src/gromacs/gmxlib/nonbonded/nb_kernel_sse2_single/nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4W4_sse2_single.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2017, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS sse2_single kernel generator.
  37  */
  38 #include "gmxpre.h"
  39
  40 #include "config.h"
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "gromacs/gmxlib/nrnb.h"
  46
  47 #include "kernelutil_x86_sse2_single.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4W4_VF_sse2_single
  51  * Electrostatics interaction: Ewald
  52  * VdW interaction:            LJEwald
  53  * Geometry:                   Water4-Water4
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4W4_VF_sse2_single
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      struct t_forcerec           * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  67      * just 0 for non-waters.
  68      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
  69      * jnr indices corresponding to data put in the four positions in the SIMD register.
  70      */
  71     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  72     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  73     int              jnrA,jnrB,jnrC,jnrD;
  74     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  75     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  76     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  77     real             rcutoff_scalar;
  78     real             *shiftvec,*fshift,*x,*f;
  79     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  80     real             scratch[4*DIM];
  81     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  82     int              vdwioffset0;
  83     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  84     int              vdwioffset1;
  85     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  86     int              vdwioffset2;
  87     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  88     int              vdwioffset3;
  89     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  90     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  91     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  92     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  93     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  94     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  95     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  96     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
  97     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
  98     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  99     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 100     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 101     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
 102     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 103     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 104     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
 105     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
 106     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
 107     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
 108     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
 109     real             *charge;
 110     int              nvdwtype;
 111     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 112     int              *vdwtype;
 113     real             *vdwparam;
 114     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
 115     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
 116     __m128           c6grid_00;
 117     __m128           c6grid_11;
 118     __m128           c6grid_12;
 119     __m128           c6grid_13;
 120     __m128           c6grid_21;
 121     __m128           c6grid_22;
 122     __m128           c6grid_23;
 123     __m128           c6grid_31;
 124     __m128           c6grid_32;
 125     __m128           c6grid_33;
 126     __m128           ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,f6A,f6B,sh_lj_ewald;
 127     real             *vdwgridparam;
 128     __m128           one_half = _mm_set1_ps(0.5);
 129     __m128           minus_one = _mm_set1_ps(-1.0);
 130     __m128i          ewitab;
 131     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 132     real             *ewtab;
 133     __m128           dummy_mask,cutoff_mask;
 134     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 135     __m128           one     = _mm_set1_ps(1.0);
 136     __m128           two     = _mm_set1_ps(2.0);
 137     x                = xx[0];
 138     f                = ff[0];
 139
 140     nri              = nlist->nri;
 141     iinr             = nlist->iinr;
 142     jindex           = nlist->jindex;
 143     jjnr             = nlist->jjnr;
 144     shiftidx         = nlist->shift;
 145     gid              = nlist->gid;
 146     shiftvec         = fr->shift_vec[0];
 147     fshift           = fr->fshift[0];
 148     facel            = _mm_set1_ps(fr->ic->epsfac);
 149     charge           = mdatoms->chargeA;
 150     nvdwtype         = fr->ntype;
 151     vdwparam         = fr->nbfp;
 152     vdwtype          = mdatoms->typeA;
 153     vdwgridparam     = fr->ljpme_c6grid;
 154     sh_lj_ewald      = _mm_set1_ps(fr->ic->sh_lj_ewald);
 155     ewclj            = _mm_set1_ps(fr->ic->ewaldcoeff_lj);
 156     ewclj2           = _mm_mul_ps(minus_one,_mm_mul_ps(ewclj,ewclj));
 157
 158     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 159     ewtab            = fr->ic->tabq_coul_FDV0;
 160     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 161     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 162
 163     /* Setup water-specific parameters */
 164     inr              = nlist->iinr[0];
 165     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 166     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 167     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
 168     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 169
 170     jq1              = _mm_set1_ps(charge[inr+1]);
 171     jq2              = _mm_set1_ps(charge[inr+2]);
 172     jq3              = _mm_set1_ps(charge[inr+3]);
 173     vdwjidx0A        = 2*vdwtype[inr+0];
 174     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
 175     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
 176     c6grid_00        = _mm_set1_ps(vdwgridparam[vdwioffset0+vdwjidx0A]);
 177     qq11             = _mm_mul_ps(iq1,jq1);
 178     qq12             = _mm_mul_ps(iq1,jq2);
 179     qq13             = _mm_mul_ps(iq1,jq3);
 180     qq21             = _mm_mul_ps(iq2,jq1);
 181     qq22             = _mm_mul_ps(iq2,jq2);
 182     qq23             = _mm_mul_ps(iq2,jq3);
 183     qq31             = _mm_mul_ps(iq3,jq1);
 184     qq32             = _mm_mul_ps(iq3,jq2);
 185     qq33             = _mm_mul_ps(iq3,jq3);
 186
 187     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 188     rcutoff_scalar   = fr->ic->rcoulomb;
 189     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 190     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 191
 192     sh_vdw_invrcut6  = _mm_set1_ps(fr->ic->sh_invrc6);
 193     rvdw             = _mm_set1_ps(fr->ic->rvdw);
 194
 195     /* Avoid stupid compiler warnings */
 196     jnrA = jnrB = jnrC = jnrD = 0;
 197     j_coord_offsetA = 0;
 198     j_coord_offsetB = 0;
 199     j_coord_offsetC = 0;
 200     j_coord_offsetD = 0;
 201
 202     outeriter        = 0;
 203     inneriter        = 0;
 204
 205     for(iidx=0;iidx<4*DIM;iidx++)
 206     {
 207         scratch[iidx] = 0.0;
 208     }
 209
 210     /* Start outer loop over neighborlists */
 211     for(iidx=0; iidx<nri; iidx++)
 212     {
 213         /* Load shift vector for this list */
 214         i_shift_offset   = DIM*shiftidx[iidx];
 215
 216         /* Load limits for loop over neighbors */
 217         j_index_start    = jindex[iidx];
 218         j_index_end      = jindex[iidx+1];
 219
 220         /* Get outer coordinate index */
 221         inr              = iinr[iidx];
 222         i_coord_offset   = DIM*inr;
 223
 224         /* Load i particle coords and add shift vector */
 225         gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
 226                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 227
 228         fix0             = _mm_setzero_ps();
 229         fiy0             = _mm_setzero_ps();
 230         fiz0             = _mm_setzero_ps();
 231         fix1             = _mm_setzero_ps();
 232         fiy1             = _mm_setzero_ps();
 233         fiz1             = _mm_setzero_ps();
 234         fix2             = _mm_setzero_ps();
 235         fiy2             = _mm_setzero_ps();
 236         fiz2             = _mm_setzero_ps();
 237         fix3             = _mm_setzero_ps();
 238         fiy3             = _mm_setzero_ps();
 239         fiz3             = _mm_setzero_ps();
 240
 241         /* Reset potential sums */
 242         velecsum         = _mm_setzero_ps();
 243         vvdwsum          = _mm_setzero_ps();
 244
 245         /* Start inner kernel loop */
 246         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 247         {
 248
 249             /* Get j neighbor index, and coordinate index */
 250             jnrA             = jjnr[jidx];
 251             jnrB             = jjnr[jidx+1];
 252             jnrC             = jjnr[jidx+2];
 253             jnrD             = jjnr[jidx+3];
 254             j_coord_offsetA  = DIM*jnrA;
 255             j_coord_offsetB  = DIM*jnrB;
 256             j_coord_offsetC  = DIM*jnrC;
 257             j_coord_offsetD  = DIM*jnrD;
 258
 259             /* load j atom coordinates */
 260             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 261                                               x+j_coord_offsetC,x+j_coord_offsetD,
 262                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 263                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 264
 265             /* Calculate displacement vector */
 266             dx00             = _mm_sub_ps(ix0,jx0);
 267             dy00             = _mm_sub_ps(iy0,jy0);
 268             dz00             = _mm_sub_ps(iz0,jz0);
 269             dx11             = _mm_sub_ps(ix1,jx1);
 270             dy11             = _mm_sub_ps(iy1,jy1);
 271             dz11             = _mm_sub_ps(iz1,jz1);
 272             dx12             = _mm_sub_ps(ix1,jx2);
 273             dy12             = _mm_sub_ps(iy1,jy2);
 274             dz12             = _mm_sub_ps(iz1,jz2);
 275             dx13             = _mm_sub_ps(ix1,jx3);
 276             dy13             = _mm_sub_ps(iy1,jy3);
 277             dz13             = _mm_sub_ps(iz1,jz3);
 278             dx21             = _mm_sub_ps(ix2,jx1);
 279             dy21             = _mm_sub_ps(iy2,jy1);
 280             dz21             = _mm_sub_ps(iz2,jz1);
 281             dx22             = _mm_sub_ps(ix2,jx2);
 282             dy22             = _mm_sub_ps(iy2,jy2);
 283             dz22             = _mm_sub_ps(iz2,jz2);
 284             dx23             = _mm_sub_ps(ix2,jx3);
 285             dy23             = _mm_sub_ps(iy2,jy3);
 286             dz23             = _mm_sub_ps(iz2,jz3);
 287             dx31             = _mm_sub_ps(ix3,jx1);
 288             dy31             = _mm_sub_ps(iy3,jy1);
 289             dz31             = _mm_sub_ps(iz3,jz1);
 290             dx32             = _mm_sub_ps(ix3,jx2);
 291             dy32             = _mm_sub_ps(iy3,jy2);
 292             dz32             = _mm_sub_ps(iz3,jz2);
 293             dx33             = _mm_sub_ps(ix3,jx3);
 294             dy33             = _mm_sub_ps(iy3,jy3);
 295             dz33             = _mm_sub_ps(iz3,jz3);
 296
 297             /* Calculate squared distance and things based on it */
 298             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 299             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 300             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 301             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 302             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 303             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 304             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 305             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 306             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 307             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 308
 309             rinv00           = sse2_invsqrt_f(rsq00);
 310             rinv11           = sse2_invsqrt_f(rsq11);
 311             rinv12           = sse2_invsqrt_f(rsq12);
 312             rinv13           = sse2_invsqrt_f(rsq13);
 313             rinv21           = sse2_invsqrt_f(rsq21);
 314             rinv22           = sse2_invsqrt_f(rsq22);
 315             rinv23           = sse2_invsqrt_f(rsq23);
 316             rinv31           = sse2_invsqrt_f(rsq31);
 317             rinv32           = sse2_invsqrt_f(rsq32);
 318             rinv33           = sse2_invsqrt_f(rsq33);
 319
 320             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 321             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
 322             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
 323             rinvsq13         = _mm_mul_ps(rinv13,rinv13);
 324             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
 325             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
 326             rinvsq23         = _mm_mul_ps(rinv23,rinv23);
 327             rinvsq31         = _mm_mul_ps(rinv31,rinv31);
 328             rinvsq32         = _mm_mul_ps(rinv32,rinv32);
 329             rinvsq33         = _mm_mul_ps(rinv33,rinv33);
 330
 331             fjx0             = _mm_setzero_ps();
 332             fjy0             = _mm_setzero_ps();
 333             fjz0             = _mm_setzero_ps();
 334             fjx1             = _mm_setzero_ps();
 335             fjy1             = _mm_setzero_ps();
 336             fjz1             = _mm_setzero_ps();
 337             fjx2             = _mm_setzero_ps();
 338             fjy2             = _mm_setzero_ps();
 339             fjz2             = _mm_setzero_ps();
 340             fjx3             = _mm_setzero_ps();
 341             fjy3             = _mm_setzero_ps();
 342             fjz3             = _mm_setzero_ps();
 343
 344             /**************************
 345              * CALCULATE INTERACTIONS *
 346              **************************/
 347
 348             if (gmx_mm_any_lt(rsq00,rcutoff2))
 349             {
 350
 351             r00              = _mm_mul_ps(rsq00,rinv00);
 352
 353             /* Analytical LJ-PME */
 354             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 355             ewcljrsq         = _mm_mul_ps(ewclj2,rsq00);
 356             ewclj6           = _mm_mul_ps(ewclj2,_mm_mul_ps(ewclj2,ewclj2));
 357             exponent         = sse2_exp_f(ewcljrsq);
 358             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
 359             poly             = _mm_mul_ps(exponent,_mm_add_ps(_mm_sub_ps(one,ewcljrsq),_mm_mul_ps(_mm_mul_ps(ewcljrsq,ewcljrsq),one_half)));
 360             /* vvdw6 = [C6 - C6grid * (1-poly)]/r6 */
 361             vvdw6            = _mm_mul_ps(_mm_sub_ps(c6_00,_mm_mul_ps(c6grid_00,_mm_sub_ps(one,poly))),rinvsix);
 362             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 363             vvdw             = _mm_sub_ps(_mm_mul_ps( _mm_sub_ps(vvdw12 , _mm_mul_ps(c12_00,_mm_mul_ps(sh_vdw_invrcut6,sh_vdw_invrcut6))), one_twelfth) ,
 364                                           _mm_mul_ps( _mm_sub_ps(vvdw6,_mm_add_ps(_mm_mul_ps(c6_00,sh_vdw_invrcut6),_mm_mul_ps(c6grid_00,sh_lj_ewald))),one_sixth));
 365             /* fvdw = vvdw12/r - (vvdw6/r + (C6grid * exponent * beta^6)/r) */
 366             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,_mm_sub_ps(vvdw6,_mm_mul_ps(_mm_mul_ps(c6grid_00,one_sixth),_mm_mul_ps(exponent,ewclj6)))),rinvsq00);
 367
 368             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
 369
 370             /* Update potential sum for this i atom from the interaction with this j atom. */
 371             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
 372             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 373
 374             fscal            = fvdw;
 375
 376             fscal            = _mm_and_ps(fscal,cutoff_mask);
 377
 378             /* Calculate temporary vectorial force */
 379             tx               = _mm_mul_ps(fscal,dx00);
 380             ty               = _mm_mul_ps(fscal,dy00);
 381             tz               = _mm_mul_ps(fscal,dz00);
