src/gromacs/gmxlib/nonbonded/nb_kernel_sse4_1_single/nb_kernel_ElecEwSw_VdwLJSw_GeomP1P1_sse4_1_single.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2017, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS sse4_1_single kernel generator.
  37  */
  38 #include "gmxpre.h"
  39
  40 #include "config.h"
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "gromacs/gmxlib/nrnb.h"
  46
  47 #include "kernelutil_x86_sse4_1_single.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwLJSw_GeomP1P1_VF_sse4_1_single
  51  * Electrostatics interaction: Ewald
  52  * VdW interaction:            LennardJones
  53  * Geometry:                   Particle-Particle
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecEwSw_VdwLJSw_GeomP1P1_VF_sse4_1_single
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      struct t_forcerec           * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  67      * just 0 for non-waters.
  68      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
  69      * jnr indices corresponding to data put in the four positions in the SIMD register.
  70      */
  71     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  72     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  73     int              jnrA,jnrB,jnrC,jnrD;
  74     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  75     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  76     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  77     real             rcutoff_scalar;
  78     real             *shiftvec,*fshift,*x,*f;
  79     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  80     real             scratch[4*DIM];
  81     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  82     int              vdwioffset0;
  83     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  84     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  85     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  86     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  87     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
  88     real             *charge;
  89     int              nvdwtype;
  90     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  91     int              *vdwtype;
  92     real             *vdwparam;
  93     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
  94     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
  95     __m128i          ewitab;
  96     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
  97     real             *ewtab;
  98     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
  99     real             rswitch_scalar,d_scalar;
 100     __m128           dummy_mask,cutoff_mask;
 101     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 102     __m128           one     = _mm_set1_ps(1.0);
 103     __m128           two     = _mm_set1_ps(2.0);
 104     x                = xx[0];
 105     f                = ff[0];
 106
 107     nri              = nlist->nri;
 108     iinr             = nlist->iinr;
 109     jindex           = nlist->jindex;
 110     jjnr             = nlist->jjnr;
 111     shiftidx         = nlist->shift;
 112     gid              = nlist->gid;
 113     shiftvec         = fr->shift_vec[0];
 114     fshift           = fr->fshift[0];
 115     facel            = _mm_set1_ps(fr->ic->epsfac);
 116     charge           = mdatoms->chargeA;
 117     nvdwtype         = fr->ntype;
 118     vdwparam         = fr->nbfp;
 119     vdwtype          = mdatoms->typeA;
 120
 121     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 122     ewtab            = fr->ic->tabq_coul_FDV0;
 123     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 124     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 125
 126     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 127     rcutoff_scalar   = fr->ic->rcoulomb;
 128     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 129     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 130
 131     rswitch_scalar   = fr->ic->rcoulomb_switch;
 132     rswitch          = _mm_set1_ps(rswitch_scalar);
 133     /* Setup switch parameters */
 134     d_scalar         = rcutoff_scalar-rswitch_scalar;
 135     d                = _mm_set1_ps(d_scalar);
 136     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
 137     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 138     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 139     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
 140     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 141     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 142
 143     /* Avoid stupid compiler warnings */
 144     jnrA = jnrB = jnrC = jnrD = 0;
 145     j_coord_offsetA = 0;
 146     j_coord_offsetB = 0;
 147     j_coord_offsetC = 0;
 148     j_coord_offsetD = 0;
 149
 150     outeriter        = 0;
 151     inneriter        = 0;
 152
 153     for(iidx=0;iidx<4*DIM;iidx++)
 154     {
 155         scratch[iidx] = 0.0;
 156     }
 157
 158     /* Start outer loop over neighborlists */
 159     for(iidx=0; iidx<nri; iidx++)
 160     {
 161         /* Load shift vector for this list */
 162         i_shift_offset   = DIM*shiftidx[iidx];
 163
 164         /* Load limits for loop over neighbors */
 165         j_index_start    = jindex[iidx];
 166         j_index_end      = jindex[iidx+1];
 167
