src/gromacs/gmxlib/nonbonded/nb_kernel_sse2_single/nb_kernel_ElecEwSw_VdwLJSw_GeomW3W3_sse2_single.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS 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/legacyheaders/types/simple.h"
  46 #include "gromacs/math/vec.h"
  47 #include "gromacs/legacyheaders/nrnb.h"
  48
  49 #include "gromacs/simd/math_x86_sse2_single.h"
  50 #include "kernelutil_x86_sse2_single.h"
  51
  52 /*
  53  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwLJSw_GeomW3W3_VF_sse2_single
  54  * Electrostatics interaction: Ewald
  55  * VdW interaction:            LennardJones
  56  * Geometry:                   Water3-Water3
  57  * Calculate force/pot:        PotentialAndForce
  58  */
  59 void
  60 nb_kernel_ElecEwSw_VdwLJSw_GeomW3W3_VF_sse2_single
  61                     (t_nblist                    * gmx_restrict       nlist,
  62                      rvec                        * gmx_restrict          xx,
  63                      rvec                        * gmx_restrict          ff,
  64                      t_forcerec                  * gmx_restrict          fr,
  65                      t_mdatoms                   * gmx_restrict     mdatoms,
  66                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  67                      t_nrnb                      * gmx_restrict        nrnb)
  68 {
  69     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  70      * just 0 for non-waters.
  71      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
  72      * jnr indices corresponding to data put in the four positions in the SIMD register.
  73      */
  74     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  75     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  76     int              jnrA,jnrB,jnrC,jnrD;
  77     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  78     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  79     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  80     real             rcutoff_scalar;
  81     real             *shiftvec,*fshift,*x,*f;
  82     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  83     real             scratch[4*DIM];
  84     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  85     int              vdwioffset0;
  86     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  87     int              vdwioffset1;
  88     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  89     int              vdwioffset2;
  90     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  91     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  92     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  93     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  94     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  95     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  96     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  97     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  98     __m128           dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
  99     __m128           dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
 100     __m128           dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 101     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 102     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 103     __m128           dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 104     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 105     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 106     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
 107     real             *charge;
 108     int              nvdwtype;
 109     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 110     int              *vdwtype;
 111     real             *vdwparam;
 112     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
 113     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
 114     __m128i          ewitab;
 115     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 116     real             *ewtab;
 117     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
 118     real             rswitch_scalar,d_scalar;
 119     __m128           dummy_mask,cutoff_mask;
 120     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 121     __m128           one     = _mm_set1_ps(1.0);
 122     __m128           two     = _mm_set1_ps(2.0);
 123     x                = xx[0];
 124     f                = ff[0];
 125
 126     nri              = nlist->nri;
 127     iinr             = nlist->iinr;
 128     jindex           = nlist->jindex;
 129     jjnr             = nlist->jjnr;
 130     shiftidx         = nlist->shift;
 131     gid              = nlist->gid;
 132     shiftvec         = fr->shift_vec[0];
 133     fshift           = fr->fshift[0];
 134     facel            = _mm_set1_ps(fr->epsfac);
 135     charge           = mdatoms->chargeA;
 136     nvdwtype         = fr->ntype;
 137     vdwparam         = fr->nbfp;
 138     vdwtype          = mdatoms->typeA;
 139
 140     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 141     ewtab            = fr->ic->tabq_coul_FDV0;
 142     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 143     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 144
 145     /* Setup water-specific parameters */
 146     inr              = nlist->iinr[0];
 147     iq0              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+0]));
 148     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 149     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 150     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 151
 152     jq0              = _mm_set1_ps(charge[inr+0]);
 153     jq1              = _mm_set1_ps(charge[inr+1]);
 154     jq2              = _mm_set1_ps(charge[inr+2]);
 155     vdwjidx0A        = 2*vdwtype[inr+0];
 156     qq00             = _mm_mul_ps(iq0,jq0);
 157     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
 158     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
 159     qq01             = _mm_mul_ps(iq0,jq1);
 160     qq02             = _mm_mul_ps(iq0,jq2);
 161     qq10             = _mm_mul_ps(iq1,jq0);
 162     qq11             = _mm_mul_ps(iq1,jq1);
 163     qq12             = _mm_mul_ps(iq1,jq2);
 164     qq20             = _mm_mul_ps(iq2,jq0);
 165     qq21             = _mm_mul_ps(iq2,jq1);
 166     qq22             = _mm_mul_ps(iq2,jq2);
 167
 168     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 169     rcutoff_scalar   = fr->rcoulomb;
 170     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 171     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 172
 173     rswitch_scalar   = fr->rcoulomb_switch;
 174     rswitch          = _mm_set1_ps(rswitch_scalar);
 175     /* Setup switch parameters */
 176     d_scalar         = rcutoff_scalar-rswitch_scalar;
 177     d                = _mm_set1_ps(d_scalar);
 178     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
 179     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 180     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 181     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
 182     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 183     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 184
 185     /* Avoid stupid compiler warnings */
 186     jnrA = jnrB = jnrC = jnrD = 0;
 187     j_coord_offsetA = 0;
 188     j_coord_offsetB = 0;
 189     j_coord_offsetC = 0;
 190     j_coord_offsetD = 0;
 191
 192     outeriter        = 0;
 193     inneriter        = 0;
 194
 195     for(iidx=0;iidx<4*DIM;iidx++)
 196     {
 197         scratch[iidx] = 0.0;
 198     }
 199
 200     /* Start outer loop over neighborlists */
 201     for(iidx=0; iidx<nri; iidx++)
 202     {
 203         /* Load shift vector for this list */
 204         i_shift_offset   = DIM*shiftidx[iidx];
 205
 206         /* Load limits for loop over neighbors */
 207         j_index_start    = jindex[iidx];
 208         j_index_end      = jindex[iidx+1];
 209
 210         /* Get outer coordinate index */
 211         inr              = iinr[iidx];
 212         i_coord_offset   = DIM*inr;
 213
 214         /* Load i particle coords and add shift vector */
 215         gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
 216                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 217
 218         fix0             = _mm_setzero_ps();
 219         fiy0             = _mm_setzero_ps();
 220         fiz0             = _mm_setzero_ps();
 221         fix1             = _mm_setzero_ps();
 222         fiy1             = _mm_setzero_ps();
 223         fiz1             = _mm_setzero_ps();
 224         fix2             = _mm_setzero_ps();
 225         fiy2             = _mm_setzero_ps();
 226         fiz2             = _mm_setzero_ps();
 227
 228         /* Reset potential sums */
 229         velecsum         = _mm_setzero_ps();
 230         vvdwsum          = _mm_setzero_ps();
 231
 232         /* Start inner kernel loop */
 233         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 234         {
 235
 236             /* Get j neighbor index, and coordinate index */
 237             jnrA             = jjnr[jidx];
 238             jnrB             = jjnr[jidx+1];
 239             jnrC             = jjnr[jidx+2];
 240             jnrD             = jjnr[jidx+3];
 241             j_coord_offsetA  = DIM*jnrA;
 242             j_coord_offsetB  = DIM*jnrB;
 243             j_coord_offsetC  = DIM*jnrC;
 244             j_coord_offsetD  = DIM*jnrD;
 245
 246             /* load j atom coordinates */
 247             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 248                                               x+j_coord_offsetC,x+j_coord_offsetD,
 249                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 250
 251             /* Calculate displacement vector */
 252             dx00             = _mm_sub_ps(ix0,jx0);
 253             dy00             = _mm_sub_ps(iy0,jy0);
 254             dz00             = _mm_sub_ps(iz0,jz0);
 255             dx01             = _mm_sub_ps(ix0,jx1);
 256             dy01             = _mm_sub_ps(iy0,jy1);
 257             dz01             = _mm_sub_ps(iz0,jz1);
 258             dx02             = _mm_sub_ps(ix0,jx2);
 259             dy02             = _mm_sub_ps(iy0,jy2);
 260             dz02             = _mm_sub_ps(iz0,jz2);
 261             dx10             = _mm_sub_ps(ix1,jx0);
 262             dy10             = _mm_sub_ps(iy1,jy0);
 263             dz10             = _mm_sub_ps(iz1,jz0);
 264             dx11             = _mm_sub_ps(ix1,jx1);
 265             dy11             = _mm_sub_ps(iy1,jy1);
 266             dz11             = _mm_sub_ps(iz1,jz1);
 267             dx12             = _mm_sub_ps(ix1,jx2);
 268             dy12             = _mm_sub_ps(iy1,jy2);
 269             dz12             = _mm_sub_ps(iz1,jz2);
 270             dx20             = _mm_sub_ps(ix2,jx0);
 271             dy20             = _mm_sub_ps(iy2,jy0);
 272             dz20             = _mm_sub_ps(iz2,jz0);
 273             dx21             = _mm_sub_ps(ix2,jx1);
 274             dy21             = _mm_sub_ps(iy2,jy1);
 275             dz21             = _mm_sub_ps(iz2,jz1);
 276             dx22             = _mm_sub_ps(ix2,jx2);
 277             dy22             = _mm_sub_ps(iy2,jy2);
 278             dz22             = _mm_sub_ps(iz2,jz2);
 279
 280             /* Calculate squared distance and things based on it */
 281             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 282             rsq01            = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
 283             rsq02            = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
 284             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
 285             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 286             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 287             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
 288             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 289             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 290
 291             rinv00           = gmx_mm_invsqrt_ps(rsq00);
 292             rinv01           = gmx_mm_invsqrt_ps(rsq01);
 293             rinv02           = gmx_mm_invsqrt_ps(rsq02);
 294             rinv10           = gmx_mm_invsqrt_ps(rsq10);
 295             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 296             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 297             rinv20           = gmx_mm_invsqrt_ps(rsq20);
 298             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 299             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 300
 301             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 302             rinvsq01         = _mm_mul_ps(rinv01,rinv01);
 303             rinvsq02         = _mm_mul_ps(rinv02,rinv02);
 304             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
 305             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
 306             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
 307             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
 308             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
 309             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
 310
 311             fjx0             = _mm_setzero_ps();
 312             fjy0             = _mm_setzero_ps();
 313             fjz0             = _mm_setzero_ps();
 314             fjx1             = _mm_setzero_ps();
 315             fjy1             = _mm_setzero_ps();
 316             fjz1             = _mm_setzero_ps();
 317             fjx2             = _mm_setzero_ps();
 318             fjy2             = _mm_setzero_ps();
 319             fjz2             = _mm_setzero_ps();
 320
 321             /**************************
 322              * CALCULATE INTERACTIONS *
 323              **************************/
 324
 325             if (gmx_mm_any_lt(rsq00,rcutoff2))
 326             {
 327
 328             r00              = _mm_mul_ps(rsq00,rinv00);
 329
 330             /* EWALD ELECTROSTATICS */
 331
 332             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 333             ewrt             = _mm_mul_ps(r00,ewtabscale);
 334             ewitab           = _mm_cvttps_epi32(ewrt);
 335             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 336             ewitab           = _mm_slli_epi32(ewitab,2);
 337             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 338             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 339             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 340             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 341             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 342             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 343             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 344             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
 345             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
 346
 347             /* LENNARD-JONES DISPERSION/REPULSION */
 348
 349             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 350             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 351             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 352             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
 353             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 354
 355             d                = _mm_sub_ps(r00,rswitch);
 356             d                = _mm_max_ps(d,_mm_setzero_ps());
 357             d2               = _mm_mul_ps(d,d);
 358             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 359
 360             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 361
 362             /* Evaluate switch function */
 363             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 364             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
 365             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
 366             velec            = _mm_mul_ps(velec,sw);
 367             vvdw             = _mm_mul_ps(vvdw,sw);
 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             velec            = _mm_and_ps(velec,cutoff_mask);
 372             velecsum         = _mm_add_ps(velecsum,velec);
 373             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
 374             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 375
 376             fscal            = _mm_add_ps(felec,fvdw);
 377
 378             fscal            = _mm_and_ps(fscal,cutoff_mask);
 379
 380             /* Calculate temporary vectorial force */
 381             tx               = _mm_mul_ps(fscal,dx00);
 382             ty               = _mm_mul_ps(fscal,dy00);
 383             tz               = _mm_mul_ps(fscal,dz00);
 384
 385             /* Update vectorial force */
 386             fix0             = _mm_add_ps(fix0,tx);
 387             fiy0             = _mm_add_ps(fiy0,ty);
 388             fiz0             = _mm_add_ps(fiz0,tz);
 389
 390             fjx0             = _mm_add_ps(fjx0,tx);
 391             fjy0             = _mm_add_ps(fjy0,ty);
 392             fjz0             = _mm_add_ps(fjz0,tz);
 393
 394             }
 395
 396             /**************************
 397              * CALCULATE INTERACTIONS *
 398              **************************/
 399
 400             if (gmx_mm_any_lt(rsq01,rcutoff2))
 401             {
 402
 403             r01              = _mm_mul_ps(rsq01,rinv01);
 404
 405             /* EWALD ELECTROSTATICS */
 406
 407             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 408             ewrt             = _mm_mul_ps(r01,ewtabscale);
 409             ewitab           = _mm_cvttps_epi32(ewrt);
 410             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 411             ewitab           = _mm_slli_epi32(ewitab,2);
 412             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 413             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 414             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 415             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 416             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 417             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 418             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 419             velec            = _mm_mul_ps(qq01,_mm_sub_ps(rinv01,velec));
 420             felec            = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
 421
 422             d                = _mm_sub_ps(r01,rswitch);
 423             d                = _mm_max_ps(d,_mm_setzero_ps());
 424             d2               = _mm_mul_ps(d,d);
 425             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 426
 427             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 428
 429             /* Evaluate switch function */
 430             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 431             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv01,_mm_mul_ps(velec,dsw)) );
 432             velec            = _mm_mul_ps(velec,sw);
 433             cutoff_mask      = _mm_cmplt_ps(rsq01,rcutoff2);
 434
 435             /* Update potential sum for this i atom from the interaction with this j atom. */
 436             velec            = _mm_and_ps(velec,cutoff_mask);
 437             velecsum         = _mm_add_ps(velecsum,velec);
 438
 439             fscal            = felec;
 440
 441             fscal            = _mm_and_ps(fscal,cutoff_mask);
 442
 443             /* Calculate temporary vectorial force */
 444             tx               = _mm_mul_ps(fscal,dx01);
 445             ty               = _mm_mul_ps(fscal,dy01);
 446             tz               = _mm_mul_ps(fscal,dz01);
 447
 448             /* Update vectorial force */
 449             fix0             = _mm_add_ps(fix0,tx);
 450             fiy0             = _mm_add_ps(fiy0,ty);
 451             fiz0             = _mm_add_ps(fiz0,tz);
 452
 453             fjx1             = _mm_add_ps(fjx1,tx);
 454             fjy1             = _mm_add_ps(fjy1,ty);
 455             fjz1             = _mm_add_ps(fjz1,tz);
 456
 457             }
 458
 459             /**************************
 460              * CALCULATE INTERACTIONS *
 461              **************************/
 462
 463             if (gmx_mm_any_lt(rsq02,rcutoff2))
 464             {
 465
 466             r02              = _mm_mul_ps(rsq02,rinv02);
 467
 468             /* EWALD ELECTROSTATICS */
 469
 470             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 471             ewrt             = _mm_mul_ps(r02,ewtabscale);
 472             ewitab           = _mm_cvttps_epi32(ewrt);
 473             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 474             ewitab           = _mm_slli_epi32(ewitab,2);