 382
 383             /* Update vectorial force */
 384             fix0             = _mm_add_ps(fix0,tx);
 385             fiy0             = _mm_add_ps(fiy0,ty);
 386             fiz0             = _mm_add_ps(fiz0,tz);
 387
 388             fjx0             = _mm_add_ps(fjx0,tx);
 389             fjy0             = _mm_add_ps(fjy0,ty);
 390             fjz0             = _mm_add_ps(fjz0,tz);
 391
 392             }
 393
 394             /**************************
 395              * CALCULATE INTERACTIONS *
 396              **************************/
 397
 398             if (gmx_mm_any_lt(rsq11,rcutoff2))
 399             {
 400
 401             r11              = _mm_mul_ps(rsq11,rinv11);
 402
 403             /* EWALD ELECTROSTATICS */
 404
 405             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 406             ewrt             = _mm_mul_ps(r11,ewtabscale);
 407             ewitab           = _mm_cvttps_epi32(ewrt);
 408             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 409             ewitab           = _mm_slli_epi32(ewitab,2);
 410             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 411             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 412             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 413             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 414             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 415             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 416             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 417             velec            = _mm_mul_ps(qq11,_mm_sub_ps(_mm_sub_ps(rinv11,sh_ewald),velec));
 418             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
 419
 420             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
 421
 422             /* Update potential sum for this i atom from the interaction with this j atom. */
 423             velec            = _mm_and_ps(velec,cutoff_mask);
 424             velecsum         = _mm_add_ps(velecsum,velec);
 425
 426             fscal            = felec;
 427
 428             fscal            = _mm_and_ps(fscal,cutoff_mask);
 429
 430             /* Calculate temporary vectorial force */
 431             tx               = _mm_mul_ps(fscal,dx11);
 432             ty               = _mm_mul_ps(fscal,dy11);
 433             tz               = _mm_mul_ps(fscal,dz11);
 434
 435             /* Update vectorial force */
 436             fix1             = _mm_add_ps(fix1,tx);
 437             fiy1             = _mm_add_ps(fiy1,ty);
 438             fiz1             = _mm_add_ps(fiz1,tz);
 439
 440             fjx1             = _mm_add_ps(fjx1,tx);
 441             fjy1             = _mm_add_ps(fjy1,ty);
 442             fjz1             = _mm_add_ps(fjz1,tz);
 443
 444             }
 445
 446             /**************************
 447              * CALCULATE INTERACTIONS *
 448              **************************/
 449
 450             if (gmx_mm_any_lt(rsq12,rcutoff2))
 451             {
 452
 453             r12              = _mm_mul_ps(rsq12,rinv12);
 454
 455             /* EWALD ELECTROSTATICS */
 456
 457             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 458             ewrt             = _mm_mul_ps(r12,ewtabscale);
 459             ewitab           = _mm_cvttps_epi32(ewrt);
 460             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 461             ewitab           = _mm_slli_epi32(ewitab,2);
 462             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 463             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 464             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 465             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 466             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 467             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 468             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 469             velec            = _mm_mul_ps(qq12,_mm_sub_ps(_mm_sub_ps(rinv12,sh_ewald),velec));
 470             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
 471
 472             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
 473
 474             /* Update potential sum for this i atom from the interaction with this j atom. */
 475             velec            = _mm_and_ps(velec,cutoff_mask);
 476             velecsum         = _mm_add_ps(velecsum,velec);
 477
 478             fscal            = felec;
 479
 480             fscal            = _mm_and_ps(fscal,cutoff_mask);
 481
 482             /* Calculate temporary vectorial force */
 483             tx               = _mm_mul_ps(fscal,dx12);
 484             ty               = _mm_mul_ps(fscal,dy12);
 485             tz               = _mm_mul_ps(fscal,dz12);
 486
 487             /* Update vectorial force */
 488             fix1             = _mm_add_ps(fix1,tx);
 489             fiy1             = _mm_add_ps(fiy1,ty);
 490             fiz1             = _mm_add_ps(fiz1,tz);
 491
 492             fjx2             = _mm_add_ps(fjx2,tx);
 493             fjy2             = _mm_add_ps(fjy2,ty);
 494             fjz2             = _mm_add_ps(fjz2,tz);
 495
 496             }
 497
 498             /**************************
 499              * CALCULATE INTERACTIONS *
 500              **************************/
 501
 502             if (gmx_mm_any_lt(rsq13,rcutoff2))
 503             {
 504
 505             r13              = _mm_mul_ps(rsq13,rinv13);
 506
 507             /* EWALD ELECTROSTATICS */
 508
 509             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 510             ewrt             = _mm_mul_ps(r13,ewtabscale);
 511             ewitab           = _mm_cvttps_epi32(ewrt);
 512             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 513             ewitab           = _mm_slli_epi32(ewitab,2);
 514             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 515             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 516             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 517             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 518             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 519             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 520             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 521             velec            = _mm_mul_ps(qq13,_mm_sub_ps(_mm_sub_ps(rinv13,sh_ewald),velec));
 522             felec            = _mm_mul_ps(_mm_mul_ps(qq13,rinv13),_mm_sub_ps(rinvsq13,felec));
 523
 524             cutoff_mask      = _mm_cmplt_ps(rsq13,rcutoff2);
 525
 526             /* Update potential sum for this i atom from the interaction with this j atom. */
 527             velec            = _mm_and_ps(velec,cutoff_mask);
 528             velecsum         = _mm_add_ps(velecsum,velec);
 529
 530             fscal            = felec;
 531
 532             fscal            = _mm_and_ps(fscal,cutoff_mask);
 533
 534             /* Calculate temporary vectorial force */
 535             tx               = _mm_mul_ps(fscal,dx13);
 536             ty               = _mm_mul_ps(fscal,dy13);
 537             tz               = _mm_mul_ps(fscal,dz13);
 538
 539             /* Update vectorial force */
 540             fix1             = _mm_add_ps(fix1,tx);
 541             fiy1             = _mm_add_ps(fiy1,ty);
 542             fiz1             = _mm_add_ps(fiz1,tz);
 543
 544             fjx3             = _mm_add_ps(fjx3,tx);
 545             fjy3             = _mm_add_ps(fjy3,ty);
 546             fjz3             = _mm_add_ps(fjz3,tz);
 547
 548             }
 549
 550             /**************************
 551              * CALCULATE INTERACTIONS *
 552              **************************/
 553
 554             if (gmx_mm_any_lt(rsq21,rcutoff2))
 555             {
 556
 557             r21              = _mm_mul_ps(rsq21,rinv21);
 558
 559             /* EWALD ELECTROSTATICS */
 560
 561             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 562             ewrt             = _mm_mul_ps(r21,ewtabscale);
 563             ewitab           = _mm_cvttps_epi32(ewrt);
 564             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 565             ewitab           = _mm_slli_epi32(ewitab,2);
 566             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 567             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 568             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 569             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 570             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 571             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 572             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 573             velec            = _mm_mul_ps(qq21,_mm_sub_ps(_mm_sub_ps(rinv21,sh_ewald),velec));
 574             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
 575
 576             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
 577
 578             /* Update potential sum for this i atom from the interaction with this j atom. */
 579             velec            = _mm_and_ps(velec,cutoff_mask);
 580             velecsum         = _mm_add_ps(velecsum,velec);
 581
 582             fscal            = felec;
 583
 584             fscal            = _mm_and_ps(fscal,cutoff_mask);
 585
 586             /* Calculate temporary vectorial force */
 587             tx               = _mm_mul_ps(fscal,dx21);
 588             ty               = _mm_mul_ps(fscal,dy21);
 589             tz               = _mm_mul_ps(fscal,dz21);
 590
 591             /* Update vectorial force */
 592             fix2             = _mm_add_ps(fix2,tx);
 593             fiy2             = _mm_add_ps(fiy2,ty);
 594             fiz2             = _mm_add_ps(fiz2,tz);
 595
 596             fjx1             = _mm_add_ps(fjx1,tx);
 597             fjy1             = _mm_add_ps(fjy1,ty);
 598             fjz1             = _mm_add_ps(fjz1,tz);
 599
 600             }
 601
 602             /**************************
 603              * CALCULATE INTERACTIONS *
 604              **************************/
 605
 606             if (gmx_mm_any_lt(rsq22,rcutoff2))
 607             {
 608
 609             r22              = _mm_mul_ps(rsq22,rinv22);
 610
 611             /* EWALD ELECTROSTATICS */
 612
 613             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 614             ewrt             = _mm_mul_ps(r22,ewtabscale);
 615             ewitab           = _mm_cvttps_epi32(ewrt);
 616             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 617             ewitab           = _mm_slli_epi32(ewitab,2);
 618             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 619             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 620             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 621             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 622             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 623             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 624             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 625             velec            = _mm_mul_ps(qq22,_mm_sub_ps(_mm_sub_ps(rinv22,sh_ewald),velec));
 626             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
 627
 628             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
 629
 630             /* Update potential sum for this i atom from the interaction with this j atom. */
 631             velec            = _mm_and_ps(velec,cutoff_mask);
 632             velecsum         = _mm_add_ps(velecsum,velec);
 633
 634             fscal            = felec;
 635
 636             fscal            = _mm_and_ps(fscal,cutoff_mask);
 637
 638             /* Calculate temporary vectorial force */
 639             tx               = _mm_mul_ps(fscal,dx22);
 640             ty               = _mm_mul_ps(fscal,dy22);
 641             tz               = _mm_mul_ps(fscal,dz22);
 642
 643             /* Update vectorial force */
 644             fix2             = _mm_add_ps(fix2,tx);
 645             fiy2             = _mm_add_ps(fiy2,ty);
 646             fiz2             = _mm_add_ps(fiz2,tz);
 647
 648             fjx2             = _mm_add_ps(fjx2,tx);
 649             fjy2             = _mm_add_ps(fjy2,ty);
 650             fjz2             = _mm_add_ps(fjz2,tz);
 651
 652             }
 653
 654             /**************************
 655              * CALCULATE INTERACTIONS *
 656              **************************/
 657
 658             if (gmx_mm_any_lt(rsq23,rcutoff2))
 659             {
 660
 661             r23              = _mm_mul_ps(rsq23,rinv23);
 662
 663             /* EWALD ELECTROSTATICS */
 664
 665             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 666             ewrt             = _mm_mul_ps(r23,ewtabscale);
 667             ewitab           = _mm_cvttps_epi32(ewrt);
 668             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 669             ewitab           = _mm_slli_epi32(ewitab,2);
 670             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 671             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 672             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 673             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 674             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 675             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 676             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 677             velec            = _mm_mul_ps(qq23,_mm_sub_ps(_mm_sub_ps(rinv23,sh_ewald),velec));
 678             felec            = _mm_mul_ps(_mm_mul_ps(qq23,rinv23),_mm_sub_ps(rinvsq23,felec));
 679
 680             cutoff_mask      = _mm_cmplt_ps(rsq23,rcutoff2);
 681
 682             /* Update potential sum for this i atom from the interaction with this j atom. */
 683             velec            = _mm_and_ps(velec,cutoff_mask);
 684             velecsum         = _mm_add_ps(velecsum,velec);
 685
 686             fscal            = felec;
 687
 688             fscal            = _mm_and_ps(fscal,cutoff_mask);
 689
 690             /* Calculate temporary vectorial force */
 691             tx               = _mm_mul_ps(fscal,dx23);
 692             ty               = _mm_mul_ps(fscal,dy23);
 693             tz               = _mm_mul_ps(fscal,dz23);
 694
 695             /* Update vectorial force */
 696             fix2             = _mm_add_ps(fix2,tx);
 697             fiy2             = _mm_add_ps(fiy2,ty);
 698             fiz2             = _mm_add_ps(fiz2,tz);
 699
 700             fjx3             = _mm_add_ps(fjx3,tx);
 701             fjy3             = _mm_add_ps(fjy3,ty);
 702             fjz3             = _mm_add_ps(fjz3,tz);
 703
 704             }
 705
 706             /**************************
 707              * CALCULATE INTERACTIONS *
 708              **************************/
 709
 710             if (gmx_mm_any_lt(rsq31,rcutoff2))
 711             {
 712
 713             r31              = _mm_mul_ps(rsq31,rinv31);
 714
 715             /* EWALD ELECTROSTATICS */
 716
 717             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 718             ewrt             = _mm_mul_ps(r31,ewtabscale);
 719             ewitab           = _mm_cvttps_epi32(ewrt);
 720             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 721             ewitab           = _mm_slli_epi32(ewitab,2);
 722             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 723             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 724             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 725             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 726             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 727             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 728             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 729             velec            = _mm_mul_ps(qq31,_mm_sub_ps(_mm_sub_ps(rinv31,sh_ewald),velec));
 730             felec            = _mm_mul_ps(_mm_mul_ps(qq31,rinv31),_mm_sub_ps(rinvsq31,felec));
 731
 732             cutoff_mask      = _mm_cmplt_ps(rsq31,rcutoff2);
 733
 734             /* Update potential sum for this i atom from the interaction with this j atom. */
 735             velec            = _mm_and_ps(velec,cutoff_mask);
 736             velecsum         = _mm_add_ps(velecsum,velec);
 737
 738             fscal            = felec;
 739
 740             fscal            = _mm_and_ps(fscal,cutoff_mask);
 741
 742             /* Calculate temporary vectorial force */