 168         /* Get outer coordinate index */
 169         inr              = iinr[iidx];
 170         i_coord_offset   = DIM*inr;
 171
 172         /* Load i particle coords and add shift vector */
 173         gmx_mm_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 174
 175         fix0             = _mm_setzero_ps();
 176         fiy0             = _mm_setzero_ps();
 177         fiz0             = _mm_setzero_ps();
 178
 179         /* Load parameters for i particles */
 180         iq0              = _mm_mul_ps(facel,_mm_load1_ps(charge+inr+0));
 181         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 182
 183         /* Reset potential sums */
 184         velecsum         = _mm_setzero_ps();
 185         vvdwsum          = _mm_setzero_ps();
 186
 187         /* Start inner kernel loop */
 188         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 189         {
 190
 191             /* Get j neighbor index, and coordinate index */
 192             jnrA             = jjnr[jidx];
 193             jnrB             = jjnr[jidx+1];
 194             jnrC             = jjnr[jidx+2];
 195             jnrD             = jjnr[jidx+3];
 196             j_coord_offsetA  = DIM*jnrA;
 197             j_coord_offsetB  = DIM*jnrB;
 198             j_coord_offsetC  = DIM*jnrC;
 199             j_coord_offsetD  = DIM*jnrD;
 200
 201             /* load j atom coordinates */
 202             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 203                                               x+j_coord_offsetC,x+j_coord_offsetD,
 204                                               &jx0,&jy0,&jz0);
 205
 206             /* Calculate displacement vector */
 207             dx00             = _mm_sub_ps(ix0,jx0);
 208             dy00             = _mm_sub_ps(iy0,jy0);
 209             dz00             = _mm_sub_ps(iz0,jz0);
 210
 211             /* Calculate squared distance and things based on it */
 212             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 213
 214             rinv00           = sse41_invsqrt_f(rsq00);
 215
 216             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 217
 218             /* Load parameters for j particles */
 219             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 220                                                               charge+jnrC+0,charge+jnrD+0);
 221             vdwjidx0A        = 2*vdwtype[jnrA+0];
 222             vdwjidx0B        = 2*vdwtype[jnrB+0];
 223             vdwjidx0C        = 2*vdwtype[jnrC+0];
 224             vdwjidx0D        = 2*vdwtype[jnrD+0];
 225
 226             /**************************
 227              * CALCULATE INTERACTIONS *
 228              **************************/
 229
 230             if (gmx_mm_any_lt(rsq00,rcutoff2))
 231             {
 232
 233             r00              = _mm_mul_ps(rsq00,rinv00);
 234
 235             /* Compute parameters for interactions between i and j atoms */
 236             qq00             = _mm_mul_ps(iq0,jq0);
 237             gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
 238                                          vdwparam+vdwioffset0+vdwjidx0B,
 239                                          vdwparam+vdwioffset0+vdwjidx0C,
 240                                          vdwparam+vdwioffset0+vdwjidx0D,
 241                                          &c6_00,&c12_00);
 242
 243             /* EWALD ELECTROSTATICS */
 244
 245             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 246             ewrt             = _mm_mul_ps(r00,ewtabscale);
 247             ewitab           = _mm_cvttps_epi32(ewrt);
 248             eweps            = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR));
 249             ewitab           = _mm_slli_epi32(ewitab,2);
 250             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 251             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 252             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 253             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 254             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 255             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 256             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 257             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
 258             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
 259
 260             /* LENNARD-JONES DISPERSION/REPULSION */
 261
 262             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 263             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 264             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 265             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
 266             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 267
 268             d                = _mm_sub_ps(r00,rswitch);
 269             d                = _mm_max_ps(d,_mm_setzero_ps());
 270             d2               = _mm_mul_ps(d,d);
 271             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)))))));
 272
 273             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 274
 275             /* Evaluate switch function */
 276             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 277             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