 475             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 476             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 477             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 478             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 479             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 480             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 481             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 482             velec            = _mm_mul_ps(qq02,_mm_sub_ps(rinv02,velec));
 483             felec            = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
 484
 485             d                = _mm_sub_ps(r02,rswitch);
 486             d                = _mm_max_ps(d,_mm_setzero_ps());
 487             d2               = _mm_mul_ps(d,d);
 488             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 489
 490             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 491
 492             /* Evaluate switch function */
 493             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 494             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv02,_mm_mul_ps(velec,dsw)) );
 495             velec            = _mm_mul_ps(velec,sw);
 496             cutoff_mask      = _mm_cmplt_ps(rsq02,rcutoff2);
 497
 498             /* Update potential sum for this i atom from the interaction with this j atom. */
 499             velec            = _mm_and_ps(velec,cutoff_mask);
 500             velecsum         = _mm_add_ps(velecsum,velec);
 501
 502             fscal            = felec;
 503
 504             fscal            = _mm_and_ps(fscal,cutoff_mask);
 505
 506             /* Calculate temporary vectorial force */
 507             tx               = _mm_mul_ps(fscal,dx02);
 508             ty               = _mm_mul_ps(fscal,dy02);
 509             tz               = _mm_mul_ps(fscal,dz02);
 510
 511             /* Update vectorial force */
 512             fix0             = _mm_add_ps(fix0,tx);
 513             fiy0             = _mm_add_ps(fiy0,ty);
 514             fiz0             = _mm_add_ps(fiz0,tz);
 515
 516             fjx2             = _mm_add_ps(fjx2,tx);
 517             fjy2             = _mm_add_ps(fjy2,ty);
 518             fjz2             = _mm_add_ps(fjz2,tz);
 519
 520             }
 521
 522             /**************************
 523              * CALCULATE INTERACTIONS *
 524              **************************/
 525
 526             if (gmx_mm_any_lt(rsq10,rcutoff2))
 527             {
 528
 529             r10              = _mm_mul_ps(rsq10,rinv10);
 530
 531             /* EWALD ELECTROSTATICS */
 532
 533             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 534             ewrt             = _mm_mul_ps(r10,ewtabscale);
 535             ewitab           = _mm_cvttps_epi32(ewrt);
 536             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 537             ewitab           = _mm_slli_epi32(ewitab,2);
 538             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 539             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 540             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 541             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 542             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 543             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 544             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 545             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
 546             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
 547
 548             d                = _mm_sub_ps(r10,rswitch);
 549             d                = _mm_max_ps(d,_mm_setzero_ps());
 550             d2               = _mm_mul_ps(d,d);
 551             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 552
 553             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 554
 555             /* Evaluate switch function */
 556             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 557             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
 558             velec            = _mm_mul_ps(velec,sw);
 559             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
 560
 561             /* Update potential sum for this i atom from the interaction with this j atom. */
 562             velec            = _mm_and_ps(velec,cutoff_mask);
 563             velecsum         = _mm_add_ps(velecsum,velec);
 564
 565             fscal            = felec;
 566
 567             fscal            = _mm_and_ps(fscal,cutoff_mask);
 568
 569             /* Calculate temporary vectorial force */
 570             tx               = _mm_mul_ps(fscal,dx10);
 571             ty               = _mm_mul_ps(fscal,dy10);
 572             tz               = _mm_mul_ps(fscal,dz10);
 573
 574             /* Update vectorial force */
 575             fix1             = _mm_add_ps(fix1,tx);
 576             fiy1             = _mm_add_ps(fiy1,ty);
 577             fiz1             = _mm_add_ps(fiz1,tz);
 578
 579             fjx0             = _mm_add_ps(fjx0,tx);
 580             fjy0             = _mm_add_ps(fjy0,ty);
 581             fjz0             = _mm_add_ps(fjz0,tz);
 582
 583             }
 584
 585             /**************************
 586              * CALCULATE INTERACTIONS *
 587              **************************/
 588
 589             if (gmx_mm_any_lt(rsq11,rcutoff2))
 590             {
 591
 592             r11              = _mm_mul_ps(rsq11,rinv11);
 593
 594             /* EWALD ELECTROSTATICS */
 595
 596             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 597             ewrt             = _mm_mul_ps(r11,ewtabscale);
 598             ewitab           = _mm_cvttps_epi32(ewrt);
 599             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 600             ewitab           = _mm_slli_epi32(ewitab,2);
 601             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 602             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 603             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 604             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 605             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 606             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 607             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 608             velec            = _mm_mul_ps(qq11,_mm_sub_ps(rinv11,velec));
 609             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
 610
 611             d                = _mm_sub_ps(r11,rswitch);
 612             d                = _mm_max_ps(d,_mm_setzero_ps());
 613             d2               = _mm_mul_ps(d,d);
 614             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 615
 616             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 617
 618             /* Evaluate switch function */
 619             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 620             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv11,_mm_mul_ps(velec,dsw)) );
 621             velec            = _mm_mul_ps(velec,sw);
 622             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
 623
 624             /* Update potential sum for this i atom from the interaction with this j atom. */
 625             velec            = _mm_and_ps(velec,cutoff_mask);
 626             velecsum         = _mm_add_ps(velecsum,velec);
 627
 628             fscal            = felec;
 629
 630             fscal            = _mm_and_ps(fscal,cutoff_mask);
 631
 632             /* Calculate temporary vectorial force */
 633             tx               = _mm_mul_ps(fscal,dx11);
 634             ty               = _mm_mul_ps(fscal,dy11);
 635             tz               = _mm_mul_ps(fscal,dz11);
 636
 637             /* Update vectorial force */
 638             fix1             = _mm_add_ps(fix1,tx);
 639             fiy1             = _mm_add_ps(fiy1,ty);
 640             fiz1             = _mm_add_ps(fiz1,tz);
 641
 642             fjx1             = _mm_add_ps(fjx1,tx);
 643             fjy1             = _mm_add_ps(fjy1,ty);
 644             fjz1             = _mm_add_ps(fjz1,tz);
 645
 646             }
 647
 648             /**************************
 649              * CALCULATE INTERACTIONS *
 650              **************************/
 651
 652             if (gmx_mm_any_lt(rsq12,rcutoff2))
 653             {
 654
 655             r12              = _mm_mul_ps(rsq12,rinv12);
 656
 657             /* EWALD ELECTROSTATICS */
 658
 659             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 660             ewrt             = _mm_mul_ps(r12,ewtabscale);
 661             ewitab           = _mm_cvttps_epi32(ewrt);
 662             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 663             ewitab           = _mm_slli_epi32(ewitab,2);
 664             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 665             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 666             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 667             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 668             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 669             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 670             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 671             velec            = _mm_mul_ps(qq12,_mm_sub_ps(rinv12,velec));
 672             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
 673
 674             d                = _mm_sub_ps(r12,rswitch);
 675             d                = _mm_max_ps(d,_mm_setzero_ps());
 676             d2               = _mm_mul_ps(d,d);
 677             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 678
 679             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 680
 681             /* Evaluate switch function */
 682             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 683             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv12,_mm_mul_ps(velec,dsw)) );
 684             velec            = _mm_mul_ps(velec,sw);
 685             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
 686
 687             /* Update potential sum for this i atom from the interaction with this j atom. */
 688             velec            = _mm_and_ps(velec,cutoff_mask);
 689             velecsum         = _mm_add_ps(velecsum,velec);
 690
 691             fscal            = felec;
 692
 693             fscal            = _mm_and_ps(fscal,cutoff_mask);
 694
 695             /* Calculate temporary vectorial force */
 696             tx               = _mm_mul_ps(fscal,dx12);
 697             ty               = _mm_mul_ps(fscal,dy12);
 698             tz               = _mm_mul_ps(fscal,dz12);
 699
 700             /* Update vectorial force */
 701             fix1             = _mm_add_ps(fix1,tx);
 702             fiy1             = _mm_add_ps(fiy1,ty);
 703             fiz1             = _mm_add_ps(fiz1,tz);
 704
 705             fjx2             = _mm_add_ps(fjx2,tx);
 706             fjy2             = _mm_add_ps(fjy2,ty);
 707             fjz2             = _mm_add_ps(fjz2,tz);
 708
 709             }
 710
 711             /**************************
 712              * CALCULATE INTERACTIONS *
 713              **************************/
 714
 715             if (gmx_mm_any_lt(rsq20,rcutoff2))
 716             {
 717
 718             r20              = _mm_mul_ps(rsq20,rinv20);
 719
 720             /* EWALD ELECTROSTATICS */
 721
 722             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 723             ewrt             = _mm_mul_ps(r20,ewtabscale);
 724             ewitab           = _mm_cvttps_epi32(ewrt);
 725             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 726             ewitab           = _mm_slli_epi32(ewitab,2);
 727             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 728             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 729             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 730             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 731             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 732             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 733             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 734             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
 735             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
 736
 737             d                = _mm_sub_ps(r20,rswitch);
 738             d                = _mm_max_ps(d,_mm_setzero_ps());
 739             d2               = _mm_mul_ps(d,d);
 740             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 741
 742             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 743
 744             /* Evaluate switch function */
 745             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 746             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
 747             velec            = _mm_mul_ps(velec,sw);
 748             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
 749
 750             /* Update potential sum for this i atom from the interaction with this j atom. */
 751             velec            = _mm_and_ps(velec,cutoff_mask);
 752             velecsum         = _mm_add_ps(velecsum,velec);
 753
 754             fscal            = felec;
 755
 756             fscal            = _mm_and_ps(fscal,cutoff_mask);
 757
 758             /* Calculate temporary vectorial force */
 759             tx               = _mm_mul_ps(fscal,dx20);
 760             ty               = _mm_mul_ps(fscal,dy20);
 761             tz               = _mm_mul_ps(fscal,dz20);
 762
 763             /* Update vectorial force */
 764             fix2             = _mm_add_ps(fix2,tx);
 765             fiy2             = _mm_add_ps(fiy2,ty);
 766             fiz2             = _mm_add_ps(fiz2,tz);
 767
 768             fjx0             = _mm_add_ps(fjx0,tx);
 769             fjy0             = _mm_add_ps(fjy0,ty);
 770             fjz0             = _mm_add_ps(fjz0,tz);
 771
 772             }
 773
 774             /**************************
 775              * CALCULATE INTERACTIONS *
 776              **************************/
 777
 778             if (gmx_mm_any_lt(rsq21,rcutoff2))
 779             {
 780
 781             r21              = _mm_mul_ps(rsq21,rinv21);
 782
 783             /* EWALD ELECTROSTATICS */
 784
 785             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 786             ewrt             = _mm_mul_ps(r21,ewtabscale);
 787             ewitab           = _mm_cvttps_epi32(ewrt);
 788             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 789             ewitab           = _mm_slli_epi32(ewitab,2);
 790             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 791             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 792             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 793             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 794             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 795             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 796             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 797             velec            = _mm_mul_ps(qq21,_mm_sub_ps(rinv21,velec));
 798             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
 799
 800             d                = _mm_sub_ps(r21,rswitch);
 801             d                = _mm_max_ps(d,_mm_setzero_ps());
 802             d2               = _mm_mul_ps(d,d);
 803             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 804
 805             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 806
 807             /* Evaluate switch function */
 808             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 809             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv21,_mm_mul_ps(velec,dsw)) );
 810             velec            = _mm_mul_ps(velec,sw);
 811             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
 812
 813             /* Update potential sum for this i atom from the interaction with this j atom. */
 814             velec            = _mm_and_ps(velec,cutoff_mask);
 815             velecsum         = _mm_add_ps(velecsum,velec);
 816
 817             fscal            = felec;
 818
 819             fscal            = _mm_and_ps(fscal,cutoff_mask);
 820
 821             /* Calculate temporary vectorial force */
 822             tx               = _mm_mul_ps(fscal,dx21);
 823             ty               = _mm_mul_ps(fscal,dy21);
 824             tz               = _mm_mul_ps(fscal,dz21);
 825
 826             /* Update vectorial force */
 827             fix2             = _mm_add_ps(fix2,tx);
 828             fiy2             = _mm_add_ps(fiy2,ty);
 829             fiz2             = _mm_add_ps(fiz2,tz);
 830
 831             fjx1             = _mm_add_ps(fjx1,tx);
 832             fjy1             = _mm_add_ps(fjy1,ty);
 833             fjz1             = _mm_add_ps(fjz1,tz);
 834
 835             }
 836
 837             /**************************
 838              * CALCULATE INTERACTIONS *
 839              **************************/
 840
 841             if (gmx_mm_any_lt(rsq22,rcutoff2))
 842             {
 843
 844             r22              = _mm_mul_ps(rsq22,rinv22);
 845
 846             /* EWALD ELECTROSTATICS */
 847
 848             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 849             ewrt             = _mm_mul_ps(r22,ewtabscale);
 850             ewitab           = _mm_cvttps_epi32(ewrt);
 851             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 852             ewitab           = _mm_slli_epi32(ewitab,2);
 853             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 854             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 855             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 856             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 857             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 858             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 859             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 860             velec            = _mm_mul_ps(qq22,_mm_sub_ps(rinv22,velec));
 861             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
 862
 863             d                = _mm_sub_ps(r22,rswitch);
 864             d                = _mm_max_ps(d,_mm_setzero_ps());
 865             d2               = _mm_mul_ps(d,d);
 866             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
 867
 868             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 869
 870             /* Evaluate switch function */
 871             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 872             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv22,_mm_mul_ps(velec,dsw)) );
 873             velec            = _mm_mul_ps(velec,sw);
 874             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
 875
 876             /* Update potential sum for this i atom from the interaction with this j atom. */
 877             velec            = _mm_and_ps(velec,cutoff_mask);
 878             velecsum         = _mm_add_ps(velecsum,velec);
 879
 880             fscal            = felec;
 881
 882             fscal            = _mm_and_ps(fscal,cutoff_mask);
 883
 884             /* Calculate temporary vectorial force */
 885             tx               = _mm_mul_ps(fscal,dx22);
 886             ty               = _mm_mul_ps(fscal,dy22);
 887             tz               = _mm_mul_ps(fscal,dz22);
 888
 889             /* Update vectorial force */
 890             fix2             = _mm_add_ps(fix2,tx);
 891             fiy2             = _mm_add_ps(fiy2,ty);
 892             fiz2             = _mm_add_ps(fiz2,tz);
 893
 894             fjx2             = _mm_add_ps(fjx2,tx);
 895             fjy2             = _mm_add_ps(fjy2,ty);
 896             fjz2             = _mm_add_ps(fjz2,tz);
 897
 898             }
 899
 900             fjptrA             = f+j_coord_offsetA;
 901             fjptrB             = f+j_coord_offsetB;
 902             fjptrC             = f+j_coord_offsetC;
 903             fjptrD             = f+j_coord_offsetD;
 904
 905             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
 906                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
 907
 908             /* Inner loop uses 603 flops */
 909         }
 910
 911         if(jidx<j_index_end)
 912         {
 913
 914             /* Get j neighbor index, and coordinate index */
 915             jnrlistA         = jjnr[jidx];
 916             jnrlistB         = jjnr[jidx+1];
 917             jnrlistC         = jjnr[jidx+2];
 918             jnrlistD         = jjnr[jidx+3];
 919             /* Sign of each element will be negative for non-real atoms.
