src/gmxlib/nonbonded/nb_kernel_sse2_single/nb_kernel_ElecEwSw_VdwNone_GeomW4P1_sse2_single.c

   1 /*
   2  * Note: this file was generated by the Gromacs sse2_single kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_sse2_single.h"
  34 #include "kernelutil_x86_sse2_single.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwNone_GeomW4P1_VF_sse2_single
  38  * Electrostatics interaction: Ewald
  39  * VdW interaction:            None
  40  * Geometry:                   Water4-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecEwSw_VdwNone_GeomW4P1_VF_sse2_single
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB,jnrC,jnrD;
  61     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  62     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  63     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  64     real             rcutoff_scalar;
  65     real             *shiftvec,*fshift,*x,*f;
  66     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  67     real             scratch[4*DIM];
  68     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  69     int              vdwioffset1;
  70     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  71     int              vdwioffset2;
  72     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  73     int              vdwioffset3;
  74     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  75     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  76     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  77     __m128           dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  78     __m128           dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  79     __m128           dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
  80     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
  81     real             *charge;
  82     __m128i          ewitab;
  83     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
  84     real             *ewtab;
  85     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
  86     real             rswitch_scalar,d_scalar;
  87     __m128           dummy_mask,cutoff_mask;
  88     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
  89     __m128           one     = _mm_set1_ps(1.0);
  90     __m128           two     = _mm_set1_ps(2.0);
  91     x                = xx[0];
  92     f                = ff[0];
  93
  94     nri              = nlist->nri;
  95     iinr             = nlist->iinr;
  96     jindex           = nlist->jindex;
  97     jjnr             = nlist->jjnr;
  98     shiftidx         = nlist->shift;
  99     gid              = nlist->gid;
 100     shiftvec         = fr->shift_vec[0];
 101     fshift           = fr->fshift[0];
 102     facel            = _mm_set1_ps(fr->epsfac);
 103     charge           = mdatoms->chargeA;
 104
 105     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 106     ewtab            = fr->ic->tabq_coul_FDV0;
 107     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 108     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 109
 110     /* Setup water-specific parameters */
 111     inr              = nlist->iinr[0];
 112     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 113     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 114     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
 115
 116     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 117     rcutoff_scalar   = fr->rcoulomb;
 118     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 119     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 120
 121     rswitch_scalar   = fr->rcoulomb_switch;
 122     rswitch          = _mm_set1_ps(rswitch_scalar);
 123     /* Setup switch parameters */
 124     d_scalar         = rcutoff_scalar-rswitch_scalar;
 125     d                = _mm_set1_ps(d_scalar);
 126     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
 127     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 128     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 129     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
 130     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 131     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 132
 133     /* Avoid stupid compiler warnings */
 134     jnrA = jnrB = jnrC = jnrD = 0;
 135     j_coord_offsetA = 0;
 136     j_coord_offsetB = 0;
 137     j_coord_offsetC = 0;
 138     j_coord_offsetD = 0;
 139
 140     outeriter        = 0;
 141     inneriter        = 0;
 142
 143     for(iidx=0;iidx<4*DIM;iidx++)
 144     {
 145         scratch[iidx] = 0.0;
 146     }
 147
 148     /* Start outer loop over neighborlists */
 149     for(iidx=0; iidx<nri; iidx++)
 150     {
 151         /* Load shift vector for this list */
 152         i_shift_offset   = DIM*shiftidx[iidx];
 153
 154         /* Load limits for loop over neighbors */
 155         j_index_start    = jindex[iidx];
 156         j_index_end      = jindex[iidx+1];
 157
 158         /* Get outer coordinate index */
 159         inr              = iinr[iidx];
 160         i_coord_offset   = DIM*inr;
 161
 162         /* Load i particle coords and add shift vector */
 163         gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
 164                                                  &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 165
 166         fix1             = _mm_setzero_ps();
 167         fiy1             = _mm_setzero_ps();
 168         fiz1             = _mm_setzero_ps();
 169         fix2             = _mm_setzero_ps();
 170         fiy2             = _mm_setzero_ps();
 171         fiz2             = _mm_setzero_ps();
 172         fix3             = _mm_setzero_ps();
 173         fiy3             = _mm_setzero_ps();
 174         fiz3             = _mm_setzero_ps();
 175
 176         /* Reset potential sums */
 177         velecsum         = _mm_setzero_ps();
 178
 179         /* Start inner kernel loop */
 180         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 181         {
 182
 183             /* Get j neighbor index, and coordinate index */
 184             jnrA             = jjnr[jidx];
 185             jnrB             = jjnr[jidx+1];
 186             jnrC             = jjnr[jidx+2];
 187             jnrD             = jjnr[jidx+3];
 188             j_coord_offsetA  = DIM*jnrA;
 189             j_coord_offsetB  = DIM*jnrB;
 190             j_coord_offsetC  = DIM*jnrC;
 191             j_coord_offsetD  = DIM*jnrD;
 192
 193             /* load j atom coordinates */
 194             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 195                                               x+j_coord_offsetC,x+j_coord_offsetD,
 196                                               &jx0,&jy0,&jz0);
 197
 198             /* Calculate displacement vector */
 199             dx10             = _mm_sub_ps(ix1,jx0);
 200             dy10             = _mm_sub_ps(iy1,jy0);
 201             dz10             = _mm_sub_ps(iz1,jz0);
 202             dx20             = _mm_sub_ps(ix2,jx0);
 203             dy20             = _mm_sub_ps(iy2,jy0);
 204             dz20             = _mm_sub_ps(iz2,jz0);
 205             dx30             = _mm_sub_ps(ix3,jx0);
 206             dy30             = _mm_sub_ps(iy3,jy0);
 207             dz30             = _mm_sub_ps(iz3,jz0);
 208
 209             /* Calculate squared distance and things based on it */
 210             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
 211             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
 212             rsq30            = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
 213
 214             rinv10           = gmx_mm_invsqrt_ps(rsq10);
 215             rinv20           = gmx_mm_invsqrt_ps(rsq20);
 216             rinv30           = gmx_mm_invsqrt_ps(rsq30);
 217
 218             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
 219             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
 220             rinvsq30         = _mm_mul_ps(rinv30,rinv30);
 221
 222             /* Load parameters for j particles */
 223             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 224                                                               charge+jnrC+0,charge+jnrD+0);
 225
 226             fjx0             = _mm_setzero_ps();
 227             fjy0             = _mm_setzero_ps();