 278             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
 279             velec            = _mm_mul_ps(velec,sw);
 280             vvdw             = _mm_mul_ps(vvdw,sw);
 281             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
 282
 283             /* Update potential sum for this i atom from the interaction with this j atom. */
 284             velec            = _mm_and_ps(velec,cutoff_mask);
 285             velecsum         = _mm_add_ps(velecsum,velec);
 286             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
 287             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 288
 289             fscal            = _mm_add_ps(felec,fvdw);
 290
 291             fscal            = _mm_and_ps(fscal,cutoff_mask);
 292
 293             /* Calculate temporary vectorial force */
 294             tx               = _mm_mul_ps(fscal,dx00);
 295             ty               = _mm_mul_ps(fscal,dy00);
 296             tz               = _mm_mul_ps(fscal,dz00);
 297
 298             /* Update vectorial force */
 299             fix0             = _mm_add_ps(fix0,tx);
 300             fiy0             = _mm_add_ps(fiy0,ty);
 301             fiz0             = _mm_add_ps(fiz0,tz);
 302
 303             fjptrA             = f+j_coord_offsetA;
 304             fjptrB             = f+j_coord_offsetB;
 305             fjptrC             = f+j_coord_offsetC;
 306             fjptrD             = f+j_coord_offsetD;
 307             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,tx,ty,tz);
 308
 309             }
 310
 311             /* Inner loop uses 83 flops */
 312         }
 313
 314         if(jidx<j_index_end)
 315         {
 316
 317             /* Get j neighbor index, and coordinate index */
 318             jnrlistA         = jjnr[jidx];
 319             jnrlistB         = jjnr[jidx+1];
 320             jnrlistC         = jjnr[jidx+2];
 321             jnrlistD         = jjnr[jidx+3];
 322             /* Sign of each element will be negative for non-real atoms.
 323              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 324              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 325              */
 326             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 327             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 328             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 329             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 330             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 331             j_coord_offsetA  = DIM*jnrA;
 332             j_coord_offsetB  = DIM*jnrB;
 333             j_coord_offsetC  = DIM*jnrC;
 334             j_coord_offsetD  = DIM*jnrD;
 335
 336             /* load j atom coordinates */
 337             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 338                                               x+j_coord_offsetC,x+j_coord_offsetD,
 339                                               &jx0,&jy0,&jz0);
 340
 341             /* Calculate displacement vector */
 342             dx00             = _mm_sub_ps(ix0,jx0);
 343             dy00             = _mm_sub_ps(iy0,jy0);
 344             dz00             = _mm_sub_ps(iz0,jz0);
 345
 346             /* Calculate squared distance and things based on it */
 347             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 348
 349             rinv00           = sse41_invsqrt_f(rsq00);
 350
 351             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 352
 353             /* Load parameters for j particles */
 354             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 355                                                               charge+jnrC+0,charge+jnrD+0);
 356             vdwjidx0A        = 2*vdwtype[jnrA+0];
 357             vdwjidx0B        = 2*vdwtype[jnrB+0];
 358             vdwjidx0C        = 2*vdwtype[jnrC+0];
 359             vdwjidx0D        = 2*vdwtype[jnrD+0];
 360
 361             /**************************
 362              * CALCULATE INTERACTIONS *
 363              **************************/
 364
 365             if (gmx_mm_any_lt(rsq00,rcutoff2))
 366             {
 367
 368             r00              = _mm_mul_ps(rsq00,rinv00);
 369             r00              = _mm_andnot_ps(dummy_mask,r00);
 370
 371             /* Compute parameters for interactions between i and j atoms */
 372             qq00             = _mm_mul_ps(iq0,jq0);
 373             gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
 374                                          vdwparam+vdwioffset0+vdwjidx0B,
 375                                          vdwparam+vdwioffset0+vdwjidx0C,
 376                                          vdwparam+vdwioffset0+vdwjidx0D,
 377                                          &c6_00,&c12_00);
 378
 379             /* EWALD ELECTROSTATICS */
 380
 381             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 382             ewrt             = _mm_mul_ps(r00,ewtabscale);
 383             ewitab           = _mm_cvttps_epi32(ewrt);
 384             eweps            = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR));
 385             ewitab           = _mm_slli_epi32(ewitab,2);