 743             tx               = _mm_mul_ps(fscal,dx31);
 744             ty               = _mm_mul_ps(fscal,dy31);
 745             tz               = _mm_mul_ps(fscal,dz31);
 746
 747             /* Update vectorial force */
 748             fix3             = _mm_add_ps(fix3,tx);
 749             fiy3             = _mm_add_ps(fiy3,ty);
 750             fiz3             = _mm_add_ps(fiz3,tz);
 751
 752             fjx1             = _mm_add_ps(fjx1,tx);
 753             fjy1             = _mm_add_ps(fjy1,ty);
 754             fjz1             = _mm_add_ps(fjz1,tz);
 755
 756             }
 757
 758             /**************************
 759              * CALCULATE INTERACTIONS *
 760              **************************/
 761
 762             if (gmx_mm_any_lt(rsq32,rcutoff2))
 763             {
 764
 765             r32              = _mm_mul_ps(rsq32,rinv32);
 766
 767             /* EWALD ELECTROSTATICS */
 768
 769             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 770             ewrt             = _mm_mul_ps(r32,ewtabscale);
 771             ewitab           = _mm_cvttps_epi32(ewrt);
 772             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 773             ewitab           = _mm_slli_epi32(ewitab,2);
 774             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 775             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 776             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 777             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 778             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 779             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 780             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 781             velec            = _mm_mul_ps(qq32,_mm_sub_ps(_mm_sub_ps(rinv32,sh_ewald),velec));
 782             felec            = _mm_mul_ps(_mm_mul_ps(qq32,rinv32),_mm_sub_ps(rinvsq32,felec));
 783
 784             cutoff_mask      = _mm_cmplt_ps(rsq32,rcutoff2);
 785
 786             /* Update potential sum for this i atom from the interaction with this j atom. */
 787             velec            = _mm_and_ps(velec,cutoff_mask);
 788             velecsum         = _mm_add_ps(velecsum,velec);
 789
 790             fscal            = felec;
 791
 792             fscal            = _mm_and_ps(fscal,cutoff_mask);
 793
 794             /* Calculate temporary vectorial force */
 795             tx               = _mm_mul_ps(fscal,dx32);
 796             ty               = _mm_mul_ps(fscal,dy32);
 797             tz               = _mm_mul_ps(fscal,dz32);
 798
 799             /* Update vectorial force */
 800             fix3             = _mm_add_ps(fix3,tx);
 801             fiy3             = _mm_add_ps(fiy3,ty);
 802             fiz3             = _mm_add_ps(fiz3,tz);
 803
 804             fjx2             = _mm_add_ps(fjx2,tx);
 805             fjy2             = _mm_add_ps(fjy2,ty);
 806             fjz2             = _mm_add_ps(fjz2,tz);
 807
 808             }
 809
 810             /**************************
 811              * CALCULATE INTERACTIONS *
 812              **************************/
 813
 814             if (gmx_mm_any_lt(rsq33,rcutoff2))
 815             {
 816
 817             r33              = _mm_mul_ps(rsq33,rinv33);
 818
 819             /* EWALD ELECTROSTATICS */
 820
 821             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 822             ewrt             = _mm_mul_ps(r33,ewtabscale);
 823             ewitab           = _mm_cvttps_epi32(ewrt);
 824             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 825             ewitab           = _mm_slli_epi32(ewitab,2);
 826             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 827             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 828             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 829             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 830             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 831             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 832             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 833             velec            = _mm_mul_ps(qq33,_mm_sub_ps(_mm_sub_ps(rinv33,sh_ewald),velec));
 834             felec            = _mm_mul_ps(_mm_mul_ps(qq33,rinv33),_mm_sub_ps(rinvsq33,felec));
 835
 836             cutoff_mask      = _mm_cmplt_ps(rsq33,rcutoff2);
 837
 838             /* Update potential sum for this i atom from the interaction with this j atom. */
 839             velec            = _mm_and_ps(velec,cutoff_mask);
 840             velecsum         = _mm_add_ps(velecsum,velec);
 841
 842             fscal            = felec;
 843
 844             fscal            = _mm_and_ps(fscal,cutoff_mask);
 845
 846             /* Calculate temporary vectorial force */
 847             tx               = _mm_mul_ps(fscal,dx33);
 848             ty               = _mm_mul_ps(fscal,dy33);
 849             tz               = _mm_mul_ps(fscal,dz33);
 850
 851             /* Update vectorial force */
 852             fix3             = _mm_add_ps(fix3,tx);
 853             fiy3             = _mm_add_ps(fiy3,ty);
 854             fiz3             = _mm_add_ps(fiz3,tz);
 855
 856             fjx3             = _mm_add_ps(fjx3,tx);
 857             fjy3             = _mm_add_ps(fjy3,ty);
 858             fjz3             = _mm_add_ps(fjz3,tz);
 859
 860             }
 861
 862             fjptrA             = f+j_coord_offsetA;
 863             fjptrB             = f+j_coord_offsetB;
 864             fjptrC             = f+j_coord_offsetC;
 865             fjptrD             = f+j_coord_offsetD;
 866
 867             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
 868                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
 869                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 870
 871             /* Inner loop uses 479 flops */
 872         }
 873
 874         if(jidx<j_index_end)
 875         {
 876
 877             /* Get j neighbor index, and coordinate index */
 878             jnrlistA         = jjnr[jidx];
 879             jnrlistB         = jjnr[jidx+1];
 880             jnrlistC         = jjnr[jidx+2];
 881             jnrlistD         = jjnr[jidx+3];
 882             /* Sign of each element will be negative for non-real atoms.
 883              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 884              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 885              */
 886             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 887             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 888             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 889             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 890             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 891             j_coord_offsetA  = DIM*jnrA;
 892             j_coord_offsetB  = DIM*jnrB;
 893             j_coord_offsetC  = DIM*jnrC;
 894             j_coord_offsetD  = DIM*jnrD;
 895
 896             /* load j atom coordinates */
 897             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 898                                               x+j_coord_offsetC,x+j_coord_offsetD,
 899                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 900                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 901
 902             /* Calculate displacement vector */
 903             dx00             = _mm_sub_ps(ix0,jx0);
 904             dy00             = _mm_sub_ps(iy0,jy0);
 905             dz00             = _mm_sub_ps(iz0,jz0);
 906             dx11             = _mm_sub_ps(ix1,jx1);
 907             dy11             = _mm_sub_ps(iy1,jy1);
 908             dz11             = _mm_sub_ps(iz1,jz1);
 909             dx12             = _mm_sub_ps(ix1,jx2);
 910             dy12             = _mm_sub_ps(iy1,jy2);
 911             dz12             = _mm_sub_ps(iz1,jz2);
 912             dx13             = _mm_sub_ps(ix1,jx3);
 913             dy13             = _mm_sub_ps(iy1,jy3);
 914             dz13             = _mm_sub_ps(iz1,jz3);
 915             dx21             = _mm_sub_ps(ix2,jx1);
 916             dy21             = _mm_sub_ps(iy2,jy1);
 917             dz21             = _mm_sub_ps(iz2,jz1);
 918             dx22             = _mm_sub_ps(ix2,jx2);
 919             dy22             = _mm_sub_ps(iy2,jy2);
 920             dz22             = _mm_sub_ps(iz2,jz2);
 921             dx23             = _mm_sub_ps(ix2,jx3);
 922             dy23             = _mm_sub_ps(iy2,jy3);
 923             dz23             = _mm_sub_ps(iz2,jz3);
 924             dx31             = _mm_sub_ps(ix3,jx1);
 925             dy31             = _mm_sub_ps(iy3,jy1);
 926             dz31             = _mm_sub_ps(iz3,jz1);
 927             dx32             = _mm_sub_ps(ix3,jx2);
 928             dy32             = _mm_sub_ps(iy3,jy2);
 929             dz32             = _mm_sub_ps(iz3,jz2);
 930             dx33             = _mm_sub_ps(ix3,jx3);
 931             dy33             = _mm_sub_ps(iy3,jy3);
 932             dz33             = _mm_sub_ps(iz3,jz3);
 933
 934             /* Calculate squared distance and things based on it */
 935             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 936             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 937             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 938             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 939             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 940             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 941             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 942             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 943             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 944             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 945
 946             rinv00           = sse2_invsqrt_f(rsq00);
 947             rinv11           = sse2_invsqrt_f(rsq11);
 948             rinv12           = sse2_invsqrt_f(rsq12);
 949             rinv13           = sse2_invsqrt_f(rsq13);
 950             rinv21           = sse2_invsqrt_f(rsq21);
 951             rinv22           = sse2_invsqrt_f(rsq22);
 952             rinv23           = sse2_invsqrt_f(rsq23);
 953             rinv31           = sse2_invsqrt_f(rsq31);
 954             rinv32           = sse2_invsqrt_f(rsq32);
 955             rinv33           = sse2_invsqrt_f(rsq33);
 956
 957             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 958             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
 959             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
 960             rinvsq13         = _mm_mul_ps(rinv13,rinv13);
 961             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
 962             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
 963             rinvsq23         = _mm_mul_ps(rinv23,rinv23);
 964             rinvsq31         = _mm_mul_ps(rinv31,rinv31);
 965             rinvsq32         = _mm_mul_ps(rinv32,rinv32);
 966             rinvsq33         = _mm_mul_ps(rinv33,rinv33);
 967
 968             fjx0             = _mm_setzero_ps();
 969             fjy0             = _mm_setzero_ps();
 970             fjz0             = _mm_setzero_ps();
 971             fjx1             = _mm_setzero_ps();
 972             fjy1             = _mm_setzero_ps();
 973             fjz1             = _mm_setzero_ps();
 974             fjx2             = _mm_setzero_ps();
 975             fjy2             = _mm_setzero_ps();
 976             fjz2             = _mm_setzero_ps();
 977             fjx3             = _mm_setzero_ps();
 978             fjy3             = _mm_setzero_ps();
 979             fjz3             = _mm_setzero_ps();
 980
 981             /**************************
 982              * CALCULATE INTERACTIONS *
 983              **************************/
 984
 985             if (gmx_mm_any_lt(rsq00,rcutoff2))
 986             {
 987
 988             r00              = _mm_mul_ps(rsq00,rinv00);
 989             r00              = _mm_andnot_ps(dummy_mask,r00);
 990
 991             /* Analytical LJ-PME */
 992             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 993             ewcljrsq         = _mm_mul_ps(ewclj2,rsq00);
 994             ewclj6           = _mm_mul_ps(ewclj2,_mm_mul_ps(ewclj2,ewclj2));
 995             exponent         = sse2_exp_f(ewcljrsq);
 996             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
 997             poly             = _mm_mul_ps(exponent,_mm_add_ps(_mm_sub_ps(one,ewcljrsq),_mm_mul_ps(_mm_mul_ps(ewcljrsq,ewcljrsq),one_half)));
 998             /* vvdw6 = [C6 - C6grid * (1-poly)]/r6 */
 999             vvdw6            = _mm_mul_ps(_mm_sub_ps(c6_00,_mm_mul_ps(c6grid_00,_mm_sub_ps(one,poly))),rinvsix);
1000             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
1001             vvdw             = _mm_sub_ps(_mm_mul_ps( _mm_sub_ps(vvdw12 , _mm_mul_ps(c12_00,_mm_mul_ps(sh_vdw_invrcut6,sh_vdw_invrcut6))), one_twelfth) ,
1002                                           _mm_mul_ps( _mm_sub_ps(vvdw6,_mm_add_ps(_mm_mul_ps(c6_00,sh_vdw_invrcut6),_mm_mul_ps(c6grid_00,sh_lj_ewald))),one_sixth));
1003             /* fvdw = vvdw12/r - (vvdw6/r + (C6grid * exponent * beta^6)/r) */
1004             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,_mm_sub_ps(vvdw6,_mm_mul_ps(_mm_mul_ps(c6grid_00,one_sixth),_mm_mul_ps(exponent,ewclj6)))),rinvsq00);
1005
1006             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
1007
1008             /* Update potential sum for this i atom from the interaction with this j atom. */
1009             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
1010             vvdw             = _mm_andnot_ps(dummy_mask,vvdw);
1011             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
1012
1013             fscal            = fvdw;
1014
1015             fscal            = _mm_and_ps(fscal,cutoff_mask);
1016
1017             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1018
1019             /* Calculate temporary vectorial force */
1020             tx               = _mm_mul_ps(fscal,dx00);
1021             ty               = _mm_mul_ps(fscal,dy00);
1022             tz               = _mm_mul_ps(fscal,dz00);
1023
1024             /* Update vectorial force */
1025             fix0             = _mm_add_ps(fix0,tx);
1026             fiy0             = _mm_add_ps(fiy0,ty);
1027             fiz0             = _mm_add_ps(fiz0,tz);
1028
1029             fjx0             = _mm_add_ps(fjx0,tx);
1030             fjy0             = _mm_add_ps(fjy0,ty);
1031             fjz0             = _mm_add_ps(fjz0,tz);
1032
1033             }
1034
1035             /**************************
1036              * CALCULATE INTERACTIONS *
1037              **************************/
1038
1039             if (gmx_mm_any_lt(rsq11,rcutoff2))
1040             {
1041
1042             r11              = _mm_mul_ps(rsq11,rinv11);
1043             r11              = _mm_andnot_ps(dummy_mask,r11);
1044
1045             /* EWALD ELECTROSTATICS */
1046
1047             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1048             ewrt             = _mm_mul_ps(r11,ewtabscale);
1049             ewitab           = _mm_cvttps_epi32(ewrt);
1050             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1051             ewitab           = _mm_slli_epi32(ewitab,2);
1052             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1053             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1054             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1055             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1056             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1057             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1058             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1059             velec            = _mm_mul_ps(qq11,_mm_sub_ps(_mm_sub_ps(rinv11,sh_ewald),velec));
1060             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
1061
1062             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
1063
1064             /* Update potential sum for this i atom from the interaction with this j atom. */
1065             velec            = _mm_and_ps(velec,cutoff_mask);
1066             velec            = _mm_andnot_ps(dummy_mask,velec);
1067             velecsum         = _mm_add_ps(velecsum,velec);
1068
1069             fscal            = felec;
1070
1071             fscal            = _mm_and_ps(fscal,cutoff_mask);
1072
1073             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1074