 920              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 921              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 922              */
 923             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 924             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 925             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 926             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 927             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 928             j_coord_offsetA  = DIM*jnrA;
 929             j_coord_offsetB  = DIM*jnrB;
 930             j_coord_offsetC  = DIM*jnrC;
 931             j_coord_offsetD  = DIM*jnrD;
 932
 933             /* load j atom coordinates */
 934             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 935                                               x+j_coord_offsetC,x+j_coord_offsetD,
 936                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
 937
 938             /* Calculate displacement vector */
 939             dx00             = _mm_sub_ps(ix0,jx0);
 940             dy00             = _mm_sub_ps(iy0,jy0);
 941             dz00             = _mm_sub_ps(iz0,jz0);
 942             dx01             = _mm_sub_ps(ix0,jx1);
 943             dy01             = _mm_sub_ps(iy0,jy1);
 944             dz01             = _mm_sub_ps(iz0,jz1);
 945             dx02             = _mm_sub_ps(ix0,jx2);
 946             dy02             = _mm_sub_ps(iy0,jy2);
 947             dz02             = _mm_sub_ps(iz0,jz2);
 948             dx10             = _mm_sub_ps(ix1,jx0);
 949             dy10             = _mm_sub_ps(iy1,jy0);
 950             dz10             = _mm_sub_ps(iz1,jz0);
 951             dx11             = _mm_sub_ps(ix1,jx1);
 952             dy11             = _mm_sub_ps(iy1,jy1);
 953             dz11             = _mm_sub_ps(iz1,jz1);
 954             dx12             = _mm_sub_ps(ix1,jx2);
 955             dy12             = _mm_sub_ps(iy1,jy2);
 956             dz12             = _mm_sub_ps(iz1,jz2);
 957             dx20             = _mm_sub_ps(ix2,jx0);
 958             dy20             = _mm_sub_ps(iy2,jy0);
 959             dz20             = _mm_sub_ps(iz2,jz0);
 960             dx21             = _mm_sub_ps(ix2,jx1);
 961             dy21             = _mm_sub_ps(iy2,jy1);
 962             dz21             = _mm_sub_ps(iz2,jz1);
 963             dx22             = _mm_sub_ps(ix2,jx2);
 964             dy22             = _mm_sub_ps(iy2,jy2);
 965             dz22             = _mm_sub_ps(iz2,jz2);
 966
 967             /* Calculate squared distance and things based on it */
 968             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 969             rsq01            = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
 970             rsq02            = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
 971             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
 972             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 973             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 974             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
 975             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 976             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 977
 978             rinv00           = gmx_mm_invsqrt_ps(rsq00);
 979             rinv01           = gmx_mm_invsqrt_ps(rsq01);
 980             rinv02           = gmx_mm_invsqrt_ps(rsq02);
 981             rinv10           = gmx_mm_invsqrt_ps(rsq10);
 982             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 983             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 984             rinv20           = gmx_mm_invsqrt_ps(rsq20);
 985             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 986             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 987
 988             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 989             rinvsq01         = _mm_mul_ps(rinv01,rinv01);
 990             rinvsq02         = _mm_mul_ps(rinv02,rinv02);
 991             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
 992             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
 993             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
 994             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
 995             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
 996             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
 997
 998             fjx0             = _mm_setzero_ps();
 999             fjy0             = _mm_setzero_ps();
1000             fjz0             = _mm_setzero_ps();
1001             fjx1             = _mm_setzero_ps();
1002             fjy1             = _mm_setzero_ps();
1003             fjz1             = _mm_setzero_ps();
1004             fjx2             = _mm_setzero_ps();
1005             fjy2             = _mm_setzero_ps();
1006             fjz2             = _mm_setzero_ps();
1007
1008             /**************************
1009              * CALCULATE INTERACTIONS *
1010              **************************/
1011
1012             if (gmx_mm_any_lt(rsq00,rcutoff2))
1013             {
1014
1015             r00              = _mm_mul_ps(rsq00,rinv00);
1016             r00              = _mm_andnot_ps(dummy_mask,r00);
1017
1018             /* EWALD ELECTROSTATICS */
1019
1020             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1021             ewrt             = _mm_mul_ps(r00,ewtabscale);
1022             ewitab           = _mm_cvttps_epi32(ewrt);
1023             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1024             ewitab           = _mm_slli_epi32(ewitab,2);
1025             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1026             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1027             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1028             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1029             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1030             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1031             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1032             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
1033             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
1034
1035             /* LENNARD-JONES DISPERSION/REPULSION */
1036
1037             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
1038             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
1039             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
1040             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
1041             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
1042
1043             d                = _mm_sub_ps(r00,rswitch);
1044             d                = _mm_max_ps(d,_mm_setzero_ps());
1045             d2               = _mm_mul_ps(d,d);
1046             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1047
1048             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1049
1050             /* Evaluate switch function */
1051             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1052             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
1053             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
1054             velec            = _mm_mul_ps(velec,sw);
1055             vvdw             = _mm_mul_ps(vvdw,sw);
1056             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
1057
1058             /* Update potential sum for this i atom from the interaction with this j atom. */
1059             velec            = _mm_and_ps(velec,cutoff_mask);
1060             velec            = _mm_andnot_ps(dummy_mask,velec);
1061             velecsum         = _mm_add_ps(velecsum,velec);
1062             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
1063             vvdw             = _mm_andnot_ps(dummy_mask,vvdw);
1064             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
1065
1066             fscal            = _mm_add_ps(felec,fvdw);
1067
1068             fscal            = _mm_and_ps(fscal,cutoff_mask);
1069
1070             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1071
1072             /* Calculate temporary vectorial force */
1073             tx               = _mm_mul_ps(fscal,dx00);
1074             ty               = _mm_mul_ps(fscal,dy00);
1075             tz               = _mm_mul_ps(fscal,dz00);
1076
1077             /* Update vectorial force */
1078             fix0             = _mm_add_ps(fix0,tx);
1079             fiy0             = _mm_add_ps(fiy0,ty);
1080             fiz0             = _mm_add_ps(fiz0,tz);
1081
1082             fjx0             = _mm_add_ps(fjx0,tx);
1083             fjy0             = _mm_add_ps(fjy0,ty);
1084             fjz0             = _mm_add_ps(fjz0,tz);
1085
1086             }
1087
1088             /**************************
1089              * CALCULATE INTERACTIONS *
1090              **************************/
1091
1092             if (gmx_mm_any_lt(rsq01,rcutoff2))
1093             {
1094
1095             r01              = _mm_mul_ps(rsq01,rinv01);
1096             r01              = _mm_andnot_ps(dummy_mask,r01);
1097
1098             /* EWALD ELECTROSTATICS */
1099
1100             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1101             ewrt             = _mm_mul_ps(r01,ewtabscale);
1102             ewitab           = _mm_cvttps_epi32(ewrt);
1103             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1104             ewitab           = _mm_slli_epi32(ewitab,2);
1105             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1106             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1107             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1108             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1109             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1110             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1111             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1112             velec            = _mm_mul_ps(qq01,_mm_sub_ps(rinv01,velec));
1113             felec            = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
1114
1115             d                = _mm_sub_ps(r01,rswitch);
1116             d                = _mm_max_ps(d,_mm_setzero_ps());
1117             d2               = _mm_mul_ps(d,d);
1118             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1119
1120             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1121
1122             /* Evaluate switch function */
1123             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1124             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv01,_mm_mul_ps(velec,dsw)) );
1125             velec            = _mm_mul_ps(velec,sw);
1126             cutoff_mask      = _mm_cmplt_ps(rsq01,rcutoff2);
1127
1128             /* Update potential sum for this i atom from the interaction with this j atom. */
1129             velec            = _mm_and_ps(velec,cutoff_mask);
1130             velec            = _mm_andnot_ps(dummy_mask,velec);
1131             velecsum         = _mm_add_ps(velecsum,velec);
1132
1133             fscal            = felec;
1134
1135             fscal            = _mm_and_ps(fscal,cutoff_mask);
1136
1137             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1138
1139             /* Calculate temporary vectorial force */
1140             tx               = _mm_mul_ps(fscal,dx01);
1141             ty               = _mm_mul_ps(fscal,dy01);
1142             tz               = _mm_mul_ps(fscal,dz01);
1143
1144             /* Update vectorial force */
1145             fix0             = _mm_add_ps(fix0,tx);
1146             fiy0             = _mm_add_ps(fiy0,ty);
1147             fiz0             = _mm_add_ps(fiz0,tz);
1148
1149             fjx1             = _mm_add_ps(fjx1,tx);
1150             fjy1             = _mm_add_ps(fjy1,ty);
1151             fjz1             = _mm_add_ps(fjz1,tz);
1152
1153             }
1154
1155             /**************************
1156              * CALCULATE INTERACTIONS *
1157              **************************/
1158
1159             if (gmx_mm_any_lt(rsq02,rcutoff2))
1160             {
1161
1162             r02              = _mm_mul_ps(rsq02,rinv02);
1163             r02              = _mm_andnot_ps(dummy_mask,r02);
1164
1165             /* EWALD ELECTROSTATICS */
1166
1167             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1168             ewrt             = _mm_mul_ps(r02,ewtabscale);
1169             ewitab           = _mm_cvttps_epi32(ewrt);
1170             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1171             ewitab           = _mm_slli_epi32(ewitab,2);
1172             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1173             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1174             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1175             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1176             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1177             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1178             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1179             velec            = _mm_mul_ps(qq02,_mm_sub_ps(rinv02,velec));
1180             felec            = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
1181
1182             d                = _mm_sub_ps(r02,rswitch);
1183             d                = _mm_max_ps(d,_mm_setzero_ps());
1184             d2               = _mm_mul_ps(d,d);
1185             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1186
1187             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1188
1189             /* Evaluate switch function */
1190             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1191             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv02,_mm_mul_ps(velec,dsw)) );
1192             velec            = _mm_mul_ps(velec,sw);
1193             cutoff_mask      = _mm_cmplt_ps(rsq02,rcutoff2);
1194
1195             /* Update potential sum for this i atom from the interaction with this j atom. */
1196             velec            = _mm_and_ps(velec,cutoff_mask);
1197             velec            = _mm_andnot_ps(dummy_mask,velec);
1198             velecsum         = _mm_add_ps(velecsum,velec);
1199
1200             fscal            = felec;
1201
1202             fscal            = _mm_and_ps(fscal,cutoff_mask);
1203
1204             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1205
1206             /* Calculate temporary vectorial force */
1207             tx               = _mm_mul_ps(fscal,dx02);
1208             ty               = _mm_mul_ps(fscal,dy02);
1209             tz               = _mm_mul_ps(fscal,dz02);
1210
1211             /* Update vectorial force */
1212             fix0             = _mm_add_ps(fix0,tx);
1213             fiy0             = _mm_add_ps(fiy0,ty);
1214             fiz0             = _mm_add_ps(fiz0,tz);
1215
1216             fjx2             = _mm_add_ps(fjx2,tx);
1217             fjy2             = _mm_add_ps(fjy2,ty);
1218             fjz2             = _mm_add_ps(fjz2,tz);
1219
1220             }
1221
1222             /**************************
1223              * CALCULATE INTERACTIONS *
1224              **************************/
1225
1226             if (gmx_mm_any_lt(rsq10,rcutoff2))
1227             {
1228
1229             r10              = _mm_mul_ps(rsq10,rinv10);
1230             r10              = _mm_andnot_ps(dummy_mask,r10);
1231
1232             /* EWALD ELECTROSTATICS */
1233
1234             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1235             ewrt             = _mm_mul_ps(r10,ewtabscale);
1236             ewitab           = _mm_cvttps_epi32(ewrt);
1237             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1238             ewitab           = _mm_slli_epi32(ewitab,2);
1239             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1240             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1241             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1242             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1243             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1244             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1245             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1246             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
1247             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
1248
1249             d                = _mm_sub_ps(r10,rswitch);
1250             d                = _mm_max_ps(d,_mm_setzero_ps());
1251             d2               = _mm_mul_ps(d,d);
1252             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1253
1254             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1255
1256             /* Evaluate switch function */
1257             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1258             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
1259             velec            = _mm_mul_ps(velec,sw);
1260             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
1261
1262             /* Update potential sum for this i atom from the interaction with this j atom. */
1263             velec            = _mm_and_ps(velec,cutoff_mask);
1264             velec            = _mm_andnot_ps(dummy_mask,velec);
1265             velecsum         = _mm_add_ps(velecsum,velec);
1266
1267             fscal            = felec;
1268
1269             fscal            = _mm_and_ps(fscal,cutoff_mask);
1270
1271             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1272
1273             /* Calculate temporary vectorial force */
1274             tx               = _mm_mul_ps(fscal,dx10);
1275             ty               = _mm_mul_ps(fscal,dy10);
1276             tz               = _mm_mul_ps(fscal,dz10);
1277
1278             /* Update vectorial force */
1279             fix1             = _mm_add_ps(fix1,tx);
1280             fiy1             = _mm_add_ps(fiy1,ty);
1281             fiz1             = _mm_add_ps(fiz1,tz);
1282
1283             fjx0             = _mm_add_ps(fjx0,tx);
1284             fjy0             = _mm_add_ps(fjy0,ty);