 386             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 387             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 388             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 389             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 390             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 391             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 392             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 393             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
 394             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
 395
 396             /* LENNARD-JONES DISPERSION/REPULSION */
 397
 398             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 399             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 400             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 401             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
 402             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 403
 404             d                = _mm_sub_ps(r00,rswitch);
 405             d                = _mm_max_ps(d,_mm_setzero_ps());
 406             d2               = _mm_mul_ps(d,d);
 407             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)))))));
 408
 409             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 410
 411             /* Evaluate switch function */
 412             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 413             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
 414             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
 415             velec            = _mm_mul_ps(velec,sw);
 416             vvdw             = _mm_mul_ps(vvdw,sw);
 417             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
 418
 419             /* Update potential sum for this i atom from the interaction with this j atom. */
 420             velec            = _mm_and_ps(velec,cutoff_mask);
 421             velec            = _mm_andnot_ps(dummy_mask,velec);
 422             velecsum         = _mm_add_ps(velecsum,velec);
 423             vvdw             = _mm_and_ps(vvdw,cutoff_mask);
 424             vvdw             = _mm_andnot_ps(dummy_mask,vvdw);
 425             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 426
 427             fscal            = _mm_add_ps(felec,fvdw);
 428
 429             fscal            = _mm_and_ps(fscal,cutoff_mask);
 430
 431             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 432
 433             /* Calculate temporary vectorial force */
 434             tx               = _mm_mul_ps(fscal,dx00);
 435             ty               = _mm_mul_ps(fscal,dy00);
 436             tz               = _mm_mul_ps(fscal,dz00);
 437
 438             /* Update vectorial force */
 439             fix0             = _mm_add_ps(fix0,tx);
 440             fiy0             = _mm_add_ps(fiy0,ty);
 441             fiz0             = _mm_add_ps(fiz0,tz);
 442
 443             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 444             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 445             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 446             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 447             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,tx,ty,tz);
 448
 449             }
 450
 451             /* Inner loop uses 84 flops */
 452         }
 453
 454         /* End of innermost loop */
 455
 456         gmx_mm_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
 457                                               f+i_coord_offset,fshift+i_shift_offset);
 458
 459         ggid                        = gid[iidx];
 460         /* Update potential energies */
 461         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
 462         gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
 463
 464         /* Increment number of inner iterations */
 465         inneriter                  += j_index_end - j_index_start;
 466
 467         /* Outer loop uses 9 flops */
 468     }
 469
 470     /* Increment number of outer iterations */
 471     outeriter        += nri;
 472
 473     /* Update outer/inner flops */
 474
 475     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*9 + inneriter*84);
 476 }
 477 /*
 478  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwLJSw_GeomP1P1_F_sse4_1_single
 479  * Electrostatics interaction: Ewald
 480  * VdW interaction:            LennardJones
 481  * Geometry:                   Particle-Particle
 482  * Calculate force/pot:        Force
 483  */
 484 void
 485 nb_kernel_ElecEwSw_VdwLJSw_GeomP1P1_F_sse4_1_single
 486                     (t_nblist                    * gmx_restrict       nlist,
 487                      rvec                        * gmx_restrict          xx,
 488                      rvec                        * gmx_restrict          ff,
 489                      struct t_forcerec           * gmx_restrict          fr,
 490                      t_mdatoms                   * gmx_restrict     mdatoms,
 491                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 492                      t_nrnb                      * gmx_restrict        nrnb)
 493 {
 494     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 495      * just 0 for non-waters.
 496      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
 497      * jnr indices corresponding to data put in the four positions in the SIMD register.