1075             /* Calculate temporary vectorial force */
1076             tx               = _mm_mul_ps(fscal,dx11);
1077             ty               = _mm_mul_ps(fscal,dy11);
1078             tz               = _mm_mul_ps(fscal,dz11);
1079
1080             /* Update vectorial force */
1081             fix1             = _mm_add_ps(fix1,tx);
1082             fiy1             = _mm_add_ps(fiy1,ty);
1083             fiz1             = _mm_add_ps(fiz1,tz);
1084
1085             fjx1             = _mm_add_ps(fjx1,tx);
1086             fjy1             = _mm_add_ps(fjy1,ty);
1087             fjz1             = _mm_add_ps(fjz1,tz);
1088
1089             }
1090
1091             /**************************
1092              * CALCULATE INTERACTIONS *
1093              **************************/
1094
1095             if (gmx_mm_any_lt(rsq12,rcutoff2))
1096             {
1097
1098             r12              = _mm_mul_ps(rsq12,rinv12);
1099             r12              = _mm_andnot_ps(dummy_mask,r12);
1100
1101             /* EWALD ELECTROSTATICS */
1102
1103             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1104             ewrt             = _mm_mul_ps(r12,ewtabscale);
1105             ewitab           = _mm_cvttps_epi32(ewrt);
1106             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1107             ewitab           = _mm_slli_epi32(ewitab,2);
1108             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1109             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1110             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1111             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1112             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1113             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1114             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1115             velec            = _mm_mul_ps(qq12,_mm_sub_ps(_mm_sub_ps(rinv12,sh_ewald),velec));
1116             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
1117
1118             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
1119
1120             /* Update potential sum for this i atom from the interaction with this j atom. */
1121             velec            = _mm_and_ps(velec,cutoff_mask);
1122             velec            = _mm_andnot_ps(dummy_mask,velec);
1123             velecsum         = _mm_add_ps(velecsum,velec);
1124
1125             fscal            = felec;
1126
1127             fscal            = _mm_and_ps(fscal,cutoff_mask);
1128
1129             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1130
1131             /* Calculate temporary vectorial force */
1132             tx               = _mm_mul_ps(fscal,dx12);
1133             ty               = _mm_mul_ps(fscal,dy12);
1134             tz               = _mm_mul_ps(fscal,dz12);
1135
1136             /* Update vectorial force */
1137             fix1             = _mm_add_ps(fix1,tx);
1138             fiy1             = _mm_add_ps(fiy1,ty);
1139             fiz1             = _mm_add_ps(fiz1,tz);
1140
1141             fjx2             = _mm_add_ps(fjx2,tx);
1142             fjy2             = _mm_add_ps(fjy2,ty);
1143             fjz2             = _mm_add_ps(fjz2,tz);
1144
1145             }
1146
1147             /**************************
1148              * CALCULATE INTERACTIONS *
1149              **************************/
1150
1151             if (gmx_mm_any_lt(rsq13,rcutoff2))
1152             {
1153
1154             r13              = _mm_mul_ps(rsq13,rinv13);
1155             r13              = _mm_andnot_ps(dummy_mask,r13);
1156
1157             /* EWALD ELECTROSTATICS */
1158
1159             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1160             ewrt             = _mm_mul_ps(r13,ewtabscale);
1161             ewitab           = _mm_cvttps_epi32(ewrt);
1162             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1163             ewitab           = _mm_slli_epi32(ewitab,2);
1164             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1165             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1166             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1167             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1168             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1169             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1170             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1171             velec            = _mm_mul_ps(qq13,_mm_sub_ps(_mm_sub_ps(rinv13,sh_ewald),velec));
1172             felec            = _mm_mul_ps(_mm_mul_ps(qq13,rinv13),_mm_sub_ps(rinvsq13,felec));
1173
1174             cutoff_mask      = _mm_cmplt_ps(rsq13,rcutoff2);
1175
1176             /* Update potential sum for this i atom from the interaction with this j atom. */
1177             velec            = _mm_and_ps(velec,cutoff_mask);
1178             velec            = _mm_andnot_ps(dummy_mask,velec);
1179             velecsum         = _mm_add_ps(velecsum,velec);
1180
1181             fscal            = felec;
1182
1183             fscal            = _mm_and_ps(fscal,cutoff_mask);
1184
1185             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1186
1187             /* Calculate temporary vectorial force */
1188             tx               = _mm_mul_ps(fscal,dx13);
1189             ty               = _mm_mul_ps(fscal,dy13);
1190             tz               = _mm_mul_ps(fscal,dz13);
1191
1192             /* Update vectorial force */
1193             fix1             = _mm_add_ps(fix1,tx);
1194             fiy1             = _mm_add_ps(fiy1,ty);
1195             fiz1             = _mm_add_ps(fiz1,tz);
1196
1197             fjx3             = _mm_add_ps(fjx3,tx);
1198             fjy3             = _mm_add_ps(fjy3,ty);
1199             fjz3             = _mm_add_ps(fjz3,tz);
1200
1201             }
1202
1203             /**************************
1204              * CALCULATE INTERACTIONS *
1205              **************************/
1206
1207             if (gmx_mm_any_lt(rsq21,rcutoff2))
1208             {
1209
1210             r21              = _mm_mul_ps(rsq21,rinv21);
1211             r21              = _mm_andnot_ps(dummy_mask,r21);
1212
1213             /* EWALD ELECTROSTATICS */
1214
1215             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1216             ewrt             = _mm_mul_ps(r21,ewtabscale);
1217             ewitab           = _mm_cvttps_epi32(ewrt);
1218             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1219             ewitab           = _mm_slli_epi32(ewitab,2);
1220             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1221             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1222             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1223             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1224             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1225             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1226             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1227             velec            = _mm_mul_ps(qq21,_mm_sub_ps(_mm_sub_ps(rinv21,sh_ewald),velec));
1228             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
1229
1230             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
1231
1232             /* Update potential sum for this i atom from the interaction with this j atom. */
1233             velec            = _mm_and_ps(velec,cutoff_mask);
1234             velec            = _mm_andnot_ps(dummy_mask,velec);
1235             velecsum         = _mm_add_ps(velecsum,velec);
1236
1237             fscal            = felec;
1238
1239             fscal            = _mm_and_ps(fscal,cutoff_mask);
1240
1241             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1242
1243             /* Calculate temporary vectorial force */
1244             tx               = _mm_mul_ps(fscal,dx21);
1245             ty               = _mm_mul_ps(fscal,dy21);
1246             tz               = _mm_mul_ps(fscal,dz21);
1247
1248             /* Update vectorial force */
1249             fix2             = _mm_add_ps(fix2,tx);
1250             fiy2             = _mm_add_ps(fiy2,ty);
1251             fiz2             = _mm_add_ps(fiz2,tz);
1252
1253             fjx1             = _mm_add_ps(fjx1,tx);
1254             fjy1             = _mm_add_ps(fjy1,ty);
1255             fjz1             = _mm_add_ps(fjz1,tz);
1256
1257             }
1258
1259             /**************************
1260              * CALCULATE INTERACTIONS *
1261              **************************/
1262
1263             if (gmx_mm_any_lt(rsq22,rcutoff2))
1264             {
1265
1266             r22              = _mm_mul_ps(rsq22,rinv22);
1267             r22              = _mm_andnot_ps(dummy_mask,r22);
1268
1269             /* EWALD ELECTROSTATICS */
1270
1271             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1272             ewrt             = _mm_mul_ps(r22,ewtabscale);
1273             ewitab           = _mm_cvttps_epi32(ewrt);
1274             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1275             ewitab           = _mm_slli_epi32(ewitab,2);
1276             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1277             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1278             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1279             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1280             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1281             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1282             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1283             velec            = _mm_mul_ps(qq22,_mm_sub_ps(_mm_sub_ps(rinv22,sh_ewald),velec));
1284             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
1285
1286             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
1287
1288             /* Update potential sum for this i atom from the interaction with this j atom. */
1289             velec            = _mm_and_ps(velec,cutoff_mask);
1290             velec            = _mm_andnot_ps(dummy_mask,velec);
1291             velecsum         = _mm_add_ps(velecsum,velec);
1292
1293             fscal            = felec;
1294
1295             fscal            = _mm_and_ps(fscal,cutoff_mask);
1296
1297             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1298
1299             /* Calculate temporary vectorial force */
1300             tx               = _mm_mul_ps(fscal,dx22);
1301             ty               = _mm_mul_ps(fscal,dy22);
1302             tz               = _mm_mul_ps(fscal,dz22);
1303
1304             /* Update vectorial force */
1305             fix2             = _mm_add_ps(fix2,tx);
1306             fiy2             = _mm_add_ps(fiy2,ty);
1307             fiz2             = _mm_add_ps(fiz2,tz);
1308
1309             fjx2             = _mm_add_ps(fjx2,tx);
1310             fjy2             = _mm_add_ps(fjy2,ty);
1311             fjz2             = _mm_add_ps(fjz2,tz);
1312
1313             }
1314
1315             /**************************
1316              * CALCULATE INTERACTIONS *
1317              **************************/
1318
1319             if (gmx_mm_any_lt(rsq23,rcutoff2))
1320             {
1321
1322             r23              = _mm_mul_ps(rsq23,rinv23);
1323             r23              = _mm_andnot_ps(dummy_mask,r23);
1324
1325             /* EWALD ELECTROSTATICS */
1326
1327             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1328             ewrt             = _mm_mul_ps(r23,ewtabscale);
1329             ewitab           = _mm_cvttps_epi32(ewrt);
1330             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1331             ewitab           = _mm_slli_epi32(ewitab,2);
1332             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1333             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1334             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1335             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1336             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1337             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1338             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1339             velec            = _mm_mul_ps(qq23,_mm_sub_ps(_mm_sub_ps(rinv23,sh_ewald),velec));
1340             felec            = _mm_mul_ps(_mm_mul_ps(qq23,rinv23),_mm_sub_ps(rinvsq23,felec));
1341
1342             cutoff_mask      = _mm_cmplt_ps(rsq23,rcutoff2);
1343
1344             /* Update potential sum for this i atom from the interaction with this j atom. */
1345             velec            = _mm_and_ps(velec,cutoff_mask);
1346             velec            = _mm_andnot_ps(dummy_mask,velec);
1347             velecsum         = _mm_add_ps(velecsum,velec);
1348
1349             fscal            = felec;
1350
1351             fscal            = _mm_and_ps(fscal,cutoff_mask);
1352
1353             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1354
1355             /* Calculate temporary vectorial force */
1356             tx               = _mm_mul_ps(fscal,dx23);
1357             ty               = _mm_mul_ps(fscal,dy23);
1358             tz               = _mm_mul_ps(fscal,dz23);
1359
1360             /* Update vectorial force */
1361             fix2             = _mm_add_ps(fix2,tx);
1362             fiy2             = _mm_add_ps(fiy2,ty);
1363             fiz2             = _mm_add_ps(fiz2,tz);
1364
1365             fjx3             = _mm_add_ps(fjx3,tx);
1366             fjy3             = _mm_add_ps(fjy3,ty);
1367             fjz3             = _mm_add_ps(fjz3,tz);
1368
1369             }
1370
1371             /**************************
1372              * CALCULATE INTERACTIONS *
1373              **************************/
1374
1375             if (gmx_mm_any_lt(rsq31,rcutoff2))
1376             {
1377
1378             r31              = _mm_mul_ps(rsq31,rinv31);
1379             r31              = _mm_andnot_ps(dummy_mask,r31);
1380
1381             /* EWALD ELECTROSTATICS */
1382
1383             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1384             ewrt             = _mm_mul_ps(r31,ewtabscale);
1385             ewitab           = _mm_cvttps_epi32(ewrt);
1386             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1387             ewitab           = _mm_slli_epi32(ewitab,2);
1388             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1389             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1390             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1391             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1392             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1393             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1394             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1395             velec            = _mm_mul_ps(qq31,_mm_sub_ps(_mm_sub_ps(rinv31,sh_ewald),velec));
1396             felec            = _mm_mul_ps(_mm_mul_ps(qq31,rinv31),_mm_sub_ps(rinvsq31,felec));
1397
1398             cutoff_mask      = _mm_cmplt_ps(rsq31,rcutoff2);
1399
1400             /* Update potential sum for this i atom from the interaction with this j atom. */
1401             velec            = _mm_and_ps(velec,cutoff_mask);
1402             velec            = _mm_andnot_ps(dummy_mask,velec);
1403             velecsum         = _mm_add_ps(velecsum,velec);
1404
1405             fscal            = felec;
1406
1407             fscal            = _mm_and_ps(fscal,cutoff_mask);
1408
1409             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1410
1411             /* Calculate temporary vectorial force */
1412             tx               = _mm_mul_ps(fscal,dx31);
1413             ty               = _mm_mul_ps(fscal,dy31);
1414             tz               = _mm_mul_ps(fscal,dz31);
1415
1416             /* Update vectorial force */
1417             fix3             = _mm_add_ps(fix3,tx);
1418             fiy3             = _mm_add_ps(fiy3,ty);
1419             fiz3             = _mm_add_ps(fiz3,tz);
1420
1421             fjx1             = _mm_add_ps(fjx1,tx);
1422             fjy1             = _mm_add_ps(fjy1,ty);
1423             fjz1             = _mm_add_ps(fjz1,tz);
1424
1425             }
1426
1427             /**************************
1428              * CALCULATE INTERACTIONS *
1429              **************************/
1430
1431             if (gmx_mm_any_lt(rsq32,rcutoff2))
1432             {
1433
1434             r32              = _mm_mul_ps(rsq32,rinv32);
