(ix3,jx1);
 575             dy31             = _mm256_sub_pd(iy3,jy1);
 576             dz31             = _mm256_sub_pd(iz3,jz1);
 577             dx32             = _mm256_sub_pd(ix3,jx2);
 578             dy32             = _mm256_sub_pd(iy3,jy2);
 579             dz32             = _mm256_sub_pd(iz3,jz2);
 580             dx33             = _mm256_sub_pd(ix3,jx3);
 581             dy33             = _mm256_sub_pd(iy3,jy3);
 582             dz33             = _mm256_sub_pd(iz3,jz3);
 583
 584             /* Calculate squared distance and things based on it */
 585             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
 586             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
 587             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
 588             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
 589             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
 590             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
 591             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
 592             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
 593             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
 594
 595             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
 596             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
 597             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
 598             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
 599             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
 600             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
 601             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
 602             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
 603             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
 604
 605             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
 606             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
 607             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
 608             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
 609             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
 610             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
 611             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
 612             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
 613             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
 614
 615             fjx1             = _mm256_setzero_pd();
 616             fjy1             = _mm256_setzero_pd();
 617             fjz1             = _mm256_setzero_pd();
 618             fjx2             = _mm256_setzero_pd();
 619             fjy2             = _mm256_setzero_pd();
 620             fjz2             = _mm256_setzero_pd();
 621             fjx3             = _mm256_setzero_pd();
 622             fjy3             = _mm256_setzero_pd();
 623             fjz3             = _mm256_setzero_pd();
 624
 625             /**************************
 626              * CALCULATE INTERACTIONS *
 627              **************************/
 628
 629             /* REACTION-FIELD ELECTROSTATICS */
 630             velec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_add_pd(rinv11,_mm256_mul_pd(krf,rsq11)),crf));
 631             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
 632
 633             /* Update potential sum for this i atom from the interaction with this j atom. */
 634             velec            = _mm256_andnot_pd(dummy_mask,velec);
 635             velecsum         = _mm256_add_pd(velecsum,velec);
 636
 637             fscal            = felec;
 638
 639             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 640
 641             /* Calculate temporary vectorial force */
 642             tx               = _mm256_mul_pd(fscal,dx11);
 643             ty               = _mm256_mul_pd(fscal,dy11);
 644             tz               = _mm256_mul_pd(fscal,dz11);
 645
 646             /* Update vectorial force */
 647             fix1             = _mm256_add_pd(fix1,tx);
 648             fiy1             = _mm256_add_pd(fiy1,ty);
 649             fiz1             = _mm256_add_pd(fiz1,tz);
 650
 651             fjx1             = _mm256_add_pd(fjx1,tx);
 652             fjy1             = _mm256_add_pd(fjy1,ty);
 653             fjz1             = _mm256_add_pd(fjz1,tz);
 654
 655             /**************************
 656              * CALCULATE INTERACTIONS *
 657              **************************/
 658
 659             /* REACTION-FIELD ELECTROSTATICS */
 660             velec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_add_pd(rinv12,_mm256_mul_pd(krf,rsq12)),crf));
 661             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
 662
 663             /* Update potential sum for this i atom from the interaction with this j atom. */
 664             velec            = _mm256_andnot_pd(dummy_mask,velec);
 665             velecsum         = _mm256_add_pd(velecsum,velec);
 666
 667             fscal            = felec;
 668
 669             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 670
 671             /* Calculate temporary vectorial force */
 672             tx               = _mm256_mul_pd(fscal,dx12);
 673             ty               = _mm256_mul_pd(fscal,dy12);
 674             tz               = _mm256_mul_pd(fscal,dz12);
 675
 676             /* Update vectorial force */
 677             fix1             = _mm256_add_pd(fix1,tx);
 678             fiy1             = _mm256_add_pd(fiy1,ty);
 679             fiz1             = _mm256_add_pd(fiz1,tz);
 680
 681             fjx2             = _mm256_add_pd(fjx2,tx);
 682             fjy2             = _mm256_add_pd(fjy2,ty);
 683             fjz2             = _mm256_add_pd(fjz2,tz);
 684
 685             /**************************
 686              * CALCULATE INTERACTIONS *
 687              **************************/
 688
 689             /* REACTION-FIELD ELECTROSTATICS */
 690             velec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_add_pd(rinv13,_mm256_mul_pd(krf,rsq13)),crf));
 691             felec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
 692
 693             /* Update potential sum for this i atom from the interaction with this j atom. */
 694             velec            = _mm256_andnot_pd(dummy_mask,velec);
 695             velecsum         = _mm256_add_pd(velecsum,velec);
 696
 697             fscal            = felec;
 698
 699             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 700
 701             /* Calculate temporary vectorial force */
 702             tx               = _mm256_mul_pd(fscal,dx13);
 703             ty               = _mm256_mul_pd(fscal,dy13);
 704             tz               = _mm256_mul_pd(fscal,dz13);
 705
 706             /* Update vectorial force */
 707             fix1             = _mm256_add_pd(fix1,tx);
 708             fiy1             = _mm256_add_pd(fiy1,ty);
 709             fiz1             = _mm256_add_pd(fiz1,tz);
 710
 711             fjx3             = _mm256_add_pd(fjx3,tx);
 712             fjy3             = _mm256_add_pd(fjy3,ty);
 713             fjz3             = _mm256_add_pd(fjz3,tz);
 714
 715             /**************************
 716              * CALCULATE INTERACTIONS *
 717              **************************/
 718
 719             /* REACTION-FIELD ELECTROSTATICS */
 720             velec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_add_pd(rinv21,_mm256_mul_pd(krf,rsq21)),crf));
 721             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
 722
 723             /* Update potential sum for this i atom from the interaction with this j atom. */
