src/gromacs/gmxlib/nonbonded/nb_kernel_avx_128_fma_double/nb_kernel_ElecEwSh_VdwLJSh_GeomW4W4_avx_128_fma_double.c

   1 /*
   2  * Note: this file was generated by the Gromacs avx_128_fma_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_128_fma_double.h"
  34 #include "kernelutil_x86_avx_128_fma_double.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJSh_GeomW4W4_VF_avx_128_fma_double
  38  * Electrostatics interaction: Ewald
  39  * VdW interaction:            LennardJones
  40  * Geometry:                   Water4-Water4
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecEwSh_VdwLJSh_GeomW4W4_VF_avx_128_fma_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 refer to j loop unrolling done with SSE double precision, e.g. for the two 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;
  61     int              j_coord_offsetA,j_coord_offsetB;
  62     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  63     real             rcutoff_scalar;
  64     real             *shiftvec,*fshift,*x,*f;
  65     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  66     int              vdwioffset0;
  67     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  68     int              vdwioffset1;
  69     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  70     int              vdwioffset2;
  71     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  72     int              vdwioffset3;
  73     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  74     int              vdwjidx0A,vdwjidx0B;
  75     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  76     int              vdwjidx1A,vdwjidx1B;
  77     __m128d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  78     int              vdwjidx2A,vdwjidx2B;
  79     __m128d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  80     int              vdwjidx3A,vdwjidx3B;
  81     __m128d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
  82     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  83     __m128d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
  84     __m128d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
  85     __m128d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
  86     __m128d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
  87     __m128d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
  88     __m128d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
  89     __m128d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
  90     __m128d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
  91     __m128d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
  92     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  93     real             *charge;
  94     int              nvdwtype;
  95     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  96     int              *vdwtype;
  97     real             *vdwparam;
  98     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
  99     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 100     __m128i          ewitab;
 101     __m128d          ewtabscale,eweps,twoeweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 102     real             *ewtab;
 103     __m128d          dummy_mask,cutoff_mask;
 104     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 105     __m128d          one     = _mm_set1_pd(1.0);
 106     __m128d          two     = _mm_set1_pd(2.0);
 107     x                = xx[0];
 108     f                = ff[0];
 109
 110     nri              = nlist->nri;
 111     iinr             = nlist->iinr;
 112     jindex           = nlist->jindex;
 113     jjnr             = nlist->jjnr;
 114     shiftidx         = nlist->shift;
 115     gid              = nlist->gid;
 116     shiftvec         = fr->shift_vec[0];
 117     fshift           = fr->fshift[0];
 118     facel            = _mm_set1_pd(fr->epsfac);
 119     charge           = mdatoms->chargeA;
 120     nvdwtype         = fr->ntype;
 121     vdwparam         = fr->nbfp;
 122     vdwtype          = mdatoms->typeA;
 123
 124     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 125     ewtab            = fr->ic->tabq_coul_FDV0;
 126     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 127     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 128
 129     /* Setup water-specific parameters */
 130     inr              = nlist->iinr[0];
 131     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 132     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 133     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 134     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 135
 136     jq1              = _mm_set1_pd(charge[inr+1]);
 137     jq2              = _mm_set1_pd(charge[inr+2]);
 138     jq3              = _mm_set1_pd(charge[inr+3]);
 139     vdwjidx0A        = 2*vdwtype[inr+0];
 140     c6_00            = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
 141     c12_00           = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
 142     qq11             = _mm_mul_pd(iq1,jq1);
 143     qq12             = _mm_mul_pd(iq1,jq2);
 144     qq13             = _mm_mul_pd(iq1,jq3);
 145     qq21             = _mm_mul_pd(iq2,jq1);
 146     qq22             = _mm_mul_pd(iq2,jq2);
 147     qq23             = _mm_mul_pd(iq2,jq3);
 148     qq31             = _mm_mul_pd(iq3,jq1);
 149     qq32             = _mm_mul_pd(iq3,jq2);
 150     qq33             = _mm_mul_pd(iq3,jq3);
 151
 152     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 153     rcutoff_scalar   = fr->rcoulomb;
 154     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 155     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 156
 157     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
 158     rvdw             = _mm_set1_pd(fr->rvdw);
 159
 160     /* Avoid stupid compiler warnings */
 161     jnrA = jnrB = 0;
 162     j_coord_offsetA = 0;
 163     j_coord_offsetB = 0;
 164
 165     outeriter        = 0;
 166     inneriter        = 0;
 167
 168     /* Start outer loop over neighborlists */
 169     for(iidx=0; iidx<nri; iidx++)
 170     {
 171         /* Load shift vector for this list */
 172         i_shift_offset   = DIM*shiftidx[iidx];
 173
 174         /* Load limits for loop over neighbors */
 175         j_index_start    = jindex[iidx];
 176         j_index_end      = jindex[iidx+1];
 177
 178         /* Get outer coordinate index */
 179         inr              = iinr[iidx];
 180         i_coord_offset   = DIM*inr;
 181
 182         /* Load i particle coords and add shift vector */
 183         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 184                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 185
 186         fix0             = _mm_setzero_pd();
 187         fiy0             = _mm_setzero_pd();
 188         fiz0             = _mm_setzero_pd();
 189         fix1             = _mm_setzero_pd();
 190         fiy1             = _mm_setzero_pd();
 191         fiz1             = _mm_setzero_pd();
 192         fix2             = _mm_setzero_pd();
 193         fiy2             = _mm_setzero_pd();
 194         fiz2             = _mm_setzero_pd();
 195         fix3             = _mm_setzero_pd();
 196         fiy3             = _mm_setzero_pd();
 197         fiz3             = _mm_setzero_pd();
 198
 199         /* Reset potential sums */
 200         velecsum         = _mm_setzero_pd();
 201         vvdwsum          = _mm_setzero_pd();
 202
 203         /* Start inner kernel loop */
 204         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 205         {
 206
 207             /* Get j neighbor index, and coordinate index */
 208             jnrA             = jjnr[jidx];
 209             jnrB             = jjnr[jidx+1];
 210             j_coord_offsetA  = DIM*jnrA;
 211             j_coord_offsetB  = DIM*jnrB;
 212
 213             /* load j atom coordinates */
 214             gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 215                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 216                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 217
 218             /* Calculate displacement vector */
 219             dx00             = _mm_sub_pd(ix0,jx0);
 220             dy00             = _mm_sub_pd(iy0,jy0);
 221             dz00             = _mm_sub_pd(iz0,jz0);
 222             dx11             = _mm_sub_pd(ix1,jx1);
 223             dy11             = _mm_sub_pd(iy1,jy1);
 224             dz11             = _mm_sub_pd(iz1,jz1);
 225             dx12             = _mm_sub_pd(ix1,jx2);
 226             dy12             = _mm_sub_pd(iy1,jy2);
 227             dz12             = _mm_sub_pd(iz1,jz2);
 228             dx13             = _mm_sub_pd(ix1,jx3);
 229             dy13             = _mm_sub_pd(iy1,jy3);
 230             dz13             = _mm_sub_pd(iz1,jz3);
 231             dx21             = _mm_sub_pd(ix2,jx1);
 232             dy21             = _mm_sub_pd(iy2,jy1);
 233             dz21             = _mm_sub_pd(iz2,jz1);
 234             dx22             = _mm_sub_pd(ix2,jx2);
 235             dy22             = _mm_sub_pd(iy2,jy2);
 236             dz22             = _mm_sub_pd(iz2,jz2);
 237             dx23             = _mm_sub_pd(ix2,jx3);
 238             dy23             = _mm_sub_pd(iy2,jy3);
 239             dz23             = _mm_sub_pd(iz2,jz3);
 240             dx31             = _mm_sub_pd(ix3,jx1);
 241             dy31             = _mm_sub_pd(iy3,jy1);
 242             dz31             = _mm_sub_pd(iz3,jz1);
 243             dx32             = _mm_sub_pd(ix3,jx2);
 244             dy32             = _mm_sub_pd(iy3,jy2);
 245             dz32             = _mm_sub_pd(iz3,jz2);
 246             dx33             = _mm_sub_pd(ix3,jx3);
 247             dy33             = _mm_sub_pd(iy3,jy3);
 248             dz33             = _mm_sub_pd(iz3,jz3);
 249
 250             /* Calculate squared distance and things based on it */
 251             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 252             rsq11            = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
 253             rsq12            = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
 254             rsq13            = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
 255             rsq21            = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
 256             rsq22            = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
 257             rsq23            = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
 258             rsq31            = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
 259             rsq32            = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
 260             rsq33            = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
 261
 262             rinv11           = gmx_mm_invsqrt_pd(rsq11);
 263             rinv12           = gmx_mm_invsqrt_pd(rsq12);
 264             rinv13           = gmx_mm_invsqrt_pd(rsq13);
 265             rinv21           = gmx_mm_invsqrt_pd(rsq21);
 266             rinv22           = gmx_mm_invsqrt_pd(rsq22);
 267             rinv23           = gmx_mm_invsqrt_pd(rsq23);
 268             rinv31           = gmx_mm_invsqrt_pd(rsq31);
 269             rinv32           = gmx_mm_invsqrt_pd(rsq32);
 270             rinv33           = gmx_mm_invsqrt_pd(rsq33);
 271
 272             rinvsq00         = gmx_mm_inv_pd(rsq00);
 273             rinvsq11         = _mm_mul_pd(rinv11,rinv11);
 274             rinvsq12         = _mm_mul_pd(rinv12,rinv12);
 275             rinvsq13         = _mm_mul_pd(rinv13,rinv13);
 276             rinvsq21         = _mm_mul_pd(rinv21,rinv21);
 277             rinvsq22         = _mm_mul_pd(rinv22,rinv22);
 278             rinvsq23         = _mm_mul_pd(rinv23,rinv23);
 279             rinvsq31         = _mm_mul_pd(rinv31,rinv31);
 280             rinvsq32         = _mm_mul_pd(rinv32,rinv32);
 281             rinvsq33         = _mm_mul_pd(rinv33,rinv33);
 282
 283             fjx0             = _mm_setzero_pd();
 284             fjy0             = _mm_setzero_pd();
 285             fjz0             = _mm_setzero_pd();
 286             fjx1             = _mm_setzero_pd();
 287             fjy1             = _mm_setzero_pd();
 288             fjz1             = _mm_setzero_pd();
 289             fjx2             = _mm_setzero_pd();
 290             fjy2             = _mm_setzero_pd();
 291             fjz2             = _mm_setzero_pd();
 292             fjx3             = _mm_setzero_pd();
 293             fjy3             = _mm_setzero_pd();
 294             fjz3             = _mm_setzero_pd();
 295
 296             /**************************
 297              * CALCULATE INTERACTIONS *
 298              **************************/
 299
 300             if (gmx_mm_any_lt(rsq00,rcutoff2))
 301             {
 302
 303             /* LENNARD-JONES DISPERSION/REPULSION */
 304
 305             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 306             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 307             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 308             vvdw             = _mm_msub_pd(_mm_nmacc_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6),vvdw12),one_twelfth,
 309                                            _mm_mul_pd(_mm_nmacc_pd( c6_00,sh_vdw_invrcut6,vvdw6),one_sixth));
 310             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 311
 312             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 313
 314             /* Update potential sum for this i atom from the interaction with this j atom. */
 315             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 316             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 317
 318             fscal            = fvdw;
 319
 320             fscal            = _mm_and_pd(fscal,cutoff_mask);
 321
 322             /* Update vectorial force */
 323             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 324             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 325             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 326
 327             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 328             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 329             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 330
 331             }
 332
 333             /**************************
 334              * CALCULATE INTERACTIONS *
 335              **************************/
 336
 337             if (gmx_mm_any_lt(rsq11,rcutoff2))
 338             {
 339
 340             r11              = _mm_mul_pd(rsq11,rinv11);
 341
 342             /* EWALD ELECTROSTATICS */
 343
 344             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 345             ewrt             = _mm_mul_pd(r11,ewtabscale);
 346             ewitab           = _mm_cvttpd_epi32(ewrt);
 347 #ifdef __XOP__
 348             eweps            = _mm_frcz_pd(ewrt);
 349 #else