1435             r32              = _mm_andnot_ps(dummy_mask,r32);
1436
1437             /* EWALD ELECTROSTATICS */
1438
1439             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1440             ewrt             = _mm_mul_ps(r32,ewtabscale);
1441             ewitab           = _mm_cvttps_epi32(ewrt);
1442             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1443             ewitab           = _mm_slli_epi32(ewitab,2);
1444             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1445             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1446             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1447             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1448             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1449             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1450             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1451             velec            = _mm_mul_ps(qq32,_mm_sub_ps(_mm_sub_ps(rinv32,sh_ewald),velec));
1452             felec            = _mm_mul_ps(_mm_mul_ps(qq32,rinv32),_mm_sub_ps(rinvsq32,felec));
1453
1454             cutoff_mask      = _mm_cmplt_ps(rsq32,rcutoff2);
1455
1456             /* Update potential sum for this i atom from the interaction with this j atom. */
1457             velec            = _mm_and_ps(velec,cutoff_mask);
1458             velec            = _mm_andnot_ps(dummy_mask,velec);
1459             velecsum         = _mm_add_ps(velecsum,velec);
1460
1461             fscal            = felec;
1462
1463             fscal            = _mm_and_ps(fscal,cutoff_mask);
1464
1465             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1466
1467             /* Calculate temporary vectorial force */
1468             tx               = _mm_mul_ps(fscal,dx32);
1469             ty               = _mm_mul_ps(fscal,dy32);
1470             tz               = _mm_mul_ps(fscal,dz32);
1471
1472             /* Update vectorial force */
1473             fix3             = _mm_add_ps(fix3,tx);
1474             fiy3             = _mm_add_ps(fiy3,ty);
1475             fiz3             = _mm_add_ps(fiz3,tz);
1476
1477             fjx2             = _mm_add_ps(fjx2,tx);
1478             fjy2             = _mm_add_ps(fjy2,ty);
1479             fjz2             = _mm_add_ps(fjz2,tz);
1480
1481             }
1482
1483             /**************************
1484              * CALCULATE INTERACTIONS *
1485              **************************/
1486
1487             if (gmx_mm_any_lt(rsq33,rcutoff2))
1488             {
1489
1490             r33              = _mm_mul_ps(rsq33,rinv33);
1491             r33              = _mm_andnot_ps(dummy_mask,r33);
1492
1493             /* EWALD ELECTROSTATICS */
1494
1495             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1496             ewrt             = _mm_mul_ps(r33,ewtabscale);
1497             ewitab           = _mm_cvttps_epi32(ewrt);
1498             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1499             ewitab           = _mm_slli_epi32(ewitab,2);
1500             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1501             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1502             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1503             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1504             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1505             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1506             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1507             velec            = _mm_mul_ps(qq33,_mm_sub_ps(_mm_sub_ps(rinv33,sh_ewald),velec));
1508             felec            = _mm_mul_ps(_mm_mul_ps(qq33,rinv33),_mm_sub_ps(rinvsq33,felec));
1509
1510             cutoff_mask      = _mm_cmplt_ps(rsq33,rcutoff2);
1511
1512             /* Update potential sum for this i atom from the interaction with this j atom. */
1513             velec            = _mm_and_ps(velec,cutoff_mask);
1514             velec            = _mm_andnot_ps(dummy_mask,velec);
1515             velecsum         = _mm_add_ps(velecsum,velec);
1516
1517             fscal            = felec;
1518
1519             fscal            = _mm_and_ps(fscal,cutoff_mask);
1520
1521             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1522
1523             /* Calculate temporary vectorial force */
1524             tx               = _mm_mul_ps(fscal,dx33);
1525             ty               = _mm_mul_ps(fscal,dy33);
1526             tz               = _mm_mul_ps(fscal,dz33);
1527
1528             /* Update vectorial force */
1529             fix3             = _mm_add_ps(fix3,tx);
1530             fiy3             = _mm_add_ps(fiy3,ty);
1531             fiz3             = _mm_add_ps(fiz3,tz);
1532
1533             fjx3             = _mm_add_ps(fjx3,tx);
1534             fjy3             = _mm_add_ps(fjy3,ty);
1535             fjz3             = _mm_add_ps(fjz3,tz);
1536
1537             }
1538
1539             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1540             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1541             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1542             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1543
1544             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
1545                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1546                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1547
1548             /* Inner loop uses 489 flops */
1549         }
1550
1551         /* End of innermost loop */
1552
1553         gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1554                                               f+i_coord_offset,fshift+i_shift_offset);
1555
1556         ggid                        = gid[iidx];
1557         /* Update potential energies */
1558         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1559         gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
1560
1561         /* Increment number of inner iterations */
1562         inneriter                  += j_index_end - j_index_start;
1563
1564         /* Outer loop uses 26 flops */
1565     }
1566
1567     /* Increment number of outer iterations */
1568     outeriter        += nri;
1569
1570     /* Update outer/inner flops */
1571
1572     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*489);
1573 }
1574 /*
1575  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4W4_F_sse2_single
1576  * Electrostatics interaction: Ewald
1577  * VdW interaction:            LJEwald
1578  * Geometry:                   Water4-Water4
1579  * Calculate force/pot:        Force
1580  */
1581 void
1582 nb_kernel_ElecEwSh_VdwLJEwSh_GeomW4W4_F_sse2_single
1583                     (t_nblist                    * gmx_restrict       nlist,
1584                      rvec                        * gmx_restrict          xx,
1585                      rvec                        * gmx_restrict          ff,
1586                      struct t_forcerec           * gmx_restrict          fr,
1587                      t_mdatoms                   * gmx_restrict     mdatoms,
1588                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1589                      t_nrnb                      * gmx_restrict        nrnb)
1590 {
1591     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1592      * just 0 for non-waters.
1593      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
1594      * jnr indices corresponding to data put in the four positions in the SIMD register.
1595      */
1596     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1597     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1598     int              jnrA,jnrB,jnrC,jnrD;
1599     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1600     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1601     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1602     real             rcutoff_scalar;
1603     real             *shiftvec,*fshift,*x,*f;
1604     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1605     real             scratch[4*DIM];
1606     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1607     int              vdwioffset0;
1608     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1609     int              vdwioffset1;
1610     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1611     int              vdwioffset2;
1612     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1613     int              vdwioffset3;
1614     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1615     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1616     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1617     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1618     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1619     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1620     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1621     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1622     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1623     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1624     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1625     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1626     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1627     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1628     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1629     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1630     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1631     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1632     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1633     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
1634     real             *charge;
1635     int              nvdwtype;
1636     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1637     int              *vdwtype;
1638     real             *vdwparam;
1639     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
1640     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
1641     __m128           c6grid_00;
1642     __m128           c6grid_11;
1643     __m128           c6grid_12;
1644     __m128           c6grid_13;
1645     __m128           c6grid_21;
1646     __m128           c6grid_22;
1647     __m128           c6grid_23;
1648     __m128           c6grid_31;
1649     __m128           c6grid_32;
1650     __m128           c6grid_33;
1651     __m128           ewclj,ewclj2,ewclj6,ewcljrsq,poly,exponent,f6A,f6B,sh_lj_ewald;
1652     real             *vdwgridparam;
1653     __m128           one_half = _mm_set1_ps(0.5);
1654     __m128           minus_one = _mm_set1_ps(-1.0);
1655     __m128i          ewitab;
1656     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1657     real             *ewtab;
1658     __m128           dummy_mask,cutoff_mask;
1659     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1660     __m128           one     = _mm_set1_ps(1.0);
1661     __m128           two     = _mm_set1_ps(2.0);
1662     x                = xx[0];
1663     f                = ff[0];
1664
1665     nri              = nlist->nri;
1666     iinr             = nlist->iinr;
1667     jindex           = nlist->jindex;
1668     jjnr             = nlist->jjnr;
1669     shiftidx         = nlist->shift;
1670     gid              = nlist->gid;
1671     shiftvec         = fr->shift_vec[0];
1672     fshift           = fr->fshift[0];
1673     facel            = _mm_set1_ps(fr->ic->epsfac);
1674     charge           = mdatoms->chargeA;
1675     nvdwtype         = fr->ntype;
1676     vdwparam         = fr->nbfp;
1677     vdwtype          = mdatoms->typeA;
1678     vdwgridparam     = fr->ljpme_c6grid;
1679     sh_lj_ewald      = _mm_set1_ps(fr->ic->sh_lj_ewald);
1680     ewclj            = _mm_set1_ps(fr->ic->ewaldcoeff_lj);
1681     ewclj2           = _mm_mul_ps(minus_one,_mm_mul_ps(ewclj,ewclj));
1682
1683     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
1684     ewtab            = fr->ic->tabq_coul_F;
1685     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
1686     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
1687
1688     /* Setup water-specific parameters */
1689     inr              = nlist->iinr[0];
1690     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1691     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1692     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
1693     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
1694
1695     jq1              = _mm_set1_ps(charge[inr+1]);
1696     jq2              = _mm_set1_ps(charge[inr+2]);
1697     jq3              = _mm_set1_ps(charge[inr+3]);
1698     vdwjidx0A        = 2*vdwtype[inr+0];
1699     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
1700     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
1701     c6grid_00        = _mm_set1_ps(vdwgridparam[vdwioffset0+vdwjidx0A]);
1702     qq11             = _mm_mul_ps(iq1,jq1);
1703     qq12             = _mm_mul_ps(iq1,jq2);
1704     qq13             = _mm_mul_ps(iq1,jq3);
1705     qq21             = _mm_mul_ps(iq2,jq1);
1706     qq22             = _mm_mul_ps(iq2,jq2);
1707     qq23             = _mm_mul_ps(iq2,jq3);
1708     qq31             = _mm_mul_ps(iq3,jq1);
1709     qq32             = _mm_mul_ps(iq3,jq2);
1710     qq33             = _mm_mul_ps(iq3,jq3);
1711
1712     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1713     rcutoff_scalar   = fr->ic->rcoulomb;
1714     rcutoff          = _mm_set1_ps(rcutoff_scalar);
1715     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
1716
1717     sh_vdw_invrcut6  = _mm_set1_ps(fr->ic->sh_invrc6);
1718     rvdw             = _mm_set1_ps(fr->ic->rvdw);
1719
1720     /* Avoid stupid compiler warnings */
1721     jnrA = jnrB = jnrC = jnrD = 0;
1722     j_coord_offsetA = 0;
1723     j_coord_offsetB = 0;
1724     j_coord_offsetC = 0;
1725     j_coord_offsetD = 0;
1726
1727     outeriter        = 0;
1728     inneriter        = 0;
1729
1730     for(iidx=0;iidx<4*DIM;iidx++)
1731     {
1732         scratch[iidx] = 0.0;
1733     }
1734
1735     /* Start outer loop over neighborlists */
1736     for(iidx=0; iidx<nri; iidx++)
1737     {
1738         /* Load shift vector for this list */
1739         i_shift_offset   = DIM*shiftidx[iidx];
1740
1741         /* Load limits for loop over neighbors */
1742         j_index_start    = jindex[iidx];
1743         j_index_end      = jindex[iidx+1];
1744
1745         /* Get outer coordinate index */
1746         inr              = iinr[iidx];
1747         i_coord_offset   = DIM*inr;
1748
1749         /* Load i particle coords and add shift vector */
1750         gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
1751                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1752
1753         fix0             = _mm_setzero_ps();
1754         fiy0             = _mm_setzero_ps();
1755         fiz0             = _mm_setzero_ps();
1756         fix1             = _mm_setzero_ps();
1757         fiy1             = _mm_setzero_ps();
1758         fiz1             = _mm_setzero_ps();
1759         fix2             = _mm_setzero_ps();
1760         fiy2             = _mm_setzero_ps();
1761         fiz2             = _mm_setzero_ps();
1762         fix3             = _mm_setzero_ps();
1763         fiy3             = _mm_setzero_ps();
1764         fiz3             = _mm_setzero_ps();
1765
1766         /* Start inner kernel loop */
1767         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1768         {
1769
1770             /* Get j neighbor index, and coordinate index */
1771             jnrA             = jjnr[jidx];
1772             jnrB             = jjnr[jidx+1];
1773             jnrC             = jjnr[jidx+2];
1774             jnrD             = jjnr[jidx+3];
1775             j_coord_offsetA  = DIM*jnrA;
1776             j_coord_offsetB  = DIM*jnrB;
1777             j_coord_offsetC  = DIM*jnrC;
1778             j_coord_offsetD  = DIM*jnrD;
1779
1780             /* load j atom coordinates */
1781             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1782                                               x+j_coord_offsetC,x+j_coord_offsetD,
1783                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1784                                               &jy2,&jz2,&jx3,&jy3,&jz3);
1785
1786             /* Calculate displacement vector */
1787             dx00             = _mm_sub_ps(ix0,jx0);
1788             dy00             = _mm_sub_ps(iy0,jy0);
1789             dz00             = _mm_sub_ps(iz0,jz0);
1790             dx11             = _mm_sub_ps(ix1,jx1);
1791             dy11             = _mm_sub_ps(iy1,jy1);
1792             dz11             = _mm_sub_ps(iz1,jz1);
1793             dx12             = _mm_sub_ps(ix1,jx2);