 228             fjz0             = _mm_setzero_ps();
 229
 230             /**************************
 231              * CALCULATE INTERACTIONS *
 232              **************************/
 233
 234             if (gmx_mm_any_lt(rsq10,rcutoff2))
 235             {
 236
 237             r10              = _mm_mul_ps(rsq10,rinv10);
 238
 239             /* Compute parameters for interactions between i and j atoms */
 240             qq10             = _mm_mul_ps(iq1,jq0);
 241
 242             /* EWALD ELECTROSTATICS */
 243
 244             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 245             ewrt             = _mm_mul_ps(r10,ewtabscale);
 246             ewitab           = _mm_cvttps_epi32(ewrt);
 247             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 248             ewitab           = _mm_slli_epi32(ewitab,2);
 249             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 250             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 251             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 252             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 253             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 254             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 255             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 256             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
 257             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
 258
 259             d                = _mm_sub_ps(r10,rswitch);
 260             d                = _mm_max_ps(d,_mm_setzero_ps());
 261             d2               = _mm_mul_ps(d,d);
 262             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)))))));
 263
 264             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 265
 266             /* Evaluate switch function */
 267             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 268             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
 269             velec            = _mm_mul_ps(velec,sw);
 270             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
 271
 272             /* Update potential sum for this i atom from the interaction with this j atom. */
 273             velec            = _mm_and_ps(velec,cutoff_mask);
 274             velecsum         = _mm_add_ps(velecsum,velec);
 275
 276             fscal            = felec;
 277
 278             fscal            = _mm_and_ps(fscal,cutoff_mask);
 279
 280             /* Calculate temporary vectorial force */
 281             tx               = _mm_mul_ps(fscal,dx10);
 282             ty               = _mm_mul_ps(fscal,dy10);
 283             tz               = _mm_mul_ps(fscal,dz10);
 284
 285             /* Update vectorial force */
 286             fix1             = _mm_add_ps(fix1,tx);
 287             fiy1             = _mm_add_ps(fiy1,ty);
 288             fiz1             = _mm_add_ps(fiz1,tz);
 289
 290             fjx0             = _mm_add_ps(fjx0,tx);
 291             fjy0             = _mm_add_ps(fjy0,ty);
 292             fjz0             = _mm_add_ps(fjz0,tz);
 293
 294             }
 295
 296             /**************************
 297              * CALCULATE INTERACTIONS *
 298              **************************/
 299
 300             if (gmx_mm_any_lt(rsq20,rcutoff2))
 301             {
 302
 303             r20              = _mm_mul_ps(rsq20,rinv20);
 304
 305             /* Compute parameters for interactions between i and j atoms */
 306             qq20             = _mm_mul_ps(iq2,jq0);
 307
 308             /* EWALD ELECTROSTATICS */
 309
 310             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 311             ewrt             = _mm_mul_ps(r20,ewtabscale);
 312             ewitab           = _mm_cvttps_epi32(ewrt);
 313             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 314             ewitab           = _mm_slli_epi32(ewitab,2);
 315             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 316             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 317             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 318             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 319             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 320             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 321             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 322             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
 323             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
 324
 325             d                = _mm_sub_ps(r20,rswitch);
 326             d                = _mm_max_ps(d,_mm_setzero_ps());
 327             d2               = _mm_mul_ps(d,d);
 328             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)))))));
 329
 330             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 331
 332             /* Evaluate switch function */
 333             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 334             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
 335             velec            = _mm_mul_ps(velec,sw);
 336             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
 337
 338             /* Update potential sum for this i atom from the interaction with this j atom. */
 339             velec            = _mm_and_ps(velec,cutoff_mask);
 340             velecsum         = _mm_add_ps(velecsum,velec);
 341
 342             fscal            = felec;
 343
 344             fscal            = _mm_and_ps(fscal,cutoff_mask);
 345
 346             /* Calculate temporary vectorial force */
 347             tx               = _mm_mul_ps(fscal,dx20);
 348             ty               = _mm_mul_ps(fscal,dy20);
 349             tz               = _mm_mul_ps(fscal,dz20);
 350
 351             /* Update vectorial force */
 352             fix2             = _mm_add_ps(fix2,tx);
 353             fiy2             = _mm_add_ps(fiy2,ty);
 354             fiz2             = _mm_add_ps(fiz2,tz);
 355
 356             fjx0             = _mm_add_ps(fjx0,tx);
 357             fjy0             = _mm_add_ps(fjy0,ty);
 358             fjz0             = _mm_add_ps(fjz0,tz);
 359
 360             }
 361
 362             /**************************
 363              * CALCULATE INTERACTIONS *
 364              **************************/
 365
 366             if (gmx_mm_any_lt(rsq30,rcutoff2))
 367             {
 368
 369             r30              = _mm_mul_ps(rsq30,rinv30);
 370
 371             /* Compute parameters for interactions between i and j atoms */
 372             qq30             = _mm_mul_ps(iq3,jq0);
 373
 374             /* EWALD ELECTROSTATICS */
 375
 376             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 377             ewrt             = _mm_mul_ps(r30,ewtabscale);
 378             ewitab           = _mm_cvttps_epi32(ewrt);
 379             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 380             ewitab           = _mm_slli_epi32(ewitab,2);
 381             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 382             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 383             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 384             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 385             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 386             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 387             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 388             velec            = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec));
 389             felec            = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec));
 390
 391             d                = _mm_sub_ps(r30,rswitch);
 392             d                = _mm_max_ps(d,_mm_setzero_ps());
 393             d2               = _mm_mul_ps(d,d);
 394             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)))))));
 395
 396             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 397
 398             /* Evaluate switch function */
 399             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 400             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv30,_mm_mul_ps(velec,dsw)) );
 401             velec            = _mm_mul_ps(velec,sw);