 350             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 351 #endif
 352             twoeweps         = _mm_add_pd(eweps,eweps);
 353             ewitab           = _mm_slli_epi32(ewitab,2);
 354             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 355             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 356             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 357             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 358             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 359             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 360             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 361             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 362             velec            = _mm_mul_pd(qq11,_mm_sub_pd(_mm_sub_pd(rinv11,sh_ewald),velec));
 363             felec            = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
 364
 365             cutoff_mask      = _mm_cmplt_pd(rsq11,rcutoff2);
 366
 367             /* Update potential sum for this i atom from the interaction with this j atom. */
 368             velec            = _mm_and_pd(velec,cutoff_mask);
 369             velecsum         = _mm_add_pd(velecsum,velec);
 370
 371             fscal            = felec;
 372
 373             fscal            = _mm_and_pd(fscal,cutoff_mask);
 374
 375             /* Update vectorial force */
 376             fix1             = _mm_macc_pd(dx11,fscal,fix1);
 377             fiy1             = _mm_macc_pd(dy11,fscal,fiy1);
 378             fiz1             = _mm_macc_pd(dz11,fscal,fiz1);
 379
 380             fjx1             = _mm_macc_pd(dx11,fscal,fjx1);
 381             fjy1             = _mm_macc_pd(dy11,fscal,fjy1);
 382             fjz1             = _mm_macc_pd(dz11,fscal,fjz1);
 383
 384             }
 385
 386             /**************************
 387              * CALCULATE INTERACTIONS *
 388              **************************/
 389
 390             if (gmx_mm_any_lt(rsq12,rcutoff2))
 391             {
 392
 393             r12              = _mm_mul_pd(rsq12,rinv12);
 394
 395             /* EWALD ELECTROSTATICS */
 396
 397             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 398             ewrt             = _mm_mul_pd(r12,ewtabscale);
 399             ewitab           = _mm_cvttpd_epi32(ewrt);
 400 #ifdef __XOP__
 401             eweps            = _mm_frcz_pd(ewrt);
 402 #else
 403             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 404 #endif
 405             twoeweps         = _mm_add_pd(eweps,eweps);
 406             ewitab           = _mm_slli_epi32(ewitab,2);
 407             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 408             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 409             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 410             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 411             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 412             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 413             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 414             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 415             velec            = _mm_mul_pd(qq12,_mm_sub_pd(_mm_sub_pd(rinv12,sh_ewald),velec));
 416             felec            = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
 417
 418             cutoff_mask      = _mm_cmplt_pd(rsq12,rcutoff2);
 419
 420             /* Update potential sum for this i atom from the interaction with this j atom. */
 421             velec            = _mm_and_pd(velec,cutoff_mask);
 422             velecsum         = _mm_add_pd(velecsum,velec);
 423
 424             fscal            = felec;
 425
 426             fscal            = _mm_and_pd(fscal,cutoff_mask);
 427
 428             /* Update vectorial force */
 429             fix1             = _mm_macc_pd(dx12,fscal,fix1);
 430             fiy1             = _mm_macc_pd(dy12,fscal,fiy1);
 431             fiz1             = _mm_macc_pd(dz12,fscal,fiz1);
 432
 433             fjx2             = _mm_macc_pd(dx12,fscal,fjx2);
 434             fjy2             = _mm_macc_pd(dy12,fscal,fjy2);
 435             fjz2             = _mm_macc_pd(dz12,fscal,fjz2);
 436
 437             }
 438
 439             /**************************
 440              * CALCULATE INTERACTIONS *
 441              **************************/
 442
 443             if (gmx_mm_any_lt(rsq13,rcutoff2))
 444             {
 445
 446             r13              = _mm_mul_pd(rsq13,rinv13);
 447
 448             /* EWALD ELECTROSTATICS */
 449
 450             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 451             ewrt             = _mm_mul_pd(r13,ewtabscale);
 452             ewitab           = _mm_cvttpd_epi32(ewrt);
 453 #ifdef __XOP__
 454             eweps            = _mm_frcz_pd(ewrt);
 455 #else
 456             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 457 #endif
 458             twoeweps         = _mm_add_pd(eweps,eweps);
 459             ewitab           = _mm_slli_epi32(ewitab,2);
 460             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 461             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 462             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 463             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 464             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 465             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 466             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 467             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 468             velec            = _mm_mul_pd(qq13,_mm_sub_pd(_mm_sub_pd(rinv13,sh_ewald),velec));
 469             felec            = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
 470
 471             cutoff_mask      = _mm_cmplt_pd(rsq13,rcutoff2);
 472
 473             /* Update potential sum for this i atom from the interaction with this j atom. */
 474             velec            = _mm_and_pd(velec,cutoff_mask);
 475             velecsum         = _mm_add_pd(velecsum,velec);
 476
 477             fscal            = felec;
 478
 479             fscal            = _mm_and_pd(fscal,cutoff_mask);
 480
 481             /* Update vectorial force */
 482             fix1             = _mm_macc_pd(dx13,fscal,fix1);
 483             fiy1             = _mm_macc_pd(dy13,fscal,fiy1);
 484             fiz1             = _mm_macc_pd(dz13,fscal,fiz1);
 485
 486             fjx3             = _mm_macc_pd(dx13,fscal,fjx3);
 487             fjy3             = _mm_macc_pd(dy13,fscal,fjy3);
 488             fjz3             = _mm_macc_pd(dz13,fscal,fjz3);
 489
 490             }
 491
 492             /**************************
 493              * CALCULATE INTERACTIONS *
 494              **************************/
 495
 496             if (gmx_mm_any_lt(rsq21,rcutoff2))
 497             {
 498
 499             r21              = _mm_mul_pd(rsq21,rinv21);
 500
 501             /* EWALD ELECTROSTATICS */
 502
 503             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 504             ewrt             = _mm_mul_pd(r21,ewtabscale);
 505             ewitab           = _mm_cvttpd_epi32(ewrt);
 506 #ifdef __XOP__
 507             eweps            = _mm_frcz_pd(ewrt);
 508 #else
 509             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 510 #endif
 511             twoeweps         = _mm_add_pd(eweps,eweps);
 512             ewitab           = _mm_slli_epi32(ewitab,2);
 513             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 514             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 515             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 516             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 517             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 518             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 519             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 520             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 521             velec            = _mm_mul_pd(qq21,_mm_sub_pd(_mm_sub_pd(rinv21,sh_ewald),velec));
 522             felec            = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
 523
 524             cutoff_mask      = _mm_cmplt_pd(rsq21,rcutoff2);
 525
 526             /* Update potential sum for this i atom from the interaction with this j atom. */
 527             velec            = _mm_and_pd(velec,cutoff_mask);
 528             velecsum         = _mm_add_pd(velecsum,velec);
 529
 530             fscal            = felec;
 531
 532             fscal            = _mm_and_pd(fscal,cutoff_mask);
 533
 534             /* Update vectorial force */
 535             fix2             = _mm_macc_pd(dx21,fscal,fix2);
 536             fiy2             = _mm_macc_pd(dy21,fscal,fiy2);
 537             fiz2             = _mm_macc_pd(dz21,fscal,fiz2);
 538
 539             fjx1             = _mm_macc_pd(dx21,fscal,fjx1);
 540             fjy1             = _mm_macc_pd(dy21,fscal,fjy1);
 541             fjz1             = _mm_macc_pd(dz21,fscal,fjz1);
 542
 543             }
 544
 545             /**************************
 546              * CALCULATE INTERACTIONS *
 547              **************************/
 548
 549             if (gmx_mm_any_lt(rsq22,rcutoff2))
 550             {
 551
 552             r22              = _mm_mul_pd(rsq22,rinv22);
 553
 554             /* EWALD ELECTROSTATICS */
 555
 556             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 557             ewrt             = _mm_mul_pd(r22,ewtabscale);
 558             ewitab           = _mm_cvttpd_epi32(ewrt);
 559 #ifdef __XOP__
 560             eweps            = _mm_frcz_pd(ewrt);
 561 #else
 562             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 563 #endif
 564             twoeweps         = _mm_add_pd(eweps,eweps);
 565             ewitab           = _mm_slli_epi32(ewitab,2);
 566             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 567             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 568             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 569             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 570             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 571             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 572             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 573             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 574             velec            = _mm_mul_pd(qq22,_mm_sub_pd(_mm_sub_pd(rinv22,sh_ewald),velec));
 575             felec            = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
 576
 577             cutoff_mask      = _mm_cmplt_pd(rsq22,rcutoff2);
 578
 579             /* Update potential sum for this i atom from the interaction with this j atom. */
 580             velec            = _mm_and_pd(velec,cutoff_mask);
 581             velecsum         = _mm_add_pd(velecsum,velec);
 582
 583             fscal            = felec;
 584
 585             fscal            = _mm_and_pd(fscal,cutoff_mask);
 586
 587             /* Update vectorial force */
 588             fix2             = _mm_macc_pd(dx22,fscal,fix2);
 589             fiy2             = _mm_macc_pd(dy22,fscal,fiy2);
 590             fiz2             = _mm_macc_pd(dz22,fscal,fiz2);
 591
 592             fjx2             = _mm_macc_pd(dx22,fscal,fjx2);
 593             fjy2             = _mm_macc_pd(dy22,fscal,fjy2);
 594             fjz2             = _mm_macc_pd(dz22,fscal,fjz2);
 595
 596             }
 597
 598             /**************************
 599              * CALCULATE INTERACTIONS *
 600              **************************/
 601
 602             if (gmx_mm_any_lt(rsq23,rcutoff2))
 603             {
 604
 605             r23              = _mm_mul_pd(rsq23,rinv23);
 606
 607             /* EWALD ELECTROSTATICS */
 608
 609             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 610             ewrt             = _mm_mul_pd(r23,ewtabscale);
 611             ewitab           = _mm_cvttpd_epi32(ewrt);
 612 #ifdef __XOP__
 613             eweps            = _mm_frcz_pd(ewrt);
 614 #else
 615             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 616 #endif
 617             twoeweps         = _mm_add_pd(eweps,eweps);
 618             ewitab           = _mm_slli_epi32(ewitab,2);
 619             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 620             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 621             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 622             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 623             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 624             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 625             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 626             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 627             velec            = _mm_mul_pd(qq23,_mm_sub_pd(_mm_sub_pd(rinv23,sh_ewald),velec));
 628             felec            = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
 629
 630             cutoff_mask      = _mm_cmplt_pd(rsq23,rcutoff2);
 631
 632             /* Update potential sum for this i atom from the interaction with this j atom. */
 633             velec            = _mm_and_pd(velec,cutoff_mask);
 634             velecsum         = _mm_add_pd(velecsum,velec);
 635
 636             fscal            = felec;
 637
 638             fscal            = _mm_and_pd(fscal,cutoff_mask);
 639
 640             /* Update vectorial force */
 641             fix2             = _mm_macc_pd(dx23,fscal,fix2);
 642             fiy2             = _mm_macc_pd(dy23,fscal,fiy2);
 643             fiz2             = _mm_macc_pd(dz23,fscal,fiz2);
 644
 645             fjx3             = _mm_macc_pd(dx23,fscal,fjx3);
 646             fjy3             = _mm_macc_pd(dy23,fscal,fjy3);
 647             fjz3             = _mm_macc_pd(dz23,fscal,fjz3);
 648
 649             }
 650
 651             /**************************
 652              * CALCULATE INTERACTIONS *
 653              **************************/
 654
 655             if (gmx_mm_any_lt(rsq31,rcutoff2))
 656             {
 657
 658             r31              = _mm_mul_pd(rsq31,rinv31);
 659
 660             /* EWALD ELECTROSTATICS */
 661
 662             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 663             ewrt             = _mm_mul_pd(r31,ewtabscale);
 664             ewitab           = _mm_cvttpd_epi32(ewrt);
 665 #ifdef __XOP__
 666             eweps            = _mm_frcz_pd(ewrt);
 667 #else
 668             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 669 #endif
 670             twoeweps         = _mm_add_pd(eweps,eweps);
 671             ewitab           = _mm_slli_epi32(ewitab,2);