 498      */
 499     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 500     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 501     int              jnrA,jnrB,jnrC,jnrD;
 502     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
 503     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
 504     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 505     real             rcutoff_scalar;
 506     real             *shiftvec,*fshift,*x,*f;
 507     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
 508     real             scratch[4*DIM];
 509     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 510     int              vdwioffset0;
 511     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 512     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
 513     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 514     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 515     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
 516     real             *charge;
 517     int              nvdwtype;
 518     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 519     int              *vdwtype;
 520     real             *vdwparam;
 521     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
 522     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
 523     __m128i          ewitab;
 524     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 525     real             *ewtab;
 526     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
 527     real             rswitch_scalar,d_scalar;
 528     __m128           dummy_mask,cutoff_mask;
 529     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 530     __m128           one     = _mm_set1_ps(1.0);
 531     __m128           two     = _mm_set1_ps(2.0);
 532     x                = xx[0];
 533     f                = ff[0];
 534
 535     nri              = nlist->nri;
 536     iinr             = nlist->iinr;
 537     jindex           = nlist->jindex;
 538     jjnr             = nlist->jjnr;
 539     shiftidx         = nlist->shift;
 540     gid              = nlist->gid;
 541     shiftvec         = fr->shift_vec[0];
 542     fshift           = fr->fshift[0];
 543     facel            = _mm_set1_ps(fr->ic->epsfac);
 544     charge           = mdatoms->chargeA;
 545     nvdwtype         = fr->ntype;
 546     vdwparam         = fr->nbfp;
 547     vdwtype          = mdatoms->typeA;
 548
 549     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 550     ewtab            = fr->ic->tabq_coul_FDV0;
 551     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 552     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 553
 554     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 555     rcutoff_scalar   = fr->ic->rcoulomb;
 556     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 557     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 558
 559     rswitch_scalar   = fr->ic->rcoulomb_switch;
 560     rswitch          = _mm_set1_ps(rswitch_scalar);
 561     /* Setup switch parameters */
 562     d_scalar         = rcutoff_scalar-rswitch_scalar;
 563     d                = _mm_set1_ps(d_scalar);
 564     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
 565     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 566     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 567     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
 568     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 569     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 570
 571     /* Avoid stupid compiler warnings */
 572     jnrA = jnrB = jnrC = jnrD = 0;
 573     j_coord_offsetA = 0;
 574     j_coord_offsetB = 0;
 575     j_coord_offsetC = 0;
 576     j_coord_offsetD = 0;
 577
 578     outeriter        = 0;
 579     inneriter        = 0;
 580
 581     for(iidx=0;iidx<4*DIM;iidx++)
 582     {
 583         scratch[iidx] = 0.0;
 584     }
 585
 586     /* Start outer loop over neighborlists */
 587     for(iidx=0; iidx<nri; iidx++)
 588     {
 589         /* Load shift vector for this list */
 590         i_shift_offset   = DIM*shiftidx[iidx];
 591
 592         /* Load limits for loop over neighbors */
 593         j_index_start    = jindex[iidx];
 594         j_index_end      = jindex[iidx+1];
 595
 596         /* Get outer coordinate index */
 597         inr              = iinr[iidx];
 598         i_coord_offset   = DIM*inr;
 599
 600         /* Load i particle coords and add shift vector */
 601         gmx_mm_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 602
 603         fix0             = _mm_setzero_ps();
 604         fiy0             = _mm_setzero_ps();
 605         fiz0             = _mm_setzero_ps();
 606
 607         /* Load parameters for i particles */
 608         iq0              = _mm_mul_ps(facel,_mm_load1_ps(charge+inr+0));
 609         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 610
 611         /* Start inner kernel loop */
 612         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 613         {
 614
 615             /* Get j neighbor index, and coordinate index */
 616             jnrA             = jjnr[jidx];
 617             jnrB             = jjnr[jidx+1];
 618             jnrC             = jjnr[jidx+2];
 619             jnrD             = jjnr[jidx+3];
 620             j_coord_offsetA  = DIM*jnrA;
 621             j_coord_offsetB  = DIM*jnrB;
 622             j_coord_offsetC  = DIM*jnrC;
 623             j_coord_offsetD  = DIM*jnrD;
 624
 625             /* load j atom coordinates */