1285             fjz0             = _mm_add_ps(fjz0,tz);
1286
1287             }
1288
1289             /**************************
1290              * CALCULATE INTERACTIONS *
1291              **************************/
1292
1293             if (gmx_mm_any_lt(rsq11,rcutoff2))
1294             {
1295
1296             r11              = _mm_mul_ps(rsq11,rinv11);
1297             r11              = _mm_andnot_ps(dummy_mask,r11);
1298
1299             /* EWALD ELECTROSTATICS */
1300
1301             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1302             ewrt             = _mm_mul_ps(r11,ewtabscale);
1303             ewitab           = _mm_cvttps_epi32(ewrt);
1304             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1305             ewitab           = _mm_slli_epi32(ewitab,2);
1306             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1307             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1308             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1309             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1310             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1311             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1312             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1313             velec            = _mm_mul_ps(qq11,_mm_sub_ps(rinv11,velec));
1314             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
1315
1316             d                = _mm_sub_ps(r11,rswitch);
1317             d                = _mm_max_ps(d,_mm_setzero_ps());
1318             d2               = _mm_mul_ps(d,d);
1319             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1320
1321             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1322
1323             /* Evaluate switch function */
1324             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1325             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv11,_mm_mul_ps(velec,dsw)) );
1326             velec            = _mm_mul_ps(velec,sw);
1327             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
1328
1329             /* Update potential sum for this i atom from the interaction with this j atom. */
1330             velec            = _mm_and_ps(velec,cutoff_mask);
1331             velec            = _mm_andnot_ps(dummy_mask,velec);
1332             velecsum         = _mm_add_ps(velecsum,velec);
1333
1334             fscal            = felec;
1335
1336             fscal            = _mm_and_ps(fscal,cutoff_mask);
1337
1338             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1339
1340             /* Calculate temporary vectorial force */
1341             tx               = _mm_mul_ps(fscal,dx11);
1342             ty               = _mm_mul_ps(fscal,dy11);
1343             tz               = _mm_mul_ps(fscal,dz11);
1344
1345             /* Update vectorial force */
1346             fix1             = _mm_add_ps(fix1,tx);
1347             fiy1             = _mm_add_ps(fiy1,ty);
1348             fiz1             = _mm_add_ps(fiz1,tz);
1349
1350             fjx1             = _mm_add_ps(fjx1,tx);
1351             fjy1             = _mm_add_ps(fjy1,ty);
1352             fjz1             = _mm_add_ps(fjz1,tz);
1353
1354             }
1355
1356             /**************************
1357              * CALCULATE INTERACTIONS *
1358              **************************/
1359
1360             if (gmx_mm_any_lt(rsq12,rcutoff2))
1361             {
1362
1363             r12              = _mm_mul_ps(rsq12,rinv12);
1364             r12              = _mm_andnot_ps(dummy_mask,r12);
1365
1366             /* EWALD ELECTROSTATICS */
1367
1368             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1369             ewrt             = _mm_mul_ps(r12,ewtabscale);
1370             ewitab           = _mm_cvttps_epi32(ewrt);
1371             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1372             ewitab           = _mm_slli_epi32(ewitab,2);
1373             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1374             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1375             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1376             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1377             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1378             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1379             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1380             velec            = _mm_mul_ps(qq12,_mm_sub_ps(rinv12,velec));
1381             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
1382
1383             d                = _mm_sub_ps(r12,rswitch);
1384             d                = _mm_max_ps(d,_mm_setzero_ps());
1385             d2               = _mm_mul_ps(d,d);
1386             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1387
1388             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1389
1390             /* Evaluate switch function */
1391             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1392             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv12,_mm_mul_ps(velec,dsw)) );
1393             velec            = _mm_mul_ps(velec,sw);
1394             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
1395
1396             /* Update potential sum for this i atom from the interaction with this j atom. */
1397             velec            = _mm_and_ps(velec,cutoff_mask);
1398             velec            = _mm_andnot_ps(dummy_mask,velec);
1399             velecsum         = _mm_add_ps(velecsum,velec);
1400
1401             fscal            = felec;
1402
1403             fscal            = _mm_and_ps(fscal,cutoff_mask);
1404
1405             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1406
1407             /* Calculate temporary vectorial force */
1408             tx               = _mm_mul_ps(fscal,dx12);
1409             ty               = _mm_mul_ps(fscal,dy12);
1410             tz               = _mm_mul_ps(fscal,dz12);
1411
1412             /* Update vectorial force */
1413             fix1             = _mm_add_ps(fix1,tx);
1414             fiy1             = _mm_add_ps(fiy1,ty);
1415             fiz1             = _mm_add_ps(fiz1,tz);
1416
1417             fjx2             = _mm_add_ps(fjx2,tx);
1418             fjy2             = _mm_add_ps(fjy2,ty);
1419             fjz2             = _mm_add_ps(fjz2,tz);
1420
1421             }
1422
1423             /**************************
1424              * CALCULATE INTERACTIONS *
1425              **************************/
1426
1427             if (gmx_mm_any_lt(rsq20,rcutoff2))
1428             {
1429
1430             r20              = _mm_mul_ps(rsq20,rinv20);
1431             r20              = _mm_andnot_ps(dummy_mask,r20);
1432
1433             /* EWALD ELECTROSTATICS */
1434
1435             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1436             ewrt             = _mm_mul_ps(r20,ewtabscale);
1437             ewitab           = _mm_cvttps_epi32(ewrt);
1438             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1439             ewitab           = _mm_slli_epi32(ewitab,2);
1440             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1441             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1442             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1443             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1444             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1445             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1446             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1447             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
1448             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
1449
1450             d                = _mm_sub_ps(r20,rswitch);
1451             d                = _mm_max_ps(d,_mm_setzero_ps());
1452             d2               = _mm_mul_ps(d,d);
1453             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1454
1455             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1456
1457             /* Evaluate switch function */
1458             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1459             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
1460             velec            = _mm_mul_ps(velec,sw);
1461             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
1462
1463             /* Update potential sum for this i atom from the interaction with this j atom. */
1464             velec            = _mm_and_ps(velec,cutoff_mask);
1465             velec            = _mm_andnot_ps(dummy_mask,velec);
1466             velecsum         = _mm_add_ps(velecsum,velec);
1467
1468             fscal            = felec;
1469
1470             fscal            = _mm_and_ps(fscal,cutoff_mask);
1471
1472             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1473
1474             /* Calculate temporary vectorial force */
1475             tx               = _mm_mul_ps(fscal,dx20);
1476             ty               = _mm_mul_ps(fscal,dy20);
1477             tz               = _mm_mul_ps(fscal,dz20);
1478
1479             /* Update vectorial force */
1480             fix2             = _mm_add_ps(fix2,tx);
1481             fiy2             = _mm_add_ps(fiy2,ty);
1482             fiz2             = _mm_add_ps(fiz2,tz);
1483
1484             fjx0             = _mm_add_ps(fjx0,tx);
1485             fjy0             = _mm_add_ps(fjy0,ty);
1486             fjz0             = _mm_add_ps(fjz0,tz);
1487
1488             }
1489
1490             /**************************
1491              * CALCULATE INTERACTIONS *
1492              **************************/
1493
1494             if (gmx_mm_any_lt(rsq21,rcutoff2))
1495             {
1496
1497             r21              = _mm_mul_ps(rsq21,rinv21);
1498             r21              = _mm_andnot_ps(dummy_mask,r21);
1499
1500             /* EWALD ELECTROSTATICS */
1501
1502             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1503             ewrt             = _mm_mul_ps(r21,ewtabscale);
1504             ewitab           = _mm_cvttps_epi32(ewrt);
1505             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1506             ewitab           = _mm_slli_epi32(ewitab,2);
1507             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1508             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1509             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1510             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1511             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1512             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1513             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1514             velec            = _mm_mul_ps(qq21,_mm_sub_ps(rinv21,velec));
1515             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
1516
1517             d                = _mm_sub_ps(r21,rswitch);
1518             d                = _mm_max_ps(d,_mm_setzero_ps());
1519             d2               = _mm_mul_ps(d,d);
1520             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1521
1522             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1523
1524             /* Evaluate switch function */
1525             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1526             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv21,_mm_mul_ps(velec,dsw)) );
1527             velec            = _mm_mul_ps(velec,sw);
1528             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
1529
1530             /* Update potential sum for this i atom from the interaction with this j atom. */
1531             velec            = _mm_and_ps(velec,cutoff_mask);
1532             velec            = _mm_andnot_ps(dummy_mask,velec);
1533             velecsum         = _mm_add_ps(velecsum,velec);
1534
1535             fscal            = felec;
1536
1537             fscal            = _mm_and_ps(fscal,cutoff_mask);
1538
1539             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1540
1541             /* Calculate temporary vectorial force */
1542             tx               = _mm_mul_ps(fscal,dx21);
1543             ty               = _mm_mul_ps(fscal,dy21);
1544             tz               = _mm_mul_ps(fscal,dz21);
1545
1546             /* Update vectorial force */
1547             fix2             = _mm_add_ps(fix2,tx);
1548             fiy2             = _mm_add_ps(fiy2,ty);
1549             fiz2             = _mm_add_ps(fiz2,tz);
1550
1551             fjx1             = _mm_add_ps(fjx1,tx);
1552             fjy1             = _mm_add_ps(fjy1,ty);
1553             fjz1             = _mm_add_ps(fjz1,tz);
1554
1555             }
1556
1557             /**************************
1558              * CALCULATE INTERACTIONS *
1559              **************************/
1560
1561             if (gmx_mm_any_lt(rsq22,rcutoff2))
1562             {
1563
1564             r22              = _mm_mul_ps(rsq22,rinv22);
1565             r22              = _mm_andnot_ps(dummy_mask,r22);
1566
1567             /* EWALD ELECTROSTATICS */
1568
1569             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1570             ewrt             = _mm_mul_ps(r22,ewtabscale);
1571             ewitab           = _mm_cvttps_epi32(ewrt);
1572             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1573             ewitab           = _mm_slli_epi32(ewitab,2);
1574             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1575             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1576             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1577             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1578             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1579             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1580             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1581             velec            = _mm_mul_ps(qq22,_mm_sub_ps(rinv22,velec));
1582             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
1583
1584             d                = _mm_sub_ps(r22,rswitch);
1585             d                = _mm_max_ps(d,_mm_setzero_ps());
1586             d2               = _mm_mul_ps(d,d);
1587             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1588
1589             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1590
1591             /* Evaluate switch function */
1592             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1593             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv22,_mm_mul_ps(velec,dsw)) );
1594             velec            = _mm_mul_ps(velec,sw);
1595             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
1596
1597             /* Update potential sum for this i atom from the interaction with this j atom. */
1598             velec            = _mm_and_ps(velec,cutoff_mask);
1599             velec            = _mm_andnot_ps(dummy_mask,velec);
1600             velecsum         = _mm_add_ps(velecsum,velec);
1601
1602             fscal            = felec;
1603
1604             fscal            = _mm_and_ps(fscal,cutoff_mask);
1605
1606             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1607
1608             /* Calculate temporary vectorial force */
1609             tx               = _mm_mul_ps(fscal,dx22);
1610             ty               = _mm_mul_ps(fscal,dy22);
1611             tz               = _mm_mul_ps(fscal,dz22);
1612
1613             /* Update vectorial force */
1614             fix2             = _mm_add_ps(fix2,tx);
1615             fiy2             = _mm_add_ps(fiy2,ty);
1616             fiz2             = _mm_add_ps(fiz2,tz);
1617
1618             fjx2             = _mm_add_ps(fjx2,tx);
1619             fjy2             = _mm_add_ps(fjy2,ty);
1620             fjz2             = _mm_add_ps(fjz2,tz);
1621
1622             }
1623
1624             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1625             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1626             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1627             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1628
1629             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
1630                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
1631
1632             /* Inner loop uses 612 flops */
1633         }
1634
1635         /* End of innermost loop */
1636
1637         gmx_mm_update_iforce_3atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
1638                                               f+i_coord_offset,fshift+i_shift_offset);
1639
1640         ggid                        = gid[iidx];
1641         /* Update potential energies */
1642         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1643         gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
1644
1645         /* Increment number of inner iterations */
1646         inneriter                  += j_index_end - j_index_start;
1647
1648         /* Outer loop uses 20 flops */
1649     }
1650
1651     /* Increment number of outer iterations */
1652     outeriter        += nri;
1653
1654     /* Update outer/inner flops */
1655
1656     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_VF,outeriter*20 + inneriter*612);
1657 }
1658 /*
1659  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwLJSw_GeomW3W3_F_sse2_single
1660  * Electrostatics interaction: Ewald
1661  * VdW interaction:            LennardJones
1662  * Geometry:                   Water3-Water3
1663  * Calculate force/pot:        Force
1664  */
1665 void
1666 nb_kernel_ElecEwSw_VdwLJSw_GeomW3W3_F_sse2_single
1667                     (t_nblist                    * gmx_restrict       nlist,
1668                      rvec                        * gmx_restrict          xx,
1669                      rvec                        * gmx_restrict          ff,
1670                      t_forcerec                  * gmx_restrict          fr,
1671                      t_mdatoms                   * gmx_restrict     mdatoms,
1672                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1673                      t_nrnb                      * gmx_restrict        nrnb)
1674 {
1675     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1676      * just 0 for non-waters.