 672             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 673             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 674             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 675             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 676             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 677             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 678             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 679             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 680             velec            = _mm_mul_pd(qq31,_mm_sub_pd(_mm_sub_pd(rinv31,sh_ewald),velec));
 681             felec            = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
 682
 683             cutoff_mask      = _mm_cmplt_pd(rsq31,rcutoff2);
 684
 685             /* Update potential sum for this i atom from the interaction with this j atom. */
 686             velec            = _mm_and_pd(velec,cutoff_mask);
 687             velecsum         = _mm_add_pd(velecsum,velec);
 688
 689             fscal            = felec;
 690
 691             fscal            = _mm_and_pd(fscal,cutoff_mask);
 692
 693             /* Update vectorial force */
 694             fix3             = _mm_macc_pd(dx31,fscal,fix3);
 695             fiy3             = _mm_macc_pd(dy31,fscal,fiy3);
 696             fiz3             = _mm_macc_pd(dz31,fscal,fiz3);
 697
 698             fjx1             = _mm_macc_pd(dx31,fscal,fjx1);
 699             fjy1             = _mm_macc_pd(dy31,fscal,fjy1);
 700             fjz1             = _mm_macc_pd(dz31,fscal,fjz1);
 701
 702             }
 703
 704             /**************************
 705              * CALCULATE INTERACTIONS *
 706              **************************/
 707
 708             if (gmx_mm_any_lt(rsq32,rcutoff2))
 709             {
 710
 711             r32              = _mm_mul_pd(rsq32,rinv32);
 712
 713             /* EWALD ELECTROSTATICS */
 714
 715             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 716             ewrt             = _mm_mul_pd(r32,ewtabscale);
 717             ewitab           = _mm_cvttpd_epi32(ewrt);
 718 #ifdef __XOP__
 719             eweps            = _mm_frcz_pd(ewrt);
 720 #else
 721             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 722 #endif
 723             twoeweps         = _mm_add_pd(eweps,eweps);
 724             ewitab           = _mm_slli_epi32(ewitab,2);
 725             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 726             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 727             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 728             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 729             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 730             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 731             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 732             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 733             velec            = _mm_mul_pd(qq32,_mm_sub_pd(_mm_sub_pd(rinv32,sh_ewald),velec));
 734             felec            = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
 735
 736             cutoff_mask      = _mm_cmplt_pd(rsq32,rcutoff2);
 737
 738             /* Update potential sum for this i atom from the interaction with this j atom. */
 739             velec            = _mm_and_pd(velec,cutoff_mask);
 740             velecsum         = _mm_add_pd(velecsum,velec);
 741
 742             fscal            = felec;
 743
 744             fscal            = _mm_and_pd(fscal,cutoff_mask);
 745
 746             /* Update vectorial force */
 747             fix3             = _mm_macc_pd(dx32,fscal,fix3);
 748             fiy3             = _mm_macc_pd(dy32,fscal,fiy3);
 749             fiz3             = _mm_macc_pd(dz32,fscal,fiz3);
 750
 751             fjx2             = _mm_macc_pd(dx32,fscal,fjx2);
 752             fjy2             = _mm_macc_pd(dy32,fscal,fjy2);
 753             fjz2             = _mm_macc_pd(dz32,fscal,fjz2);
 754
 755             }
 756
 757             /**************************
 758              * CALCULATE INTERACTIONS *
 759              **************************/
 760
 761             if (gmx_mm_any_lt(rsq33,rcutoff2))
 762             {
 763
 764             r33              = _mm_mul_pd(rsq33,rinv33);
 765
 766             /* EWALD ELECTROSTATICS */
 767
 768             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 769             ewrt             = _mm_mul_pd(r33,ewtabscale);
 770             ewitab           = _mm_cvttpd_epi32(ewrt);
 771 #ifdef __XOP__
 772             eweps            = _mm_frcz_pd(ewrt);
 773 #else
 774             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 775 #endif
 776             twoeweps         = _mm_add_pd(eweps,eweps);
 777             ewitab           = _mm_slli_epi32(ewitab,2);
 778             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 779             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 780             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 781             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 782             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 783             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 784             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 785             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 786             velec            = _mm_mul_pd(qq33,_mm_sub_pd(_mm_sub_pd(rinv33,sh_ewald),velec));
 787             felec            = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
 788
 789             cutoff_mask      = _mm_cmplt_pd(rsq33,rcutoff2);
 790
 791             /* Update potential sum for this i atom from the interaction with this j atom. */
 792             velec            = _mm_and_pd(velec,cutoff_mask);
 793             velecsum         = _mm_add_pd(velecsum,velec);
 794
 795             fscal            = felec;
 796
 797             fscal            = _mm_and_pd(fscal,cutoff_mask);
 798
 799             /* Update vectorial force */
 800             fix3             = _mm_macc_pd(dx33,fscal,fix3);
 801             fiy3             = _mm_macc_pd(dy33,fscal,fiy3);
 802             fiz3             = _mm_macc_pd(dz33,fscal,fiz3);
 803
 804             fjx3             = _mm_macc_pd(dx33,fscal,fjx3);
 805             fjy3             = _mm_macc_pd(dy33,fscal,fjy3);
 806             fjz3             = _mm_macc_pd(dz33,fscal,fjz3);
 807
 808             }
 809
 810             gmx_mm_decrement_4rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 811
 812             /* Inner loop uses 488 flops */
 813         }
 814
 815         if(jidx<j_index_end)
 816         {
 817
 818             jnrA             = jjnr[jidx];
 819             j_coord_offsetA  = DIM*jnrA;
 820
 821             /* load j atom coordinates */
 822             gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 823                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 824                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 825
 826             /* Calculate displacement vector */
 827             dx00             = _mm_sub_pd(ix0,jx0);
 828             dy00             = _mm_sub_pd(iy0,jy0);
 829             dz00             = _mm_sub_pd(iz0,jz0);
 830             dx11             = _mm_sub_pd(ix1,jx1);
 831             dy11             = _mm_sub_pd(iy1,jy1);
 832             dz11             = _mm_sub_pd(iz1,jz1);
 833             dx12             = _mm_sub_pd(ix1,jx2);
 834             dy12             = _mm_sub_pd(iy1,jy2);
 835             dz12             = _mm_sub_pd(iz1,jz2);
 836             dx13             = _mm_sub_pd(ix1,jx3);
 837             dy13             = _mm_sub_pd(iy1,jy3);
 838             dz13             = _mm_sub_pd(iz1,jz3);
 839             dx21             = _mm_sub_pd(ix2,jx1);
 840             dy21             = _mm_sub_pd(iy2,jy1);
 841             dz21             = _mm_sub_pd(iz2,jz1);
 842             dx22             = _mm_sub_pd(ix2,jx2);
 843             dy22             = _mm_sub_pd(iy2,jy2);
 844             dz22             = _mm_sub_pd(iz2,jz2);
 845             dx23             = _mm_sub_pd(ix2,jx3);
 846             dy23             = _mm_sub_pd(iy2,jy3);
 847             dz23             = _mm_sub_pd(iz2,jz3);
 848             dx31             = _mm_sub_pd(ix3,jx1);
 849             dy31             = _mm_sub_pd(iy3,jy1);
 850             dz31             = _mm_sub_pd(iz3,jz1);
 851             dx32             = _mm_sub_pd(ix3,jx2);
 852             dy32             = _mm_sub_pd(iy3,jy2);
 853             dz32             = _mm_sub_pd(iz3,jz2);
 854             dx33             = _mm_sub_pd(ix3,jx3);
 855             dy33             = _mm_sub_pd(iy3,jy3);
 856             dz33             = _mm_sub_pd(iz3,jz3);
 857
 858             /* Calculate squared distance and things based on it */
 859             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 860             rsq11            = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
 861             rsq12            = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
 862             rsq13            = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
 863             rsq21            = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
 864             rsq22            = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
 865             rsq23            = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
 866             rsq31            = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
 867             rsq32            = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
 868             rsq33            = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
 869
 870             rinv11           = gmx_mm_invsqrt_pd(rsq11);
 871             rinv12           = gmx_mm_invsqrt_pd(rsq12);
 872             rinv13           = gmx_mm_invsqrt_pd(rsq13);
 873             rinv21           = gmx_mm_invsqrt_pd(rsq21);
 874             rinv22           = gmx_mm_invsqrt_pd(rsq22);
 875             rinv23           = gmx_mm_invsqrt_pd(rsq23);
 876             rinv31           = gmx_mm_invsqrt_pd(rsq31);
 877             rinv32           = gmx_mm_invsqrt_pd(rsq32);
 878             rinv33           = gmx_mm_invsqrt_pd(rsq33);
 879
 880             rinvsq00         = gmx_mm_inv_pd(rsq00);
 881             rinvsq11         = _mm_mul_pd(rinv11,rinv11);
 882             rinvsq12         = _mm_mul_pd(rinv12,rinv12);
 883             rinvsq13         = _mm_mul_pd(rinv13,rinv13);
 884             rinvsq21         = _mm_mul_pd(rinv21,rinv21);
 885             rinvsq22         = _mm_mul_pd(rinv22,rinv22);
 886             rinvsq23         = _mm_mul_pd(rinv23,rinv23);
 887             rinvsq31         = _mm_mul_pd(rinv31,rinv31);
 888             rinvsq32         = _mm_mul_pd(rinv32,rinv32);
 889             rinvsq33         = _mm_mul_pd(rinv33,rinv33);
 890
 891             fjx0             = _mm_setzero_pd();
 892             fjy0             = _mm_setzero_pd();
 893             fjz0             = _mm_setzero_pd();
 894             fjx1             = _mm_setzero_pd();
 895             fjy1             = _mm_setzero_pd();
 896             fjz1             = _mm_setzero_pd();
 897             fjx2             = _mm_setzero_pd();
 898             fjy2             = _mm_setzero_pd();
 899             fjz2             = _mm_setzero_pd();
 900             fjx3             = _mm_setzero_pd();
 901             fjy3             = _mm_setzero_pd();
 902             fjz3             = _mm_setzero_pd();
 903
 904             /**************************
 905              * CALCULATE INTERACTIONS *
 906              **************************/
 907
 908             if (gmx_mm_any_lt(rsq00,rcutoff2))
 909             {
 910
 911             /* LENNARD-JONES DISPERSION/REPULSION */
 912
 913             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 914             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 915             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 916             vvdw             = _mm_msub_pd(_mm_nmacc_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6),vvdw12),one_twelfth,
 917                                            _mm_mul_pd(_mm_nmacc_pd( c6_00,sh_vdw_invrcut6,vvdw6),one_sixth));
 918             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 919
 920             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 921
 922             /* Update potential sum for this i atom from the interaction with this j atom. */
 923             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 924             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 925             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 926
 927             fscal            = fvdw;
 928
 929             fscal            = _mm_and_pd(fscal,cutoff_mask);
 930
 931             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 932
 933             /* Update vectorial force */
 934             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 935             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 936             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 937
 938             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 939             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 940             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 941
 942             }
 943
 944             /**************************
 945              * CALCULATE INTERACTIONS *
 946              **************************/
 947
 948             if (gmx_mm_any_lt(rsq11,rcutoff2))
 949             {
 950
 951             r11              = _mm_mul_pd(rsq11,rinv11);
 952
 953             /* EWALD ELECTROSTATICS */
 954
 955             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 956             ewrt             = _mm_mul_pd(r11,ewtabscale);
 957             ewitab           = _mm_cvttpd_epi32(ewrt);
 958 #ifdef __XOP__
 959             eweps            = _mm_frcz_pd(ewrt);
 960 #else
 961             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 962 #endif
 963             twoeweps         = _mm_add_pd(eweps,eweps);
 964             ewitab           = _mm_slli_epi32(ewitab,2);
 965             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 966             ewtabD           = _mm_setzero_pd();
 967             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 968             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 969             ewtabFn          = _mm_setzero_pd();
 970             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 971             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 972             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 973             velec            = _mm_mul_pd(qq11,_mm_sub_pd(_mm_sub_pd(rinv11,sh_ewald),velec));
