src/gmxlib/nonbonded/nb_kernel_sse2_single/nb_kernel_ElecCSTab_VdwNone_GeomW4W4_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_ElecCSTab_VdwNone_GeomW4W4_VF_sse2_single
  38  * Electrostatics interaction: CubicSplineTable
  39  * VdW interaction:            None
  40  * Geometry:                   Water4-Water4
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecCSTab_VdwNone_GeomW4W4_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              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  62     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  63     real             shX,shY,shZ,rcutoff_scalar;
  64     real             *shiftvec,*fshift,*x,*f;
  65     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  66     int              vdwioffset1;
  67     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  68     int              vdwioffset2;
  69     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  70     int              vdwioffset3;
  71     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  72     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  73     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  74     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  75     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  76     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
  77     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
  78     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
  79     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
  80     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
  81     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
  82     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
  83     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
  84     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
  85     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
  86     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
  87     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
  88     real             *charge;
  89     __m128i          vfitab;
  90     __m128i          ifour       = _mm_set1_epi32(4);
  91     __m128           rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
  92     real             *vftab;
  93     __m128           dummy_mask,cutoff_mask;
  94     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
  95     __m128           one     = _mm_set1_ps(1.0);
  96     __m128           two     = _mm_set1_ps(2.0);
  97     x                = xx[0];
  98     f                = ff[0];
  99
 100     nri              = nlist->nri;
 101     iinr             = nlist->iinr;
 102     jindex           = nlist->jindex;
 103     jjnr             = nlist->jjnr;
 104     shiftidx         = nlist->shift;
 105     gid              = nlist->gid;
 106     shiftvec         = fr->shift_vec[0];
 107     fshift           = fr->fshift[0];
 108     facel            = _mm_set1_ps(fr->epsfac);
 109     charge           = mdatoms->chargeA;
 110
 111     vftab            = kernel_data->table_elec->data;
 112     vftabscale       = _mm_set1_ps(kernel_data->table_elec->scale);
 113
 114     /* Setup water-specific parameters */
 115     inr              = nlist->iinr[0];
 116     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 117     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 118     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
 119
 120     jq1              = _mm_set1_ps(charge[inr+1]);
 121     jq2              = _mm_set1_ps(charge[inr+2]);
 122     jq3              = _mm_set1_ps(charge[inr+3]);
 123     qq11             = _mm_mul_ps(iq1,jq1);
 124     qq12             = _mm_mul_ps(iq1,jq2);
 125     qq13             = _mm_mul_ps(iq1,jq3);
 126     qq21             = _mm_mul_ps(iq2,jq1);
 127     qq22             = _mm_mul_ps(iq2,jq2);
 128     qq23             = _mm_mul_ps(iq2,jq3);
 129     qq31             = _mm_mul_ps(iq3,jq1);
 130     qq32             = _mm_mul_ps(iq3,jq2);
 131     qq33             = _mm_mul_ps(iq3,jq3);
 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     /* Start outer loop over neighborlists */
 144     for(iidx=0; iidx<nri; iidx++)
 145     {
 146         /* Load shift vector for this list */
 147         i_shift_offset   = DIM*shiftidx[iidx];
 148         shX              = shiftvec[i_shift_offset+XX];
 149         shY              = shiftvec[i_shift_offset+YY];
 150         shZ              = shiftvec[i_shift_offset+ZZ];
 151
 152         /* Load limits for loop over neighbors */
 153         j_index_start    = jindex[iidx];
 154         j_index_end      = jindex[iidx+1];
 155
 156         /* Get outer coordinate index */
 157         inr              = iinr[iidx];
 158         i_coord_offset   = DIM*inr;
 159
 160         /* Load i particle coords and add shift vector */
 161         ix1              = _mm_set1_ps(shX + x[i_coord_offset+DIM*1+XX]);
 162         iy1              = _mm_set1_ps(shY + x[i_coord_offset+DIM*1+YY]);
 163         iz1              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*1+ZZ]);
 164         ix2              = _mm_set1_ps(shX + x[i_coord_offset+DIM*2+XX]);
 165         iy2              = _mm_set1_ps(shY + x[i_coord_offset+DIM*2+YY]);
 166         iz2              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*2+ZZ]);
 167         ix3              = _mm_set1_ps(shX + x[i_coord_offset+DIM*3+XX]);