1677      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
1678      * jnr indices corresponding to data put in the four positions in the SIMD register.
1679      */
1680     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1681     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1682     int              jnrA,jnrB,jnrC,jnrD;
1683     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1684     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1685     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1686     real             rcutoff_scalar;
1687     real             *shiftvec,*fshift,*x,*f;
1688     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1689     real             scratch[4*DIM];
1690     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1691     int              vdwioffset0;
1692     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1693     int              vdwioffset1;
1694     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1695     int              vdwioffset2;
1696     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1697     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1698     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1699     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1700     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1701     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1702     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1703     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1704     __m128           dx01,dy01,dz01,rsq01,rinv01,rinvsq01,r01,qq01,c6_01,c12_01;
1705     __m128           dx02,dy02,dz02,rsq02,rinv02,rinvsq02,r02,qq02,c6_02,c12_02;
1706     __m128           dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
1707     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1708     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1709     __m128           dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
1710     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1711     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1712     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
1713     real             *charge;
1714     int              nvdwtype;
1715     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1716     int              *vdwtype;
1717     real             *vdwparam;
1718     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
1719     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
1720     __m128i          ewitab;
1721     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
1722     real             *ewtab;
1723     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
1724     real             rswitch_scalar,d_scalar;
1725     __m128           dummy_mask,cutoff_mask;
1726     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1727     __m128           one     = _mm_set1_ps(1.0);
1728     __m128           two     = _mm_set1_ps(2.0);
1729     x                = xx[0];
1730     f                = ff[0];
1731
1732     nri              = nlist->nri;
1733     iinr             = nlist->iinr;
1734     jindex           = nlist->jindex;
1735     jjnr             = nlist->jjnr;
1736     shiftidx         = nlist->shift;
1737     gid              = nlist->gid;
1738     shiftvec         = fr->shift_vec[0];
1739     fshift           = fr->fshift[0];
1740     facel            = _mm_set1_ps(fr->epsfac);
1741     charge           = mdatoms->chargeA;
1742     nvdwtype         = fr->ntype;
1743     vdwparam         = fr->nbfp;
1744     vdwtype          = mdatoms->typeA;
1745
1746     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
1747     ewtab            = fr->ic->tabq_coul_FDV0;
1748     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
1749     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
1750
1751     /* Setup water-specific parameters */
1752     inr              = nlist->iinr[0];
1753     iq0              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+0]));
1754     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1755     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1756     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
1757
1758     jq0              = _mm_set1_ps(charge[inr+0]);
1759     jq1              = _mm_set1_ps(charge[inr+1]);
1760     jq2              = _mm_set1_ps(charge[inr+2]);
1761     vdwjidx0A        = 2*vdwtype[inr+0];
1762     qq00             = _mm_mul_ps(iq0,jq0);
1763     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
1764     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
1765     qq01             = _mm_mul_ps(iq0,jq1);
1766     qq02             = _mm_mul_ps(iq0,jq2);
1767     qq10             = _mm_mul_ps(iq1,jq0);
1768     qq11             = _mm_mul_ps(iq1,jq1);
1769     qq12             = _mm_mul_ps(iq1,jq2);
1770     qq20             = _mm_mul_ps(iq2,jq0);
1771     qq21             = _mm_mul_ps(iq2,jq1);
1772     qq22             = _mm_mul_ps(iq2,jq2);
1773
1774     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
1775     rcutoff_scalar   = fr->rcoulomb;
1776     rcutoff          = _mm_set1_ps(rcutoff_scalar);
1777     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
1778
1779     rswitch_scalar   = fr->rcoulomb_switch;
1780     rswitch          = _mm_set1_ps(rswitch_scalar);
1781     /* Setup switch parameters */
1782     d_scalar         = rcutoff_scalar-rswitch_scalar;
1783     d                = _mm_set1_ps(d_scalar);
1784     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
1785     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
1786     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
1787     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
1788     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
1789     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
1790
1791     /* Avoid stupid compiler warnings */
1792     jnrA = jnrB = jnrC = jnrD = 0;
1793     j_coord_offsetA = 0;
1794     j_coord_offsetB = 0;
1795     j_coord_offsetC = 0;
1796     j_coord_offsetD = 0;
1797
1798     outeriter        = 0;
1799     inneriter        = 0;
1800
1801     for(iidx=0;iidx<4*DIM;iidx++)
1802     {
1803         scratch[iidx] = 0.0;
1804     }
1805
1806     /* Start outer loop over neighborlists */
1807     for(iidx=0; iidx<nri; iidx++)
1808     {
1809         /* Load shift vector for this list */
1810         i_shift_offset   = DIM*shiftidx[iidx];
1811
1812         /* Load limits for loop over neighbors */
1813         j_index_start    = jindex[iidx];
1814         j_index_end      = jindex[iidx+1];
1815
1816         /* Get outer coordinate index */
1817         inr              = iinr[iidx];
1818         i_coord_offset   = DIM*inr;
1819
1820         /* Load i particle coords and add shift vector */
1821         gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
1822                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
1823
1824         fix0             = _mm_setzero_ps();
1825         fiy0             = _mm_setzero_ps();
1826         fiz0             = _mm_setzero_ps();
1827         fix1             = _mm_setzero_ps();
1828         fiy1             = _mm_setzero_ps();
1829         fiz1             = _mm_setzero_ps();
1830         fix2             = _mm_setzero_ps();
1831         fiy2             = _mm_setzero_ps();
1832         fiz2             = _mm_setzero_ps();
1833
1834         /* Start inner kernel loop */
1835         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1836         {
1837
1838             /* Get j neighbor index, and coordinate index */
1839             jnrA             = jjnr[jidx];
1840             jnrB             = jjnr[jidx+1];
1841             jnrC             = jjnr[jidx+2];
1842             jnrD             = jjnr[jidx+3];
1843             j_coord_offsetA  = DIM*jnrA;
1844             j_coord_offsetB  = DIM*jnrB;
1845             j_coord_offsetC  = DIM*jnrC;
1846             j_coord_offsetD  = DIM*jnrD;
1847
1848             /* load j atom coordinates */
1849             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1850                                               x+j_coord_offsetC,x+j_coord_offsetD,
1851                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
1852
1853             /* Calculate displacement vector */
1854             dx00             = _mm_sub_ps(ix0,jx0);
1855             dy00             = _mm_sub_ps(iy0,jy0);
1856             dz00             = _mm_sub_ps(iz0,jz0);
1857             dx01             = _mm_sub_ps(ix0,jx1);
1858             dy01             = _mm_sub_ps(iy0,jy1);
1859             dz01             = _mm_sub_ps(iz0,jz1);
1860             dx02             = _mm_sub_ps(ix0,jx2);
1861             dy02             = _mm_sub_ps(iy0,jy2);
1862             dz02             = _mm_sub_ps(iz0,jz2);
1863             dx10             = _mm_sub_ps(ix1,jx0);
1864             dy10             = _mm_sub_ps(iy1,jy0);
1865             dz10             = _mm_sub_ps(iz1,jz0);
1866             dx11             = _mm_sub_ps(ix1,jx1);
1867             dy11             = _mm_sub_ps(iy1,jy1);
1868             dz11             = _mm_sub_ps(iz1,jz1);
1869             dx12             = _mm_sub_ps(ix1,jx2);
1870             dy12             = _mm_sub_ps(iy1,jy2);
1871             dz12             = _mm_sub_ps(iz1,jz2);
1872             dx20             = _mm_sub_ps(ix2,jx0);
1873             dy20             = _mm_sub_ps(iy2,jy0);
1874             dz20             = _mm_sub_ps(iz2,jz0);
1875             dx21             = _mm_sub_ps(ix2,jx1);
1876             dy21             = _mm_sub_ps(iy2,jy1);
1877             dz21             = _mm_sub_ps(iz2,jz1);
1878             dx22             = _mm_sub_ps(ix2,jx2);
1879             dy22             = _mm_sub_ps(iy2,jy2);
1880             dz22             = _mm_sub_ps(iz2,jz2);
1881
1882             /* Calculate squared distance and things based on it */
1883             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
1884             rsq01            = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
1885             rsq02            = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
1886             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
1887             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1888             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1889             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
1890             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1891             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1892
1893             rinv00           = gmx_mm_invsqrt_ps(rsq00);
1894             rinv01           = gmx_mm_invsqrt_ps(rsq01);
1895             rinv02           = gmx_mm_invsqrt_ps(rsq02);
1896             rinv10           = gmx_mm_invsqrt_ps(rsq10);
1897             rinv11           = gmx_mm_invsqrt_ps(rsq11);
1898             rinv12           = gmx_mm_invsqrt_ps(rsq12);
1899             rinv20           = gmx_mm_invsqrt_ps(rsq20);
1900             rinv21           = gmx_mm_invsqrt_ps(rsq21);
1901             rinv22           = gmx_mm_invsqrt_ps(rsq22);
1902
1903             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
1904             rinvsq01         = _mm_mul_ps(rinv01,rinv01);
1905             rinvsq02         = _mm_mul_ps(rinv02,rinv02);
1906             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
1907             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
1908             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
1909             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
1910             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
1911             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
1912
1913             fjx0             = _mm_setzero_ps();
1914             fjy0             = _mm_setzero_ps();
1915             fjz0             = _mm_setzero_ps();
1916             fjx1             = _mm_setzero_ps();
1917             fjy1             = _mm_setzero_ps();
1918             fjz1             = _mm_setzero_ps();
1919             fjx2             = _mm_setzero_ps();
1920             fjy2             = _mm_setzero_ps();
1921             fjz2             = _mm_setzero_ps();
1922
1923             /**************************
1924              * CALCULATE INTERACTIONS *
1925              **************************/
1926
1927             if (gmx_mm_any_lt(rsq00,rcutoff2))
1928             {
1929
1930             r00              = _mm_mul_ps(rsq00,rinv00);
1931
1932             /* EWALD ELECTROSTATICS */
1933
1934             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1935             ewrt             = _mm_mul_ps(r00,ewtabscale);
1936             ewitab           = _mm_cvttps_epi32(ewrt);
1937             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1938             ewitab           = _mm_slli_epi32(ewitab,2);
1939             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1940             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1941             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1942             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1943             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1944             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1945             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1946             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
1947             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
1948
1949             /* LENNARD-JONES DISPERSION/REPULSION */
1950
1951             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
1952             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
1953             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
1954             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
1955             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
1956
1957             d                = _mm_sub_ps(r00,rswitch);
1958             d                = _mm_max_ps(d,_mm_setzero_ps());
1959             d2               = _mm_mul_ps(d,d);
1960             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
1961
1962             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1963
1964             /* Evaluate switch function */
1965             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1966             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
1967             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
1968             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
1969
1970             fscal            = _mm_add_ps(felec,fvdw);
1971
1972             fscal            = _mm_and_ps(fscal,cutoff_mask);
1973
1974             /* Calculate temporary vectorial force */
1975             tx               = _mm_mul_ps(fscal,dx00);
1976             ty               = _mm_mul_ps(fscal,dy00);
1977             tz               = _mm_mul_ps(fscal,dz00);
1978
1979             /* Update vectorial force */
1980             fix0             = _mm_add_ps(fix0,tx);
1981             fiy0             = _mm_add_ps(fiy0,ty);
1982             fiz0             = _mm_add_ps(fiz0,tz);
1983
1984             fjx0             = _mm_add_ps(fjx0,tx);
1985             fjy0             = _mm_add_ps(fjy0,ty);
1986             fjz0             = _mm_add_ps(fjz0,tz);
1987
1988             }
1989
1990             /**************************
1991              * CALCULATE INTERACTIONS *
1992              **************************/
1993
1994             if (gmx_mm_any_lt(rsq01,rcutoff2))
1995             {
1996
1997             r01              = _mm_mul_ps(rsq01,rinv01);
1998
1999             /* EWALD ELECTROSTATICS */
2000
2001             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2002             ewrt             = _mm_mul_ps(r01,ewtabscale);
2003             ewitab           = _mm_cvttps_epi32(ewrt);
2004             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2005             ewitab           = _mm_slli_epi32(ewitab,2);
2006             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2007             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2008             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2009             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2010             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2011             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2012             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2013             velec            = _mm_mul_ps(qq01,_mm_sub_ps(rinv01,velec));
2014             felec            = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
2015
2016             d                = _mm_sub_ps(r01,rswitch);
2017             d                = _mm_max_ps(d,_mm_setzero_ps());
2018             d2               = _mm_mul_ps(d,d);
2019             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2020
2021             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2022
2023             /* Evaluate switch function */
2024             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2025             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv01,_mm_mul_ps(velec,dsw)) );
2026             cutoff_mask      = _mm_cmplt_ps(rsq01,rcutoff2);
2027
2028             fscal            = felec;
2029
2030             fscal            = _mm_and_ps(fscal,cutoff_mask);
2031
2032             /* Calculate temporary vectorial force */
2033             tx               = _mm_mul_ps(fscal,dx01);
2034             ty               = _mm_mul_ps(fscal,dy01);
2035             tz               = _mm_mul_ps(fscal,dz01);
2036
2037             /* Update vectorial force */
2038             fix0             = _mm_add_ps(fix0,tx);
2039             fiy0             = _mm_add_ps(fiy0,ty);
2040             fiz0             = _mm_add_ps(fiz0,tz);
2041
2042             fjx1             = _mm_add_ps(fjx1,tx);
2043             fjy1             = _mm_add_ps(fjy1,ty);
2044             fjz1             = _mm_add_ps(fjz1,tz);
2045
2046             }
2047
2048             /**************************
2049              * CALCULATE INTERACTIONS *
2050              **************************/
2051
2052             if (gmx_mm_any_lt(rsq02,rcutoff2))
2053             {
2054
2055             r02              = _mm_mul_ps(rsq02,rinv02);
2056
2057             /* EWALD ELECTROSTATICS */
2058
2059             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2060             ewrt             = _mm_mul_ps(r02,ewtabscale);
2061             ewitab           = _mm_cvttps_epi32(ewrt);
2062             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2063             ewitab           = _mm_slli_epi32(ewitab,2);