1794             dy12             = _mm_sub_ps(iy1,jy2);
1795             dz12             = _mm_sub_ps(iz1,jz2);
1796             dx13             = _mm_sub_ps(ix1,jx3);
1797             dy13             = _mm_sub_ps(iy1,jy3);
1798             dz13             = _mm_sub_ps(iz1,jz3);
1799             dx21             = _mm_sub_ps(ix2,jx1);
1800             dy21             = _mm_sub_ps(iy2,jy1);
1801             dz21             = _mm_sub_ps(iz2,jz1);
1802             dx22             = _mm_sub_ps(ix2,jx2);
1803             dy22             = _mm_sub_ps(iy2,jy2);
1804             dz22             = _mm_sub_ps(iz2,jz2);
1805             dx23             = _mm_sub_ps(ix2,jx3);
1806             dy23             = _mm_sub_ps(iy2,jy3);
1807             dz23             = _mm_sub_ps(iz2,jz3);
1808             dx31             = _mm_sub_ps(ix3,jx1);
1809             dy31             = _mm_sub_ps(iy3,jy1);
1810             dz31             = _mm_sub_ps(iz3,jz1);
1811             dx32             = _mm_sub_ps(ix3,jx2);
1812             dy32             = _mm_sub_ps(iy3,jy2);
1813             dz32             = _mm_sub_ps(iz3,jz2);
1814             dx33             = _mm_sub_ps(ix3,jx3);
1815             dy33             = _mm_sub_ps(iy3,jy3);
1816             dz33             = _mm_sub_ps(iz3,jz3);
1817
1818             /* Calculate squared distance and things based on it */
1819             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
1820             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1821             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1822             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1823             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1824             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1825             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1826             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1827             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1828             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1829
1830             rinv00           = sse2_invsqrt_f(rsq00);
1831             rinv11           = sse2_invsqrt_f(rsq11);
1832             rinv12           = sse2_invsqrt_f(rsq12);
1833             rinv13           = sse2_invsqrt_f(rsq13);
1834             rinv21           = sse2_invsqrt_f(rsq21);
1835             rinv22           = sse2_invsqrt_f(rsq22);
1836             rinv23           = sse2_invsqrt_f(rsq23);
1837             rinv31           = sse2_invsqrt_f(rsq31);
1838             rinv32           = sse2_invsqrt_f(rsq32);
1839             rinv33           = sse2_invsqrt_f(rsq33);
1840
1841             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
1842             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
1843             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
1844             rinvsq13         = _mm_mul_ps(rinv13,rinv13);
1845             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
1846             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
1847             rinvsq23         = _mm_mul_ps(rinv23,rinv23);
1848             rinvsq31         = _mm_mul_ps(rinv31,rinv31);
1849             rinvsq32         = _mm_mul_ps(rinv32,rinv32);
1850             rinvsq33         = _mm_mul_ps(rinv33,rinv33);
1851
1852             fjx0             = _mm_setzero_ps();
1853             fjy0             = _mm_setzero_ps();
1854             fjz0             = _mm_setzero_ps();
1855             fjx1             = _mm_setzero_ps();
1856             fjy1             = _mm_setzero_ps();
1857             fjz1             = _mm_setzero_ps();
1858             fjx2             = _mm_setzero_ps();
1859             fjy2             = _mm_setzero_ps();
1860             fjz2             = _mm_setzero_ps();
1861             fjx3             = _mm_setzero_ps();
1862             fjy3             = _mm_setzero_ps();
1863             fjz3             = _mm_setzero_ps();
1864
1865             /**************************
1866              * CALCULATE INTERACTIONS *
1867              **************************/
1868
1869             if (gmx_mm_any_lt(rsq00,rcutoff2))
1870             {
1871
1872             r00              = _mm_mul_ps(rsq00,rinv00);
1873
1874             /* Analytical LJ-PME */
1875             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
1876             ewcljrsq         = _mm_mul_ps(ewclj2,rsq00);
1877             ewclj6           = _mm_mul_ps(ewclj2,_mm_mul_ps(ewclj2,ewclj2));
1878             exponent         = sse2_exp_f(ewcljrsq);
1879             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
1880             poly             = _mm_mul_ps(exponent,_mm_add_ps(_mm_sub_ps(one,ewcljrsq),_mm_mul_ps(_mm_mul_ps(ewcljrsq,ewcljrsq),one_half)));
1881             /* f6A = 6 * C6grid * (1 - poly) */
1882             f6A              = _mm_mul_ps(c6grid_00,_mm_sub_ps(one,poly));
1883             /* f6B = C6grid * exponent * beta^6 */
1884             f6B              = _mm_mul_ps(_mm_mul_ps(c6grid_00,one_sixth),_mm_mul_ps(exponent,ewclj6));
1885             /* fvdw = 12*C12/r13 - ((6*C6 - f6A)/r6 + f6B)/r */
1886             fvdw              = _mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_sub_ps(_mm_mul_ps(c12_00,rinvsix),_mm_sub_ps(c6_00,f6A)),rinvsix),f6B),rinvsq00);
1887
1888             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
1889
1890             fscal            = fvdw;
1891
1892             fscal            = _mm_and_ps(fscal,cutoff_mask);
1893
1894             /* Calculate temporary vectorial force */
1895             tx               = _mm_mul_ps(fscal,dx00);
1896             ty               = _mm_mul_ps(fscal,dy00);
1897             tz               = _mm_mul_ps(fscal,dz00);
1898
1899             /* Update vectorial force */
1900             fix0             = _mm_add_ps(fix0,tx);
1901             fiy0             = _mm_add_ps(fiy0,ty);
1902             fiz0             = _mm_add_ps(fiz0,tz);
1903
1904             fjx0             = _mm_add_ps(fjx0,tx);
1905             fjy0             = _mm_add_ps(fjy0,ty);
1906             fjz0             = _mm_add_ps(fjz0,tz);
1907
1908             }
1909
1910             /**************************
1911              * CALCULATE INTERACTIONS *
1912              **************************/
1913
1914             if (gmx_mm_any_lt(rsq11,rcutoff2))
1915             {
1916
1917             r11              = _mm_mul_ps(rsq11,rinv11);
1918
1919             /* EWALD ELECTROSTATICS */
1920
1921             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1922             ewrt             = _mm_mul_ps(r11,ewtabscale);
1923             ewitab           = _mm_cvttps_epi32(ewrt);
1924             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1925             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1926                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1927                                          &ewtabF,&ewtabFn);
1928             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1929             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
1930
1931             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
1932
1933             fscal            = felec;
1934
1935             fscal            = _mm_and_ps(fscal,cutoff_mask);
1936
1937             /* Calculate temporary vectorial force */
1938             tx               = _mm_mul_ps(fscal,dx11);
1939             ty               = _mm_mul_ps(fscal,dy11);
1940             tz               = _mm_mul_ps(fscal,dz11);
1941
1942             /* Update vectorial force */
1943             fix1             = _mm_add_ps(fix1,tx);
1944             fiy1             = _mm_add_ps(fiy1,ty);
1945             fiz1             = _mm_add_ps(fiz1,tz);
1946
1947             fjx1             = _mm_add_ps(fjx1,tx);
1948             fjy1             = _mm_add_ps(fjy1,ty);
1949             fjz1             = _mm_add_ps(fjz1,tz);
1950
1951             }
1952
1953             /**************************
1954              * CALCULATE INTERACTIONS *
1955              **************************/
1956
1957             if (gmx_mm_any_lt(rsq12,rcutoff2))
1958             {
1959
1960             r12              = _mm_mul_ps(rsq12,rinv12);
1961
1962             /* EWALD ELECTROSTATICS */
1963
1964             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1965             ewrt             = _mm_mul_ps(r12,ewtabscale);
1966             ewitab           = _mm_cvttps_epi32(ewrt);
1967             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1968             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
1969                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
1970                                          &ewtabF,&ewtabFn);
1971             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
1972             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
1973
1974             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
1975
1976             fscal            = felec;
1977
1978             fscal            = _mm_and_ps(fscal,cutoff_mask);
1979
1980             /* Calculate temporary vectorial force */
1981             tx               = _mm_mul_ps(fscal,dx12);
1982             ty               = _mm_mul_ps(fscal,dy12);
1983             tz               = _mm_mul_ps(fscal,dz12);
1984
1985             /* Update vectorial force */
1986             fix1             = _mm_add_ps(fix1,tx);
1987             fiy1             = _mm_add_ps(fiy1,ty);
1988             fiz1             = _mm_add_ps(fiz1,tz);
1989
1990             fjx2             = _mm_add_ps(fjx2,tx);
1991             fjy2             = _mm_add_ps(fjy2,ty);
1992             fjz2             = _mm_add_ps(fjz2,tz);
1993
1994             }
1995
1996             /**************************
1997              * CALCULATE INTERACTIONS *
1998              **************************/
1999
2000             if (gmx_mm_any_lt(rsq13,rcutoff2))
2001             {
2002
2003             r13              = _mm_mul_ps(rsq13,rinv13);
2004
2005             /* EWALD ELECTROSTATICS */
2006
2007             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2008             ewrt             = _mm_mul_ps(r13,ewtabscale);
2009             ewitab           = _mm_cvttps_epi32(ewrt);
2010             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2011             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2012                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2013                                          &ewtabF,&ewtabFn);
2014             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2015             felec            = _mm_mul_ps(_mm_mul_ps(qq13,rinv13),_mm_sub_ps(rinvsq13,felec));
2016
2017             cutoff_mask      = _mm_cmplt_ps(rsq13,rcutoff2);
2018
2019             fscal            = felec;
2020
2021             fscal            = _mm_and_ps(fscal,cutoff_mask);
2022
2023             /* Calculate temporary vectorial force */
2024             tx               = _mm_mul_ps(fscal,dx13);
2025             ty               = _mm_mul_ps(fscal,dy13);
2026             tz               = _mm_mul_ps(fscal,dz13);
2027
2028             /* Update vectorial force */
2029             fix1             = _mm_add_ps(fix1,tx);
2030             fiy1             = _mm_add_ps(fiy1,ty);
2031             fiz1             = _mm_add_ps(fiz1,tz);
2032
2033             fjx3             = _mm_add_ps(fjx3,tx);
2034             fjy3             = _mm_add_ps(fjy3,ty);
2035             fjz3             = _mm_add_ps(fjz3,tz);
2036
2037             }
2038
2039             /**************************
2040              * CALCULATE INTERACTIONS *
2041              **************************/
2042
2043             if (gmx_mm_any_lt(rsq21,rcutoff2))
2044             {
2045
2046             r21              = _mm_mul_ps(rsq21,rinv21);
2047
2048             /* EWALD ELECTROSTATICS */
2049
2050             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2051             ewrt             = _mm_mul_ps(r21,ewtabscale);
2052             ewitab           = _mm_cvttps_epi32(ewrt);
2053             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2054             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2055                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2056                                          &ewtabF,&ewtabFn);
2057             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2058             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
2059
2060             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
2061
2062             fscal            = felec;
2063
2064             fscal            = _mm_and_ps(fscal,cutoff_mask);
2065
2066             /* Calculate temporary vectorial force */
2067             tx               = _mm_mul_ps(fscal,dx21);
2068             ty               = _mm_mul_ps(fscal,dy21);
2069             tz               = _mm_mul_ps(fscal,dz21);
2070
2071             /* Update vectorial force */
2072             fix2             = _mm_add_ps(fix2,tx);
2073             fiy2             = _mm_add_ps(fiy2,ty);
2074             fiz2             = _mm_add_ps(fiz2,tz);
2075
2076             fjx1             = _mm_add_ps(fjx1,tx);
2077             fjy1             = _mm_add_ps(fjy1,ty);
2078             fjz1             = _mm_add_ps(fjz1,tz);
2079
2080             }
2081
2082             /**************************
2083              * CALCULATE INTERACTIONS *
2084              **************************/
2085
2086             if (gmx_mm_any_lt(rsq22,rcutoff2))
2087             {
2088
2089             r22              = _mm_mul_ps(rsq22,rinv22);
2090
2091             /* EWALD ELECTROSTATICS */
2092
2093             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2094             ewrt             = _mm_mul_ps(r22,ewtabscale);
2095             ewitab           = _mm_cvttps_epi32(ewrt);
2096             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2097             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2098                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2099                                          &ewtabF,&ewtabFn);
2100             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2101             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
2102
2103             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
2104
2105             fscal            = felec;
2106
2107             fscal            = _mm_and_ps(fscal,cutoff_mask);
2108
2109             /* Calculate temporary vectorial force */
2110             tx               = _mm_mul_ps(fscal,dx22);
2111             ty               = _mm_mul_ps(fscal,dy22);
2112             tz               = _mm_mul_ps(fscal,dz22);
2113
2114             /* Update vectorial force */
2115             fix2             = _mm_add_ps(fix2,tx);
2116             fiy2             = _mm_add_ps(fiy2,ty);
2117             fiz2             = _mm_add_ps(fiz2,tz);
2118
2119             fjx2             = _mm_add_ps(fjx2,tx);
2120             fjy2             = _mm_add_ps(fjy2,ty);
2121             fjz2             = _mm_add_ps(fjz2,tz);
2122
2123             }
2124
2125             /**************************
2126              * CALCULATE INTERACTIONS *
2127              **************************/
2128
2129             if (gmx_mm_any_lt(rsq23,rcutoff2))
2130             {
2131
2132             r23              = _mm_mul_ps(rsq23,rinv23);
2133
2134             /* EWALD ELECTROSTATICS */
2135
2136             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2137             ewrt             = _mm_mul_ps(r23,ewtabscale);
2138             ewitab           = _mm_cvttps_epi32(ewrt);
2139             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2140             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2141                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2142                                          &ewtabF,&ewtabFn);
2143             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2144             felec            = _mm_mul_ps(_mm_mul_ps(qq23,rinv23),_mm_sub_ps(rinvsq23,felec));
2145
2146             cutoff_mask      = _mm_cmplt_ps(rsq23,rcutoff2);
2147
2148             fscal            = felec;
2149
2150             fscal            = _mm_and_ps(fscal,cutoff_mask);
2151
2152             /* Calculate temporary vectorial force */
2153             tx               = _mm_mul_ps(fscal,dx23);
2154             ty               = _mm_mul_ps(fscal,dy23);
2155             tz               = _mm_mul_ps(fscal,dz23);
2156
2157             /* Update vectorial force */
2158             fix2             = _mm_add_ps(fix2,tx);