 402             cutoff_mask      = _mm_cmplt_ps(rsq30,rcutoff2);
 403
 404             /* Update potential sum for this i atom from the interaction with this j atom. */
 405             velec            = _mm_and_ps(velec,cutoff_mask);
 406             velecsum         = _mm_add_ps(velecsum,velec);
 407
 408             fscal            = felec;
 409
 410             fscal            = _mm_and_ps(fscal,cutoff_mask);
 411
 412             /* Calculate temporary vectorial force */
 413             tx               = _mm_mul_ps(fscal,dx30);
 414             ty               = _mm_mul_ps(fscal,dy30);
 415             tz               = _mm_mul_ps(fscal,dz30);
 416
 417             /* Update vectorial force */
 418             fix3             = _mm_add_ps(fix3,tx);
 419             fiy3             = _mm_add_ps(fiy3,ty);
 420             fiz3             = _mm_add_ps(fiz3,tz);
 421
 422             fjx0             = _mm_add_ps(fjx0,tx);
 423             fjy0             = _mm_add_ps(fjy0,ty);
 424             fjz0             = _mm_add_ps(fjz0,tz);
 425
 426             }
 427
 428             fjptrA             = f+j_coord_offsetA;
 429             fjptrB             = f+j_coord_offsetB;
 430             fjptrC             = f+j_coord_offsetC;
 431             fjptrD             = f+j_coord_offsetD;
 432
 433             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
 434
 435             /* Inner loop uses 195 flops */
 436         }
 437
 438         if(jidx<j_index_end)
 439         {
 440
 441             /* Get j neighbor index, and coordinate index */
 442             jnrlistA         = jjnr[jidx];
 443             jnrlistB         = jjnr[jidx+1];
 444             jnrlistC         = jjnr[jidx+2];
 445             jnrlistD         = jjnr[jidx+3];
 446             /* Sign of each element will be negative for non-real atoms.
 447              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 448              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 449              */
 450             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 451             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 452             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 453             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 454             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 455             j_coord_offsetA  = DIM*jnrA;
 456             j_coord_offsetB  = DIM*jnrB;
 457             j_coord_offsetC  = DIM*jnrC;
 458             j_coord_offsetD  = DIM*jnrD;
 459
 460             /* load j atom coordinates */
 461             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 462                                               x+j_coord_offsetC,x+j_coord_offsetD,
 463                                               &jx0,&jy0,&jz0);
 464
 465             /* Calculate displacement vector */
 466             dx10             = _mm_sub_ps(ix1,jx0);
 467             dy10             = _mm_sub_ps(iy1,jy0);
 468             dz10             = _mm_sub_ps(iz1,jz0);
 469             dx20             = _mm_sub_ps(ix2,jx0);
 470             dy20             = _mm_sub_ps(iy2,jy0);
 471             dz20             = _mm_sub_ps(iz2,jz0);
 472             dx30             = _mm_sub_ps(ix3,jx0);
 473             dy30             = _mm_sub_ps(iy3,jy0);
 474             dz30             = _mm_sub_ps(iz3,jz0);
 475
 476             /* Calculate squared distance and things based on it */
 477             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
 478             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
 479             rsq30            = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
 480
 481             rinv10           = gmx_mm_invsqrt_ps(rsq10);
 482             rinv20           = gmx_mm_invsqrt_ps(rsq20);
 483             rinv30           = gmx_mm_invsqrt_ps(rsq30);
 484
 485             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
 486             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
 487             rinvsq30         = _mm_mul_ps(rinv30,rinv30);
 488
 489             /* Load parameters for j particles */
 490             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 491                                                               charge+jnrC+0,charge+jnrD+0);
 492
 493             fjx0             = _mm_setzero_ps();
 494             fjy0             = _mm_setzero_ps();
 495             fjz0             = _mm_setzero_ps();
 496
 497             /**************************
 498              * CALCULATE INTERACTIONS *
 499              **************************/
 500
 501             if (gmx_mm_any_lt(rsq10,rcutoff2))
 502             {
 503
 504             r10              = _mm_mul_ps(rsq10,rinv10);
 505             r10              = _mm_andnot_ps(dummy_mask,r10);
 506
 507             /* Compute parameters for interactions between i and j atoms */
 508             qq10             = _mm_mul_ps(iq1,jq0);
 509
 510             /* EWALD ELECTROSTATICS */
 511
 512             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 513             ewrt             = _mm_mul_ps(r10,ewtabscale);
 514             ewitab           = _mm_cvttps_epi32(ewrt);
 515             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 516             ewitab           = _mm_slli_epi32(ewitab,2);
 517             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 518             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 519             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 520             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 521             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 522             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 523             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 524             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
 525             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
 526
 527             d                = _mm_sub_ps(r10,rswitch);
 528             d                = _mm_max_ps(d,_mm_setzero_ps());
 529             d2               = _mm_mul_ps(d,d);
 530             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)))))));
 531
 532             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 533
 534             /* Evaluate switch function */
 535             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 536             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
 537             velec            = _mm_mul_ps(velec,sw);
 538             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
 539
 540             /* Update potential sum for this i atom from the interaction with this j atom. */
 541             velec            = _mm_and_ps(velec,cutoff_mask);
 542             velec            = _mm_andnot_ps(dummy_mask,velec);
 543             velecsum         = _mm_add_ps(velecsum,velec);
 544
 545             fscal            = felec;
 546
 547             fscal            = _mm_and_ps(fscal,cutoff_mask);
 548
 549             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 550
 551             /* Calculate temporary vectorial force */
 552             tx               = _mm_mul_ps(fscal,dx10);
 553             ty               = _mm_mul_ps(fscal,dy10);
 554             tz               = _mm_mul_ps(fscal,dz10);
 555
 556             /* Update vectorial force */
 557             fix1             = _mm_add_ps(fix1,tx);
 558             fiy1             = _mm_add_ps(fiy1,ty);
 559             fiz1             = _mm_add_ps(fiz1,tz);
 560
 561             fjx0             = _mm_add_ps(fjx0,tx);
 562             fjy0             = _mm_add_ps(fjy0,ty);
 563             fjz0             = _mm_add_ps(fjz0,tz);
 564
 565             }
 566
 567             /**************************
 568              * CALCULATE INTERACTIONS *
 569              **************************/
 570
 571             if (gmx_mm_any_lt(rsq20,rcutoff2))
 572             {
 573
 574             r20              = _mm_mul_ps(rsq20,rinv20);
 575             r20              = _mm_andnot_ps(dummy_mask,r20);
 576
 577             /* Compute parameters for interactions between i and j atoms */
 578             qq20             = _mm_mul_ps(iq2,jq0);
 579
 580             /* EWALD ELECTROSTATICS */