 974             felec            = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
 975
 976             cutoff_mask      = _mm_cmplt_pd(rsq11,rcutoff2);
 977
 978             /* Update potential sum for this i atom from the interaction with this j atom. */
 979             velec            = _mm_and_pd(velec,cutoff_mask);
 980             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 981             velecsum         = _mm_add_pd(velecsum,velec);
 982
 983             fscal            = felec;
 984
 985             fscal            = _mm_and_pd(fscal,cutoff_mask);
 986
 987             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 988
 989             /* Update vectorial force */
 990             fix1             = _mm_macc_pd(dx11,fscal,fix1);
 991             fiy1             = _mm_macc_pd(dy11,fscal,fiy1);
 992             fiz1             = _mm_macc_pd(dz11,fscal,fiz1);
 993
 994             fjx1             = _mm_macc_pd(dx11,fscal,fjx1);
 995             fjy1             = _mm_macc_pd(dy11,fscal,fjy1);
 996             fjz1             = _mm_macc_pd(dz11,fscal,fjz1);
 997
 998             }
 999
1000             /**************************
1001              * CALCULATE INTERACTIONS *
1002              **************************/
1003
1004             if (gmx_mm_any_lt(rsq12,rcutoff2))
1005             {
1006
1007             r12              = _mm_mul_pd(rsq12,rinv12);
1008
1009             /* EWALD ELECTROSTATICS */
1010
1011             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1012             ewrt             = _mm_mul_pd(r12,ewtabscale);
1013             ewitab           = _mm_cvttpd_epi32(ewrt);
1014 #ifdef __XOP__
1015             eweps            = _mm_frcz_pd(ewrt);
1016 #else
1017             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1018 #endif
1019             twoeweps         = _mm_add_pd(eweps,eweps);
1020             ewitab           = _mm_slli_epi32(ewitab,2);
1021             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1022             ewtabD           = _mm_setzero_pd();
1023             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1024             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1025             ewtabFn          = _mm_setzero_pd();
1026             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1027             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1028             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1029             velec            = _mm_mul_pd(qq12,_mm_sub_pd(_mm_sub_pd(rinv12,sh_ewald),velec));
1030             felec            = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
1031
1032             cutoff_mask      = _mm_cmplt_pd(rsq12,rcutoff2);
1033
1034             /* Update potential sum for this i atom from the interaction with this j atom. */
1035             velec            = _mm_and_pd(velec,cutoff_mask);
1036             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1037             velecsum         = _mm_add_pd(velecsum,velec);
1038
1039             fscal            = felec;
1040
1041             fscal            = _mm_and_pd(fscal,cutoff_mask);
1042
1043             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1044
1045             /* Update vectorial force */
1046             fix1             = _mm_macc_pd(dx12,fscal,fix1);
1047             fiy1             = _mm_macc_pd(dy12,fscal,fiy1);
1048             fiz1             = _mm_macc_pd(dz12,fscal,fiz1);
1049
1050             fjx2             = _mm_macc_pd(dx12,fscal,fjx2);
1051             fjy2             = _mm_macc_pd(dy12,fscal,fjy2);
1052             fjz2             = _mm_macc_pd(dz12,fscal,fjz2);
1053
1054             }
1055
1056             /**************************
1057              * CALCULATE INTERACTIONS *
1058              **************************/
1059
1060             if (gmx_mm_any_lt(rsq13,rcutoff2))
1061             {
1062
1063             r13              = _mm_mul_pd(rsq13,rinv13);
1064
1065             /* EWALD ELECTROSTATICS */
1066
1067             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1068             ewrt             = _mm_mul_pd(r13,ewtabscale);
1069             ewitab           = _mm_cvttpd_epi32(ewrt);
1070 #ifdef __XOP__
1071             eweps            = _mm_frcz_pd(ewrt);
1072 #else
1073             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1074 #endif
1075             twoeweps         = _mm_add_pd(eweps,eweps);
1076             ewitab           = _mm_slli_epi32(ewitab,2);
1077             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1078             ewtabD           = _mm_setzero_pd();
1079             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1080             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1081             ewtabFn          = _mm_setzero_pd();
1082             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1083             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1084             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1085             velec            = _mm_mul_pd(qq13,_mm_sub_pd(_mm_sub_pd(rinv13,sh_ewald),velec));
1086             felec            = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
1087
1088             cutoff_mask      = _mm_cmplt_pd(rsq13,rcutoff2);
1089
1090             /* Update potential sum for this i atom from the interaction with this j atom. */
1091             velec            = _mm_and_pd(velec,cutoff_mask);
1092             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1093             velecsum         = _mm_add_pd(velecsum,velec);
1094
1095             fscal            = felec;
1096
1097             fscal            = _mm_and_pd(fscal,cutoff_mask);
1098
1099             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1100
1101             /* Update vectorial force */
1102             fix1             = _mm_macc_pd(dx13,fscal,fix1);
1103             fiy1             = _mm_macc_pd(dy13,fscal,fiy1);
1104             fiz1             = _mm_macc_pd(dz13,fscal,fiz1);
1105
1106             fjx3             = _mm_macc_pd(dx13,fscal,fjx3);
1107             fjy3             = _mm_macc_pd(dy13,fscal,fjy3);
1108             fjz3             = _mm_macc_pd(dz13,fscal,fjz3);
1109
1110             }
1111
1112             /**************************
1113              * CALCULATE INTERACTIONS *
1114              **************************/
1115
1116             if (gmx_mm_any_lt(rsq21,rcutoff2))
1117             {
1118
1119             r21              = _mm_mul_pd(rsq21,rinv21);
1120
1121             /* EWALD ELECTROSTATICS */
1122
1123             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1124             ewrt             = _mm_mul_pd(r21,ewtabscale);
1125             ewitab           = _mm_cvttpd_epi32(ewrt);
1126 #ifdef __XOP__
1127             eweps            = _mm_frcz_pd(ewrt);
1128 #else
1129             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1130 #endif
1131             twoeweps         = _mm_add_pd(eweps,eweps);
1132             ewitab           = _mm_slli_epi32(ewitab,2);
1133             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1134             ewtabD           = _mm_setzero_pd();
1135             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1136             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1137             ewtabFn          = _mm_setzero_pd();
1138             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1139             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1140             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1141             velec            = _mm_mul_pd(qq21,_mm_sub_pd(_mm_sub_pd(rinv21,sh_ewald),velec));
1142             felec            = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1143
1144             cutoff_mask      = _mm_cmplt_pd(rsq21,rcutoff2);
1145
1146             /* Update potential sum for this i atom from the interaction with this j atom. */
1147             velec            = _mm_and_pd(velec,cutoff_mask);
1148             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1149             velecsum         = _mm_add_pd(velecsum,velec);
1150
1151             fscal            = felec;
1152
1153             fscal            = _mm_and_pd(fscal,cutoff_mask);
1154
1155             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1156
1157             /* Update vectorial force */
1158             fix2             = _mm_macc_pd(dx21,fscal,fix2);
1159             fiy2             = _mm_macc_pd(dy21,fscal,fiy2);
1160             fiz2             = _mm_macc_pd(dz21,fscal,fiz2);
1161
1162             fjx1             = _mm_macc_pd(dx21,fscal,fjx1);
1163             fjy1             = _mm_macc_pd(dy21,fscal,fjy1);
1164             fjz1             = _mm_macc_pd(dz21,fscal,fjz1);
1165
1166             }
1167
1168             /**************************
1169              * CALCULATE INTERACTIONS *
1170              **************************/
1171
1172             if (gmx_mm_any_lt(rsq22,rcutoff2))
1173             {
1174
1175             r22              = _mm_mul_pd(rsq22,rinv22);
1176
1177             /* EWALD ELECTROSTATICS */
1178
1179             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1180             ewrt             = _mm_mul_pd(r22,ewtabscale);
1181             ewitab           = _mm_cvttpd_epi32(ewrt);
1182 #ifdef __XOP__
1183             eweps            = _mm_frcz_pd(ewrt);
1184 #else
1185             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1186 #endif
1187             twoeweps         = _mm_add_pd(eweps,eweps);
1188             ewitab           = _mm_slli_epi32(ewitab,2);
1189             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1190             ewtabD           = _mm_setzero_pd();
1191             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1192             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1193             ewtabFn          = _mm_setzero_pd();
1194             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1195             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1196             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1197             velec            = _mm_mul_pd(qq22,_mm_sub_pd(_mm_sub_pd(rinv22,sh_ewald),velec));
1198             felec            = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1199
1200             cutoff_mask      = _mm_cmplt_pd(rsq22,rcutoff2);
1201
1202             /* Update potential sum for this i atom from the interaction with this j atom. */
1203             velec            = _mm_and_pd(velec,cutoff_mask);
1204             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1205             velecsum         = _mm_add_pd(velecsum,velec);
1206
1207             fscal            = felec;
1208
1209             fscal            = _mm_and_pd(fscal,cutoff_mask);
1210
1211             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1212
1213             /* Update vectorial force */
1214             fix2             = _mm_macc_pd(dx22,fscal,fix2);
1215             fiy2             = _mm_macc_pd(dy22,fscal,fiy2);
1216             fiz2             = _mm_macc_pd(dz22,fscal,fiz2);
1217
1218             fjx2             = _mm_macc_pd(dx22,fscal,fjx2);
1219             fjy2             = _mm_macc_pd(dy22,fscal,fjy2);
1220             fjz2             = _mm_macc_pd(dz22,fscal,fjz2);
1221
1222             }
1223
1224             /**************************
1225              * CALCULATE INTERACTIONS *
1226              **************************/
1227
1228             if (gmx_mm_any_lt(rsq23,rcutoff2))
1229             {
1230
1231             r23              = _mm_mul_pd(rsq23,rinv23);
1232
1233             /* EWALD ELECTROSTATICS */
1234
1235             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1236             ewrt             = _mm_mul_pd(r23,ewtabscale);
1237             ewitab           = _mm_cvttpd_epi32(ewrt);
1238 #ifdef __XOP__
1239             eweps            = _mm_frcz_pd(ewrt);
1240 #else
1241             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1242 #endif
1243             twoeweps         = _mm_add_pd(eweps,eweps);
1244             ewitab           = _mm_slli_epi32(ewitab,2);
1245             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1246             ewtabD           = _mm_setzero_pd();
1247             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1248             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1249             ewtabFn          = _mm_setzero_pd();
1250             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1251             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1252             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1253             velec            = _mm_mul_pd(qq23,_mm_sub_pd(_mm_sub_pd(rinv23,sh_ewald),velec));
1254             felec            = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
1255
1256             cutoff_mask      = _mm_cmplt_pd(rsq23,rcutoff2);
1257
1258             /* Update potential sum for this i atom from the interaction with this j atom. */
1259             velec            = _mm_and_pd(velec,cutoff_mask);
1260             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1261             velecsum         = _mm_add_pd(velecsum,velec);
1262
1263             fscal            = felec;
1264
1265             fscal            = _mm_and_pd(fscal,cutoff_mask);
1266
1267             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1268
1269             /* Update vectorial force */
1270             fix2             = _mm_macc_pd(dx23,fscal,fix2);
1271             fiy2             = _mm_macc_pd(dy23,fscal,fiy2);
1272             fiz2             = _mm_macc_pd(dz23,fscal,fiz2);
1273
1274             fjx3             = _mm_macc_pd(dx23,fscal,fjx3);
1275             fjy3             = _mm_macc_pd(dy23,fscal,fjy3);
1276             fjz3             = _mm_macc_pd(dz23,fscal,fjz3);
1277
1278             }
1279
1280             /**************************
1281              * CALCULATE INTERACTIONS *
1282              **************************/
1283
1284             if (gmx_mm_any_lt(rsq31,rcutoff2))
1285             {
1286
1287             r31              = _mm_mul_pd(rsq31,rinv31);
1288
1289             /* EWALD ELECTROSTATICS */
1290
1291             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1292             ewrt             = _mm_mul_pd(r31,ewtabscale);
1293             ewitab           = _mm_cvttpd_epi32(ewrt);
1294 #ifdef __XOP__
1295             eweps            = _mm_frcz_pd(ewrt);
1296 #else
1297             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1298 #endif
1299             twoeweps         = _mm_add_pd(eweps,eweps);
1300             ewitab           = _mm_slli_epi32(ewitab,2);
1301             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1302             ewtabD           = _mm_setzero_pd();