 626             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 627                                               x+j_coord_offsetC,x+j_coord_offsetD,
 628                                               &jx0,&jy0,&jz0);
 629
 630             /* Calculate displacement vector */
 631             dx00             = _mm_sub_ps(ix0,jx0);
 632             dy00             = _mm_sub_ps(iy0,jy0);
 633             dz00             = _mm_sub_ps(iz0,jz0);
 634
 635             /* Calculate squared distance and things based on it */
 636             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 637
 638             rinv00           = sse41_invsqrt_f(rsq00);
 639
 640             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 641
 642             /* Load parameters for j particles */
 643             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 644                                                               charge+jnrC+0,charge+jnrD+0);
 645             vdwjidx0A        = 2*vdwtype[jnrA+0];
 646             vdwjidx0B        = 2*vdwtype[jnrB+0];
 647             vdwjidx0C        = 2*vdwtype[jnrC+0];
 648             vdwjidx0D        = 2*vdwtype[jnrD+0];
 649
 650             /**************************
 651              * CALCULATE INTERACTIONS *
 652              **************************/
 653
 654             if (gmx_mm_any_lt(rsq00,rcutoff2))
 655             {
 656
 657             r00              = _mm_mul_ps(rsq00,rinv00);
 658
 659             /* Compute parameters for interactions between i and j atoms */
 660             qq00             = _mm_mul_ps(iq0,jq0);
 661             gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
 662                                          vdwparam+vdwioffset0+vdwjidx0B,
 663                                          vdwparam+vdwioffset0+vdwjidx0C,
 664                                          vdwparam+vdwioffset0+vdwjidx0D,
 665                                          &c6_00,&c12_00);
 666
 667             /* EWALD ELECTROSTATICS */
 668
 669             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 670             ewrt             = _mm_mul_ps(r00,ewtabscale);
 671             ewitab           = _mm_cvttps_epi32(ewrt);
 672             eweps            = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR));
 673             ewitab           = _mm_slli_epi32(ewitab,2);
 674             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 675             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 676             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 677             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 678             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 679             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 680             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 681             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
 682             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
 683
 684             /* LENNARD-JONES DISPERSION/REPULSION */
 685
 686             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 687             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 688             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 689             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
 690             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 691
 692             d                = _mm_sub_ps(r00,rswitch);
 693             d                = _mm_max_ps(d,_mm_setzero_ps());
 694             d2               = _mm_mul_ps(d,d);
 695             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)))))));
 696
 697             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 698
 699             /* Evaluate switch function */
 700             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 701             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
 702             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
 703             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
 704
 705             fscal            = _mm_add_ps(felec,fvdw);
 706
 707             fscal            = _mm_and_ps(fscal,cutoff_mask);
 708
 709             /* Calculate temporary vectorial force */
 710             tx               = _mm_mul_ps(fscal,dx00);
 711             ty               = _mm_mul_ps(fscal,dy00);
 712             tz               = _mm_mul_ps(fscal,dz00);
 713
 714             /* Update vectorial force */
 715             fix0             = _mm_add_ps(fix0,tx);
 716             fiy0             = _mm_add_ps(fiy0,ty);
 717             fiz0             = _mm_add_ps(fiz0,tz);
 718
 719             fjptrA             = f+j_coord_offsetA;
 720             fjptrB             = f+j_coord_offsetB;
 721             fjptrC             = f+j_coord_offsetC;
 722             fjptrD             = f+j_coord_offsetD;
 723             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,tx,ty,tz);
 724
 725             }
 726
 727             /* Inner loop uses 77 flops */
 728         }
 729
 730         if(jidx<j_index_end)
 731         {
 732
 733             /* Get j neighbor index, and coordinate index */
 734             jnrlistA         = jjnr[jidx];
 735             jnrlistB         = jjnr[jidx+1];
 736             jnrlistC         = jjnr[jidx+2];
 737             jnrlistD         = jjnr[jidx+3];
 738             /* Sign of each element will be negative for non-real atoms.