 724             velec            = _mm256_andnot_pd(dummy_mask,velec);
 725             velecsum         = _mm256_add_pd(velecsum,velec);
 726
 727             fscal            = felec;
 728
 729             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 730
 731             /* Calculate temporary vectorial force */
 732             tx               = _mm256_mul_pd(fscal,dx21);
 733             ty               = _mm256_mul_pd(fscal,dy21);
 734             tz               = _mm256_mul_pd(fscal,dz21);
 735
 736             /* Update vectorial force */
 737             fix2             = _mm256_add_pd(fix2,tx);
 738             fiy2             = _mm256_add_pd(fiy2,ty);
 739             fiz2             = _mm256_add_pd(fiz2,tz);
 740
 741             fjx1             = _mm256_add_pd(fjx1,tx);
 742             fjy1             = _mm256_add_pd(fjy1,ty);
 743             fjz1             = _mm256_add_pd(fjz1,tz);
 744
 745             /**************************
 746              * CALCULATE INTERACTIONS *
 747              **************************/
 748
 749             /* REACTION-FIELD ELECTROSTATICS */
 750             velec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_add_pd(rinv22,_mm256_mul_pd(krf,rsq22)),crf));
 751             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
 752
 753             /* Update potential sum for this i atom from the interaction with this j atom. */
 754             velec            = _mm256_andnot_pd(dummy_mask,velec);
 755             velecsum         = _mm256_add_pd(velecsum,velec);
 756
 757             fscal            = felec;
 758
 759             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 760
 761             /* Calculate temporary vectorial force */
 762             tx               = _mm256_mul_pd(fscal,dx22);
 763             ty               = _mm256_mul_pd(fscal,dy22);
 764             tz               = _mm256_mul_pd(fscal,dz22);
 765
 766             /* Update vectorial force */
 767             fix2             = _mm256_add_pd(fix2,tx);
 768             fiy2             = _mm256_add_pd(fiy2,ty);
 769             fiz2             = _mm256_add_pd(fiz2,tz);
 770
 771             fjx2             = _mm256_add_pd(fjx2,tx);
 772             fjy2             = _mm256_add_pd(fjy2,ty);
 773             fjz2             = _mm256_add_pd(fjz2,tz);
 774
 775             /**************************
 776              * CALCULATE INTERACTIONS *
 777              **************************/
 778
 779             /* REACTION-FIELD ELECTROSTATICS */
 780             velec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_add_pd(rinv23,_mm256_mul_pd(krf,rsq23)),crf));
 781             felec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
 782
 783             /* Update potential sum for this i atom from the interaction with this j atom. */
 784             velec            = _mm256_andnot_pd(dummy_mask,velec);
 785             velecsum         = _mm256_add_pd(velecsum,velec);
 786
 787             fscal            = felec;
 788
 789             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 790
 791             /* Calculate temporary vectorial force */
 792             tx               = _mm256_mul_pd(fscal,dx23);
 793             ty               = _mm256_mul_pd(fscal,dy23);
 794             tz               = _mm256_mul_pd(fscal,dz23);
 795
 796             /* Update vectorial force */
 797             fix2             = _mm256_add_pd(fix2,tx);
 798             fiy2             = _mm256_add_pd(fiy2,ty);
 799             fiz2             = _mm256_add_pd(fiz2,tz);
 800
 801             fjx3             = _mm256_add_pd(fjx3,tx);
 802             fjy3             = _mm256_add_pd(fjy3,ty);
 803             fjz3             = _mm256_add_pd(fjz3,tz);
 804
 805             /**************************
 806              * CALCULATE INTERACTIONS *
 807              **************************/
 808
 809             /* REACTION-FIELD ELECTROSTATICS */
 810             velec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_add_pd(rinv31,_mm256_mul_pd(krf,rsq31)),crf));
 811             felec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
 812
 813             /* Update potential sum for this i atom from the interaction with this j atom. */
 814             velec            = _mm256_andnot_pd(dummy_mask,velec);
 815             velecsum         = _mm256_add_pd(velecsum,velec);
 816
 817             fscal            = felec;
 818
 819             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 820
 821             /* Calculate temporary vectorial force */
 822             tx               = _mm256_mul_pd(fscal,dx31);
 823             ty               = _mm256_mul_pd(fscal,dy31);
 824             tz               = _mm256_mul_pd(fscal,dz31);
 825
 826             /* Update vectorial force */
 827             fix3             = _mm256_add_pd(fix3,tx);
 828             fiy3             = _mm256_add_pd(fiy3,ty);
 829             fiz3             = _mm256_add_pd(fiz3,tz);
 830
 831             fjx1             = _mm256_add_pd(fjx1,tx);
 832             fjy1             = _mm256_add_pd(fjy1,ty);
 833             fjz1             = _mm256_add_pd(fjz1,tz);
 834
 835             /**************************
 836              * CALCULATE INTERACTIONS *
 837              **************************/
 838
 839             /* REACTION-FIELD ELECTROSTATICS */
 840             velec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_add_pd(rinv32,_mm256_mul_pd(krf,rsq32)),crf));
 841             felec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
 842
 843             /* Update potential sum for this i atom from the interaction with this j atom. */
 844             velec            = _mm256_andnot_pd(dummy_mask,velec);
 845             velecsum         = _mm256_add_pd(velecsum,velec);
 846
 847             fscal            = felec;
 848
 849             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 850
 851             /* Calculate temporary vectorial force */
 852             tx               = _mm256_mul_pd(fscal,dx32);
 853             ty               = _mm256_mul_pd(fscal,dy32);
 854             tz               = _mm256_mul_pd(fscal,dz32);
 855
 856             /* Update vectorial force */
 857             fix3             = _mm256_add_pd(fix3,tx);
 858             fiy3             = _mm256_add_pd(fiy3,ty);
 859             fiz3             = _mm256_add_pd(fiz3,tz);
 860
 861             fjx2             = _mm256_add_pd(fjx2,tx);
 862             fjy2             = _mm256_add_pd(fjy2,ty);
 863             fjz2             = _mm256_add_pd(fjz2,tz);
 864
 865             /**************************
 866              * CALCULATE INTERACTIONS *
 867              **************************/
 868
 869             /* REACTION-FIELD ELECTROSTATICS */
 870             velec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_add_pd(rinv33,_mm256_mul_pd(krf,rsq33)),crf));
 871             felec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
 872
 873             /* Update potential sum for this i atom from the interaction with this j atom. */
 874             velec            = _mm256_andnot_pd(dummy_mask,velec);
 875             velecsum         = _mm256_add_pd(velecsum,velec);
 876
 877             fscal            = felec;
 878
 879             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
 880
 881             /* Calculate temporary vectorial force */
 882             tx               = _mm256_mul_pd(fscal,dx33);
 883             ty               = _mm256_mul_pd(fscal,dy33);
 884             tz               = _mm256_mul_pd(fscal,dz33);
 885
 886             /* Update vectorial force */
 887             fix3             = _mm256_add_pd(fix3,tx);
 888             fiy3             = _mm256_add_pd(fiy3,ty);