 581
 582             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 583             ewrt             = _mm_mul_ps(r20,ewtabscale);
 584             ewitab           = _mm_cvttps_epi32(ewrt);
 585             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 586             ewitab           = _mm_slli_epi32(ewitab,2);
 587             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 588             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 589             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 590             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 591             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 592             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 593             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 594             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
 595             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
 596
 597             d                = _mm_sub_ps(r20,rswitch);
 598             d                = _mm_max_ps(d,_mm_setzero_ps());
 599             d2               = _mm_mul_ps(d,d);
 600             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)))))));
 601
 602             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 603
 604             /* Evaluate switch function */
 605             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 606             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
 607             velec            = _mm_mul_ps(velec,sw);
 608             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
 609
 610             /* Update potential sum for this i atom from the interaction with this j atom. */
 611             velec            = _mm_and_ps(velec,cutoff_mask);
 612             velec            = _mm_andnot_ps(dummy_mask,velec);
 613             velecsum         = _mm_add_ps(velecsum,velec);
 614
 615             fscal            = felec;
 616
 617             fscal            = _mm_and_ps(fscal,cutoff_mask);
 618
 619             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 620
 621             /* Calculate temporary vectorial force */
 622             tx               = _mm_mul_ps(fscal,dx20);
 623             ty               = _mm_mul_ps(fscal,dy20);
 624             tz               = _mm_mul_ps(fscal,dz20);
 625
 626             /* Update vectorial force */
 627             fix2             = _mm_add_ps(fix2,tx);
 628             fiy2             = _mm_add_ps(fiy2,ty);
 629             fiz2             = _mm_add_ps(fiz2,tz);
 630
 631             fjx0             = _mm_add_ps(fjx0,tx);
 632             fjy0             = _mm_add_ps(fjy0,ty);
 633             fjz0             = _mm_add_ps(fjz0,tz);
 634
 635             }
 636
 637             /**************************
 638              * CALCULATE INTERACTIONS *
 639              **************************/
 640
 641             if (gmx_mm_any_lt(rsq30,rcutoff2))
 642             {
 643
 644             r30              = _mm_mul_ps(rsq30,rinv30);
 645             r30              = _mm_andnot_ps(dummy_mask,r30);
 646
 647             /* Compute parameters for interactions between i and j atoms */
 648             qq30             = _mm_mul_ps(iq3,jq0);
 649
 650             /* EWALD ELECTROSTATICS */
 651
 652             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 653             ewrt             = _mm_mul_ps(r30,ewtabscale);
 654             ewitab           = _mm_cvttps_epi32(ewrt);
 655             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 656             ewitab           = _mm_slli_epi32(ewitab,2);
 657             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 658             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 659             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 660             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 661             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 662             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 663             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 664             velec            = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec));
 665             felec            = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec));
 666
 667             d                = _mm_sub_ps(r30,rswitch);
 668             d                = _mm_max_ps(d,_mm_setzero_ps());
 669             d2               = _mm_mul_ps(d,d);
 670             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)))))));
 671
 672             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 673
 674             /* Evaluate switch function */
 675             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 676             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv30,_mm_mul_ps(velec,dsw)) );
 677             velec            = _mm_mul_ps(velec,sw);
 678             cutoff_mask      = _mm_cmplt_ps(rsq30,rcutoff2);
 679
 680             /* Update potential sum for this i atom from the interaction with this j atom. */
 681             velec            = _mm_and_ps(velec,cutoff_mask);
 682             velec            = _mm_andnot_ps(dummy_mask,velec);
 683             velecsum         = _mm_add_ps(velecsum,velec);
 684
 685             fscal            = felec;
 686
 687             fscal            = _mm_and_ps(fscal,cutoff_mask);
 688
 689             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 690
 691             /* Calculate temporary vectorial force */
 692             tx               = _mm_mul_ps(fscal,dx30);
 693             ty               = _mm_mul_ps(fscal,dy30);
 694             tz               = _mm_mul_ps(fscal,dz30);
 695
 696             /* Update vectorial force */
 697             fix3             = _mm_add_ps(fix3,tx);
 698             fiy3             = _mm_add_ps(fiy3,ty);
 699             fiz3             = _mm_add_ps(fiz3,tz);
 700
 701             fjx0             = _mm_add_ps(fjx0,tx);
 702             fjy0             = _mm_add_ps(fjy0,ty);
 703             fjz0             = _mm_add_ps(fjz0,tz);
 704
 705             }
 706
 707             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 708             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 709             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 710             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 711
 712             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
 713
 714             /* Inner loop uses 198 flops */
 715         }
 716
 717         /* End of innermost loop */
 718
 719         gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 720                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
 721
 722         ggid                        = gid[iidx];
 723         /* Update potential energies */
 724         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
 725
 726         /* Increment number of inner iterations */
 727         inneriter                  += j_index_end - j_index_start;
 728
 729         /* Outer loop uses 19 flops */
 730     }
 731
 732     /* Increment number of outer iterations */
 733     outeriter        += nri;
 734
 735     /* Update outer/inner flops */
 736
 737     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_VF,outeriter*19 + inneriter*198);
 738 }
 739 /*
 740  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSw_VdwNone_GeomW4P1_F_sse2_single
 741  * Electrostatics interaction: Ewald
 742  * VdW interaction:            None
 743  * Geometry:                   Water4-Particle
 744  * Calculate force/pot:        Force
 745  */
 746 void
 747 nb_kernel_ElecEwSw_VdwNone_GeomW4P1_F_sse2_single
 748                     (t_nblist * gmx_restrict                nlist,
 749                      rvec * gmx_restrict                    xx,
 750                      rvec * gmx_restrict                    ff,
 751                      t_forcerec * gmx_restrict              fr,
 752                      t_mdatoms * gmx_restrict               mdatoms,
 753                      nb_kernel_data_t * gmx_restrict        kernel_data,
 754                      t_nrnb * gmx_restrict                  nrnb)
 755 {
 756     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 757      * just 0 for non-waters.