 168         iy3              = _mm_set1_ps(shY + x[i_coord_offset+DIM*3+YY]);
 169         iz3              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*3+ZZ]);
 170
 171         fix1             = _mm_setzero_ps();
 172         fiy1             = _mm_setzero_ps();
 173         fiz1             = _mm_setzero_ps();
 174         fix2             = _mm_setzero_ps();
 175         fiy2             = _mm_setzero_ps();
 176         fiz2             = _mm_setzero_ps();
 177         fix3             = _mm_setzero_ps();
 178         fiy3             = _mm_setzero_ps();
 179         fiz3             = _mm_setzero_ps();
 180
 181         /* Reset potential sums */
 182         velecsum         = _mm_setzero_ps();
 183
 184         /* Start inner kernel loop */
 185         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 186         {
 187
 188             /* Get j neighbor index, and coordinate index */
 189             jnrA             = jjnr[jidx];
 190             jnrB             = jjnr[jidx+1];
 191             jnrC             = jjnr[jidx+2];
 192             jnrD             = jjnr[jidx+3];
 193
 194             j_coord_offsetA  = DIM*jnrA;
 195             j_coord_offsetB  = DIM*jnrB;
 196             j_coord_offsetC  = DIM*jnrC;
 197             j_coord_offsetD  = DIM*jnrD;
 198
 199             /* load j atom coordinates */
 200             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
 201                                               x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
 202                                               &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
 203
 204             /* Calculate displacement vector */
 205             dx11             = _mm_sub_ps(ix1,jx1);
 206             dy11             = _mm_sub_ps(iy1,jy1);
 207             dz11             = _mm_sub_ps(iz1,jz1);
 208             dx12             = _mm_sub_ps(ix1,jx2);
 209             dy12             = _mm_sub_ps(iy1,jy2);
 210             dz12             = _mm_sub_ps(iz1,jz2);
 211             dx13             = _mm_sub_ps(ix1,jx3);
 212             dy13             = _mm_sub_ps(iy1,jy3);
 213             dz13             = _mm_sub_ps(iz1,jz3);
 214             dx21             = _mm_sub_ps(ix2,jx1);
 215             dy21             = _mm_sub_ps(iy2,jy1);
 216             dz21             = _mm_sub_ps(iz2,jz1);
 217             dx22             = _mm_sub_ps(ix2,jx2);
 218             dy22             = _mm_sub_ps(iy2,jy2);
 219             dz22             = _mm_sub_ps(iz2,jz2);
 220             dx23             = _mm_sub_ps(ix2,jx3);
 221             dy23             = _mm_sub_ps(iy2,jy3);
 222             dz23             = _mm_sub_ps(iz2,jz3);
 223             dx31             = _mm_sub_ps(ix3,jx1);
 224             dy31             = _mm_sub_ps(iy3,jy1);
 225             dz31             = _mm_sub_ps(iz3,jz1);
 226             dx32             = _mm_sub_ps(ix3,jx2);
 227             dy32             = _mm_sub_ps(iy3,jy2);
 228             dz32             = _mm_sub_ps(iz3,jz2);
 229             dx33             = _mm_sub_ps(ix3,jx3);
 230             dy33             = _mm_sub_ps(iy3,jy3);
 231             dz33             = _mm_sub_ps(iz3,jz3);
 232
 233             /* Calculate squared distance and things based on it */
 234             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 235             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 236             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 237             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 238             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 239             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 240             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 241             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 242             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 243
 244             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 245             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 246             rinv13           = gmx_mm_invsqrt_ps(rsq13);
 247             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 248             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 249             rinv23           = gmx_mm_invsqrt_ps(rsq23);
 250             rinv31           = gmx_mm_invsqrt_ps(rsq31);
 251             rinv32           = gmx_mm_invsqrt_ps(rsq32);
 252             rinv33           = gmx_mm_invsqrt_ps(rsq33);
 253
 254             fjx1             = _mm_setzero_ps();
 255             fjy1             = _mm_setzero_ps();
 256             fjz1             = _mm_setzero_ps();
 257             fjx2             = _mm_setzero_ps();
 258             fjy2             = _mm_setzero_ps();
 259             fjz2             = _mm_setzero_ps();
 260             fjx3             = _mm_setzero_ps();
 261             fjy3             = _mm_setzero_ps();
 262             fjz3             = _mm_setzero_ps();
 263
 264             /**************************
 265              * CALCULATE INTERACTIONS *
 266              **************************/
 267
 268             r11              = _mm_mul_ps(rsq11,rinv11);
 269
 270             /* Calculate table index by multiplying r with table scale and truncate to integer */
 271             rt               = _mm_mul_ps(r11,vftabscale);