2159             fiy2             = _mm_add_ps(fiy2,ty);
2160             fiz2             = _mm_add_ps(fiz2,tz);
2161
2162             fjx3             = _mm_add_ps(fjx3,tx);
2163             fjy3             = _mm_add_ps(fjy3,ty);
2164             fjz3             = _mm_add_ps(fjz3,tz);
2165
2166             }
2167
2168             /**************************
2169              * CALCULATE INTERACTIONS *
2170              **************************/
2171
2172             if (gmx_mm_any_lt(rsq31,rcutoff2))
2173             {
2174
2175             r31              = _mm_mul_ps(rsq31,rinv31);
2176
2177             /* EWALD ELECTROSTATICS */
2178
2179             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2180             ewrt             = _mm_mul_ps(r31,ewtabscale);
2181             ewitab           = _mm_cvttps_epi32(ewrt);
2182             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2183             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2184                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2185                                          &ewtabF,&ewtabFn);
2186             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2187             felec            = _mm_mul_ps(_mm_mul_ps(qq31,rinv31),_mm_sub_ps(rinvsq31,felec));
2188
2189             cutoff_mask      = _mm_cmplt_ps(rsq31,rcutoff2);
2190
2191             fscal            = felec;
2192
2193             fscal            = _mm_and_ps(fscal,cutoff_mask);
2194
2195             /* Calculate temporary vectorial force */
2196             tx               = _mm_mul_ps(fscal,dx31);
2197             ty               = _mm_mul_ps(fscal,dy31);
2198             tz               = _mm_mul_ps(fscal,dz31);
2199
2200             /* Update vectorial force */
2201             fix3             = _mm_add_ps(fix3,tx);
2202             fiy3             = _mm_add_ps(fiy3,ty);
2203             fiz3             = _mm_add_ps(fiz3,tz);
2204
2205             fjx1             = _mm_add_ps(fjx1,tx);
2206             fjy1             = _mm_add_ps(fjy1,ty);
2207             fjz1             = _mm_add_ps(fjz1,tz);
2208
2209             }
2210
2211             /**************************
2212              * CALCULATE INTERACTIONS *
2213              **************************/
2214
2215             if (gmx_mm_any_lt(rsq32,rcutoff2))
2216             {
2217
2218             r32              = _mm_mul_ps(rsq32,rinv32);
2219
2220             /* EWALD ELECTROSTATICS */
2221
2222             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2223             ewrt             = _mm_mul_ps(r32,ewtabscale);
2224             ewitab           = _mm_cvttps_epi32(ewrt);
2225             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2226             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2227                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2228                                          &ewtabF,&ewtabFn);
2229             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2230             felec            = _mm_mul_ps(_mm_mul_ps(qq32,rinv32),_mm_sub_ps(rinvsq32,felec));
2231
2232             cutoff_mask      = _mm_cmplt_ps(rsq32,rcutoff2);
2233
2234             fscal            = felec;
2235
2236             fscal            = _mm_and_ps(fscal,cutoff_mask);
2237
2238             /* Calculate temporary vectorial force */
2239             tx               = _mm_mul_ps(fscal,dx32);
2240             ty               = _mm_mul_ps(fscal,dy32);
2241             tz               = _mm_mul_ps(fscal,dz32);
2242
2243             /* Update vectorial force */
2244             fix3             = _mm_add_ps(fix3,tx);
2245             fiy3             = _mm_add_ps(fiy3,ty);
2246             fiz3             = _mm_add_ps(fiz3,tz);
2247
2248             fjx2             = _mm_add_ps(fjx2,tx);
2249             fjy2             = _mm_add_ps(fjy2,ty);
2250             fjz2             = _mm_add_ps(fjz2,tz);
2251
2252             }
2253
2254             /**************************
2255              * CALCULATE INTERACTIONS *
2256              **************************/
2257
2258             if (gmx_mm_any_lt(rsq33,rcutoff2))
2259             {
2260
2261             r33              = _mm_mul_ps(rsq33,rinv33);
2262
2263             /* EWALD ELECTROSTATICS */
2264
2265             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2266             ewrt             = _mm_mul_ps(r33,ewtabscale);
2267             ewitab           = _mm_cvttps_epi32(ewrt);
2268             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2269             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2270                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2271                                          &ewtabF,&ewtabFn);
2272             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2273             felec            = _mm_mul_ps(_mm_mul_ps(qq33,rinv33),_mm_sub_ps(rinvsq33,felec));
2274
2275             cutoff_mask      = _mm_cmplt_ps(rsq33,rcutoff2);
2276
2277             fscal            = felec;
2278
2279             fscal            = _mm_and_ps(fscal,cutoff_mask);
2280
2281             /* Calculate temporary vectorial force */
2282             tx               = _mm_mul_ps(fscal,dx33);
2283             ty               = _mm_mul_ps(fscal,dy33);
2284             tz               = _mm_mul_ps(fscal,dz33);
2285
2286             /* Update vectorial force */
2287             fix3             = _mm_add_ps(fix3,tx);
2288             fiy3             = _mm_add_ps(fiy3,ty);
2289             fiz3             = _mm_add_ps(fiz3,tz);
2290
2291             fjx3             = _mm_add_ps(fjx3,tx);
2292             fjy3             = _mm_add_ps(fjy3,ty);
2293             fjz3             = _mm_add_ps(fjz3,tz);
2294
2295             }
2296
2297             fjptrA             = f+j_coord_offsetA;
2298             fjptrB             = f+j_coord_offsetB;
2299             fjptrC             = f+j_coord_offsetC;
2300             fjptrD             = f+j_coord_offsetD;
2301
2302             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2303                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2304                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2305
2306             /* Inner loop uses 403 flops */
2307         }
2308
2309         if(jidx<j_index_end)
2310         {
2311
2312             /* Get j neighbor index, and coordinate index */
2313             jnrlistA         = jjnr[jidx];
2314             jnrlistB         = jjnr[jidx+1];
2315             jnrlistC         = jjnr[jidx+2];
2316             jnrlistD         = jjnr[jidx+3];
2317             /* Sign of each element will be negative for non-real atoms.
2318              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
2319              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
2320              */
2321             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
2322             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
2323             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
2324             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
2325             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
2326             j_coord_offsetA  = DIM*jnrA;
2327             j_coord_offsetB  = DIM*jnrB;
2328             j_coord_offsetC  = DIM*jnrC;
2329             j_coord_offsetD  = DIM*jnrD;
2330
2331             /* load j atom coordinates */
2332             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
2333                                               x+j_coord_offsetC,x+j_coord_offsetD,
2334                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
2335                                               &jy2,&jz2,&jx3,&jy3,&jz3);
2336
2337             /* Calculate displacement vector */
2338             dx00             = _mm_sub_ps(ix0,jx0);
2339             dy00             = _mm_sub_ps(iy0,jy0);
2340             dz00             = _mm_sub_ps(iz0,jz0);
2341             dx11             = _mm_sub_ps(ix1,jx1);
2342             dy11             = _mm_sub_ps(iy1,jy1);
2343             dz11             = _mm_sub_ps(iz1,jz1);
2344             dx12             = _mm_sub_ps(ix1,jx2);
2345             dy12             = _mm_sub_ps(iy1,jy2);
2346             dz12             = _mm_sub_ps(iz1,jz2);
2347             dx13             = _mm_sub_ps(ix1,jx3);
2348             dy13             = _mm_sub_ps(iy1,jy3);
2349             dz13             = _mm_sub_ps(iz1,jz3);
2350             dx21             = _mm_sub_ps(ix2,jx1);
2351             dy21             = _mm_sub_ps(iy2,jy1);
2352             dz21             = _mm_sub_ps(iz2,jz1);
2353             dx22             = _mm_sub_ps(ix2,jx2);
2354             dy22             = _mm_sub_ps(iy2,jy2);
2355             dz22             = _mm_sub_ps(iz2,jz2);
2356             dx23             = _mm_sub_ps(ix2,jx3);
2357             dy23             = _mm_sub_ps(iy2,jy3);
2358             dz23             = _mm_sub_ps(iz2,jz3);
2359             dx31             = _mm_sub_ps(ix3,jx1);
2360             dy31             = _mm_sub_ps(iy3,jy1);
2361             dz31             = _mm_sub_ps(iz3,jz1);
2362             dx32             = _mm_sub_ps(ix3,jx2);
2363             dy32             = _mm_sub_ps(iy3,jy2);
2364             dz32             = _mm_sub_ps(iz3,jz2);
2365             dx33             = _mm_sub_ps(ix3,jx3);
2366             dy33             = _mm_sub_ps(iy3,jy3);
2367             dz33             = _mm_sub_ps(iz3,jz3);
2368
2369             /* Calculate squared distance and things based on it */
2370             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
2371             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
2372             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
2373             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
2374             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
2375             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
2376             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
2377             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
2378             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
2379             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
2380
2381             rinv00           = sse2_invsqrt_f(rsq00);
2382             rinv11           = sse2_invsqrt_f(rsq11);
2383             rinv12           = sse2_invsqrt_f(rsq12);
2384             rinv13           = sse2_invsqrt_f(rsq13);
2385             rinv21           = sse2_invsqrt_f(rsq21);
2386             rinv22           = sse2_invsqrt_f(rsq22);
2387             rinv23           = sse2_invsqrt_f(rsq23);
2388             rinv31           = sse2_invsqrt_f(rsq31);
2389             rinv32           = sse2_invsqrt_f(rsq32);
2390             rinv33           = sse2_invsqrt_f(rsq33);
2391
2392             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
2393             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
2394             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
2395             rinvsq13         = _mm_mul_ps(rinv13,rinv13);
2396             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
2397             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
2398             rinvsq23         = _mm_mul_ps(rinv23,rinv23);
2399             rinvsq31         = _mm_mul_ps(rinv31,rinv31);
2400             rinvsq32         = _mm_mul_ps(rinv32,rinv32);
2401             rinvsq33         = _mm_mul_ps(rinv33,rinv33);
2402
2403             fjx0             = _mm_setzero_ps();
2404             fjy0             = _mm_setzero_ps();
2405             fjz0             = _mm_setzero_ps();
2406             fjx1             = _mm_setzero_ps();
2407             fjy1             = _mm_setzero_ps();
2408             fjz1             = _mm_setzero_ps();
2409             fjx2             = _mm_setzero_ps();
2410             fjy2             = _mm_setzero_ps();
2411             fjz2             = _mm_setzero_ps();
2412             fjx3             = _mm_setzero_ps();
2413             fjy3             = _mm_setzero_ps();
2414             fjz3             = _mm_setzero_ps();
2415
2416             /**************************
2417              * CALCULATE INTERACTIONS *
2418              **************************/
2419
2420             if (gmx_mm_any_lt(rsq00,rcutoff2))
2421             {
2422
2423             r00              = _mm_mul_ps(rsq00,rinv00);
2424             r00              = _mm_andnot_ps(dummy_mask,r00);
2425
2426             /* Analytical LJ-PME */
2427             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
2428             ewcljrsq         = _mm_mul_ps(ewclj2,rsq00);
2429             ewclj6           = _mm_mul_ps(ewclj2,_mm_mul_ps(ewclj2,ewclj2));
2430             exponent         = sse2_exp_f(ewcljrsq);
2431             /* poly = exp(-(beta*r)^2) * (1 + (beta*r)^2 + (beta*r)^4 /2) */
2432             poly             = _mm_mul_ps(exponent,_mm_add_ps(_mm_sub_ps(one,ewcljrsq),_mm_mul_ps(_mm_mul_ps(ewcljrsq,ewcljrsq),one_half)));
2433             /* f6A = 6 * C6grid * (1 - poly) */
2434             f6A              = _mm_mul_ps(c6grid_00,_mm_sub_ps(one,poly));
2435             /* f6B = C6grid * exponent * beta^6 */
2436             f6B              = _mm_mul_ps(_mm_mul_ps(c6grid_00,one_sixth),_mm_mul_ps(exponent,ewclj6));
2437             /* fvdw = 12*C12/r13 - ((6*C6 - f6A)/r6 + f6B)/r */
2438             fvdw              = _mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_sub_ps(_mm_mul_ps(c12_00,rinvsix),_mm_sub_ps(c6_00,f6A)),rinvsix),f6B),rinvsq00);
2439
2440             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
2441
2442             fscal            = fvdw;
2443
2444             fscal            = _mm_and_ps(fscal,cutoff_mask);
2445
2446             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2447
2448             /* Calculate temporary vectorial force */
2449             tx               = _mm_mul_ps(fscal,dx00);
2450             ty               = _mm_mul_ps(fscal,dy00);
2451             tz               = _mm_mul_ps(fscal,dz00);
2452
2453             /* Update vectorial force */
2454             fix0             = _mm_add_ps(fix0,tx);
2455             fiy0             = _mm_add_ps(fiy0,ty);
2456             fiz0             = _mm_add_ps(fiz0,tz);
2457
2458             fjx0             = _mm_add_ps(fjx0,tx);
2459             fjy0             = _mm_add_ps(fjy0,ty);
2460             fjz0             = _mm_add_ps(fjz0,tz);
2461
2462             }
2463
2464             /**************************
2465              * CALCULATE INTERACTIONS *
2466              **************************/
2467
2468             if (gmx_mm_any_lt(rsq11,rcutoff2))
2469             {
2470
2471             r11              = _mm_mul_ps(rsq11,rinv11);
2472             r11              = _mm_andnot_ps(dummy_mask,r11);
2473
2474             /* EWALD ELECTROSTATICS */
2475
2476             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2477             ewrt             = _mm_mul_ps(r11,ewtabscale);
2478             ewitab           = _mm_cvttps_epi32(ewrt);
2479             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2480             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2481                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2482                                          &ewtabF,&ewtabFn);
2483             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2484             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
2485
2486             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
2487
2488             fscal            = felec;
2489
2490             fscal            = _mm_and_ps(fscal,cutoff_mask);
2491
2492             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2493
2494             /* Calculate temporary vectorial force */
2495             tx               = _mm_mul_ps(fscal,dx11);
2496             ty               = _mm_mul_ps(fscal,dy11);
2497             tz               = _mm_mul_ps(fscal,dz11);
2498
2499             /* Update vectorial force */
2500             fix1             = _mm_add_ps(fix1,tx);
2501             fiy1             = _mm_add_ps(fiy1,ty);
2502             fiz1             = _mm_add_ps(fiz1,tz);
2503
2504             fjx1             = _mm_add_ps(fjx1,tx);
2505             fjy1             = _mm_add_ps(fjy1,ty);
2506             fjz1             = _mm_add_ps(fjz1,tz);
2507
2508             }
2509
2510             /**************************
2511              * CALCULATE INTERACTIONS *
2512              **************************/
2513
2514             if (gmx_mm_any_lt(rsq12,rcutoff2))
2515             {