 889             fiz3             = _mm256_add_pd(fiz3,tz);
 890
 891             fjx3             = _mm256_add_pd(fjx3,tx);
 892             fjy3             = _mm256_add_pd(fjy3,ty);
 893             fjz3             = _mm256_add_pd(fjz3,tz);
 894
 895             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 896             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 897             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 898             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 899
 900             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
 901                                                       fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 902
 903             /* Inner loop uses 288 flops */
 904         }
 905
 906         /* End of innermost loop */
 907
 908         gmx_mm256_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 909                                                  f+i_coord_offset+DIM,fshift+i_shift_offset);
 910
 911         ggid                        = gid[iidx];
 912         /* Update potential energies */
 913         gmx_mm256_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 914
 915         /* Increment number of inner iterations */
 916         inneriter                  += j_index_end - j_index_start;
 917
 918         /* Outer loop uses 19 flops */
 919     }
 920
 921     /* Increment number of outer iterations */
 922     outeriter        += nri;
 923
 924     /* Update outer/inner flops */
 925
 926     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_VF,outeriter*19 + inneriter*288);
 927 }
 928 /*
 929  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwNone_GeomW4W4_F_avx_256_double
 930  * Electrostatics interaction: ReactionField
 931  * VdW interaction:            None
 932  * Geometry:                   Water4-Water4
 933  * Calculate force/pot:        Force
 934  */
 935 void
 936 nb_kernel_ElecRF_VdwNone_GeomW4W4_F_avx_256_double
 937                     (t_nblist * gmx_restrict                nlist,
 938                      rvec * gmx_restrict                    xx,
 939                      rvec * gmx_restrict                    ff,
 940                      t_forcerec * gmx_restrict              fr,
 941                      t_mdatoms * gmx_restrict               mdatoms,
 942                      nb_kernel_data_t * gmx_restrict        kernel_data,
 943                      t_nrnb * gmx_restrict                  nrnb)
 944 {
 945     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 946      * just 0 for non-waters.
 947      * Suffixes A,B,C,D refer to j loop unrolling done with AVX, e.g. for the four different
 948      * jnr indices corresponding to data put in the four positions in the SIMD register.
 949      */
 950     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 951     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 952     int              jnrA,jnrB,jnrC,jnrD;
 953     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
 954     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
 955     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
 956     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 957     real             rcutoff_scalar;
 958     real             *shiftvec,*fshift,*x,*f;
 959     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
 960     real             scratch[4*DIM];
 961     __m256d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 962     real *           vdwioffsetptr1;
 963     __m256d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 964     real *           vdwioffsetptr2;
 965     __m256d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 966     real *           vdwioffsetptr3;
 967     __m256d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 968     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
 969     __m256d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
 970     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
 971     __m256d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
 972     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
 973     __m256d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
 974     __m256d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 975     __m256d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 976     __m256d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
 977     __m256d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 978     __m256d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 979     __m256d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
 980     __m256d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
 981     __m256d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
 982     __m256d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
 983     __m256d          velec,felec,velecsum,facel,crf,krf,krf2;
 984     real             *charge;
 985     __m256d          dummy_mask,cutoff_mask;
 986     __m128           tmpmask0,tmpmask1;
 987     __m256d          signbit = _mm256_castsi256_pd( _mm256_set1_epi32(0x80000000) );
 988     __m256d          one     = _mm256_set1_pd(1.0);
 989     __m256d          two     = _mm256_set1_pd(2.0);
 990     x                = xx[0];
 991     f                = ff[0];
 992
 993     nri              = nlist->nri;
 994     iinr             = nlist->iinr;
 995     jindex           = nlist->jindex;
 996     jjnr             = nlist->jjnr;
 997     shiftidx         = nlist->shift;
 998     gid              = nlist->gid;
 999     shiftvec         = fr->shift_vec[0];
1000     fshift           = fr->fshift[0];
1001     facel            = _mm256_set1_pd(fr->epsfac);
1002     charge           = mdatoms->chargeA;
1003     krf              = _mm256_set1_pd(fr->ic->k_rf);
1004     krf2             = _mm256_set1_pd(fr->ic->k_rf*2.0);
1005     crf              = _mm256_set1_pd(fr->ic->c_rf);
1006
1007     /* Setup water-specific parameters */
1008     inr              = nlist->iinr[0];
1009     iq1              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+1]));
1010     iq2              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+2]));
1011     iq3              = _mm256_mul_pd(facel,_mm256_set1_pd(charge[inr+3]));
1012
1013     jq1              = _mm256_set1_pd(charge[inr+1]);
1014     jq2              = _mm256_set1_pd(charge[inr+2]);
1015     jq3              = _mm256_set1_pd(charge[inr+3]);
1016     qq11             = _mm256_mul_pd(iq1,jq1);
1017     qq12             = _mm256_mul_pd(iq1,jq2);
1018     qq13             = _mm256_mul_pd(iq1,jq3);
1019     qq21             = _mm256_mul_pd(iq2,jq1);
1020     qq22             = _mm256_mul_pd(iq2,jq2);
1021     qq23             = _mm256_mul_pd(iq2,jq3);
1022     qq31             = _mm256_mul_pd(iq3,jq1);
1023     qq32             = _mm256_mul_pd(iq3,jq2);
1024     qq33             = _mm256_mul_pd(iq3,jq3);
1025
1026     /* Avoid stupid compiler warnings */
1027     jnrA = jnrB = jnrC = jnrD = 0;
1028     j_coord_offsetA = 0;
1029     j_coord_offsetB = 0;
1030     j_coord_offsetC = 0;
1031     j_coord_offsetD = 0;
1032
1033     outeriter        = 0;
1034     inneriter        = 0;
1035
1036     for(iidx=0;iidx<4*DIM;iidx++)
1037     {
1038         scratch[iidx] = 0.0;
1039     }
1040
1041     /* Start outer loop over neighborlists */
1042     for(iidx=0; iidx<nri; iidx++)
1043     {
1044         /* Load shift vector for this list */
1045         i_shift_offset   = DIM*shiftidx[iidx];
1046
1047         /* Load limits for loop over neighbors */
1048         j_index_start    = jindex[iidx];
1049         j_index_end      = jindex[iidx+1];
1050
1051         /* Get outer coordinate index */
1052         inr              = iinr[iidx];
1053         i_coord_offset   = DIM*inr;