 272             vfitab           = _mm_cvttps_epi32(rt);
 273             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 274             vfitab           = _mm_slli_epi32(vfitab,2);
 275
 276             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 277             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 278             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 279             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 280             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 281             _MM_TRANSPOSE4_PS(Y,F,G,H);
 282             Heps             = _mm_mul_ps(vfeps,H);
 283             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 284             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 285             velec            = _mm_mul_ps(qq11,VV);
 286             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 287             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
 288
 289             /* Update potential sum for this i atom from the interaction with this j atom. */
 290             velecsum         = _mm_add_ps(velecsum,velec);
 291
 292             fscal            = felec;
 293
 294             /* Calculate temporary vectorial force */
 295             tx               = _mm_mul_ps(fscal,dx11);
 296             ty               = _mm_mul_ps(fscal,dy11);
 297             tz               = _mm_mul_ps(fscal,dz11);
 298
 299             /* Update vectorial force */
 300             fix1             = _mm_add_ps(fix1,tx);
 301             fiy1             = _mm_add_ps(fiy1,ty);
 302             fiz1             = _mm_add_ps(fiz1,tz);
 303
 304             fjx1             = _mm_add_ps(fjx1,tx);
 305             fjy1             = _mm_add_ps(fjy1,ty);
 306             fjz1             = _mm_add_ps(fjz1,tz);
 307
 308             /**************************
 309              * CALCULATE INTERACTIONS *
 310              **************************/
 311
 312             r12              = _mm_mul_ps(rsq12,rinv12);
 313
 314             /* Calculate table index by multiplying r with table scale and truncate to integer */
 315             rt               = _mm_mul_ps(r12,vftabscale);
 316             vfitab           = _mm_cvttps_epi32(rt);
 317             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 318             vfitab           = _mm_slli_epi32(vfitab,2);
 319
 320             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 321             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 322             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 323             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 324             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 325             _MM_TRANSPOSE4_PS(Y,F,G,H);
 326             Heps             = _mm_mul_ps(vfeps,H);
 327             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 328             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 329             velec            = _mm_mul_ps(qq12,VV);
 330             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 331             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
 332
 333             /* Update potential sum for this i atom from the interaction with this j atom. */
 334             velecsum         = _mm_add_ps(velecsum,velec);
 335
 336             fscal            = felec;
 337
 338             /* Calculate temporary vectorial force */
 339             tx               = _mm_mul_ps(fscal,dx12);
 340             ty               = _mm_mul_ps(fscal,dy12);
 341             tz               = _mm_mul_ps(fscal,dz12);
 342
 343             /* Update vectorial force */
 344             fix1             = _mm_add_ps(fix1,tx);
 345             fiy1             = _mm_add_ps(fiy1,ty);
 346             fiz1             = _mm_add_ps(fiz1,tz);
 347
 348             fjx2             = _mm_add_ps(fjx2,tx);
 349             fjy2             = _mm_add_ps(fjy2,ty);
 350             fjz2             = _mm_add_ps(fjz2,tz);
 351
 352             /**************************
 353              * CALCULATE INTERACTIONS *
 354              **************************/
 355
 356             r13              = _mm_mul_ps(rsq13,rinv13);
 357
 358             /* Calculate table index by multiplying r with table scale and truncate to integer */
 359             rt               = _mm_mul_ps(r13,vftabscale);
 360             vfitab           = _mm_cvttps_epi32(rt);
 361             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 362             vfitab           = _mm_slli_epi32(vfitab,2);
 363
 364             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 365             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 366             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 367             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 368             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 369             _MM_TRANSPOSE4_PS(Y,F,G,H);
 370             Heps             = _mm_mul_ps(vfeps,H);
 371             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 372             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 373             velec            = _mm_mul_ps(qq13,VV);
 374             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 375             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