2064             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2065             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2066             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2067             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2068             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2069             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2070             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2071             velec            = _mm_mul_ps(qq02,_mm_sub_ps(rinv02,velec));
2072             felec            = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
2073
2074             d                = _mm_sub_ps(r02,rswitch);
2075             d                = _mm_max_ps(d,_mm_setzero_ps());
2076             d2               = _mm_mul_ps(d,d);
2077             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2078
2079             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2080
2081             /* Evaluate switch function */
2082             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2083             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv02,_mm_mul_ps(velec,dsw)) );
2084             cutoff_mask      = _mm_cmplt_ps(rsq02,rcutoff2);
2085
2086             fscal            = felec;
2087
2088             fscal            = _mm_and_ps(fscal,cutoff_mask);
2089
2090             /* Calculate temporary vectorial force */
2091             tx               = _mm_mul_ps(fscal,dx02);
2092             ty               = _mm_mul_ps(fscal,dy02);
2093             tz               = _mm_mul_ps(fscal,dz02);
2094
2095             /* Update vectorial force */
2096             fix0             = _mm_add_ps(fix0,tx);
2097             fiy0             = _mm_add_ps(fiy0,ty);
2098             fiz0             = _mm_add_ps(fiz0,tz);
2099
2100             fjx2             = _mm_add_ps(fjx2,tx);
2101             fjy2             = _mm_add_ps(fjy2,ty);
2102             fjz2             = _mm_add_ps(fjz2,tz);
2103
2104             }
2105
2106             /**************************
2107              * CALCULATE INTERACTIONS *
2108              **************************/
2109
2110             if (gmx_mm_any_lt(rsq10,rcutoff2))
2111             {
2112
2113             r10              = _mm_mul_ps(rsq10,rinv10);
2114
2115             /* EWALD ELECTROSTATICS */
2116
2117             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2118             ewrt             = _mm_mul_ps(r10,ewtabscale);
2119             ewitab           = _mm_cvttps_epi32(ewrt);
2120             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2121             ewitab           = _mm_slli_epi32(ewitab,2);
2122             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2123             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2124             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2125             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2126             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2127             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2128             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2129             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
2130             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
2131
2132             d                = _mm_sub_ps(r10,rswitch);
2133             d                = _mm_max_ps(d,_mm_setzero_ps());
2134             d2               = _mm_mul_ps(d,d);
2135             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2136
2137             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2138
2139             /* Evaluate switch function */
2140             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2141             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
2142             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
2143
2144             fscal            = felec;
2145
2146             fscal            = _mm_and_ps(fscal,cutoff_mask);
2147
2148             /* Calculate temporary vectorial force */
2149             tx               = _mm_mul_ps(fscal,dx10);
2150             ty               = _mm_mul_ps(fscal,dy10);
2151             tz               = _mm_mul_ps(fscal,dz10);
2152
2153             /* Update vectorial force */
2154             fix1             = _mm_add_ps(fix1,tx);
2155             fiy1             = _mm_add_ps(fiy1,ty);
2156             fiz1             = _mm_add_ps(fiz1,tz);
2157
2158             fjx0             = _mm_add_ps(fjx0,tx);
2159             fjy0             = _mm_add_ps(fjy0,ty);
2160             fjz0             = _mm_add_ps(fjz0,tz);
2161
2162             }
2163
2164             /**************************
2165              * CALCULATE INTERACTIONS *
2166              **************************/
2167
2168             if (gmx_mm_any_lt(rsq11,rcutoff2))
2169             {
2170
2171             r11              = _mm_mul_ps(rsq11,rinv11);
2172
2173             /* EWALD ELECTROSTATICS */
2174
2175             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2176             ewrt             = _mm_mul_ps(r11,ewtabscale);
2177             ewitab           = _mm_cvttps_epi32(ewrt);
2178             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2179             ewitab           = _mm_slli_epi32(ewitab,2);
2180             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2181             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2182             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2183             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2184             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2185             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2186             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2187             velec            = _mm_mul_ps(qq11,_mm_sub_ps(rinv11,velec));
2188             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
2189
2190             d                = _mm_sub_ps(r11,rswitch);
2191             d                = _mm_max_ps(d,_mm_setzero_ps());
2192             d2               = _mm_mul_ps(d,d);
2193             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2194
2195             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2196
2197             /* Evaluate switch function */
2198             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2199             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv11,_mm_mul_ps(velec,dsw)) );
2200             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
2201
2202             fscal            = felec;
2203
2204             fscal            = _mm_and_ps(fscal,cutoff_mask);
2205
2206             /* Calculate temporary vectorial force */
2207             tx               = _mm_mul_ps(fscal,dx11);
2208             ty               = _mm_mul_ps(fscal,dy11);
2209             tz               = _mm_mul_ps(fscal,dz11);
2210
2211             /* Update vectorial force */
2212             fix1             = _mm_add_ps(fix1,tx);
2213             fiy1             = _mm_add_ps(fiy1,ty);
2214             fiz1             = _mm_add_ps(fiz1,tz);
2215
2216             fjx1             = _mm_add_ps(fjx1,tx);
2217             fjy1             = _mm_add_ps(fjy1,ty);
2218             fjz1             = _mm_add_ps(fjz1,tz);
2219
2220             }
2221
2222             /**************************
2223              * CALCULATE INTERACTIONS *
2224              **************************/
2225
2226             if (gmx_mm_any_lt(rsq12,rcutoff2))
2227             {
2228
2229             r12              = _mm_mul_ps(rsq12,rinv12);
2230
2231             /* EWALD ELECTROSTATICS */
2232
2233             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2234             ewrt             = _mm_mul_ps(r12,ewtabscale);
2235             ewitab           = _mm_cvttps_epi32(ewrt);
2236             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2237             ewitab           = _mm_slli_epi32(ewitab,2);
2238             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2239             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2240             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2241             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2242             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2243             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2244             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2245             velec            = _mm_mul_ps(qq12,_mm_sub_ps(rinv12,velec));
2246             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
2247
2248             d                = _mm_sub_ps(r12,rswitch);
2249             d                = _mm_max_ps(d,_mm_setzero_ps());
2250             d2               = _mm_mul_ps(d,d);
2251             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2252
2253             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2254
2255             /* Evaluate switch function */
2256             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2257             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv12,_mm_mul_ps(velec,dsw)) );
2258             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
2259
2260             fscal            = felec;
2261
2262             fscal            = _mm_and_ps(fscal,cutoff_mask);
2263
2264             /* Calculate temporary vectorial force */
2265             tx               = _mm_mul_ps(fscal,dx12);
2266             ty               = _mm_mul_ps(fscal,dy12);
2267             tz               = _mm_mul_ps(fscal,dz12);
2268
2269             /* Update vectorial force */
2270             fix1             = _mm_add_ps(fix1,tx);
2271             fiy1             = _mm_add_ps(fiy1,ty);
2272             fiz1             = _mm_add_ps(fiz1,tz);
2273
2274             fjx2             = _mm_add_ps(fjx2,tx);
2275             fjy2             = _mm_add_ps(fjy2,ty);
2276             fjz2             = _mm_add_ps(fjz2,tz);
2277
2278             }
2279
2280             /**************************
2281              * CALCULATE INTERACTIONS *
2282              **************************/
2283
2284             if (gmx_mm_any_lt(rsq20,rcutoff2))
2285             {
2286
2287             r20              = _mm_mul_ps(rsq20,rinv20);
2288
2289             /* EWALD ELECTROSTATICS */
2290
2291             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2292             ewrt             = _mm_mul_ps(r20,ewtabscale);
2293             ewitab           = _mm_cvttps_epi32(ewrt);
2294             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2295             ewitab           = _mm_slli_epi32(ewitab,2);
2296             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2297             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2298             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2299             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2300             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2301             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2302             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2303             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
2304             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
2305
2306             d                = _mm_sub_ps(r20,rswitch);
2307             d                = _mm_max_ps(d,_mm_setzero_ps());
2308             d2               = _mm_mul_ps(d,d);
2309             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2310
2311             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2312
2313             /* Evaluate switch function */
2314             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2315             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
2316             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
2317
2318             fscal            = felec;
2319
2320             fscal            = _mm_and_ps(fscal,cutoff_mask);
2321
2322             /* Calculate temporary vectorial force */
2323             tx               = _mm_mul_ps(fscal,dx20);
2324             ty               = _mm_mul_ps(fscal,dy20);
2325             tz               = _mm_mul_ps(fscal,dz20);
2326
2327             /* Update vectorial force */
2328             fix2             = _mm_add_ps(fix2,tx);
2329             fiy2             = _mm_add_ps(fiy2,ty);
2330             fiz2             = _mm_add_ps(fiz2,tz);
2331
2332             fjx0             = _mm_add_ps(fjx0,tx);
2333             fjy0             = _mm_add_ps(fjy0,ty);
2334             fjz0             = _mm_add_ps(fjz0,tz);
2335
2336             }
2337
2338             /**************************
2339              * CALCULATE INTERACTIONS *
2340              **************************/
2341
2342             if (gmx_mm_any_lt(rsq21,rcutoff2))
2343             {
2344
2345             r21              = _mm_mul_ps(rsq21,rinv21);
2346
2347             /* EWALD ELECTROSTATICS */
2348
2349             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2350             ewrt             = _mm_mul_ps(r21,ewtabscale);
2351             ewitab           = _mm_cvttps_epi32(ewrt);
2352             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2353             ewitab           = _mm_slli_epi32(ewitab,2);
2354             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2355             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2356             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2357             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2358             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2359             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2360             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2361             velec            = _mm_mul_ps(qq21,_mm_sub_ps(rinv21,velec));
2362             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
2363
2364             d                = _mm_sub_ps(r21,rswitch);
2365             d                = _mm_max_ps(d,_mm_setzero_ps());
2366             d2               = _mm_mul_ps(d,d);
2367             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2368
2369             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2370
2371             /* Evaluate switch function */
2372             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2373             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv21,_mm_mul_ps(velec,dsw)) );
2374             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
2375
2376             fscal            = felec;
2377
2378             fscal            = _mm_and_ps(fscal,cutoff_mask);
2379
2380             /* Calculate temporary vectorial force */
2381             tx               = _mm_mul_ps(fscal,dx21);
2382             ty               = _mm_mul_ps(fscal,dy21);
2383             tz               = _mm_mul_ps(fscal,dz21);
2384
2385             /* Update vectorial force */
2386             fix2             = _mm_add_ps(fix2,tx);
2387             fiy2             = _mm_add_ps(fiy2,ty);
2388             fiz2             = _mm_add_ps(fiz2,tz);
2389
2390             fjx1             = _mm_add_ps(fjx1,tx);
2391             fjy1             = _mm_add_ps(fjy1,ty);
2392             fjz1             = _mm_add_ps(fjz1,tz);
2393
2394             }
2395
2396             /**************************
2397              * CALCULATE INTERACTIONS *
2398              **************************/
2399
2400             if (gmx_mm_any_lt(rsq22,rcutoff2))
2401             {
2402
2403             r22              = _mm_mul_ps(rsq22,rinv22);
2404
2405             /* EWALD ELECTROSTATICS */
2406
2407             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2408             ewrt             = _mm_mul_ps(r22,ewtabscale);
2409             ewitab           = _mm_cvttps_epi32(ewrt);
2410             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2411             ewitab           = _mm_slli_epi32(ewitab,2);
2412             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2413             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2414             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2415             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2416             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2417             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2418             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2419             velec            = _mm_mul_ps(qq22,_mm_sub_ps(rinv22,velec));
2420             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
2421
2422             d                = _mm_sub_ps(r22,rswitch);
2423             d                = _mm_max_ps(d,_mm_setzero_ps());
2424             d2               = _mm_mul_ps(d,d);
2425             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2426
2427             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2428
2429             /* Evaluate switch function */
2430             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2431             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv22,_mm_mul_ps(velec,dsw)) );
2432             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
2433
2434             fscal            = felec;
2435
2436             fscal            = _mm_and_ps(fscal,cutoff_mask);
2437
2438             /* Calculate temporary vectorial force */
2439             tx               = _mm_mul_ps(fscal,dx22);
2440             ty               = _mm_mul_ps(fscal,dy22);
2441             tz               = _mm_mul_ps(fscal,dz22);
2442
2443             /* Update vectorial force */
2444             fix2             = _mm_add_ps(fix2,tx);
2445             fiy2             = _mm_add_ps(fiy2,ty);
2446             fiz2             = _mm_add_ps(fiz2,tz);
2447
2448             fjx2             = _mm_add_ps(fjx2,tx);
2449             fjy2             = _mm_add_ps(fjy2,ty);
2450             fjz2             = _mm_add_ps(fjz2,tz);
2451
2452             }
2453
2454             fjptrA             = f+j_coord_offsetA;
2455             fjptrB             = f+j_coord_offsetB;
2456             fjptrC             = f+j_coord_offsetC;
2457             fjptrD             = f+j_coord_offsetD;
2458
2459             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2460                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
2461
2462             /* Inner loop uses 573 flops */
2463         }
2464
2465         if(jidx<j_index_end)
2466         {
2467
2468             /* Get j neighbor index, and coordinate index */
2469             jnrlistA         = jjnr[jidx];
2470             jnrlistB         = jjnr[jidx+1];
2471             jnrlistC         = jjnr[jidx+2];
2472             jnrlistD         = jjnr[jidx+3];
2473             /* Sign of each element will be negative for non-real atoms.