1054
1055         /* Load i particle coords and add shift vector */
1056         gmx_mm256_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
1057                                                     &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1058
1059         fix1             = _mm256_setzero_pd();
1060         fiy1             = _mm256_setzero_pd();
1061         fiz1             = _mm256_setzero_pd();
1062         fix2             = _mm256_setzero_pd();
1063         fiy2             = _mm256_setzero_pd();
1064         fiz2             = _mm256_setzero_pd();
1065         fix3             = _mm256_setzero_pd();
1066         fiy3             = _mm256_setzero_pd();
1067         fiz3             = _mm256_setzero_pd();
1068
1069         /* Start inner kernel loop */
1070         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1071         {
1072
1073             /* Get j neighbor index, and coordinate index */
1074             jnrA             = jjnr[jidx];
1075             jnrB             = jjnr[jidx+1];
1076             jnrC             = jjnr[jidx+2];
1077             jnrD             = jjnr[jidx+3];
1078             j_coord_offsetA  = DIM*jnrA;
1079             j_coord_offsetB  = DIM*jnrB;
1080             j_coord_offsetC  = DIM*jnrC;
1081             j_coord_offsetD  = DIM*jnrD;
1082
1083             /* load j atom coordinates */
1084             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1085                                                  x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1086                                                  &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1087
1088             /* Calculate displacement vector */
1089             dx11             = _mm256_sub_pd(ix1,jx1);
1090             dy11             = _mm256_sub_pd(iy1,jy1);
1091             dz11             = _mm256_sub_pd(iz1,jz1);
1092             dx12             = _mm256_sub_pd(ix1,jx2);
1093             dy12             = _mm256_sub_pd(iy1,jy2);
1094             dz12             = _mm256_sub_pd(iz1,jz2);
1095             dx13             = _mm256_sub_pd(ix1,jx3);
1096             dy13             = _mm256_sub_pd(iy1,jy3);
1097             dz13             = _mm256_sub_pd(iz1,jz3);
1098             dx21             = _mm256_sub_pd(ix2,jx1);
1099             dy21             = _mm256_sub_pd(iy2,jy1);
1100             dz21             = _mm256_sub_pd(iz2,jz1);
1101             dx22             = _mm256_sub_pd(ix2,jx2);
1102             dy22             = _mm256_sub_pd(iy2,jy2);
1103             dz22             = _mm256_sub_pd(iz2,jz2);
1104             dx23             = _mm256_sub_pd(ix2,jx3);
1105             dy23             = _mm256_sub_pd(iy2,jy3);
1106             dz23             = _mm256_sub_pd(iz2,jz3);
1107             dx31             = _mm256_sub_pd(ix3,jx1);
1108             dy31             = _mm256_sub_pd(iy3,jy1);
1109             dz31             = _mm256_sub_pd(iz3,jz1);
1110             dx32             = _mm256_sub_pd(ix3,jx2);
1111             dy32             = _mm256_sub_pd(iy3,jy2);
1112             dz32             = _mm256_sub_pd(iz3,jz2);
1113             dx33             = _mm256_sub_pd(ix3,jx3);
1114             dy33             = _mm256_sub_pd(iy3,jy3);
1115             dz33             = _mm256_sub_pd(iz3,jz3);
1116
1117             /* Calculate squared distance and things based on it */
1118             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1119             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1120             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
1121             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1122             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1123             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
1124             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
1125             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
1126             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
1127
1128             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1129             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1130             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
1131             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1132             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1133             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
1134             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
1135             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
1136             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
1137
1138             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1139             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1140             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
1141             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1142             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1143             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
1144             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
1145             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
1146             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
1147
1148             fjx1             = _mm256_setzero_pd();
1149             fjy1             = _mm256_setzero_pd();
1150             fjz1             = _mm256_setzero_pd();
1151             fjx2             = _mm256_setzero_pd();
1152             fjy2             = _mm256_setzero_pd();
1153             fjz2             = _mm256_setzero_pd();
1154             fjx3             = _mm256_setzero_pd();
1155             fjy3             = _mm256_setzero_pd();
1156             fjz3             = _mm256_setzero_pd();
1157
1158             /**************************
1159              * CALCULATE INTERACTIONS *
1160              **************************/
1161
1162             /* REACTION-FIELD ELECTROSTATICS */
1163             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1164
1165             fscal            = felec;
1166
1167             /* Calculate temporary vectorial force */
1168             tx               = _mm256_mul_pd(fscal,dx11);
1169             ty               = _mm256_mul_pd(fscal,dy11);
1170             tz               = _mm256_mul_pd(fscal,dz11);
1171
1172             /* Update vectorial force */
1173             fix1             = _mm256_add_pd(fix1,tx);
1174             fiy1             = _mm256_add_pd(fiy1,ty);
1175             fiz1             = _mm256_add_pd(fiz1,tz);
1176
1177             fjx1             = _mm256_add_pd(fjx1,tx);
1178             fjy1             = _mm256_add_pd(fjy1,ty);
1179             fjz1             = _mm256_add_pd(fjz1,tz);
1180
1181             /**************************
1182              * CALCULATE INTERACTIONS *
1183              **************************/
1184
1185             /* REACTION-FIELD ELECTROSTATICS */
1186             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1187
1188             fscal            = felec;
1189
1190             /* Calculate temporary vectorial force */
1191             tx               = _mm256_mul_pd(fscal,dx12);
1192             ty               = _mm256_mul_pd(fscal,dy12);
1193             tz               = _mm256_mul_pd(fscal,dz12);
1194
1195             /* Update vectorial force */
1196             fix1             = _mm256_add_pd(fix1,tx);
1197             fiy1             = _mm256_add_pd(fiy1,ty);
1198             fiz1             = _mm256_add_pd(fiz1,tz);
1199
1200             fjx2             = _mm256_add_pd(fjx2,tx);
1201             fjy2             = _mm256_add_pd(fjy2,ty);
1202             fjz2             = _mm256_add_pd(fjz2,tz);
1203