1303             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1304             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1305             ewtabFn          = _mm_setzero_pd();
1306             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1307             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1308             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1309             velec            = _mm_mul_pd(qq31,_mm_sub_pd(_mm_sub_pd(rinv31,sh_ewald),velec));
1310             felec            = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
1311
1312             cutoff_mask      = _mm_cmplt_pd(rsq31,rcutoff2);
1313
1314             /* Update potential sum for this i atom from the interaction with this j atom. */
1315             velec            = _mm_and_pd(velec,cutoff_mask);
1316             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1317             velecsum         = _mm_add_pd(velecsum,velec);
1318
1319             fscal            = felec;
1320
1321             fscal            = _mm_and_pd(fscal,cutoff_mask);
1322
1323             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1324
1325             /* Update vectorial force */
1326             fix3             = _mm_macc_pd(dx31,fscal,fix3);
1327             fiy3             = _mm_macc_pd(dy31,fscal,fiy3);
1328             fiz3             = _mm_macc_pd(dz31,fscal,fiz3);
1329
1330             fjx1             = _mm_macc_pd(dx31,fscal,fjx1);
1331             fjy1             = _mm_macc_pd(dy31,fscal,fjy1);
1332             fjz1             = _mm_macc_pd(dz31,fscal,fjz1);
1333
1334             }
1335
1336             /**************************
1337              * CALCULATE INTERACTIONS *
1338              **************************/
1339
1340             if (gmx_mm_any_lt(rsq32,rcutoff2))
1341             {
1342
1343             r32              = _mm_mul_pd(rsq32,rinv32);
1344
1345             /* EWALD ELECTROSTATICS */
1346
1347             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1348             ewrt             = _mm_mul_pd(r32,ewtabscale);
1349             ewitab           = _mm_cvttpd_epi32(ewrt);
1350 #ifdef __XOP__
1351             eweps            = _mm_frcz_pd(ewrt);
1352 #else
1353             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1354 #endif
1355             twoeweps         = _mm_add_pd(eweps,eweps);
1356             ewitab           = _mm_slli_epi32(ewitab,2);
1357             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1358             ewtabD           = _mm_setzero_pd();
1359             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1360             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1361             ewtabFn          = _mm_setzero_pd();
1362             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1363             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1364             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1365             velec            = _mm_mul_pd(qq32,_mm_sub_pd(_mm_sub_pd(rinv32,sh_ewald),velec));
1366             felec            = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
1367
1368             cutoff_mask      = _mm_cmplt_pd(rsq32,rcutoff2);
1369
1370             /* Update potential sum for this i atom from the interaction with this j atom. */
1371             velec            = _mm_and_pd(velec,cutoff_mask);
1372             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1373             velecsum         = _mm_add_pd(velecsum,velec);
1374
1375             fscal            = felec;
1376
1377             fscal            = _mm_and_pd(fscal,cutoff_mask);
1378
1379             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1380
1381             /* Update vectorial force */
1382             fix3             = _mm_macc_pd(dx32,fscal,fix3);
1383             fiy3             = _mm_macc_pd(dy32,fscal,fiy3);
1384             fiz3             = _mm_macc_pd(dz32,fscal,fiz3);
1385
1386             fjx2             = _mm_macc_pd(dx32,fscal,fjx2);
1387             fjy2             = _mm_macc_pd(dy32,fscal,fjy2);
1388             fjz2             = _mm_macc_pd(dz32,fscal,fjz2);
1389
1390             }
1391
1392             /**************************
1393              * CALCULATE INTERACTIONS *
1394              **************************/
1395
1396             if (gmx_mm_any_lt(rsq33,rcutoff2))
1397             {
1398
1399             r33              = _mm_mul_pd(rsq33,rinv33);
1400
1401             /* EWALD ELECTROSTATICS */
1402
1403             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1404             ewrt             = _mm_mul_pd(r33,ewtabscale);
1405             ewitab           = _mm_cvttpd_epi32(ewrt);
1406 #ifdef __XOP__
1407             eweps            = _mm_frcz_pd(ewrt);
1408 #else
1409             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1410 #endif
1411             twoeweps         = _mm_add_pd(eweps,eweps);
1412             ewitab           = _mm_slli_epi32(ewitab,2);
1413             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
1414             ewtabD           = _mm_setzero_pd();
1415             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
1416             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
1417             ewtabFn          = _mm_setzero_pd();
1418             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
1419             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
1420             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
1421             velec            = _mm_mul_pd(qq33,_mm_sub_pd(_mm_sub_pd(rinv33,sh_ewald),velec));
1422             felec            = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
1423
1424             cutoff_mask      = _mm_cmplt_pd(rsq33,rcutoff2);
1425
1426             /* Update potential sum for this i atom from the interaction with this j atom. */
1427             velec            = _mm_and_pd(velec,cutoff_mask);
1428             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
1429             velecsum         = _mm_add_pd(velecsum,velec);
1430
1431             fscal            = felec;
1432
1433             fscal            = _mm_and_pd(fscal,cutoff_mask);
1434
1435             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1436
1437             /* Update vectorial force */
1438             fix3             = _mm_macc_pd(dx33,fscal,fix3);
1439             fiy3             = _mm_macc_pd(dy33,fscal,fiy3);
1440             fiz3             = _mm_macc_pd(dz33,fscal,fiz3);
1441
1442             fjx3             = _mm_macc_pd(dx33,fscal,fjx3);
1443             fjy3             = _mm_macc_pd(dy33,fscal,fjy3);
1444             fjz3             = _mm_macc_pd(dz33,fscal,fjz3);
1445
1446             }
1447
1448             gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1449
1450             /* Inner loop uses 488 flops */
1451         }
1452
1453         /* End of innermost loop */
1454
1455         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1456                                               f+i_coord_offset,fshift+i_shift_offset);
1457
1458         ggid                        = gid[iidx];
1459         /* Update potential energies */
1460         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
1461         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
1462
1463         /* Increment number of inner iterations */
1464         inneriter                  += j_index_end - j_index_start;
1465
1466         /* Outer loop uses 26 flops */
1467     }
1468
1469     /* Increment number of outer iterations */
1470     outeriter        += nri;
1471
1472     /* Update outer/inner flops */
1473
1474     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*488);
1475 }
1476 /*
1477  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJSh_GeomW4W4_F_avx_128_fma_double
1478  * Electrostatics interaction: Ewald
1479  * VdW interaction:            LennardJones
1480  * Geometry:                   Water4-Water4
1481  * Calculate force/pot:        Force
1482  */
1483 void
1484 nb_kernel_ElecEwSh_VdwLJSh_GeomW4W4_F_avx_128_fma_double
1485                     (t_nblist * gmx_restrict                nlist,
1486                      rvec * gmx_restrict                    xx,
1487                      rvec * gmx_restrict                    ff,
1488                      t_forcerec * gmx_restrict              fr,
1489                      t_mdatoms * gmx_restrict               mdatoms,
1490                      nb_kernel_data_t * gmx_restrict        kernel_data,
1491                      t_nrnb * gmx_restrict                  nrnb)
1492 {
1493     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1494      * just 0 for non-waters.
1495      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
1496      * jnr indices corresponding to data put in the four positions in the SIMD register.
1497      */
1498     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1499     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1500     int              jnrA,jnrB;
1501     int              j_coord_offsetA,j_coord_offsetB;
1502     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1503     real             rcutoff_scalar;
1504     real             *shiftvec,*fshift,*x,*f;
1505     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1506     int              vdwioffset0;
1507     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1508     int              vdwioffset1;
1509     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1510     int              vdwioffset2;
1511     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1512     int              vdwioffset3;
1513     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1514     int              vdwjidx0A,vdwjidx0B;
1515     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1516     int              vdwjidx1A,vdwjidx1B;
1517     __m128d          jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1518     int              vdwjidx2A,vdwjidx2B;
1519     __m128d          jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1520     int              vdwjidx3A,vdwjidx3B;
1521     __m128d          jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1522     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1523     __m128d          dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1524     __m128d          dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1525     __m128d          dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1526     __m128d          dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1527     __m128d          dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1528     __m128d          dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1529     __m128d          dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1530     __m128d          dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1531     __m128d          dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1532     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
1533     real             *charge;
1534     int              nvdwtype;
1535     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1536     int              *vdwtype;
1537     real             *vdwparam;
1538     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
1539     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
1540     __m128i          ewitab;
1541     __m128d          ewtabscale,eweps,twoeweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1542     real             *ewtab;
1543     __m128d          dummy_mask,cutoff_mask;
1544     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
1545     __m128d          one     = _mm_set1_pd(1.0);
1546     __m128d          two     = _mm_set1_pd(2.0);
1547     x                = xx[0];
1548     f                = ff[0];
1549
1550     nri              = nlist->nri;
1551     iinr             = nlist->iinr;
1552     jindex           = nlist->jindex;
1553     jjnr             = nlist->jjnr;
1554     shiftidx         = nlist->shift;
1555     gid              = nlist->gid;
1556     shiftvec         = fr->shift_vec[0];
1557     fshift           = fr->fshift[0];
1558     facel            = _mm_set1_pd(fr->epsfac);
1559     charge           = mdatoms->chargeA;
1560     nvdwtype         = fr->ntype;
1561     vdwparam         = fr->nbfp;
1562     vdwtype          = mdatoms->typeA;
1563
1564     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
1565     ewtab            = fr->ic->tabq_coul_F;
1566     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
1567     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
1568
1569     /* Setup water-specific parameters */
1570     inr              = nlist->iinr[0];
1571     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
1572     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
1573     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
1574     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
1575
1576     jq1              = _mm_set1_pd(charge[inr+1]);
1577     jq2              = _mm_set1_pd(charge[inr+2]);
1578     jq3              = _mm_set1_pd(charge[inr+3]);
1579     vdwjidx0A        = 2*vdwtype[inr+0];
1580     c6_00            = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A]);
1581     c12_00           = _mm_set1_pd(vdwparam[vdwioffset0+vdwjidx0A+1]);
1582     qq11             = _mm_mul_pd(iq1,jq1);
1583     qq12             = _mm_mul_pd(iq1,jq2);
1584     qq13             = _mm_mul_pd(iq1,jq3);
1585     qq21             = _mm_mul_pd(iq2,jq1);
1586     qq22             = _mm_mul_pd(iq2,jq2);
1587     qq23             = _mm_mul_pd(iq2,jq3);
1588     qq31             = _mm_mul_pd(iq3,jq1);
1589     qq32             = _mm_mul_pd(iq3,jq2);
1590     qq33             = _mm_mul_pd(iq3,jq3);
1591
1592     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1593     rcutoff_scalar   = fr->rcoulomb;
1594     rcutoff          = _mm_set1_pd(rcutoff_scalar);
1595     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
1596
1597     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
1598     rvdw             = _mm_set1_pd(fr->rvdw);
1599
1600     /* Avoid stupid compiler warnings */
1601     jnrA = jnrB = 0;
1602     j_coord_offsetA = 0;
1603     j_coord_offsetB = 0;
1604
1605     outeriter        = 0;
1606     inneriter        = 0;
1607
1608     /* Start outer loop over neighborlists */
1609     for(iidx=0; iidx<nri; iidx++)
1610     {
1611         /* Load shift vector for this list */
1612         i_shift_offset   = DIM*shiftidx[iidx];
1613
1614         /* Load limits for loop over neighbors */
1615         j_index_start    = jindex[iidx];
1616         j_index_end      = jindex[iidx+1];
1617
1618         /* Get outer coordinate index */
1619         inr              = iinr[iidx];
1620         i_coord_offset   = DIM*inr;
1621
1622         /* Load i particle coords and add shift vector */
1623         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