2474              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
2475              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
2476              */
2477             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
2478             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
2479             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
2480             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
2481             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
2482             j_coord_offsetA  = DIM*jnrA;
2483             j_coord_offsetB  = DIM*jnrB;
2484             j_coord_offsetC  = DIM*jnrC;
2485             j_coord_offsetD  = DIM*jnrD;
2486
2487             /* load j atom coordinates */
2488             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
2489                                               x+j_coord_offsetC,x+j_coord_offsetD,
2490                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,&jy2,&jz2);
2491
2492             /* Calculate displacement vector */
2493             dx00             = _mm_sub_ps(ix0,jx0);
2494             dy00             = _mm_sub_ps(iy0,jy0);
2495             dz00             = _mm_sub_ps(iz0,jz0);
2496             dx01             = _mm_sub_ps(ix0,jx1);
2497             dy01             = _mm_sub_ps(iy0,jy1);
2498             dz01             = _mm_sub_ps(iz0,jz1);
2499             dx02             = _mm_sub_ps(ix0,jx2);
2500             dy02             = _mm_sub_ps(iy0,jy2);
2501             dz02             = _mm_sub_ps(iz0,jz2);
2502             dx10             = _mm_sub_ps(ix1,jx0);
2503             dy10             = _mm_sub_ps(iy1,jy0);
2504             dz10             = _mm_sub_ps(iz1,jz0);
2505             dx11             = _mm_sub_ps(ix1,jx1);
2506             dy11             = _mm_sub_ps(iy1,jy1);
2507             dz11             = _mm_sub_ps(iz1,jz1);
2508             dx12             = _mm_sub_ps(ix1,jx2);
2509             dy12             = _mm_sub_ps(iy1,jy2);
2510             dz12             = _mm_sub_ps(iz1,jz2);
2511             dx20             = _mm_sub_ps(ix2,jx0);
2512             dy20             = _mm_sub_ps(iy2,jy0);
2513             dz20             = _mm_sub_ps(iz2,jz0);
2514             dx21             = _mm_sub_ps(ix2,jx1);
2515             dy21             = _mm_sub_ps(iy2,jy1);
2516             dz21             = _mm_sub_ps(iz2,jz1);
2517             dx22             = _mm_sub_ps(ix2,jx2);
2518             dy22             = _mm_sub_ps(iy2,jy2);
2519             dz22             = _mm_sub_ps(iz2,jz2);
2520
2521             /* Calculate squared distance and things based on it */
2522             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
2523             rsq01            = gmx_mm_calc_rsq_ps(dx01,dy01,dz01);
2524             rsq02            = gmx_mm_calc_rsq_ps(dx02,dy02,dz02);
2525             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
2526             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
2527             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
2528             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
2529             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
2530             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
2531
2532             rinv00           = gmx_mm_invsqrt_ps(rsq00);
2533             rinv01           = gmx_mm_invsqrt_ps(rsq01);
2534             rinv02           = gmx_mm_invsqrt_ps(rsq02);
2535             rinv10           = gmx_mm_invsqrt_ps(rsq10);
2536             rinv11           = gmx_mm_invsqrt_ps(rsq11);
2537             rinv12           = gmx_mm_invsqrt_ps(rsq12);
2538             rinv20           = gmx_mm_invsqrt_ps(rsq20);
2539             rinv21           = gmx_mm_invsqrt_ps(rsq21);
2540             rinv22           = gmx_mm_invsqrt_ps(rsq22);
2541
2542             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
2543             rinvsq01         = _mm_mul_ps(rinv01,rinv01);
2544             rinvsq02         = _mm_mul_ps(rinv02,rinv02);
2545             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
2546             rinvsq11         = _mm_mul_ps(rinv11,rinv11);
2547             rinvsq12         = _mm_mul_ps(rinv12,rinv12);
2548             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
2549             rinvsq21         = _mm_mul_ps(rinv21,rinv21);
2550             rinvsq22         = _mm_mul_ps(rinv22,rinv22);
2551
2552             fjx0             = _mm_setzero_ps();
2553             fjy0             = _mm_setzero_ps();
2554             fjz0             = _mm_setzero_ps();
2555             fjx1             = _mm_setzero_ps();
2556             fjy1             = _mm_setzero_ps();
2557             fjz1             = _mm_setzero_ps();
2558             fjx2             = _mm_setzero_ps();
2559             fjy2             = _mm_setzero_ps();
2560             fjz2             = _mm_setzero_ps();
2561
2562             /**************************
2563              * CALCULATE INTERACTIONS *
2564              **************************/
2565
2566             if (gmx_mm_any_lt(rsq00,rcutoff2))
2567             {
2568
2569             r00              = _mm_mul_ps(rsq00,rinv00);
2570             r00              = _mm_andnot_ps(dummy_mask,r00);
2571
2572             /* EWALD ELECTROSTATICS */
2573
2574             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2575             ewrt             = _mm_mul_ps(r00,ewtabscale);
2576             ewitab           = _mm_cvttps_epi32(ewrt);
2577             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2578             ewitab           = _mm_slli_epi32(ewitab,2);
2579             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2580             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2581             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2582             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2583             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2584             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2585             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2586             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
2587             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
2588
2589             /* LENNARD-JONES DISPERSION/REPULSION */
2590
2591             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
2592             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
2593             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
2594             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
2595             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
2596
2597             d                = _mm_sub_ps(r00,rswitch);
2598             d                = _mm_max_ps(d,_mm_setzero_ps());
2599             d2               = _mm_mul_ps(d,d);
2600             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2601
2602             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2603
2604             /* Evaluate switch function */
2605             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2606             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
2607             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
2608             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
2609
2610             fscal            = _mm_add_ps(felec,fvdw);
2611
2612             fscal            = _mm_and_ps(fscal,cutoff_mask);
2613
2614             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2615
2616             /* Calculate temporary vectorial force */
2617             tx               = _mm_mul_ps(fscal,dx00);
2618             ty               = _mm_mul_ps(fscal,dy00);
2619             tz               = _mm_mul_ps(fscal,dz00);
2620
2621             /* Update vectorial force */
2622             fix0             = _mm_add_ps(fix0,tx);
2623             fiy0             = _mm_add_ps(fiy0,ty);
2624             fiz0             = _mm_add_ps(fiz0,tz);
2625
2626             fjx0             = _mm_add_ps(fjx0,tx);
2627             fjy0             = _mm_add_ps(fjy0,ty);
2628             fjz0             = _mm_add_ps(fjz0,tz);
2629
2630             }
2631
2632             /**************************
2633              * CALCULATE INTERACTIONS *
2634              **************************/
2635
2636             if (gmx_mm_any_lt(rsq01,rcutoff2))
2637             {
2638
2639             r01              = _mm_mul_ps(rsq01,rinv01);
2640             r01              = _mm_andnot_ps(dummy_mask,r01);
2641
2642             /* EWALD ELECTROSTATICS */
2643
2644             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2645             ewrt             = _mm_mul_ps(r01,ewtabscale);
2646             ewitab           = _mm_cvttps_epi32(ewrt);
2647             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2648             ewitab           = _mm_slli_epi32(ewitab,2);
2649             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2650             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2651             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2652             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2653             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2654             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2655             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2656             velec            = _mm_mul_ps(qq01,_mm_sub_ps(rinv01,velec));
2657             felec            = _mm_mul_ps(_mm_mul_ps(qq01,rinv01),_mm_sub_ps(rinvsq01,felec));
2658
2659             d                = _mm_sub_ps(r01,rswitch);
2660             d                = _mm_max_ps(d,_mm_setzero_ps());
2661             d2               = _mm_mul_ps(d,d);
2662             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2663
2664             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2665
2666             /* Evaluate switch function */
2667             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2668             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv01,_mm_mul_ps(velec,dsw)) );
2669             cutoff_mask      = _mm_cmplt_ps(rsq01,rcutoff2);
2670
2671             fscal            = felec;
2672
2673             fscal            = _mm_and_ps(fscal,cutoff_mask);
2674
2675             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2676
2677             /* Calculate temporary vectorial force */
2678             tx               = _mm_mul_ps(fscal,dx01);
2679             ty               = _mm_mul_ps(fscal,dy01);
2680             tz               = _mm_mul_ps(fscal,dz01);
2681
2682             /* Update vectorial force */
2683             fix0             = _mm_add_ps(fix0,tx);
2684             fiy0             = _mm_add_ps(fiy0,ty);
2685             fiz0             = _mm_add_ps(fiz0,tz);
2686
2687             fjx1             = _mm_add_ps(fjx1,tx);
2688             fjy1             = _mm_add_ps(fjy1,ty);
2689             fjz1             = _mm_add_ps(fjz1,tz);
2690
2691             }
2692
2693             /**************************
2694              * CALCULATE INTERACTIONS *
2695              **************************/
2696
2697             if (gmx_mm_any_lt(rsq02,rcutoff2))
2698             {
2699
2700             r02              = _mm_mul_ps(rsq02,rinv02);
2701             r02              = _mm_andnot_ps(dummy_mask,r02);
2702
2703             /* EWALD ELECTROSTATICS */
2704
2705             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2706             ewrt             = _mm_mul_ps(r02,ewtabscale);
2707             ewitab           = _mm_cvttps_epi32(ewrt);
2708             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2709             ewitab           = _mm_slli_epi32(ewitab,2);
2710             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2711             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2712             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2713             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2714             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2715             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2716             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2717             velec            = _mm_mul_ps(qq02,_mm_sub_ps(rinv02,velec));
2718             felec            = _mm_mul_ps(_mm_mul_ps(qq02,rinv02),_mm_sub_ps(rinvsq02,felec));
2719
2720             d                = _mm_sub_ps(r02,rswitch);
2721             d                = _mm_max_ps(d,_mm_setzero_ps());
2722             d2               = _mm_mul_ps(d,d);
2723             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2724
2725             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2726
2727             /* Evaluate switch function */
2728             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2729             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv02,_mm_mul_ps(velec,dsw)) );
2730             cutoff_mask      = _mm_cmplt_ps(rsq02,rcutoff2);
2731
2732             fscal            = felec;
2733
2734             fscal            = _mm_and_ps(fscal,cutoff_mask);
2735
2736             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2737
2738             /* Calculate temporary vectorial force */
2739             tx               = _mm_mul_ps(fscal,dx02);
2740             ty               = _mm_mul_ps(fscal,dy02);
2741             tz               = _mm_mul_ps(fscal,dz02);
2742
2743             /* Update vectorial force */
2744             fix0             = _mm_add_ps(fix0,tx);
2745             fiy0             = _mm_add_ps(fiy0,ty);
2746             fiz0             = _mm_add_ps(fiz0,tz);
2747
2748             fjx2             = _mm_add_ps(fjx2,tx);
2749             fjy2             = _mm_add_ps(fjy2,ty);
2750             fjz2             = _mm_add_ps(fjz2,tz);
2751
2752             }
2753
2754             /**************************
2755              * CALCULATE INTERACTIONS *
2756              **************************/
2757
2758             if (gmx_mm_any_lt(rsq10,rcutoff2))
2759             {
2760
2761             r10              = _mm_mul_ps(rsq10,rinv10);
2762             r10              = _mm_andnot_ps(dummy_mask,r10);
2763
2764             /* EWALD ELECTROSTATICS */
2765
2766             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2767             ewrt             = _mm_mul_ps(r10,ewtabscale);
2768             ewitab           = _mm_cvttps_epi32(ewrt);
2769             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2770             ewitab           = _mm_slli_epi32(ewitab,2);
2771             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2772             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2773             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2774             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2775             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2776             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2777             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2778             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
2779             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
2780
2781             d                = _mm_sub_ps(r10,rswitch);
2782             d                = _mm_max_ps(d,_mm_setzero_ps());
2783             d2               = _mm_mul_ps(d,d);
2784             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2785
2786             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2787
2788             /* Evaluate switch function */
2789             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2790             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
2791             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
2792
2793             fscal            = felec;
2794
2795             fscal            = _mm_and_ps(fscal,cutoff_mask);
2796
2797             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2798
2799             /* Calculate temporary vectorial force */