 758      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
 759      * jnr indices corresponding to data put in the four positions in the SIMD register.
 760      */
 761     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 762     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 763     int              jnrA,jnrB,jnrC,jnrD;
 764     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
 765     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
 766     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 767     real             rcutoff_scalar;
 768     real             *shiftvec,*fshift,*x,*f;
 769     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
 770     real             scratch[4*DIM];
 771     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 772     int              vdwioffset1;
 773     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 774     int              vdwioffset2;
 775     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 776     int              vdwioffset3;
 777     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 778     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
 779     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 780     __m128           dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 781     __m128           dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 782     __m128           dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
 783     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
 784     real             *charge;
 785     __m128i          ewitab;
 786     __m128           ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 787     real             *ewtab;
 788     __m128           rswitch,swV3,swV4,swV5,swF2,swF3,swF4,d,d2,sw,dsw;
 789     real             rswitch_scalar,d_scalar;
 790     __m128           dummy_mask,cutoff_mask;
 791     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 792     __m128           one     = _mm_set1_ps(1.0);
 793     __m128           two     = _mm_set1_ps(2.0);
 794     x                = xx[0];
 795     f                = ff[0];
 796
 797     nri              = nlist->nri;
 798     iinr             = nlist->iinr;
 799     jindex           = nlist->jindex;
 800     jjnr             = nlist->jjnr;
 801     shiftidx         = nlist->shift;
 802     gid              = nlist->gid;
 803     shiftvec         = fr->shift_vec[0];
 804     fshift           = fr->fshift[0];
 805     facel            = _mm_set1_ps(fr->epsfac);
 806     charge           = mdatoms->chargeA;
 807
 808     sh_ewald         = _mm_set1_ps(fr->ic->sh_ewald);
 809     ewtab            = fr->ic->tabq_coul_FDV0;
 810     ewtabscale       = _mm_set1_ps(fr->ic->tabq_scale);
 811     ewtabhalfspace   = _mm_set1_ps(0.5/fr->ic->tabq_scale);
 812
 813     /* Setup water-specific parameters */
 814     inr              = nlist->iinr[0];
 815     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 816     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 817     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
 818
 819     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 820     rcutoff_scalar   = fr->rcoulomb;
 821     rcutoff          = _mm_set1_ps(rcutoff_scalar);
 822     rcutoff2         = _mm_mul_ps(rcutoff,rcutoff);
 823
 824     rswitch_scalar   = fr->rcoulomb_switch;
 825     rswitch          = _mm_set1_ps(rswitch_scalar);
 826     /* Setup switch parameters */
 827     d_scalar         = rcutoff_scalar-rswitch_scalar;
 828     d                = _mm_set1_ps(d_scalar);
 829     swV3             = _mm_set1_ps(-10.0/(d_scalar*d_scalar*d_scalar));
 830     swV4             = _mm_set1_ps( 15.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 831     swV5             = _mm_set1_ps( -6.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 832     swF2             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar));
 833     swF3             = _mm_set1_ps( 60.0/(d_scalar*d_scalar*d_scalar*d_scalar));
 834     swF4             = _mm_set1_ps(-30.0/(d_scalar*d_scalar*d_scalar*d_scalar*d_scalar));
 835
 836     /* Avoid stupid compiler warnings */
 837     jnrA = jnrB = jnrC = jnrD = 0;
 838     j_coord_offsetA = 0;
 839     j_coord_offsetB = 0;
 840     j_coord_offsetC = 0;
 841     j_coord_offsetD = 0;
 842
 843     outeriter        = 0;
 844     inneriter        = 0;
 845
 846     for(iidx=0;iidx<4*DIM;iidx++)
 847     {
 848         scratch[iidx] = 0.0;
 849     }
 850
 851     /* Start outer loop over neighborlists */
 852     for(iidx=0; iidx<nri; iidx++)
 853     {
 854         /* Load shift vector for this list */
 855         i_shift_offset   = DIM*shiftidx[iidx];
 856
 857         /* Load limits for loop over neighbors */
 858         j_index_start    = jindex[iidx];
 859         j_index_end      = jindex[iidx+1];
 860
 861         /* Get outer coordinate index */
 862         inr              = iinr[iidx];
 863         i_coord_offset   = DIM*inr;
 864
 865         /* Load i particle coords and add shift vector */
 866         gmx_mm_load_shift_and_3rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset+DIM,
 867                                                  &ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 868
 869         fix1             = _mm_setzero_ps();
 870         fiy1             = _mm_setzero_ps();
 871         fiz1             = _mm_setzero_ps();
 872         fix2             = _mm_setzero_ps();
 873         fiy2             = _mm_setzero_ps();
 874         fiz2             = _mm_setzero_ps();
 875         fix3             = _mm_setzero_ps();
 876         fiy3             = _mm_setzero_ps();
 877         fiz3             = _mm_setzero_ps();
 878
 879         /* Start inner kernel loop */
 880         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 881         {
 882
 883             /* Get j neighbor index, and coordinate index */
 884             jnrA             = jjnr[jidx];
 885             jnrB             = jjnr[jidx+1];
 886             jnrC             = jjnr[jidx+2];
 887             jnrD             = jjnr[jidx+3];
 888             j_coord_offsetA  = DIM*jnrA;
 889             j_coord_offsetB  = DIM*jnrB;
 890             j_coord_offsetC  = DIM*jnrC;
 891             j_coord_offsetD  = DIM*jnrD;
 892
 893             /* load j atom coordinates */
 894             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 895                                               x+j_coord_offsetC,x+j_coord_offsetD,
 896                                               &jx0,&jy0,&jz0);
 897
 898             /* Calculate displacement vector */
 899             dx10             = _mm_sub_ps(ix1,jx0);
 900             dy10             = _mm_sub_ps(iy1,jy0);
 901             dz10             = _mm_sub_ps(iz1,jz0);
 902             dx20             = _mm_sub_ps(ix2,jx0);
 903             dy20             = _mm_sub_ps(iy2,jy0);
 904             dz20             = _mm_sub_ps(iz2,jz0);
 905             dx30             = _mm_sub_ps(ix3,jx0);
 906             dy30             = _mm_sub_ps(iy3,jy0);
 907             dz30             = _mm_sub_ps(iz3,jz0);
 908
 909             /* Calculate squared distance and things based on it */
 910             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
 911             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
 912             rsq30            = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
 913
 914             rinv10           = gmx_mm_invsqrt_ps(rsq10);
 915             rinv20           = gmx_mm_invsqrt_ps(rsq20);
 916             rinv30           = gmx_mm_invsqrt_ps(rsq30);
 917