2516
2517             r12              = _mm_mul_ps(rsq12,rinv12);
2518             r12              = _mm_andnot_ps(dummy_mask,r12);
2519
2520             /* EWALD ELECTROSTATICS */
2521
2522             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2523             ewrt             = _mm_mul_ps(r12,ewtabscale);
2524             ewitab           = _mm_cvttps_epi32(ewrt);
2525             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2526             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2527                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2528                                          &ewtabF,&ewtabFn);
2529             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2530             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
2531
2532             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
2533
2534             fscal            = felec;
2535
2536             fscal            = _mm_and_ps(fscal,cutoff_mask);
2537
2538             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2539
2540             /* Calculate temporary vectorial force */
2541             tx               = _mm_mul_ps(fscal,dx12);
2542             ty               = _mm_mul_ps(fscal,dy12);
2543             tz               = _mm_mul_ps(fscal,dz12);
2544
2545             /* Update vectorial force */
2546             fix1             = _mm_add_ps(fix1,tx);
2547             fiy1             = _mm_add_ps(fiy1,ty);
2548             fiz1             = _mm_add_ps(fiz1,tz);
2549
2550             fjx2             = _mm_add_ps(fjx2,tx);
2551             fjy2             = _mm_add_ps(fjy2,ty);
2552             fjz2             = _mm_add_ps(fjz2,tz);
2553
2554             }
2555
2556             /**************************
2557              * CALCULATE INTERACTIONS *
2558              **************************/
2559
2560             if (gmx_mm_any_lt(rsq13,rcutoff2))
2561             {
2562
2563             r13              = _mm_mul_ps(rsq13,rinv13);
2564             r13              = _mm_andnot_ps(dummy_mask,r13);
2565
2566             /* EWALD ELECTROSTATICS */
2567
2568             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2569             ewrt             = _mm_mul_ps(r13,ewtabscale);
2570             ewitab           = _mm_cvttps_epi32(ewrt);
2571             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2572             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2573                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2574                                          &ewtabF,&ewtabFn);
2575             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2576             felec            = _mm_mul_ps(_mm_mul_ps(qq13,rinv13),_mm_sub_ps(rinvsq13,felec));
2577
2578             cutoff_mask      = _mm_cmplt_ps(rsq13,rcutoff2);
2579
2580             fscal            = felec;
2581
2582             fscal            = _mm_and_ps(fscal,cutoff_mask);
2583
2584             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2585
2586             /* Calculate temporary vectorial force */
2587             tx               = _mm_mul_ps(fscal,dx13);
2588             ty               = _mm_mul_ps(fscal,dy13);
2589             tz               = _mm_mul_ps(fscal,dz13);
2590
2591             /* Update vectorial force */
2592             fix1             = _mm_add_ps(fix1,tx);
2593             fiy1             = _mm_add_ps(fiy1,ty);
2594             fiz1             = _mm_add_ps(fiz1,tz);
2595
2596             fjx3             = _mm_add_ps(fjx3,tx);
2597             fjy3             = _mm_add_ps(fjy3,ty);
2598             fjz3             = _mm_add_ps(fjz3,tz);
2599
2600             }
2601
2602             /**************************
2603              * CALCULATE INTERACTIONS *
2604              **************************/
2605
2606             if (gmx_mm_any_lt(rsq21,rcutoff2))
2607             {
2608
2609             r21              = _mm_mul_ps(rsq21,rinv21);
2610             r21              = _mm_andnot_ps(dummy_mask,r21);
2611
2612             /* EWALD ELECTROSTATICS */
2613
2614             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2615             ewrt             = _mm_mul_ps(r21,ewtabscale);
2616             ewitab           = _mm_cvttps_epi32(ewrt);
2617             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2618             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2619                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2620                                          &ewtabF,&ewtabFn);
2621             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2622             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
2623
2624             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
2625
2626             fscal            = felec;
2627
2628             fscal            = _mm_and_ps(fscal,cutoff_mask);
2629
2630             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2631
2632             /* Calculate temporary vectorial force */
2633             tx               = _mm_mul_ps(fscal,dx21);
2634             ty               = _mm_mul_ps(fscal,dy21);
2635             tz               = _mm_mul_ps(fscal,dz21);
2636
2637             /* Update vectorial force */
2638             fix2             = _mm_add_ps(fix2,tx);
2639             fiy2             = _mm_add_ps(fiy2,ty);
2640             fiz2             = _mm_add_ps(fiz2,tz);
2641
2642             fjx1             = _mm_add_ps(fjx1,tx);
2643             fjy1             = _mm_add_ps(fjy1,ty);
2644             fjz1             = _mm_add_ps(fjz1,tz);
2645
2646             }
2647
2648             /**************************
2649              * CALCULATE INTERACTIONS *
2650              **************************/
2651
2652             if (gmx_mm_any_lt(rsq22,rcutoff2))
2653             {
2654
2655             r22              = _mm_mul_ps(rsq22,rinv22);
2656             r22              = _mm_andnot_ps(dummy_mask,r22);
2657
2658             /* EWALD ELECTROSTATICS */
2659
2660             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2661             ewrt             = _mm_mul_ps(r22,ewtabscale);
2662             ewitab           = _mm_cvttps_epi32(ewrt);
2663             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2664             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2665                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2666                                          &ewtabF,&ewtabFn);
2667             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2668             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
2669
2670             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
2671
2672             fscal            = felec;
2673
2674             fscal            = _mm_and_ps(fscal,cutoff_mask);
2675
2676             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2677
2678             /* Calculate temporary vectorial force */
2679             tx               = _mm_mul_ps(fscal,dx22);
2680             ty               = _mm_mul_ps(fscal,dy22);
2681             tz               = _mm_mul_ps(fscal,dz22);
2682
2683             /* Update vectorial force */
2684             fix2             = _mm_add_ps(fix2,tx);
2685             fiy2             = _mm_add_ps(fiy2,ty);
2686             fiz2             = _mm_add_ps(fiz2,tz);
2687
2688             fjx2             = _mm_add_ps(fjx2,tx);
2689             fjy2             = _mm_add_ps(fjy2,ty);
2690             fjz2             = _mm_add_ps(fjz2,tz);
2691
2692             }
2693
2694             /**************************
2695              * CALCULATE INTERACTIONS *
2696              **************************/
2697
2698             if (gmx_mm_any_lt(rsq23,rcutoff2))
2699             {
2700
2701             r23              = _mm_mul_ps(rsq23,rinv23);
2702             r23              = _mm_andnot_ps(dummy_mask,r23);
2703
2704             /* EWALD ELECTROSTATICS */
2705
2706             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2707             ewrt             = _mm_mul_ps(r23,ewtabscale);
2708             ewitab           = _mm_cvttps_epi32(ewrt);
2709             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2710             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2711                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2712                                          &ewtabF,&ewtabFn);
2713             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2714             felec            = _mm_mul_ps(_mm_mul_ps(qq23,rinv23),_mm_sub_ps(rinvsq23,felec));
2715
2716             cutoff_mask      = _mm_cmplt_ps(rsq23,rcutoff2);
2717
2718             fscal            = felec;
2719
2720             fscal            = _mm_and_ps(fscal,cutoff_mask);
2721
2722             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2723
2724             /* Calculate temporary vectorial force */
2725             tx               = _mm_mul_ps(fscal,dx23);
2726             ty               = _mm_mul_ps(fscal,dy23);
2727             tz               = _mm_mul_ps(fscal,dz23);
2728
2729             /* Update vectorial force */
2730             fix2             = _mm_add_ps(fix2,tx);
2731             fiy2             = _mm_add_ps(fiy2,ty);
2732             fiz2             = _mm_add_ps(fiz2,tz);
2733
2734             fjx3             = _mm_add_ps(fjx3,tx);
2735             fjy3             = _mm_add_ps(fjy3,ty);
2736             fjz3             = _mm_add_ps(fjz3,tz);
2737
2738             }
2739
2740             /**************************
2741              * CALCULATE INTERACTIONS *
2742              **************************/
2743
2744             if (gmx_mm_any_lt(rsq31,rcutoff2))
2745             {
2746
2747             r31              = _mm_mul_ps(rsq31,rinv31);
2748             r31              = _mm_andnot_ps(dummy_mask,r31);
2749
2750             /* EWALD ELECTROSTATICS */
2751
2752             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2753             ewrt             = _mm_mul_ps(r31,ewtabscale);
2754             ewitab           = _mm_cvttps_epi32(ewrt);
2755             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2756             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2757                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2758                                          &ewtabF,&ewtabFn);
2759             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2760             felec            = _mm_mul_ps(_mm_mul_ps(qq31,rinv31),_mm_sub_ps(rinvsq31,felec));
2761
2762             cutoff_mask      = _mm_cmplt_ps(rsq31,rcutoff2);
2763
2764             fscal            = felec;
2765
2766             fscal            = _mm_and_ps(fscal,cutoff_mask);
2767
2768             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2769
2770             /* Calculate temporary vectorial force */
2771             tx               = _mm_mul_ps(fscal,dx31);
2772             ty               = _mm_mul_ps(fscal,dy31);
2773             tz               = _mm_mul_ps(fscal,dz31);
2774
2775             /* Update vectorial force */
2776             fix3             = _mm_add_ps(fix3,tx);
2777             fiy3             = _mm_add_ps(fiy3,ty);
2778             fiz3             = _mm_add_ps(fiz3,tz);
2779
2780             fjx1             = _mm_add_ps(fjx1,tx);
2781             fjy1             = _mm_add_ps(fjy1,ty);
2782             fjz1             = _mm_add_ps(fjz1,tz);
2783
2784             }
2785
2786             /**************************
2787              * CALCULATE INTERACTIONS *
2788              **************************/
2789
2790             if (gmx_mm_any_lt(rsq32,rcutoff2))
2791             {
2792
2793             r32              = _mm_mul_ps(rsq32,rinv32);
2794             r32              = _mm_andnot_ps(dummy_mask,r32);
2795
2796             /* EWALD ELECTROSTATICS */
2797
2798             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2799             ewrt             = _mm_mul_ps(r32,ewtabscale);
2800             ewitab           = _mm_cvttps_epi32(ewrt);
2801             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2802             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2803                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2804                                          &ewtabF,&ewtabFn);
2805             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2806             felec            = _mm_mul_ps(_mm_mul_ps(qq32,rinv32),_mm_sub_ps(rinvsq32,felec));
2807
2808             cutoff_mask      = _mm_cmplt_ps(rsq32,rcutoff2);
2809
2810             fscal            = felec;
2811
2812             fscal            = _mm_and_ps(fscal,cutoff_mask);
2813
2814             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2815
2816             /* Calculate temporary vectorial force */
2817             tx               = _mm_mul_ps(fscal,dx32);
2818             ty               = _mm_mul_ps(fscal,dy32);
2819             tz               = _mm_mul_ps(fscal,dz32);
2820
2821             /* Update vectorial force */
2822             fix3             = _mm_add_ps(fix3,tx);
2823             fiy3             = _mm_add_ps(fiy3,ty);
2824             fiz3             = _mm_add_ps(fiz3,tz);
2825
2826             fjx2             = _mm_add_ps(fjx2,tx);
2827             fjy2             = _mm_add_ps(fjy2,ty);
2828             fjz2             = _mm_add_ps(fjz2,tz);
2829
2830             }
2831
2832             /**************************
2833              * CALCULATE INTERACTIONS *
2834              **************************/
2835
2836             if (gmx_mm_any_lt(rsq33,rcutoff2))
2837             {
2838
2839             r33              = _mm_mul_ps(rsq33,rinv33);
2840             r33              = _mm_andnot_ps(dummy_mask,r33);
2841
2842             /* EWALD ELECTROSTATICS */
2843
2844             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2845             ewrt             = _mm_mul_ps(r33,ewtabscale);
2846             ewitab           = _mm_cvttps_epi32(ewrt);
2847             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2848             gmx_mm_load_4pair_swizzle_ps(ewtab+gmx_mm_extract_epi32(ewitab,0),ewtab+gmx_mm_extract_epi32(ewitab,1),
2849                                          ewtab+gmx_mm_extract_epi32(ewitab,2),ewtab+gmx_mm_extract_epi32(ewitab,3),
2850                                          &ewtabF,&ewtabFn);
2851             felec            = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn));
2852             felec            = _mm_mul_ps(_mm_mul_ps(qq33,rinv33),_mm_sub_ps(rinvsq33,felec));
2853
2854             cutoff_mask      = _mm_cmplt_ps(rsq33,rcutoff2);
2855
2856             fscal            = felec;
2857
2858             fscal            = _mm_and_ps(fscal,cutoff_mask);
2859
2860             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2861
2862             /* Calculate temporary vectorial force */
2863             tx               = _mm_mul_ps(fscal,dx33);
2864             ty               = _mm_mul_ps(fscal,dy33);
2865             tz               = _mm_mul_ps(fscal,dz33);
2866
2867             /* Update vectorial force */
2868             fix3             = _mm_add_ps(fix3,tx);
2869             fiy3             = _mm_add_ps(fiy3,ty);
2870             fiz3             = _mm_add_ps(fiz3,tz);
2871
2872             fjx3             = _mm_add_ps(fjx3,tx);
2873             fjy3             = _mm_add_ps(fjy3,ty);
2874             fjz3             = _mm_add_ps(fjz3,tz);
2875
2876             }
2877
2878             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2879             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2880             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2881             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2882
2883             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2884                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2885                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2886
2887             /* Inner loop uses 413 flops */
2888         }
2889
2890         /* End of innermost loop */
2891
2892         gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2893                                               f+i_coord_offset,fshift+i_shift_offset);
2894
2895         /* Increment number of inner iterations */
2896         inneriter                  += j_index_end - j_index_start;
2897
2898         /* Outer loop uses 24 flops */
2899     }
2900
2901     /* Increment number of outer iterations */
2902     outeriter        += nri;
2903
2904     /* Update outer/inner flops */
2905
2906     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*413);
2907 }