1204             /**************************
1205              * CALCULATE INTERACTIONS *
1206              **************************/
1207
1208             /* REACTION-FIELD ELECTROSTATICS */
1209             felec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
1210
1211             fscal            = felec;
1212
1213             /* Calculate temporary vectorial force */
1214             tx               = _mm256_mul_pd(fscal,dx13);
1215             ty               = _mm256_mul_pd(fscal,dy13);
1216             tz               = _mm256_mul_pd(fscal,dz13);
1217
1218             /* Update vectorial force */
1219             fix1             = _mm256_add_pd(fix1,tx);
1220             fiy1             = _mm256_add_pd(fiy1,ty);
1221             fiz1             = _mm256_add_pd(fiz1,tz);
1222
1223             fjx3             = _mm256_add_pd(fjx3,tx);
1224             fjy3             = _mm256_add_pd(fjy3,ty);
1225             fjz3             = _mm256_add_pd(fjz3,tz);
1226
1227             /**************************
1228              * CALCULATE INTERACTIONS *
1229              **************************/
1230
1231             /* REACTION-FIELD ELECTROSTATICS */
1232             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1233
1234             fscal            = felec;
1235
1236             /* Calculate temporary vectorial force */
1237             tx               = _mm256_mul_pd(fscal,dx21);
1238             ty               = _mm256_mul_pd(fscal,dy21);
1239             tz               = _mm256_mul_pd(fscal,dz21);
1240
1241             /* Update vectorial force */
1242             fix2             = _mm256_add_pd(fix2,tx);
1243             fiy2             = _mm256_add_pd(fiy2,ty);
1244             fiz2             = _mm256_add_pd(fiz2,tz);
1245
1246             fjx1             = _mm256_add_pd(fjx1,tx);
1247             fjy1             = _mm256_add_pd(fjy1,ty);
1248             fjz1             = _mm256_add_pd(fjz1,tz);
1249
1250             /**************************
1251              * CALCULATE INTERACTIONS *
1252              **************************/
1253
1254             /* REACTION-FIELD ELECTROSTATICS */
1255             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1256
1257             fscal            = felec;
1258
1259             /* Calculate temporary vectorial force */
1260             tx               = _mm256_mul_pd(fscal,dx22);
1261             ty               = _mm256_mul_pd(fscal,dy22);
1262             tz               = _mm256_mul_pd(fscal,dz22);
1263
1264             /* Update vectorial force */
1265             fix2             = _mm256_add_pd(fix2,tx);
1266             fiy2             = _mm256_add_pd(fiy2,ty);
1267             fiz2             = _mm256_add_pd(fiz2,tz);
1268
1269             fjx2             = _mm256_add_pd(fjx2,tx);
1270             fjy2             = _mm256_add_pd(fjy2,ty);
1271             fjz2             = _mm256_add_pd(fjz2,tz);
1272
1273             /**************************
1274              * CALCULATE INTERACTIONS *
1275              **************************/
1276
1277             /* REACTION-FIELD ELECTROSTATICS */
1278             felec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
1279
1280             fscal            = felec;
1281
1282             /* Calculate temporary vectorial force */
1283             tx               = _mm256_mul_pd(fscal,dx23);
1284             ty               = _mm256_mul_pd(fscal,dy23);
1285             tz               = _mm256_mul_pd(fscal,dz23);
1286
1287             /* Update vectorial force */
1288             fix2             = _mm256_add_pd(fix2,tx);
1289             fiy2             = _mm256_add_pd(fiy2,ty);
1290             fiz2             = _mm256_add_pd(fiz2,tz);
1291
1292             fjx3             = _mm256_add_pd(fjx3,tx);
1293             fjy3             = _mm256_add_pd(fjy3,ty);
1294             fjz3             = _mm256_add_pd(fjz3,tz);
1295
1296             /**************************
1297              * CALCULATE INTERACTIONS *
1298              **************************/
1299
1300             /* REACTION-FIELD ELECTROSTATICS */
1301             felec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
1302
1303             fscal            = felec;
1304
1305             /* Calculate temporary vectorial force */
1306             tx               = _mm256_mul_pd(fscal,dx31);
1307             ty               = _mm256_mul_pd(fscal,dy31);
1308             tz               = _mm256_mul_pd(fscal,dz31);
1309
1310             /* Update vectorial force */
1311             fix3             = _mm256_add_pd(fix3,tx);
1312             fiy3             = _mm256_add_pd(fiy3,ty);
1313             fiz3             = _mm256_add_pd(fiz3,tz);
1314
1315             fjx1             = _mm256_add_pd(fjx1,tx);
1316             fjy1             = _mm256_add_pd(fjy1,ty);
1317             fjz1             = _mm256_add_pd(fjz1,tz);
1318
1319             /**************************
1320              * CALCULATE INTERACTIONS *
1321              **************************/
1322
1323             /* REACTION-FIELD ELECTROSTATICS */
1324             felec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
1325
1326             fscal            = felec;
1327
1328             /* Calculate temporary vectorial force */
1329             tx               = _mm256_mul_pd(fscal,dx32);
1330             ty               = _mm256_mul_pd(fscal,dy32);
1331             tz               = _mm256_mul_pd(fscal,dz32);
1332
1333             /* Update vectorial force */
1334             fix3             = _mm256_add_pd(fix3,tx);
1335             fiy3             = _mm256_add_pd(fiy3,ty);
1336             fiz3             = _mm256_add_pd(fiz3,tz);
1337
1338             fjx2             = _mm256_add_pd(fjx2,tx);
1339             fjy2             = _mm256_add_pd(fjy2,ty);
1340             fjz2             = _mm256_add_pd(fjz2,tz);
1341
1342             /**************************
1343              * CALCULATE INTERACTIONS *
1344              **************************/
1345
1346             /* REACTION-FIELD ELECTROSTATICS */
1347             felec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
1348
1349             fscal            = felec;
1350
1351             /* Calculate temporary vectorial force */
1352             tx               = _mm256_mul_pd(fscal,dx33);
1353             ty               = _mm256_mul_pd(fscal,dy33);
1354             tz               = _mm256_mul_pd(fscal,dz33);
1355
1356             /* Update vectorial force */
1357             fix3             = _mm256_add_pd(fix3,tx);
1358             fiy3             = _mm256_add_pd(fiy3,ty);
1359             fiz3             = _mm256_add_pd(fiz3,tz);
1360
1361             fjx3             = _mm256_add_pd(fjx3,tx);
1362             fjy3             = _mm256_add_pd(fjy3,ty);
1363             fjz3             = _mm256_add_pd(fjz3,tz);
1364
1365             fjptrA             = f+j_coord_offsetA;
1366             fjptrB             = f+j_coord_offsetB;
1367             fjptrC             = f+j_coord_offsetC;
1368             fjptrD             = f+j_coord_offsetD;
1369
1370             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
1371                                                       fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1372
1373             /* Inner loop uses 243 flops */
1374         }
1375
1376         if(jidx<j_index_end)
1377         {
1378
1379             /* Get j neighbor index, and coordinate index */
1380             jnrlistA         = jjnr[jidx];
1381             jnrlistB         = jjnr[jidx+1];
1382             jnrlistC         = jjnr[jidx+2];
1383             jnrlistD         = jjnr[jidx+3];
1384             /* Sign of each element will be negative for non-real atoms.