 918             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
 919             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
 920             rinvsq30         = _mm_mul_ps(rinv30,rinv30);
 921
 922             /* Load parameters for j particles */
 923             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 924                                                               charge+jnrC+0,charge+jnrD+0);
 925
 926             fjx0             = _mm_setzero_ps();
 927             fjy0             = _mm_setzero_ps();
 928             fjz0             = _mm_setzero_ps();
 929
 930             /**************************
 931              * CALCULATE INTERACTIONS *
 932              **************************/
 933
 934             if (gmx_mm_any_lt(rsq10,rcutoff2))
 935             {
 936
 937             r10              = _mm_mul_ps(rsq10,rinv10);
 938
 939             /* Compute parameters for interactions between i and j atoms */
 940             qq10             = _mm_mul_ps(iq1,jq0);
 941
 942             /* EWALD ELECTROSTATICS */
 943
 944             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 945             ewrt             = _mm_mul_ps(r10,ewtabscale);
 946             ewitab           = _mm_cvttps_epi32(ewrt);
 947             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
 948             ewitab           = _mm_slli_epi32(ewitab,2);
 949             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
 950             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
 951             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
 952             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
 953             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
 954             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
 955             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
 956             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
 957             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
 958
 959             d                = _mm_sub_ps(r10,rswitch);
 960             d                = _mm_max_ps(d,_mm_setzero_ps());
 961             d2               = _mm_mul_ps(d,d);
 962             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)))))));
 963
 964             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
 965
 966             /* Evaluate switch function */
 967             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
 968             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
 969             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
 970
 971             fscal            = felec;
 972
 973             fscal            = _mm_and_ps(fscal,cutoff_mask);
 974
 975             /* Calculate temporary vectorial force */
 976             tx               = _mm_mul_ps(fscal,dx10);
 977             ty               = _mm_mul_ps(fscal,dy10);
 978             tz               = _mm_mul_ps(fscal,dz10);
 979
 980             /* Update vectorial force */
 981             fix1             = _mm_add_ps(fix1,tx);
 982             fiy1             = _mm_add_ps(fiy1,ty);
 983             fiz1             = _mm_add_ps(fiz1,tz);
 984
 985             fjx0             = _mm_add_ps(fjx0,tx);
 986             fjy0             = _mm_add_ps(fjy0,ty);
 987             fjz0             = _mm_add_ps(fjz0,tz);
 988
 989             }
 990
 991             /**************************
 992              * CALCULATE INTERACTIONS *
 993              **************************/
 994
 995             if (gmx_mm_any_lt(rsq20,rcutoff2))
 996             {
 997
 998             r20              = _mm_mul_ps(rsq20,rinv20);
 999
1000             /* Compute parameters for interactions between i and j atoms */
1001             qq20             = _mm_mul_ps(iq2,jq0);
1002
1003             /* EWALD ELECTROSTATICS */
1004
1005             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1006             ewrt             = _mm_mul_ps(r20,ewtabscale);
1007             ewitab           = _mm_cvttps_epi32(ewrt);
1008             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1009             ewitab           = _mm_slli_epi32(ewitab,2);
1010             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1011             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1012             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1013             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1014             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1015             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1016             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1017             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
1018             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
1019
1020             d                = _mm_sub_ps(r20,rswitch);
1021             d                = _mm_max_ps(d,_mm_setzero_ps());
1022             d2               = _mm_mul_ps(d,d);
1023             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)))))));
1024
1025             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1026
1027             /* Evaluate switch function */
1028             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1029             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
1030             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
1031
1032             fscal            = felec;
1033
1034             fscal            = _mm_and_ps(fscal,cutoff_mask);
1035
1036             /* Calculate temporary vectorial force */
1037             tx               = _mm_mul_ps(fscal,dx20);
1038             ty               = _mm_mul_ps(fscal,dy20);
1039             tz               = _mm_mul_ps(fscal,dz20);
1040
1041             /* Update vectorial force */
1042             fix2             = _mm_add_ps(fix2,tx);
1043             fiy2             = _mm_add_ps(fiy2,ty);
1044             fiz2             = _mm_add_ps(fiz2,tz);
1045
1046             fjx0             = _mm_add_ps(fjx0,tx);
1047             fjy0             = _mm_add_ps(fjy0,ty);
1048             fjz0             = _mm_add_ps(fjz0,tz);
1049
1050             }
1051
1052             /**************************
1053              * CALCULATE INTERACTIONS *
1054              **************************/
1055
1056             if (gmx_mm_any_lt(rsq30,rcutoff2))
1057             {
1058
1059             r30              = _mm_mul_ps(rsq30,rinv30);
1060
1061             /* Compute parameters for interactions between i and j atoms */
1062             qq30             = _mm_mul_ps(iq3,jq0);
1063
1064             /* EWALD ELECTROSTATICS */
1065
1066             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1067             ewrt             = _mm_mul_ps(r30,ewtabscale);
1068             ewitab           = _mm_cvttps_epi32(ewrt);
1069             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1070             ewitab           = _mm_slli_epi32(ewitab,2);
1071             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1072             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1073             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1074             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1075             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1076             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1077             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1078             velec            = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec));