2800             tx               = _mm_mul_ps(fscal,dx10);
2801             ty               = _mm_mul_ps(fscal,dy10);
2802             tz               = _mm_mul_ps(fscal,dz10);
2803
2804             /* Update vectorial force */
2805             fix1             = _mm_add_ps(fix1,tx);
2806             fiy1             = _mm_add_ps(fiy1,ty);
2807             fiz1             = _mm_add_ps(fiz1,tz);
2808
2809             fjx0             = _mm_add_ps(fjx0,tx);
2810             fjy0             = _mm_add_ps(fjy0,ty);
2811             fjz0             = _mm_add_ps(fjz0,tz);
2812
2813             }
2814
2815             /**************************
2816              * CALCULATE INTERACTIONS *
2817              **************************/
2818
2819             if (gmx_mm_any_lt(rsq11,rcutoff2))
2820             {
2821
2822             r11              = _mm_mul_ps(rsq11,rinv11);
2823             r11              = _mm_andnot_ps(dummy_mask,r11);
2824
2825             /* EWALD ELECTROSTATICS */
2826
2827             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2828             ewrt             = _mm_mul_ps(r11,ewtabscale);
2829             ewitab           = _mm_cvttps_epi32(ewrt);
2830             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2831             ewitab           = _mm_slli_epi32(ewitab,2);
2832             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2833             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2834             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2835             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2836             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2837             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2838             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2839             velec            = _mm_mul_ps(qq11,_mm_sub_ps(rinv11,velec));
2840             felec            = _mm_mul_ps(_mm_mul_ps(qq11,rinv11),_mm_sub_ps(rinvsq11,felec));
2841
2842             d                = _mm_sub_ps(r11,rswitch);
2843             d                = _mm_max_ps(d,_mm_setzero_ps());
2844             d2               = _mm_mul_ps(d,d);
2845             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2846
2847             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2848
2849             /* Evaluate switch function */
2850             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2851             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv11,_mm_mul_ps(velec,dsw)) );
2852             cutoff_mask      = _mm_cmplt_ps(rsq11,rcutoff2);
2853
2854             fscal            = felec;
2855
2856             fscal            = _mm_and_ps(fscal,cutoff_mask);
2857
2858             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2859
2860             /* Calculate temporary vectorial force */
2861             tx               = _mm_mul_ps(fscal,dx11);
2862             ty               = _mm_mul_ps(fscal,dy11);
2863             tz               = _mm_mul_ps(fscal,dz11);
2864
2865             /* Update vectorial force */
2866             fix1             = _mm_add_ps(fix1,tx);
2867             fiy1             = _mm_add_ps(fiy1,ty);
2868             fiz1             = _mm_add_ps(fiz1,tz);
2869
2870             fjx1             = _mm_add_ps(fjx1,tx);
2871             fjy1             = _mm_add_ps(fjy1,ty);
2872             fjz1             = _mm_add_ps(fjz1,tz);
2873
2874             }
2875
2876             /**************************
2877              * CALCULATE INTERACTIONS *
2878              **************************/
2879
2880             if (gmx_mm_any_lt(rsq12,rcutoff2))
2881             {
2882
2883             r12              = _mm_mul_ps(rsq12,rinv12);
2884             r12              = _mm_andnot_ps(dummy_mask,r12);
2885
2886             /* EWALD ELECTROSTATICS */
2887
2888             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2889             ewrt             = _mm_mul_ps(r12,ewtabscale);
2890             ewitab           = _mm_cvttps_epi32(ewrt);
2891             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2892             ewitab           = _mm_slli_epi32(ewitab,2);
2893             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2894             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2895             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2896             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2897             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2898             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2899             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2900             velec            = _mm_mul_ps(qq12,_mm_sub_ps(rinv12,velec));
2901             felec            = _mm_mul_ps(_mm_mul_ps(qq12,rinv12),_mm_sub_ps(rinvsq12,felec));
2902
2903             d                = _mm_sub_ps(r12,rswitch);
2904             d                = _mm_max_ps(d,_mm_setzero_ps());
2905             d2               = _mm_mul_ps(d,d);
2906             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2907
2908             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2909
2910             /* Evaluate switch function */
2911             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2912             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv12,_mm_mul_ps(velec,dsw)) );
2913             cutoff_mask      = _mm_cmplt_ps(rsq12,rcutoff2);
2914
2915             fscal            = felec;
2916
2917             fscal            = _mm_and_ps(fscal,cutoff_mask);
2918
2919             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2920
2921             /* Calculate temporary vectorial force */
2922             tx               = _mm_mul_ps(fscal,dx12);
2923             ty               = _mm_mul_ps(fscal,dy12);
2924             tz               = _mm_mul_ps(fscal,dz12);
2925
2926             /* Update vectorial force */
2927             fix1             = _mm_add_ps(fix1,tx);
2928             fiy1             = _mm_add_ps(fiy1,ty);
2929             fiz1             = _mm_add_ps(fiz1,tz);
2930
2931             fjx2             = _mm_add_ps(fjx2,tx);
2932             fjy2             = _mm_add_ps(fjy2,ty);
2933             fjz2             = _mm_add_ps(fjz2,tz);
2934
2935             }
2936
2937             /**************************
2938              * CALCULATE INTERACTIONS *
2939              **************************/
2940
2941             if (gmx_mm_any_lt(rsq20,rcutoff2))
2942             {
2943
2944             r20              = _mm_mul_ps(rsq20,rinv20);
2945             r20              = _mm_andnot_ps(dummy_mask,r20);
2946
2947             /* EWALD ELECTROSTATICS */
2948
2949             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
2950             ewrt             = _mm_mul_ps(r20,ewtabscale);
2951             ewitab           = _mm_cvttps_epi32(ewrt);
2952             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
2953             ewitab           = _mm_slli_epi32(ewitab,2);
2954             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
2955             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
2956             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
2957             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
2958             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
2959             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
2960             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
2961             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
2962             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
2963
2964             d                = _mm_sub_ps(r20,rswitch);
2965             d                = _mm_max_ps(d,_mm_setzero_ps());
2966             d2               = _mm_mul_ps(d,d);
2967             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
2968
2969             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
2970
2971             /* Evaluate switch function */
2972             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
2973             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
2974             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
2975
2976             fscal            = felec;
2977
2978             fscal            = _mm_and_ps(fscal,cutoff_mask);
2979
2980             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2981
2982             /* Calculate temporary vectorial force */
2983             tx               = _mm_mul_ps(fscal,dx20);
2984             ty               = _mm_mul_ps(fscal,dy20);
2985             tz               = _mm_mul_ps(fscal,dz20);
2986
2987             /* Update vectorial force */
2988             fix2             = _mm_add_ps(fix2,tx);
2989             fiy2             = _mm_add_ps(fiy2,ty);
2990             fiz2             = _mm_add_ps(fiz2,tz);
2991
2992             fjx0             = _mm_add_ps(fjx0,tx);
2993             fjy0             = _mm_add_ps(fjy0,ty);
2994             fjz0             = _mm_add_ps(fjz0,tz);
2995
2996             }
2997
2998             /**************************
2999              * CALCULATE INTERACTIONS *
3000              **************************/
3001
3002             if (gmx_mm_any_lt(rsq21,rcutoff2))
3003             {
3004
3005             r21              = _mm_mul_ps(rsq21,rinv21);
3006             r21              = _mm_andnot_ps(dummy_mask,r21);
3007
3008             /* EWALD ELECTROSTATICS */
3009
3010             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
3011             ewrt             = _mm_mul_ps(r21,ewtabscale);
3012             ewitab           = _mm_cvttps_epi32(ewrt);
3013             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
3014             ewitab           = _mm_slli_epi32(ewitab,2);
3015             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
3016             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
3017             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
3018             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
3019             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
3020             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
3021             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
3022             velec            = _mm_mul_ps(qq21,_mm_sub_ps(rinv21,velec));
3023             felec            = _mm_mul_ps(_mm_mul_ps(qq21,rinv21),_mm_sub_ps(rinvsq21,felec));
3024
3025             d                = _mm_sub_ps(r21,rswitch);
3026             d                = _mm_max_ps(d,_mm_setzero_ps());
3027             d2               = _mm_mul_ps(d,d);
3028             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
3029
3030             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
3031
3032             /* Evaluate switch function */
3033             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
3034             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv21,_mm_mul_ps(velec,dsw)) );
3035             cutoff_mask      = _mm_cmplt_ps(rsq21,rcutoff2);
3036
3037             fscal            = felec;
3038
3039             fscal            = _mm_and_ps(fscal,cutoff_mask);
3040
3041             fscal            = _mm_andnot_ps(dummy_mask,fscal);
3042
3043             /* Calculate temporary vectorial force */
3044             tx               = _mm_mul_ps(fscal,dx21);
3045             ty               = _mm_mul_ps(fscal,dy21);
3046             tz               = _mm_mul_ps(fscal,dz21);
3047
3048             /* Update vectorial force */
3049             fix2             = _mm_add_ps(fix2,tx);
3050             fiy2             = _mm_add_ps(fiy2,ty);
3051             fiz2             = _mm_add_ps(fiz2,tz);
3052
3053             fjx1             = _mm_add_ps(fjx1,tx);
3054             fjy1             = _mm_add_ps(fjy1,ty);
3055             fjz1             = _mm_add_ps(fjz1,tz);
3056
3057             }
3058
3059             /**************************
3060              * CALCULATE INTERACTIONS *
3061              **************************/
3062
3063             if (gmx_mm_any_lt(rsq22,rcutoff2))
3064             {
3065
3066             r22              = _mm_mul_ps(rsq22,rinv22);
3067             r22              = _mm_andnot_ps(dummy_mask,r22);
3068
3069             /* EWALD ELECTROSTATICS */
3070
3071             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
3072             ewrt             = _mm_mul_ps(r22,ewtabscale);
3073             ewitab           = _mm_cvttps_epi32(ewrt);
3074             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
3075             ewitab           = _mm_slli_epi32(ewitab,2);
3076             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
3077             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
3078             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
3079             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
3080             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
3081             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
3082             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
3083             velec            = _mm_mul_ps(qq22,_mm_sub_ps(rinv22,velec));
3084             felec            = _mm_mul_ps(_mm_mul_ps(qq22,rinv22),_mm_sub_ps(rinvsq22,felec));
3085
3086             d                = _mm_sub_ps(r22,rswitch);
3087             d                = _mm_max_ps(d,_mm_setzero_ps());
3088             d2               = _mm_mul_ps(d,d);
3089             sw               = _mm_add_ps(one,_mm_mul_ps(d2,_mm_mul_ps(d,_mm_add_ps(swV3,_mm_mul_ps(d,_mm_add_ps(swV4,_mm_mul_ps(d,swV5)))))));
3090
3091             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
3092
3093             /* Evaluate switch function */
3094             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
3095             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv22,_mm_mul_ps(velec,dsw)) );
3096             cutoff_mask      = _mm_cmplt_ps(rsq22,rcutoff2);
3097
3098             fscal            = felec;
3099
3100             fscal            = _mm_and_ps(fscal,cutoff_mask);
3101
3102             fscal            = _mm_andnot_ps(dummy_mask,fscal);
3103
3104             /* Calculate temporary vectorial force */
3105             tx               = _mm_mul_ps(fscal,dx22);
3106             ty               = _mm_mul_ps(fscal,dy22);
3107             tz               = _mm_mul_ps(fscal,dz22);
3108
3109             /* Update vectorial force */
3110             fix2             = _mm_add_ps(fix2,tx);
3111             fiy2             = _mm_add_ps(fiy2,ty);
3112             fiz2             = _mm_add_ps(fiz2,tz);
3113
3114             fjx2             = _mm_add_ps(fjx2,tx);
3115             fjy2             = _mm_add_ps(fjy2,ty);
3116             fjz2             = _mm_add_ps(fjz2,tz);
3117
3118             }
3119
3120             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
3121             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
3122             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
3123             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
3124
3125             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
3126                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,fjx2,fjy2,fjz2);
3127
3128             /* Inner loop uses 582 flops */
3129         }
3130
3131         /* End of innermost loop */
3132
3133         gmx_mm_update_iforce_3atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
3134                                               f+i_coord_offset,fshift+i_shift_offset);
3135
3136         /* Increment number of inner iterations */
3137         inneriter                  += j_index_end - j_index_start;
3138
3139         /* Outer loop uses 18 flops */
3140     }
3141
3142     /* Increment number of outer iterations */
3143     outeriter        += nri;
3144
3145     /* Update outer/inner flops */
3146
3147     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3W3_F,outeriter*18 + inneriter*582);
3148 }