1624                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1625
1626         fix0             = _mm_setzero_pd();
1627         fiy0             = _mm_setzero_pd();
1628         fiz0             = _mm_setzero_pd();
1629         fix1             = _mm_setzero_pd();
1630         fiy1             = _mm_setzero_pd();
1631         fiz1             = _mm_setzero_pd();
1632         fix2             = _mm_setzero_pd();
1633         fiy2             = _mm_setzero_pd();
1634         fiz2             = _mm_setzero_pd();
1635         fix3             = _mm_setzero_pd();
1636         fiy3             = _mm_setzero_pd();
1637         fiz3             = _mm_setzero_pd();
1638
1639         /* Start inner kernel loop */
1640         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
1641         {
1642
1643             /* Get j neighbor index, and coordinate index */
1644             jnrA             = jjnr[jidx];
1645             jnrB             = jjnr[jidx+1];
1646             j_coord_offsetA  = DIM*jnrA;
1647             j_coord_offsetB  = DIM*jnrB;
1648
1649             /* load j atom coordinates */
1650             gmx_mm_load_4rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
1651                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1652                                               &jy2,&jz2,&jx3,&jy3,&jz3);
1653
1654             /* Calculate displacement vector */
1655             dx00             = _mm_sub_pd(ix0,jx0);
1656             dy00             = _mm_sub_pd(iy0,jy0);
1657             dz00             = _mm_sub_pd(iz0,jz0);
1658             dx11             = _mm_sub_pd(ix1,jx1);
1659             dy11             = _mm_sub_pd(iy1,jy1);
1660             dz11             = _mm_sub_pd(iz1,jz1);
1661             dx12             = _mm_sub_pd(ix1,jx2);
1662             dy12             = _mm_sub_pd(iy1,jy2);
1663             dz12             = _mm_sub_pd(iz1,jz2);
1664             dx13             = _mm_sub_pd(ix1,jx3);
1665             dy13             = _mm_sub_pd(iy1,jy3);
1666             dz13             = _mm_sub_pd(iz1,jz3);
1667             dx21             = _mm_sub_pd(ix2,jx1);
1668             dy21             = _mm_sub_pd(iy2,jy1);
1669             dz21             = _mm_sub_pd(iz2,jz1);
1670             dx22             = _mm_sub_pd(ix2,jx2);
1671             dy22             = _mm_sub_pd(iy2,jy2);
1672             dz22             = _mm_sub_pd(iz2,jz2);
1673             dx23             = _mm_sub_pd(ix2,jx3);
1674             dy23             = _mm_sub_pd(iy2,jy3);
1675             dz23             = _mm_sub_pd(iz2,jz3);
1676             dx31             = _mm_sub_pd(ix3,jx1);
1677             dy31             = _mm_sub_pd(iy3,jy1);
1678             dz31             = _mm_sub_pd(iz3,jz1);
1679             dx32             = _mm_sub_pd(ix3,jx2);
1680             dy32             = _mm_sub_pd(iy3,jy2);
1681             dz32             = _mm_sub_pd(iz3,jz2);
1682             dx33             = _mm_sub_pd(ix3,jx3);
1683             dy33             = _mm_sub_pd(iy3,jy3);
1684             dz33             = _mm_sub_pd(iz3,jz3);
1685
1686             /* Calculate squared distance and things based on it */
1687             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1688             rsq11            = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
1689             rsq12            = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
1690             rsq13            = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
1691             rsq21            = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
1692             rsq22            = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
1693             rsq23            = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
1694             rsq31            = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
1695             rsq32            = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
1696             rsq33            = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
1697
1698             rinv11           = gmx_mm_invsqrt_pd(rsq11);
1699             rinv12           = gmx_mm_invsqrt_pd(rsq12);
1700             rinv13           = gmx_mm_invsqrt_pd(rsq13);
1701             rinv21           = gmx_mm_invsqrt_pd(rsq21);
1702             rinv22           = gmx_mm_invsqrt_pd(rsq22);
1703             rinv23           = gmx_mm_invsqrt_pd(rsq23);
1704             rinv31           = gmx_mm_invsqrt_pd(rsq31);
1705             rinv32           = gmx_mm_invsqrt_pd(rsq32);
1706             rinv33           = gmx_mm_invsqrt_pd(rsq33);
1707
1708             rinvsq00         = gmx_mm_inv_pd(rsq00);
1709             rinvsq11         = _mm_mul_pd(rinv11,rinv11);
1710             rinvsq12         = _mm_mul_pd(rinv12,rinv12);
1711             rinvsq13         = _mm_mul_pd(rinv13,rinv13);
1712             rinvsq21         = _mm_mul_pd(rinv21,rinv21);
1713             rinvsq22         = _mm_mul_pd(rinv22,rinv22);
1714             rinvsq23         = _mm_mul_pd(rinv23,rinv23);
1715             rinvsq31         = _mm_mul_pd(rinv31,rinv31);
1716             rinvsq32         = _mm_mul_pd(rinv32,rinv32);
1717             rinvsq33         = _mm_mul_pd(rinv33,rinv33);
1718
1719             fjx0             = _mm_setzero_pd();
1720             fjy0             = _mm_setzero_pd();
1721             fjz0             = _mm_setzero_pd();
1722             fjx1             = _mm_setzero_pd();
1723             fjy1             = _mm_setzero_pd();
1724             fjz1             = _mm_setzero_pd();
1725             fjx2             = _mm_setzero_pd();
1726             fjy2             = _mm_setzero_pd();
1727             fjz2             = _mm_setzero_pd();
1728             fjx3             = _mm_setzero_pd();
1729             fjy3             = _mm_setzero_pd();
1730             fjz3             = _mm_setzero_pd();
1731
1732             /**************************
1733              * CALCULATE INTERACTIONS *
1734              **************************/
1735
1736             if (gmx_mm_any_lt(rsq00,rcutoff2))
1737             {
1738
1739             /* LENNARD-JONES DISPERSION/REPULSION */
1740
1741             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1742             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
1743
1744             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
1745
1746             fscal            = fvdw;
1747
1748             fscal            = _mm_and_pd(fscal,cutoff_mask);
1749
1750             /* Update vectorial force */
1751             fix0             = _mm_macc_pd(dx00,fscal,fix0);
1752             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
1753             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
1754
1755             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
1756             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
1757             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
1758
1759             }
1760
1761             /**************************
1762              * CALCULATE INTERACTIONS *
1763              **************************/
1764
1765             if (gmx_mm_any_lt(rsq11,rcutoff2))
1766             {
1767
1768             r11              = _mm_mul_pd(rsq11,rinv11);
1769
1770             /* EWALD ELECTROSTATICS */
1771
1772             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1773             ewrt             = _mm_mul_pd(r11,ewtabscale);
1774             ewitab           = _mm_cvttpd_epi32(ewrt);
1775 #ifdef __XOP__
1776             eweps            = _mm_frcz_pd(ewrt);
1777 #else
1778             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1779 #endif
1780             twoeweps         = _mm_add_pd(eweps,eweps);
1781             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1782                                          &ewtabF,&ewtabFn);
1783             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1784             felec            = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
1785
1786             cutoff_mask      = _mm_cmplt_pd(rsq11,rcutoff2);
1787
1788             fscal            = felec;
1789
1790             fscal            = _mm_and_pd(fscal,cutoff_mask);
1791
1792             /* Update vectorial force */
1793             fix1             = _mm_macc_pd(dx11,fscal,fix1);
1794             fiy1             = _mm_macc_pd(dy11,fscal,fiy1);
1795             fiz1             = _mm_macc_pd(dz11,fscal,fiz1);
1796
1797             fjx1             = _mm_macc_pd(dx11,fscal,fjx1);
1798             fjy1             = _mm_macc_pd(dy11,fscal,fjy1);
1799             fjz1             = _mm_macc_pd(dz11,fscal,fjz1);
1800
1801             }
1802
1803             /**************************
1804              * CALCULATE INTERACTIONS *
1805              **************************/
1806
1807             if (gmx_mm_any_lt(rsq12,rcutoff2))
1808             {
1809
1810             r12              = _mm_mul_pd(rsq12,rinv12);
1811
1812             /* EWALD ELECTROSTATICS */
1813
1814             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1815             ewrt             = _mm_mul_pd(r12,ewtabscale);
1816             ewitab           = _mm_cvttpd_epi32(ewrt);
1817 #ifdef __XOP__
1818             eweps            = _mm_frcz_pd(ewrt);
1819 #else
1820             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1821 #endif
1822             twoeweps         = _mm_add_pd(eweps,eweps);
1823             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1824                                          &ewtabF,&ewtabFn);
1825             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1826             felec            = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
1827
1828             cutoff_mask      = _mm_cmplt_pd(rsq12,rcutoff2);
1829
1830             fscal            = felec;
1831
1832             fscal            = _mm_and_pd(fscal,cutoff_mask);
1833
1834             /* Update vectorial force */
1835             fix1             = _mm_macc_pd(dx12,fscal,fix1);
1836             fiy1             = _mm_macc_pd(dy12,fscal,fiy1);
1837             fiz1             = _mm_macc_pd(dz12,fscal,fiz1);
1838
1839             fjx2             = _mm_macc_pd(dx12,fscal,fjx2);
1840             fjy2             = _mm_macc_pd(dy12,fscal,fjy2);
1841             fjz2             = _mm_macc_pd(dz12,fscal,fjz2);
1842
1843             }
1844
1845             /**************************
1846              * CALCULATE INTERACTIONS *
1847              **************************/
1848
1849             if (gmx_mm_any_lt(rsq13,rcutoff2))
1850             {
1851
1852             r13              = _mm_mul_pd(rsq13,rinv13);
1853
1854             /* EWALD ELECTROSTATICS */
1855
1856             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1857             ewrt             = _mm_mul_pd(r13,ewtabscale);
1858             ewitab           = _mm_cvttpd_epi32(ewrt);
1859 #ifdef __XOP__
1860             eweps            = _mm_frcz_pd(ewrt);
1861 #else
1862             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1863 #endif
1864             twoeweps         = _mm_add_pd(eweps,eweps);
1865             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1866                                          &ewtabF,&ewtabFn);
1867             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1868             felec            = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
1869
1870             cutoff_mask      = _mm_cmplt_pd(rsq13,rcutoff2);
1871
1872             fscal            = felec;
1873
1874             fscal            = _mm_and_pd(fscal,cutoff_mask);
1875
1876             /* Update vectorial force */
1877             fix1             = _mm_macc_pd(dx13,fscal,fix1);
1878             fiy1             = _mm_macc_pd(dy13,fscal,fiy1);
1879             fiz1             = _mm_macc_pd(dz13,fscal,fiz1);
1880
1881             fjx3             = _mm_macc_pd(dx13,fscal,fjx3);
1882             fjy3             = _mm_macc_pd(dy13,fscal,fjy3);
1883             fjz3             = _mm_macc_pd(dz13,fscal,fjz3);
1884
1885             }
1886
1887             /**************************
1888              * CALCULATE INTERACTIONS *
1889              **************************/
1890
1891             if (gmx_mm_any_lt(rsq21,rcutoff2))
1892             {
1893
1894             r21              = _mm_mul_pd(rsq21,rinv21);
1895
1896             /* EWALD ELECTROSTATICS */
1897
1898             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1899             ewrt             = _mm_mul_pd(r21,ewtabscale);
1900             ewitab           = _mm_cvttpd_epi32(ewrt);
1901 #ifdef __XOP__
1902             eweps            = _mm_frcz_pd(ewrt);
1903 #else
1904             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1905 #endif
1906             twoeweps         = _mm_add_pd(eweps,eweps);
1907             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1908                                          &ewtabF,&ewtabFn);
1909             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1910             felec            = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
1911
1912             cutoff_mask      = _mm_cmplt_pd(rsq21,rcutoff2);
1913
1914             fscal            = felec;
1915
1916             fscal            = _mm_and_pd(fscal,cutoff_mask);
1917
1918             /* Update vectorial force */
1919             fix2             = _mm_macc_pd(dx21,fscal,fix2);
1920             fiy2             = _mm_macc_pd(dy21,fscal,fiy2);
1921             fiz2             = _mm_macc_pd(dz21,fscal,fiz2);
1922
1923             fjx1             = _mm_macc_pd(dx21,fscal,fjx1);
1924             fjy1             = _mm_macc_pd(dy21,fscal,fjy1);
1925             fjz1             = _mm_macc_pd(dz21,fscal,fjz1);
1926
1927             }
1928
1929             /**************************
1930              * CALCULATE INTERACTIONS *
1931              **************************/
1932
1933             if (gmx_mm_any_lt(rsq22,rcutoff2))
1934             {
1935
1936             r22              = _mm_mul_pd(rsq22,rinv22);
1937
1938             /* EWALD ELECTROSTATICS */
1939
1940             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1941             ewrt             = _mm_mul_pd(r22,ewtabscale);
1942             ewitab           = _mm_cvttpd_epi32(ewrt);
1943 #ifdef __XOP__
1944             eweps            = _mm_frcz_pd(ewrt);
1945 #else
1946             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1947 #endif
1948             twoeweps         = _mm_add_pd(eweps,eweps);
1949             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1950                                          &ewtabF,&ewtabFn);
1951             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1952             felec            = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
1953
1954             cutoff_mask      = _mm_cmplt_pd(rsq22,rcutoff2);
1955
1956             fscal            = felec;