1079             felec            = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec));
1080
1081             d                = _mm_sub_ps(r30,rswitch);
1082             d                = _mm_max_ps(d,_mm_setzero_ps());
1083             d2               = _mm_mul_ps(d,d);
1084             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)))))));
1085
1086             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1087
1088             /* Evaluate switch function */
1089             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1090             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv30,_mm_mul_ps(velec,dsw)) );
1091             cutoff_mask      = _mm_cmplt_ps(rsq30,rcutoff2);
1092
1093             fscal            = felec;
1094
1095             fscal            = _mm_and_ps(fscal,cutoff_mask);
1096
1097             /* Calculate temporary vectorial force */
1098             tx               = _mm_mul_ps(fscal,dx30);
1099             ty               = _mm_mul_ps(fscal,dy30);
1100             tz               = _mm_mul_ps(fscal,dz30);
1101
1102             /* Update vectorial force */
1103             fix3             = _mm_add_ps(fix3,tx);
1104             fiy3             = _mm_add_ps(fiy3,ty);
1105             fiz3             = _mm_add_ps(fiz3,tz);
1106
1107             fjx0             = _mm_add_ps(fjx0,tx);
1108             fjy0             = _mm_add_ps(fjy0,ty);
1109             fjz0             = _mm_add_ps(fjz0,tz);
1110
1111             }
1112
1113             fjptrA             = f+j_coord_offsetA;
1114             fjptrB             = f+j_coord_offsetB;
1115             fjptrC             = f+j_coord_offsetC;
1116             fjptrD             = f+j_coord_offsetD;
1117
1118             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
1119
1120             /* Inner loop uses 186 flops */
1121         }
1122
1123         if(jidx<j_index_end)
1124         {
1125
1126             /* Get j neighbor index, and coordinate index */
1127             jnrlistA         = jjnr[jidx];
1128             jnrlistB         = jjnr[jidx+1];
1129             jnrlistC         = jjnr[jidx+2];
1130             jnrlistD         = jjnr[jidx+3];
1131             /* Sign of each element will be negative for non-real atoms.
1132              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1133              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
1134              */
1135             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1136             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1137             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1138             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1139             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1140             j_coord_offsetA  = DIM*jnrA;
1141             j_coord_offsetB  = DIM*jnrB;
1142             j_coord_offsetC  = DIM*jnrC;
1143             j_coord_offsetD  = DIM*jnrD;
1144
1145             /* load j atom coordinates */
1146             gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1147                                               x+j_coord_offsetC,x+j_coord_offsetD,
1148                                               &jx0,&jy0,&jz0);
1149
1150             /* Calculate displacement vector */
1151             dx10             = _mm_sub_ps(ix1,jx0);
1152             dy10             = _mm_sub_ps(iy1,jy0);
1153             dz10             = _mm_sub_ps(iz1,jz0);
1154             dx20             = _mm_sub_ps(ix2,jx0);
1155             dy20             = _mm_sub_ps(iy2,jy0);
1156             dz20             = _mm_sub_ps(iz2,jz0);
1157             dx30             = _mm_sub_ps(ix3,jx0);
1158             dy30             = _mm_sub_ps(iy3,jy0);
1159             dz30             = _mm_sub_ps(iz3,jz0);
1160
1161             /* Calculate squared distance and things based on it */
1162             rsq10            = gmx_mm_calc_rsq_ps(dx10,dy10,dz10);
1163             rsq20            = gmx_mm_calc_rsq_ps(dx20,dy20,dz20);
1164             rsq30            = gmx_mm_calc_rsq_ps(dx30,dy30,dz30);
1165
1166             rinv10           = gmx_mm_invsqrt_ps(rsq10);
1167             rinv20           = gmx_mm_invsqrt_ps(rsq20);
1168             rinv30           = gmx_mm_invsqrt_ps(rsq30);
1169
1170             rinvsq10         = _mm_mul_ps(rinv10,rinv10);
1171             rinvsq20         = _mm_mul_ps(rinv20,rinv20);
1172             rinvsq30         = _mm_mul_ps(rinv30,rinv30);
1173
1174             /* Load parameters for j particles */
1175             jq0              = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
1176                                                               charge+jnrC+0,charge+jnrD+0);
1177
1178             fjx0             = _mm_setzero_ps();
1179             fjy0             = _mm_setzero_ps();
1180             fjz0             = _mm_setzero_ps();
1181
1182             /**************************
1183              * CALCULATE INTERACTIONS *
1184              **************************/
1185
1186             if (gmx_mm_any_lt(rsq10,rcutoff2))
1187             {
1188
1189             r10              = _mm_mul_ps(rsq10,rinv10);
1190             r10              = _mm_andnot_ps(dummy_mask,r10);
1191
1192             /* Compute parameters for interactions between i and j atoms */
1193             qq10             = _mm_mul_ps(iq1,jq0);
1194
1195             /* EWALD ELECTROSTATICS */
1196
1197             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1198             ewrt             = _mm_mul_ps(r10,ewtabscale);
1199             ewitab           = _mm_cvttps_epi32(ewrt);
1200             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1201             ewitab           = _mm_slli_epi32(ewitab,2);
1202             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1203             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1204             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1205             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1206             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1207             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1208             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1209             velec            = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec));
1210             felec            = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec));
1211
1212             d                = _mm_sub_ps(r10,rswitch);
1213             d                = _mm_max_ps(d,_mm_setzero_ps());
1214             d2               = _mm_mul_ps(d,d);
1215             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)))))));
1216
1217             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1218
1219             /* Evaluate switch function */
1220             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1221             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv10,_mm_mul_ps(velec,dsw)) );
1222             cutoff_mask      = _mm_cmplt_ps(rsq10,rcutoff2);
1223
1224             fscal            = felec;
1225
1226             fscal            = _mm_and_ps(fscal,cutoff_mask);
1227
1228             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1229
1230             /* Calculate temporary vectorial force */
1231             tx               = _mm_mul_ps(fscal,dx10);
1232             ty               = _mm_mul_ps(fscal,dy10);
1233             tz               = _mm_mul_ps(fscal,dz10);
1234
1235             /* Update vectorial force */
1236             fix1             = _mm_add_ps(fix1,tx);
1237             fiy1             = _mm_add_ps(fiy1,ty);