1385              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1386              * so use it as val = _mm_andnot_pd(mask,val) to clear dummy entries.
1387              */
1388             tmpmask0 = gmx_mm_castsi128_pd(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1389
1390             tmpmask1 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(3,3,2,2));
1391             tmpmask0 = _mm_permute_ps(tmpmask0,_GMX_MM_PERMUTE(1,1,0,0));
1392             dummy_mask = _mm256_castps_pd(gmx_mm256_set_m128(tmpmask1,tmpmask0));
1393
1394             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1395             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1396             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1397             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1398             j_coord_offsetA  = DIM*jnrA;
1399             j_coord_offsetB  = DIM*jnrB;
1400             j_coord_offsetC  = DIM*jnrC;
1401             j_coord_offsetD  = DIM*jnrD;
1402
1403             /* load j atom coordinates */
1404             gmx_mm256_load_3rvec_4ptr_swizzle_pd(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1405                                                  x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1406                                                  &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1407
1408             /* Calculate displacement vector */
1409             dx11             = _mm256_sub_pd(ix1,jx1);
1410             dy11             = _mm256_sub_pd(iy1,jy1);
1411             dz11             = _mm256_sub_pd(iz1,jz1);
1412             dx12             = _mm256_sub_pd(ix1,jx2);
1413             dy12             = _mm256_sub_pd(iy1,jy2);
1414             dz12             = _mm256_sub_pd(iz1,jz2);
1415             dx13             = _mm256_sub_pd(ix1,jx3);
1416             dy13             = _mm256_sub_pd(iy1,jy3);
1417             dz13             = _mm256_sub_pd(iz1,jz3);
1418             dx21             = _mm256_sub_pd(ix2,jx1);
1419             dy21             = _mm256_sub_pd(iy2,jy1);
1420             dz21             = _mm256_sub_pd(iz2,jz1);
1421             dx22             = _mm256_sub_pd(ix2,jx2);
1422             dy22             = _mm256_sub_pd(iy2,jy2);
1423             dz22             = _mm256_sub_pd(iz2,jz2);
1424             dx23             = _mm256_sub_pd(ix2,jx3);
1425             dy23             = _mm256_sub_pd(iy2,jy3);
1426             dz23             = _mm256_sub_pd(iz2,jz3);
1427             dx31             = _mm256_sub_pd(ix3,jx1);
1428             dy31             = _mm256_sub_pd(iy3,jy1);
1429             dz31             = _mm256_sub_pd(iz3,jz1);
1430             dx32             = _mm256_sub_pd(ix3,jx2);
1431             dy32             = _mm256_sub_pd(iy3,jy2);
1432             dz32             = _mm256_sub_pd(iz3,jz2);
1433             dx33             = _mm256_sub_pd(ix3,jx3);
1434             dy33             = _mm256_sub_pd(iy3,jy3);
1435             dz33             = _mm256_sub_pd(iz3,jz3);
1436
1437             /* Calculate squared distance and things based on it */
1438             rsq11            = gmx_mm256_calc_rsq_pd(dx11,dy11,dz11);
1439             rsq12            = gmx_mm256_calc_rsq_pd(dx12,dy12,dz12);
1440             rsq13            = gmx_mm256_calc_rsq_pd(dx13,dy13,dz13);
1441             rsq21            = gmx_mm256_calc_rsq_pd(dx21,dy21,dz21);
1442             rsq22            = gmx_mm256_calc_rsq_pd(dx22,dy22,dz22);
1443             rsq23            = gmx_mm256_calc_rsq_pd(dx23,dy23,dz23);
1444             rsq31            = gmx_mm256_calc_rsq_pd(dx31,dy31,dz31);
1445             rsq32            = gmx_mm256_calc_rsq_pd(dx32,dy32,dz32);
1446             rsq33            = gmx_mm256_calc_rsq_pd(dx33,dy33,dz33);
1447
1448             rinv11           = gmx_mm256_invsqrt_pd(rsq11);
1449             rinv12           = gmx_mm256_invsqrt_pd(rsq12);
1450             rinv13           = gmx_mm256_invsqrt_pd(rsq13);
1451             rinv21           = gmx_mm256_invsqrt_pd(rsq21);
1452             rinv22           = gmx_mm256_invsqrt_pd(rsq22);
1453             rinv23           = gmx_mm256_invsqrt_pd(rsq23);
1454             rinv31           = gmx_mm256_invsqrt_pd(rsq31);
1455             rinv32           = gmx_mm256_invsqrt_pd(rsq32);
1456             rinv33           = gmx_mm256_invsqrt_pd(rsq33);
1457
1458             rinvsq11         = _mm256_mul_pd(rinv11,rinv11);
1459             rinvsq12         = _mm256_mul_pd(rinv12,rinv12);
1460             rinvsq13         = _mm256_mul_pd(rinv13,rinv13);
1461             rinvsq21         = _mm256_mul_pd(rinv21,rinv21);
1462             rinvsq22         = _mm256_mul_pd(rinv22,rinv22);
1463             rinvsq23         = _mm256_mul_pd(rinv23,rinv23);
1464             rinvsq31         = _mm256_mul_pd(rinv31,rinv31);
1465             rinvsq32         = _mm256_mul_pd(rinv32,rinv32);
1466             rinvsq33         = _mm256_mul_pd(rinv33,rinv33);
1467
1468             fjx1             = _mm256_setzero_pd();
1469             fjy1             = _mm256_setzero_pd();
1470             fjz1             = _mm256_setzero_pd();
1471             fjx2             = _mm256_setzero_pd();
1472             fjy2             = _mm256_setzero_pd();
1473             fjz2             = _mm256_setzero_pd();
1474             fjx3             = _mm256_setzero_pd();
1475             fjy3             = _mm256_setzero_pd();
1476             fjz3             = _mm256_setzero_pd();
1477
1478             /**************************