1957
1958             fscal            = _mm_and_pd(fscal,cutoff_mask);
1959
1960             /* Update vectorial force */
1961             fix2             = _mm_macc_pd(dx22,fscal,fix2);
1962             fiy2             = _mm_macc_pd(dy22,fscal,fiy2);
1963             fiz2             = _mm_macc_pd(dz22,fscal,fiz2);
1964
1965             fjx2             = _mm_macc_pd(dx22,fscal,fjx2);
1966             fjy2             = _mm_macc_pd(dy22,fscal,fjy2);
1967             fjz2             = _mm_macc_pd(dz22,fscal,fjz2);
1968
1969             }
1970
1971             /**************************
1972              * CALCULATE INTERACTIONS *
1973              **************************/
1974
1975             if (gmx_mm_any_lt(rsq23,rcutoff2))
1976             {
1977
1978             r23              = _mm_mul_pd(rsq23,rinv23);
1979
1980             /* EWALD ELECTROSTATICS */
1981
1982             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1983             ewrt             = _mm_mul_pd(r23,ewtabscale);
1984             ewitab           = _mm_cvttpd_epi32(ewrt);
1985 #ifdef __XOP__
1986             eweps            = _mm_frcz_pd(ewrt);
1987 #else
1988             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1989 #endif
1990             twoeweps         = _mm_add_pd(eweps,eweps);
1991             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1992                                          &ewtabF,&ewtabFn);
1993             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1994             felec            = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
1995
1996             cutoff_mask      = _mm_cmplt_pd(rsq23,rcutoff2);
1997
1998             fscal            = felec;
1999
2000             fscal            = _mm_and_pd(fscal,cutoff_mask);
2001
2002             /* Update vectorial force */
2003             fix2             = _mm_macc_pd(dx23,fscal,fix2);
2004             fiy2             = _mm_macc_pd(dy23,fscal,fiy2);
2005             fiz2             = _mm_macc_pd(dz23,fscal,fiz2);
2006
2007             fjx3             = _mm_macc_pd(dx23,fscal,fjx3);
2008             fjy3             = _mm_macc_pd(dy23,fscal,fjy3);
2009             fjz3             = _mm_macc_pd(dz23,fscal,fjz3);
2010
2011             }
2012
2013             /**************************
2014              * CALCULATE INTERACTIONS *
2015              **************************/
2016
2017             if (gmx_mm_any_lt(rsq31,rcutoff2))
2018             {
2019
2020             r31              = _mm_mul_pd(rsq31,rinv31);
2021
2022             /* EWALD ELECTROSTATICS */
2023
2024             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2025             ewrt             = _mm_mul_pd(r31,ewtabscale);
2026             ewitab           = _mm_cvttpd_epi32(ewrt);
2027 #ifdef __XOP__
2028             eweps            = _mm_frcz_pd(ewrt);
2029 #else
2030             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2031 #endif
2032             twoeweps         = _mm_add_pd(eweps,eweps);
2033             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
2034                                          &ewtabF,&ewtabFn);
2035             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2036             felec            = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
2037
2038             cutoff_mask      = _mm_cmplt_pd(rsq31,rcutoff2);
2039
2040             fscal            = felec;
2041
2042             fscal            = _mm_and_pd(fscal,cutoff_mask);
2043
2044             /* Update vectorial force */
2045             fix3             = _mm_macc_pd(dx31,fscal,fix3);
2046             fiy3             = _mm_macc_pd(dy31,fscal,fiy3);
2047             fiz3             = _mm_macc_pd(dz31,fscal,fiz3);
2048
2049             fjx1             = _mm_macc_pd(dx31,fscal,fjx1);
2050             fjy1             = _mm_macc_pd(dy31,fscal,fjy1);
2051             fjz1             = _mm_macc_pd(dz31,fscal,fjz1);
2052
2053             }
2054
2055             /**************************
2056              * CALCULATE INTERACTIONS *
2057              **************************/
2058
2059             if (gmx_mm_any_lt(rsq32,rcutoff2))
2060             {
2061
2062             r32              = _mm_mul_pd(rsq32,rinv32);
2063
2064             /* EWALD ELECTROSTATICS */
2065
2066             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2067             ewrt             = _mm_mul_pd(r32,ewtabscale);
2068             ewitab           = _mm_cvttpd_epi32(ewrt);
2069 #ifdef __XOP__
2070             eweps            = _mm_frcz_pd(ewrt);
2071 #else
2072             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2073 #endif
2074             twoeweps         = _mm_add_pd(eweps,eweps);
2075             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
2076                                          &ewtabF,&ewtabFn);
2077             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2078             felec            = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
2079
2080             cutoff_mask      = _mm_cmplt_pd(rsq32,rcutoff2);
2081
2082             fscal            = felec;
2083
2084             fscal            = _mm_and_pd(fscal,cutoff_mask);
2085
2086             /* Update vectorial force */
2087             fix3             = _mm_macc_pd(dx32,fscal,fix3);
2088             fiy3             = _mm_macc_pd(dy32,fscal,fiy3);
2089             fiz3             = _mm_macc_pd(dz32,fscal,fiz3);
2090
2091             fjx2             = _mm_macc_pd(dx32,fscal,fjx2);
2092             fjy2             = _mm_macc_pd(dy32,fscal,fjy2);
2093             fjz2             = _mm_macc_pd(dz32,fscal,fjz2);
2094
2095             }
2096
2097             /**************************
2098              * CALCULATE INTERACTIONS *
2099              **************************/
2100
2101             if (gmx_mm_any_lt(rsq33,rcutoff2))
2102             {
2103
2104             r33              = _mm_mul_pd(rsq33,rinv33);
2105
2106             /* EWALD ELECTROSTATICS */
2107
2108             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2109             ewrt             = _mm_mul_pd(r33,ewtabscale);
2110             ewitab           = _mm_cvttpd_epi32(ewrt);
2111 #ifdef __XOP__
2112             eweps            = _mm_frcz_pd(ewrt);
2113 #else
2114             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2115 #endif
2116             twoeweps         = _mm_add_pd(eweps,eweps);
2117             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
2118                                          &ewtabF,&ewtabFn);
2119             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2120             felec            = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
2121
2122             cutoff_mask      = _mm_cmplt_pd(rsq33,rcutoff2);
2123
2124             fscal            = felec;
2125
2126             fscal            = _mm_and_pd(fscal,cutoff_mask);
2127
2128             /* Update vectorial force */
2129             fix3             = _mm_macc_pd(dx33,fscal,fix3);
2130             fiy3             = _mm_macc_pd(dy33,fscal,fiy3);
2131             fiz3             = _mm_macc_pd(dz33,fscal,fiz3);
2132
2133             fjx3             = _mm_macc_pd(dx33,fscal,fjx3);
2134             fjy3             = _mm_macc_pd(dy33,fscal,fjy3);
2135             fjz3             = _mm_macc_pd(dz33,fscal,fjz3);
2136
2137             }
2138
2139             gmx_mm_decrement_4rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2140
2141             /* Inner loop uses 414 flops */
2142         }
2143
2144         if(jidx<j_index_end)
2145         {
2146
2147             jnrA             = jjnr[jidx];
2148             j_coord_offsetA  = DIM*jnrA;
2149
2150             /* load j atom coordinates */
2151             gmx_mm_load_4rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
2152                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
2153                                               &jy2,&jz2,&jx3,&jy3,&jz3);
2154
2155             /* Calculate displacement vector */
2156             dx00             = _mm_sub_pd(ix0,jx0);
2157             dy00             = _mm_sub_pd(iy0,jy0);
2158             dz00             = _mm_sub_pd(iz0,jz0);
2159             dx11             = _mm_sub_pd(ix1,jx1);
2160             dy11             = _mm_sub_pd(iy1,jy1);
2161             dz11             = _mm_sub_pd(iz1,jz1);
2162             dx12             = _mm_sub_pd(ix1,jx2);
2163             dy12             = _mm_sub_pd(iy1,jy2);
2164             dz12             = _mm_sub_pd(iz1,jz2);
2165             dx13             = _mm_sub_pd(ix1,jx3);
2166             dy13             = _mm_sub_pd(iy1,jy3);
2167             dz13             = _mm_sub_pd(iz1,jz3);
2168             dx21             = _mm_sub_pd(ix2,jx1);
2169             dy21             = _mm_sub_pd(iy2,jy1);
2170             dz21             = _mm_sub_pd(iz2,jz1);
2171             dx22             = _mm_sub_pd(ix2,jx2);
2172             dy22             = _mm_sub_pd(iy2,jy2);
2173             dz22             = _mm_sub_pd(iz2,jz2);
2174             dx23             = _mm_sub_pd(ix2,jx3);
2175             dy23             = _mm_sub_pd(iy2,jy3);
2176             dz23             = _mm_sub_pd(iz2,jz3);
2177             dx31             = _mm_sub_pd(ix3,jx1);
2178             dy31             = _mm_sub_pd(iy3,jy1);
2179             dz31             = _mm_sub_pd(iz3,jz1);
2180             dx32             = _mm_sub_pd(ix3,jx2);
2181             dy32             = _mm_sub_pd(iy3,jy2);
2182             dz32             = _mm_sub_pd(iz3,jz2);
2183             dx33             = _mm_sub_pd(ix3,jx3);
2184             dy33             = _mm_sub_pd(iy3,jy3);
2185             dz33             = _mm_sub_pd(iz3,jz3);
2186
2187             /* Calculate squared distance and things based on it */
2188             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
2189             rsq11            = gmx_mm_calc_rsq_pd(dx11,dy11,dz11);
2190             rsq12            = gmx_mm_calc_rsq_pd(dx12,dy12,dz12);
2191             rsq13            = gmx_mm_calc_rsq_pd(dx13,dy13,dz13);
2192             rsq21            = gmx_mm_calc_rsq_pd(dx21,dy21,dz21);
2193             rsq22            = gmx_mm_calc_rsq_pd(dx22,dy22,dz22);
2194             rsq23            = gmx_mm_calc_rsq_pd(dx23,dy23,dz23);
2195             rsq31            = gmx_mm_calc_rsq_pd(dx31,dy31,dz31);
2196             rsq32            = gmx_mm_calc_rsq_pd(dx32,dy32,dz32);
2197             rsq33            = gmx_mm_calc_rsq_pd(dx33,dy33,dz33);
2198
2199             rinv11           = gmx_mm_invsqrt_pd(rsq11);
2200             rinv12           = gmx_mm_invsqrt_pd(rsq12);
2201             rinv13           = gmx_mm_invsqrt_pd(rsq13);
2202             rinv21           = gmx_mm_invsqrt_pd(rsq21);
2203             rinv22           = gmx_mm_invsqrt_pd(rsq22);
2204             rinv23           = gmx_mm_invsqrt_pd(rsq23);
2205             rinv31           = gmx_mm_invsqrt_pd(rsq31);
2206             rinv32           = gmx_mm_invsqrt_pd(rsq32);
2207             rinv33           = gmx_mm_invsqrt_pd(rsq33);
2208
2209             rinvsq00         = gmx_mm_inv_pd(rsq00);
2210             rinvsq11         = _mm_mul_pd(rinv11,rinv11);
2211             rinvsq12         = _mm_mul_pd(rinv12,rinv12);
2212             rinvsq13         = _mm_mul_pd(rinv13,rinv13);
2213             rinvsq21         = _mm_mul_pd(rinv21,rinv21);
2214             rinvsq22         = _mm_mul_pd(rinv22,rinv22);
2215             rinvsq23         = _mm_mul_pd(rinv23,rinv23);
2216             rinvsq31         = _mm_mul_pd(rinv31,rinv31);
2217             rinvsq32         = _mm_mul_pd(rinv32,rinv32);
2218             rinvsq33         = _mm_mul_pd(rinv33,rinv33);
2219
2220             fjx0             = _mm_setzero_pd();
2221             fjy0             = _mm_setzero_pd();
2222             fjz0             = _mm_setzero_pd();
2223             fjx1             = _mm_setzero_pd();
2224             fjy1             = _mm_setzero_pd();
2225             fjz1             = _mm_setzero_pd();
2226             fjx2             = _mm_setzero_pd();
2227             fjy2             = _mm_setzero_pd();
2228             fjz2             = _mm_setzero_pd();
2229             fjx3             = _mm_setzero_pd();
2230             fjy3             = _mm_setzero_pd();
2231             fjz3             = _mm_setzero_pd();
2232
2233             /**************************
2234              * CALCULATE INTERACTIONS *
2235              **************************/
2236
2237             if (gmx_mm_any_lt(rsq00,rcutoff2))
2238             {
2239
2240             /* LENNARD-JONES DISPERSION/REPULSION */
2241
2242             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
2243             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
2244
2245             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
2246
2247             fscal            = fvdw;
2248
2249             fscal            = _mm_and_pd(fscal,cutoff_mask);
2250
2251             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2252
2253             /* Update vectorial force */
2254             fix0             = _mm_macc_pd(dx00,fscal,fix0);
2255             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
2256             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
2257
2258             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
2259             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
2260             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
2261
2262             }
2263
2264             /**************************
2265              * CALCULATE INTERACTIONS *
2266              **************************/
2267
2268             if (gmx_mm_any_lt(rsq11,rcutoff2))
2269             {
2270
2271             r11              = _mm_mul_pd(rsq11,rinv11);
2272
2273             /* EWALD ELECTROSTATICS */
2274
2275             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2276             ewrt             = _mm_mul_pd(r11,ewtabscale);
2277             ewitab           = _mm_cvttpd_epi32(ewrt);
2278 #ifdef __XOP__
2279             eweps            = _mm_frcz_pd(ewrt);
2280 #else
2281             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2282 #endif
2283             twoeweps         = _mm_add_pd(eweps,eweps);
2284             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2285             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2286             felec            = _mm_mul_pd(_mm_mul_pd(qq11,rinv11),_mm_sub_pd(rinvsq11,felec));
2287
2288             cutoff_mask      = _mm_cmplt_pd(rsq11,rcutoff2);
2289
2290             fscal            = felec;
2291
2292             fscal            = _mm_and_pd(fscal,cutoff_mask);
2293
2294             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2295
2296             /* Update vectorial force */
2297             fix1             = _mm_macc_pd(dx11,fscal,fix1);