1238             fiz1             = _mm_add_ps(fiz1,tz);
1239
1240             fjx0             = _mm_add_ps(fjx0,tx);
1241             fjy0             = _mm_add_ps(fjy0,ty);
1242             fjz0             = _mm_add_ps(fjz0,tz);
1243
1244             }
1245
1246             /**************************
1247              * CALCULATE INTERACTIONS *
1248              **************************/
1249
1250             if (gmx_mm_any_lt(rsq20,rcutoff2))
1251             {
1252
1253             r20              = _mm_mul_ps(rsq20,rinv20);
1254             r20              = _mm_andnot_ps(dummy_mask,r20);
1255
1256             /* Compute parameters for interactions between i and j atoms */
1257             qq20             = _mm_mul_ps(iq2,jq0);
1258
1259             /* EWALD ELECTROSTATICS */
1260
1261             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1262             ewrt             = _mm_mul_ps(r20,ewtabscale);
1263             ewitab           = _mm_cvttps_epi32(ewrt);
1264             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1265             ewitab           = _mm_slli_epi32(ewitab,2);
1266             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1267             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1268             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1269             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1270             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1271             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1272             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1273             velec            = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec));
1274             felec            = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec));
1275
1276             d                = _mm_sub_ps(r20,rswitch);
1277             d                = _mm_max_ps(d,_mm_setzero_ps());
1278             d2               = _mm_mul_ps(d,d);
1279             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)))))));
1280
1281             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1282
1283             /* Evaluate switch function */
1284             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1285             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv20,_mm_mul_ps(velec,dsw)) );
1286             cutoff_mask      = _mm_cmplt_ps(rsq20,rcutoff2);
1287
1288             fscal            = felec;
1289
1290             fscal            = _mm_and_ps(fscal,cutoff_mask);
1291
1292             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1293
1294             /* Calculate temporary vectorial force */
1295             tx               = _mm_mul_ps(fscal,dx20);
1296             ty               = _mm_mul_ps(fscal,dy20);
1297             tz               = _mm_mul_ps(fscal,dz20);
1298
1299             /* Update vectorial force */
1300             fix2             = _mm_add_ps(fix2,tx);
1301             fiy2             = _mm_add_ps(fiy2,ty);
1302             fiz2             = _mm_add_ps(fiz2,tz);
1303
1304             fjx0             = _mm_add_ps(fjx0,tx);
1305             fjy0             = _mm_add_ps(fjy0,ty);
1306             fjz0             = _mm_add_ps(fjz0,tz);
1307
1308             }
1309
1310             /**************************
1311              * CALCULATE INTERACTIONS *
1312              **************************/
1313
1314             if (gmx_mm_any_lt(rsq30,rcutoff2))
1315             {
1316
1317             r30              = _mm_mul_ps(rsq30,rinv30);
1318             r30              = _mm_andnot_ps(dummy_mask,r30);
1319
1320             /* Compute parameters for interactions between i and j atoms */
1321             qq30             = _mm_mul_ps(iq3,jq0);
1322
1323             /* EWALD ELECTROSTATICS */
1324
1325             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1326             ewrt             = _mm_mul_ps(r30,ewtabscale);
1327             ewitab           = _mm_cvttps_epi32(ewrt);
1328             eweps            = _mm_sub_ps(ewrt,_mm_cvtepi32_ps(ewitab));
1329             ewitab           = _mm_slli_epi32(ewitab,2);
1330             ewtabF           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0) );
1331             ewtabD           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1) );
1332             ewtabV           = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2) );
1333             ewtabFn          = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3) );
1334             _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn);
1335             felec            = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD));
1336             velec            = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec)));
1337             velec            = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec));
1338             felec            = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec));
1339
1340             d                = _mm_sub_ps(r30,rswitch);
1341             d                = _mm_max_ps(d,_mm_setzero_ps());
1342             d2               = _mm_mul_ps(d,d);
1343             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)))))));
1344
1345             dsw              = _mm_mul_ps(d2,_mm_add_ps(swF2,_mm_mul_ps(d,_mm_add_ps(swF3,_mm_mul_ps(d,swF4)))));
1346
1347             /* Evaluate switch function */
1348             /* fscal'=f'/r=-(v*sw)'/r=-(v'*sw+v*dsw)/r=-v'*sw/r-v*dsw/r=fscal*sw-v*dsw/r */
1349             felec            = _mm_sub_ps( _mm_mul_ps(felec,sw) , _mm_mul_ps(rinv30,_mm_mul_ps(velec,dsw)) );
1350             cutoff_mask      = _mm_cmplt_ps(rsq30,rcutoff2);
1351
1352             fscal            = felec;
1353
1354             fscal            = _mm_and_ps(fscal,cutoff_mask);
1355
1356             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1357
1358             /* Calculate temporary vectorial force */
1359             tx               = _mm_mul_ps(fscal,dx30);
1360             ty               = _mm_mul_ps(fscal,dy30);
1361             tz               = _mm_mul_ps(fscal,dz30);
1362
1363             /* Update vectorial force */
1364             fix3             = _mm_add_ps(fix3,tx);
1365             fiy3             = _mm_add_ps(fiy3,ty);
1366             fiz3             = _mm_add_ps(fiz3,tz);
1367
1368             fjx0             = _mm_add_ps(fjx0,tx);
1369             fjy0             = _mm_add_ps(fjy0,ty);
1370             fjz0             = _mm_add_ps(fjz0,tz);
1371
1372             }
1373
1374             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1375             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1376             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1377             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1378
1379             gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0);
1380
1381             /* Inner loop uses 189 flops */
1382         }
1383
1384         /* End of innermost loop */
1385
1386         gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1387                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
1388
1389         /* Increment number of inner iterations */
1390         inneriter                  += j_index_end - j_index_start;
1391
1392         /* Outer loop uses 18 flops */
1393     }
1394
1395     /* Increment number of outer iterations */
1396     outeriter        += nri;
1397
1398     /* Update outer/inner flops */
1399
1400     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_F,outeriter*18 + inneriter*189);
1401 }