1479              * CALCULATE INTERACTIONS *
1480              **************************/
1481
1482             /* REACTION-FIELD ELECTROSTATICS */
1483             felec            = _mm256_mul_pd(qq11,_mm256_sub_pd(_mm256_mul_pd(rinv11,rinvsq11),krf2));
1484
1485             fscal            = felec;
1486
1487             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1488
1489             /* Calculate temporary vectorial force */
1490             tx               = _mm256_mul_pd(fscal,dx11);
1491             ty               = _mm256_mul_pd(fscal,dy11);
1492             tz               = _mm256_mul_pd(fscal,dz11);
1493
1494             /* Update vectorial force */
1495             fix1             = _mm256_add_pd(fix1,tx);
1496             fiy1             = _mm256_add_pd(fiy1,ty);
1497             fiz1             = _mm256_add_pd(fiz1,tz);
1498
1499             fjx1             = _mm256_add_pd(fjx1,tx);
1500             fjy1             = _mm256_add_pd(fjy1,ty);
1501             fjz1             = _mm256_add_pd(fjz1,tz);
1502
1503             /**************************
1504              * CALCULATE INTERACTIONS *
1505              **************************/
1506
1507             /* REACTION-FIELD ELECTROSTATICS */
1508             felec            = _mm256_mul_pd(qq12,_mm256_sub_pd(_mm256_mul_pd(rinv12,rinvsq12),krf2));
1509
1510             fscal            = felec;
1511
1512             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1513
1514             /* Calculate temporary vectorial force */
1515             tx               = _mm256_mul_pd(fscal,dx12);
1516             ty               = _mm256_mul_pd(fscal,dy12);
1517             tz               = _mm256_mul_pd(fscal,dz12);
1518
1519             /* Update vectorial force */
1520             fix1             = _mm256_add_pd(fix1,tx);
1521             fiy1             = _mm256_add_pd(fiy1,ty);
1522             fiz1             = _mm256_add_pd(fiz1,tz);
1523
1524             fjx2             = _mm256_add_pd(fjx2,tx);
1525             fjy2             = _mm256_add_pd(fjy2,ty);
1526             fjz2             = _mm256_add_pd(fjz2,tz);
1527
1528             /**************************
1529              * CALCULATE INTERACTIONS *
1530              **************************/
1531
1532             /* REACTION-FIELD ELECTROSTATICS */
1533             felec            = _mm256_mul_pd(qq13,_mm256_sub_pd(_mm256_mul_pd(rinv13,rinvsq13),krf2));
1534
1535             fscal            = felec;
1536
1537             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1538
1539             /* Calculate temporary vectorial force */
1540             tx               = _mm256_mul_pd(fscal,dx13);
1541             ty               = _mm256_mul_pd(fscal,dy13);
1542             tz               = _mm256_mul_pd(fscal,dz13);
1543
1544             /* Update vectorial force */
1545             fix1             = _mm256_add_pd(fix1,tx);
1546             fiy1             = _mm256_add_pd(fiy1,ty);
1547             fiz1             = _mm256_add_pd(fiz1,tz);
1548
1549             fjx3             = _mm256_add_pd(fjx3,tx);
1550             fjy3             = _mm256_add_pd(fjy3,ty);
1551             fjz3             = _mm256_add_pd(fjz3,tz);
1552
1553             /**************************
1554              * CALCULATE INTERACTIONS *
1555              **************************/
1556
1557             /* REACTION-FIELD ELECTROSTATICS */
1558             felec            = _mm256_mul_pd(qq21,_mm256_sub_pd(_mm256_mul_pd(rinv21,rinvsq21),krf2));
1559
1560             fscal            = felec;
1561
1562             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1563
1564             /* Calculate temporary vectorial force */
1565             tx               = _mm256_mul_pd(fscal,dx21);
1566             ty               = _mm256_mul_pd(fscal,dy21);
1567             tz               = _mm256_mul_pd(fscal,dz21);
1568
1569             /* Update vectorial force */
1570             fix2             = _mm256_add_pd(fix2,tx);
1571             fiy2             = _mm256_add_pd(fiy2,ty);
1572             fiz2             = _mm256_add_pd(fiz2,tz);
1573
1574             fjx1             = _mm256_add_pd(fjx1,tx);
1575             fjy1             = _mm256_add_pd(fjy1,ty);
1576             fjz1             = _mm256_add_pd(fjz1,tz);
1577
1578             /**************************
1579              * CALCULATE INTERACTIONS *
1580              **************************/
1581
1582             /* REACTION-FIELD ELECTROSTATICS */
1583             felec            = _mm256_mul_pd(qq22,_mm256_sub_pd(_mm256_mul_pd(rinv22,rinvsq22),krf2));
1584
1585             fscal            = felec;
1586
1587             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1588
1589             /* Calculate temporary vectorial force */
1590             tx               = _mm256_mul_pd(fscal,dx22);
1591             ty               = _mm256_mul_pd(fscal,dy22);
1592             tz               = _mm256_mul_pd(fscal,dz22);
1593
1594             /* Update vectorial force */
1595             fix2             = _mm256_add_pd(fix2,tx);
1596             fiy2             = _mm256_add_pd(fiy2,ty);
1597             fiz2             = _mm256_add_pd(fiz2,tz);
1598
1599             fjx2             = _mm256_add_pd(fjx2,tx);
1600             fjy2             = _mm256_add_pd(fjy2,ty);
1601             fjz2             = _mm256_add_pd(fjz2,tz);
1602
1603             /**************************
1604              * CALCULATE INTERACTIONS *