 739              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 740              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 741              */
 742             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 743             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 744             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 745             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 746             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 747             j_coord_offsetA  = DIM*jnrA;
 748             j_coord_offsetB  = DIM*jnrB;
 749             j_coord_offsetC  = DIM*jnrC;
 750             j_coord_offsetD  = DIM*jnrD;
 751
 752             /* load j atom coordinates */
 753             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 754                                               x+j_coord_offsetC,x+j_coord_offsetD,
 755                                               &jx0,&jy0,&jz0);
 756
 757             /* Calculate displacement vector */
 758             dx00             = _mm_sub_ps(ix0,jx0);
 759             dy00             = _mm_sub_ps(iy0,jy0);
 760             dz00             = _mm_sub_ps(iz0,jz0);
 761
 762             /* Calculate squared distance and things based on it */
 763             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 764
 765             rinv00           = sse41_invsqrt_f(rsq00);
 766
 767             rinvsq00         = _mm_mul_ps(rinv00,rinv00);
 768
 769             /* Load parameters for j particles */
 770             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 771                                                               charge+jnrC+0,charge+jnrD+0);
 772             vdwjidx0A        = 2*vdwtype[jnrA+0];
 773             vdwjidx0B        = 2*vdwtype[jnrB+0];
 774             vdwjidx0C        = 2*vdwtype[jnrC+0];
 775             vdwjidx0D        = 2*vdwtype[jnrD+0];
 776
 777             /**************************
 778              * CALCULATE INTERACTIONS *
 779              **************************/
 780
 781             if (gmx_mm_any_lt(rsq00,rcutoff2))
 782             {
 783
 784             r00              = _mm_mul_ps(rsq00,rinv00);
 785             r00              = _mm_andnot_ps(dummy_mask,r00);
 786
 787             /* Compute parameters for interactions between i and j atoms */
 788             qq00             = _mm_mul_ps(iq0,jq0);
 789             gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A,
 790                                          vdwparam+vdwioffset0+vdwjidx0B,
 791                                          vdwparam+vdwioffset0+vdwjidx0C,
 792                                          vdwparam+vdwioffset0+vdwjidx0D,
 793                                          &c6_00,&c12_00);
 794
 795             /* EWALD ELECTROSTATICS */
 796
 797             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 798             ewrt             = _mm_mul_ps(r00,ewtabscale);
 799             ewitab           = _mm_cvttps_epi32(ewrt);
 800             eweps            = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR));
 801             ewitab           = _mm_slli_epi32(ewitab,2);
 802             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 803             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 804             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 805             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 806             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 807             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 808             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 809             velec            = _mm_mul_ps(qq00,_mm_sub_ps(rinv00,velec));
 810             felec            = _mm_mul_ps(_mm_mul_ps(qq00,rinv00),_mm_sub_ps(rinvsq00,felec));
 811
 812             /* LENNARD-JONES DISPERSION/REPULSION */
 813
 814             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 815             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 816             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 817             vvdw             = _mm_sub_ps( _mm_mul_ps(vvdw12,one_twelfth) , _mm_mul_ps(vvdw6,one_sixth) );
 818             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 819
 820             d                = _mm_sub_ps(r00,rswitch);
 821             d                = _mm_max_ps(d,_mm_setzero_ps());
 822             d2               = _mm_mul_ps(d,d);
 823             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)))))));
 824
 825             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 826
 827             /* Evaluate switch function */
 828             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 829             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(velec,dsw)) );
 830             fvdw             = _mm_sub_ps( _mm_mul_ps(fvdw,sw) , _mm_mul_ps(rinv00,_mm_mul_ps(vvdw,dsw)) );
 831             cutoff_mask      = _mm_cmplt_ps(rsq00,rcutoff2);
 832
 833             fscal            = _mm_add_ps(felec,fvdw);
 834
 835             fscal            = _mm_and_ps(fscal,cutoff_mask);
 836
 837             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 838
 839             /* Calculate temporary vectorial force */
 840             tx               = _mm_mul_ps(fscal,dx00);
 841             ty               = _mm_mul_ps(fscal,dy00);
 842             tz               = _mm_mul_ps(fscal,dz00);
 843
 844             /* Update vectorial force */
 845             fix0             = _mm_add_ps(fix0,tx);
 846             fiy0             = _mm_add_ps(fiy0,ty);
 847             fiz0             = _mm_add_ps(fiz0,tz);
 848
 849             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 850             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 851             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 852             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 853             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,tx,ty,tz);
 854
 855             }
 856
 857             /* Inner loop uses 78 flops */
 858         }
 859
 860         /* End of innermost loop */
 861
 862         gmx_mm_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
 863                                               f+i_coord_offset,fshift+i_shift_offset);
 864
 865         /* Increment number of inner iterations */
 866         inneriter                  += j_index_end - j_index_start;
 867
 868         /* Outer loop uses 7 flops */
 869     }
 870
 871     /* Increment number of outer iterations */
 872     outeriter        += nri;
 873
 874     /* Update outer/inner flops */
 875
 876     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*7 + inneriter*78);
 877 }