2298             fiy1             = _mm_macc_pd(dy11,fscal,fiy1);
2299             fiz1             = _mm_macc_pd(dz11,fscal,fiz1);
2300
2301             fjx1             = _mm_macc_pd(dx11,fscal,fjx1);
2302             fjy1             = _mm_macc_pd(dy11,fscal,fjy1);
2303             fjz1             = _mm_macc_pd(dz11,fscal,fjz1);
2304
2305             }
2306
2307             /**************************
2308              * CALCULATE INTERACTIONS *
2309              **************************/
2310
2311             if (gmx_mm_any_lt(rsq12,rcutoff2))
2312             {
2313
2314             r12              = _mm_mul_pd(rsq12,rinv12);
2315
2316             /* EWALD ELECTROSTATICS */
2317
2318             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2319             ewrt             = _mm_mul_pd(r12,ewtabscale);
2320             ewitab           = _mm_cvttpd_epi32(ewrt);
2321 #ifdef __XOP__
2322             eweps            = _mm_frcz_pd(ewrt);
2323 #else
2324             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2325 #endif
2326             twoeweps         = _mm_add_pd(eweps,eweps);
2327             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2328             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2329             felec            = _mm_mul_pd(_mm_mul_pd(qq12,rinv12),_mm_sub_pd(rinvsq12,felec));
2330
2331             cutoff_mask      = _mm_cmplt_pd(rsq12,rcutoff2);
2332
2333             fscal            = felec;
2334
2335             fscal            = _mm_and_pd(fscal,cutoff_mask);
2336
2337             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2338
2339             /* Update vectorial force */
2340             fix1             = _mm_macc_pd(dx12,fscal,fix1);
2341             fiy1             = _mm_macc_pd(dy12,fscal,fiy1);
2342             fiz1             = _mm_macc_pd(dz12,fscal,fiz1);
2343
2344             fjx2             = _mm_macc_pd(dx12,fscal,fjx2);
2345             fjy2             = _mm_macc_pd(dy12,fscal,fjy2);
2346             fjz2             = _mm_macc_pd(dz12,fscal,fjz2);
2347
2348             }
2349
2350             /**************************
2351              * CALCULATE INTERACTIONS *
2352              **************************/
2353
2354             if (gmx_mm_any_lt(rsq13,rcutoff2))
2355             {
2356
2357             r13              = _mm_mul_pd(rsq13,rinv13);
2358
2359             /* EWALD ELECTROSTATICS */
2360
2361             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2362             ewrt             = _mm_mul_pd(r13,ewtabscale);
2363             ewitab           = _mm_cvttpd_epi32(ewrt);
2364 #ifdef __XOP__
2365             eweps            = _mm_frcz_pd(ewrt);
2366 #else
2367             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2368 #endif
2369             twoeweps         = _mm_add_pd(eweps,eweps);
2370             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2371             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2372             felec            = _mm_mul_pd(_mm_mul_pd(qq13,rinv13),_mm_sub_pd(rinvsq13,felec));
2373
2374             cutoff_mask      = _mm_cmplt_pd(rsq13,rcutoff2);
2375
2376             fscal            = felec;
2377
2378             fscal            = _mm_and_pd(fscal,cutoff_mask);
2379
2380             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2381
2382             /* Update vectorial force */
2383             fix1             = _mm_macc_pd(dx13,fscal,fix1);
2384             fiy1             = _mm_macc_pd(dy13,fscal,fiy1);
2385             fiz1             = _mm_macc_pd(dz13,fscal,fiz1);
2386
2387             fjx3             = _mm_macc_pd(dx13,fscal,fjx3);
2388             fjy3             = _mm_macc_pd(dy13,fscal,fjy3);
2389             fjz3             = _mm_macc_pd(dz13,fscal,fjz3);
2390
2391             }
2392
2393             /**************************
2394              * CALCULATE INTERACTIONS *
2395              **************************/
2396
2397             if (gmx_mm_any_lt(rsq21,rcutoff2))
2398             {
2399
2400             r21              = _mm_mul_pd(rsq21,rinv21);
2401
2402             /* EWALD ELECTROSTATICS */
2403
2404             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2405             ewrt             = _mm_mul_pd(r21,ewtabscale);
2406             ewitab           = _mm_cvttpd_epi32(ewrt);
2407 #ifdef __XOP__
2408             eweps            = _mm_frcz_pd(ewrt);
2409 #else
2410             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2411 #endif
2412             twoeweps         = _mm_add_pd(eweps,eweps);
2413             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2414             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2415             felec            = _mm_mul_pd(_mm_mul_pd(qq21,rinv21),_mm_sub_pd(rinvsq21,felec));
2416
2417             cutoff_mask      = _mm_cmplt_pd(rsq21,rcutoff2);
2418
2419             fscal            = felec;
2420
2421             fscal            = _mm_and_pd(fscal,cutoff_mask);
2422
2423             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2424
2425             /* Update vectorial force */
2426             fix2             = _mm_macc_pd(dx21,fscal,fix2);
2427             fiy2             = _mm_macc_pd(dy21,fscal,fiy2);
2428             fiz2             = _mm_macc_pd(dz21,fscal,fiz2);
2429
2430             fjx1             = _mm_macc_pd(dx21,fscal,fjx1);
2431             fjy1             = _mm_macc_pd(dy21,fscal,fjy1);
2432             fjz1             = _mm_macc_pd(dz21,fscal,fjz1);
2433
2434             }
2435
2436             /**************************
2437              * CALCULATE INTERACTIONS *
2438              **************************/
2439
2440             if (gmx_mm_any_lt(rsq22,rcutoff2))
2441             {
2442
2443             r22              = _mm_mul_pd(rsq22,rinv22);
2444
2445             /* EWALD ELECTROSTATICS */
2446
2447             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2448             ewrt             = _mm_mul_pd(r22,ewtabscale);
2449             ewitab           = _mm_cvttpd_epi32(ewrt);
2450 #ifdef __XOP__
2451             eweps            = _mm_frcz_pd(ewrt);
2452 #else
2453             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2454 #endif
2455             twoeweps         = _mm_add_pd(eweps,eweps);
2456             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2457             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2458             felec            = _mm_mul_pd(_mm_mul_pd(qq22,rinv22),_mm_sub_pd(rinvsq22,felec));
2459
2460             cutoff_mask      = _mm_cmplt_pd(rsq22,rcutoff2);
2461
2462             fscal            = felec;
2463
2464             fscal            = _mm_and_pd(fscal,cutoff_mask);
2465
2466             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2467
2468             /* Update vectorial force */
2469             fix2             = _mm_macc_pd(dx22,fscal,fix2);
2470             fiy2             = _mm_macc_pd(dy22,fscal,fiy2);
2471             fiz2             = _mm_macc_pd(dz22,fscal,fiz2);
2472
2473             fjx2             = _mm_macc_pd(dx22,fscal,fjx2);
2474             fjy2             = _mm_macc_pd(dy22,fscal,fjy2);
2475             fjz2             = _mm_macc_pd(dz22,fscal,fjz2);
2476
2477             }
2478
2479             /**************************
2480              * CALCULATE INTERACTIONS *
2481              **************************/
2482
2483             if (gmx_mm_any_lt(rsq23,rcutoff2))
2484             {
2485
2486             r23              = _mm_mul_pd(rsq23,rinv23);
2487
2488             /* EWALD ELECTROSTATICS */
2489
2490             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2491             ewrt             = _mm_mul_pd(r23,ewtabscale);
2492             ewitab           = _mm_cvttpd_epi32(ewrt);
2493 #ifdef __XOP__
2494             eweps            = _mm_frcz_pd(ewrt);
2495 #else
2496             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2497 #endif
2498             twoeweps         = _mm_add_pd(eweps,eweps);
2499             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2500             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2501             felec            = _mm_mul_pd(_mm_mul_pd(qq23,rinv23),_mm_sub_pd(rinvsq23,felec));
2502
2503             cutoff_mask      = _mm_cmplt_pd(rsq23,rcutoff2);
2504
2505             fscal            = felec;
2506
2507             fscal            = _mm_and_pd(fscal,cutoff_mask);
2508
2509             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2510
2511             /* Update vectorial force */
2512             fix2             = _mm_macc_pd(dx23,fscal,fix2);
2513             fiy2             = _mm_macc_pd(dy23,fscal,fiy2);
2514             fiz2             = _mm_macc_pd(dz23,fscal,fiz2);
2515
2516             fjx3             = _mm_macc_pd(dx23,fscal,fjx3);
2517             fjy3             = _mm_macc_pd(dy23,fscal,fjy3);
2518             fjz3             = _mm_macc_pd(dz23,fscal,fjz3);
2519
2520             }
2521
2522             /**************************
2523              * CALCULATE INTERACTIONS *
2524              **************************/
2525
2526             if (gmx_mm_any_lt(rsq31,rcutoff2))
2527             {
2528
2529             r31              = _mm_mul_pd(rsq31,rinv31);
2530
2531             /* EWALD ELECTROSTATICS */
2532
2533             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2534             ewrt             = _mm_mul_pd(r31,ewtabscale);
2535             ewitab           = _mm_cvttpd_epi32(ewrt);
2536 #ifdef __XOP__
2537             eweps            = _mm_frcz_pd(ewrt);
2538 #else
2539             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2540 #endif
2541             twoeweps         = _mm_add_pd(eweps,eweps);
2542             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2543             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2544             felec            = _mm_mul_pd(_mm_mul_pd(qq31,rinv31),_mm_sub_pd(rinvsq31,felec));
2545
2546             cutoff_mask      = _mm_cmplt_pd(rsq31,rcutoff2);
2547
2548             fscal            = felec;
2549
2550             fscal            = _mm_and_pd(fscal,cutoff_mask);
2551
2552             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2553
2554             /* Update vectorial force */
2555             fix3             = _mm_macc_pd(dx31,fscal,fix3);
2556             fiy3             = _mm_macc_pd(dy31,fscal,fiy3);
2557             fiz3             = _mm_macc_pd(dz31,fscal,fiz3);
2558
2559             fjx1             = _mm_macc_pd(dx31,fscal,fjx1);
2560             fjy1             = _mm_macc_pd(dy31,fscal,fjy1);
2561             fjz1             = _mm_macc_pd(dz31,fscal,fjz1);
2562
2563             }
2564
2565             /**************************
2566              * CALCULATE INTERACTIONS *
2567              **************************/
2568
2569             if (gmx_mm_any_lt(rsq32,rcutoff2))
2570             {
2571
2572             r32              = _mm_mul_pd(rsq32,rinv32);
2573
2574             /* EWALD ELECTROSTATICS */
2575
2576             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2577             ewrt             = _mm_mul_pd(r32,ewtabscale);
2578             ewitab           = _mm_cvttpd_epi32(ewrt);
2579 #ifdef __XOP__
2580             eweps            = _mm_frcz_pd(ewrt);
2581 #else
2582             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2583 #endif
2584             twoeweps         = _mm_add_pd(eweps,eweps);
2585             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2586             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2587             felec            = _mm_mul_pd(_mm_mul_pd(qq32,rinv32),_mm_sub_pd(rinvsq32,felec));
2588
2589             cutoff_mask      = _mm_cmplt_pd(rsq32,rcutoff2);
2590
2591             fscal            = felec;
2592
2593             fscal            = _mm_and_pd(fscal,cutoff_mask);
2594
2595             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2596
2597             /* Update vectorial force */
2598             fix3             = _mm_macc_pd(dx32,fscal,fix3);
2599             fiy3             = _mm_macc_pd(dy32,fscal,fiy3);
2600             fiz3             = _mm_macc_pd(dz32,fscal,fiz3);
2601
2602             fjx2             = _mm_macc_pd(dx32,fscal,fjx2);
2603             fjy2             = _mm_macc_pd(dy32,fscal,fjy2);
2604             fjz2             = _mm_macc_pd(dz32,fscal,fjz2);
2605
2606             }
2607
2608             /**************************
2609              * CALCULATE INTERACTIONS *
2610              **************************/
2611
2612             if (gmx_mm_any_lt(rsq33,rcutoff2))
2613             {
2614
2615             r33              = _mm_mul_pd(rsq33,rinv33);
2616
2617             /* EWALD ELECTROSTATICS */
2618
2619             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2620             ewrt             = _mm_mul_pd(r33,ewtabscale);
2621             ewitab           = _mm_cvttpd_epi32(ewrt);
2622 #ifdef __XOP__
2623             eweps            = _mm_frcz_pd(ewrt);
2624 #else
2625             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
2626 #endif
2627             twoeweps         = _mm_add_pd(eweps,eweps);
2628             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
2629             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
2630             felec            = _mm_mul_pd(_mm_mul_pd(qq33,rinv33),_mm_sub_pd(rinvsq33,felec));
2631
2632             cutoff_mask      = _mm_cmplt_pd(rsq33,rcutoff2);
2633
2634             fscal            = felec;
2635
2636             fscal            = _mm_and_pd(fscal,cutoff_mask);
2637
2638             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
2639
2640             /* Update vectorial force */
2641             fix3             = _mm_macc_pd(dx33,fscal,fix3);
2642             fiy3             = _mm_macc_pd(dy33,fscal,fiy3);
2643             fiz3             = _mm_macc_pd(dz33,fscal,fiz3);
2644
2645             fjx3             = _mm_macc_pd(dx33,fscal,fjx3);
2646             fjy3             = _mm_macc_pd(dy33,fscal,fjy3);
2647             fjz3             = _mm_macc_pd(dz33,fscal,fjz3);
2648
2649             }
2650
2651             gmx_mm_decrement_4rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2652
2653             /* Inner loop uses 414 flops */
2654         }
2655
2656         /* End of innermost loop */
2657
2658         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2659                                               f+i_coord_offset,fshift+i_shift_offset);
2660
2661         /* Increment number of inner iterations */
2662         inneriter                  += j_index_end - j_index_start;
2663
2664         /* Outer loop uses 24 flops */
2665     }
2666
2667     /* Increment number of outer iterations */
2668     outeriter        += nri;
2669
2670     /* Update outer/inner flops */
2671
2672     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*414);
2673 }