1605              **************************/
1606
1607             /* REACTION-FIELD ELECTROSTATICS */
1608             felec            = _mm256_mul_pd(qq23,_mm256_sub_pd(_mm256_mul_pd(rinv23,rinvsq23),krf2));
1609
1610             fscal            = felec;
1611
1612             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1613
1614             /* Calculate temporary vectorial force */
1615             tx               = _mm256_mul_pd(fscal,dx23);
1616             ty               = _mm256_mul_pd(fscal,dy23);
1617             tz               = _mm256_mul_pd(fscal,dz23);
1618
1619             /* Update vectorial force */
1620             fix2             = _mm256_add_pd(fix2,tx);
1621             fiy2             = _mm256_add_pd(fiy2,ty);
1622             fiz2             = _mm256_add_pd(fiz2,tz);
1623
1624             fjx3             = _mm256_add_pd(fjx3,tx);
1625             fjy3             = _mm256_add_pd(fjy3,ty);
1626             fjz3             = _mm256_add_pd(fjz3,tz);
1627
1628             /**************************
1629              * CALCULATE INTERACTIONS *
1630              **************************/
1631
1632             /* REACTION-FIELD ELECTROSTATICS */
1633             felec            = _mm256_mul_pd(qq31,_mm256_sub_pd(_mm256_mul_pd(rinv31,rinvsq31),krf2));
1634
1635             fscal            = felec;
1636
1637             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1638
1639             /* Calculate temporary vectorial force */
1640             tx               = _mm256_mul_pd(fscal,dx31);
1641             ty               = _mm256_mul_pd(fscal,dy31);
1642             tz               = _mm256_mul_pd(fscal,dz31);
1643
1644             /* Update vectorial force */
1645             fix3             = _mm256_add_pd(fix3,tx);
1646             fiy3             = _mm256_add_pd(fiy3,ty);
1647             fiz3             = _mm256_add_pd(fiz3,tz);
1648
1649             fjx1             = _mm256_add_pd(fjx1,tx);
1650             fjy1             = _mm256_add_pd(fjy1,ty);
1651             fjz1             = _mm256_add_pd(fjz1,tz);
1652
1653             /**************************
1654              * CALCULATE INTERACTIONS *
1655              **************************/
1656
1657             /* REACTION-FIELD ELECTROSTATICS */
1658             felec            = _mm256_mul_pd(qq32,_mm256_sub_pd(_mm256_mul_pd(rinv32,rinvsq32),krf2));
1659
1660             fscal            = felec;
1661
1662             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1663
1664             /* Calculate temporary vectorial force */
1665             tx               = _mm256_mul_pd(fscal,dx32);
1666             ty               = _mm256_mul_pd(fscal,dy32);
1667             tz               = _mm256_mul_pd(fscal,dz32);
1668
1669             /* Update vectorial force */
1670             fix3             = _mm256_add_pd(fix3,tx);
1671             fiy3             = _mm256_add_pd(fiy3,ty);
1672             fiz3             = _mm256_add_pd(fiz3,tz);
1673
1674             fjx2             = _mm256_add_pd(fjx2,tx);
1675             fjy2             = _mm256_add_pd(fjy2,ty);
1676             fjz2             = _mm256_add_pd(fjz2,tz);
1677
1678             /**************************
1679              * CALCULATE INTERACTIONS *
1680              **************************/
1681
1682             /* REACTION-FIELD ELECTROSTATICS */
1683             felec            = _mm256_mul_pd(qq33,_mm256_sub_pd(_mm256_mul_pd(rinv33,rinvsq33),krf2));
1684
1685             fscal            = felec;
1686
1687             fscal            = _mm256_andnot_pd(dummy_mask,fscal);
1688
1689             /* Calculate temporary vectorial force */
1690             tx               = _mm256_mul_pd(fscal,dx33);
1691             ty               = _mm256_mul_pd(fscal,dy33);
1692             tz               = _mm256_mul_pd(fscal,dz33);
1693
1694             /* Update vectorial force */
1695             fix3             = _mm256_add_pd(fix3,tx);
1696             fiy3             = _mm256_add_pd(fiy3,ty);
1697             fiz3             = _mm256_add_pd(fiz3,tz);
1698
1699             fjx3             = _mm256_add_pd(fjx3,tx);
1700             fjy3             = _mm256_add_pd(fjy3,ty);
1701             fjz3             = _mm256_add_pd(fjz3,tz);
1702
1703             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1704             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1705             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1706             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1707
1708             gmx_mm256_decrement_3rvec_4ptr_swizzle_pd(fjptrA+DIM,fjptrB+DIM,fjptrC+DIM,fjptrD+DIM,
1709                                                       fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1710
1711             /* Inner loop uses 243 flops */
1712         }
1713
1714         /* End of innermost loop */
1715
1716         gmx_mm256_update_iforce_3atom_swizzle_pd(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1717                                                  f+i_coord_offset+DIM,fshift+i_shift_offset);
1718
1719         /* Increment number of inner iterations */
1720         inneriter                  += j_index_end - j_index_start;
1721
1722         /* Outer loop uses 18 flops */
1723     }
1724
1725     /* Increment number of outer iterations */
1726     outeriter        += nri;
1727
1728     /* Update outer/inner flops */
1729
1730     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_F,outeriter*18 + inneriter*243);
1731 }