 376
 377             /* Update potential sum for this i atom from the interaction with this j atom. */
 378             velecsum         = _mm_add_ps(velecsum,velec);
 379
 380             fscal            = felec;
 381
 382             /* Calculate temporary vectorial force */
 383             tx               = _mm_mul_ps(fscal,dx13);
 384             ty               = _mm_mul_ps(fscal,dy13);
 385             tz               = _mm_mul_ps(fscal,dz13);
 386
 387             /* Update vectorial force */
 388             fix1             = _mm_add_ps(fix1,tx);
 389             fiy1             = _mm_add_ps(fiy1,ty);
 390             fiz1             = _mm_add_ps(fiz1,tz);
 391
 392             fjx3             = _mm_add_ps(fjx3,tx);
 393             fjy3             = _mm_add_ps(fjy3,ty);
 394             fjz3             = _mm_add_ps(fjz3,tz);
 395
 396             /**************************
 397              * CALCULATE INTERACTIONS *
 398              **************************/
 399
 400             r21              = _mm_mul_ps(rsq21,rinv21);
 401
 402             /* Calculate table index by multiplying r with table scale and truncate to integer */
 403             rt               = _mm_mul_ps(r21,vftabscale);
 404             vfitab           = _mm_cvttps_epi32(rt);
 405             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 406             vfitab           = _mm_slli_epi32(vfitab,2);
 407
 408             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 409             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 410             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 411             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 412             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 413             _MM_TRANSPOSE4_PS(Y,F,G,H);
 414             Heps             = _mm_mul_ps(vfeps,H);
 415             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 416             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 417             velec            = _mm_mul_ps(qq21,VV);
 418             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 419             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
 420
 421             /* Update potential sum for this i atom from the interaction with this j atom. */
 422             velecsum         = _mm_add_ps(velecsum,velec);
 423
 424             fscal            = felec;
 425
 426             /* Calculate temporary vectorial force */
 427             tx               = _mm_mul_ps(fscal,dx21);
 428             ty               = _mm_mul_ps(fscal,dy21);
 429             tz               = _mm_mul_ps(fscal,dz21);
 430
 431             /* Update vectorial force */
 432             fix2             = _mm_add_ps(fix2,tx);
 433             fiy2             = _mm_add_ps(fiy2,ty);
 434             fiz2             = _mm_add_ps(fiz2,tz);
 435
 436             fjx1             = _mm_add_ps(fjx1,tx);
 437             fjy1             = _mm_add_ps(fjy1,ty);
 438             fjz1             = _mm_add_ps(fjz1,tz);
 439
 440             /**************************
 441              * CALCULATE INTERACTIONS *
 442              **************************/
 443
 444             r22              = _mm_mul_ps(rsq22,rinv22);
 445
 446             /* Calculate table index by multiplying r with table scale and truncate to integer */
 447             rt               = _mm_mul_ps(r22,vftabscale);
 448             vfitab           = _mm_cvttps_epi32(rt);
 449             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 450             vfitab           = _mm_slli_epi32(vfitab,2);
 451
 452             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 453             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 454             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 455             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 456             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 457             _MM_TRANSPOSE4_PS(Y,F,G,H);
 458             Heps             = _mm_mul_ps(vfeps,H);
 459             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 460             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 461             velec            = _mm_mul_ps(qq22,VV);
 462             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 463             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
 464
 465             /* Update potential sum for this i atom from the interaction with this j atom. */
 466             velecsum         = _mm_add_ps(velecsum,velec);
 467
 468             fscal            = felec;
 469
 470             /* Calculate temporary vectorial force */
 471             tx               = _mm_mul_ps(fscal,dx22);
 472             ty               = _mm_mul_ps(fscal,dy22);
 473             tz               = _mm_mul_ps(fscal,dz22);
 474
 475             /* Update vectorial force */
 476             fix2             = _mm_add_ps(fix2,tx);
 477             fiy2             = _mm_add_ps(fiy2,ty);
 478             fiz2             = _mm_add_ps(fiz2,tz);
 479
 480             fjx2             = _mm_add_ps(fjx2,tx);
 481             fjy2             = _mm_add_ps(fjy2,ty);
 482             fjz2             = _mm_add_ps(fjz2,tz);
 483
 484             /**************************
 485              * CALCULATE INTERACTIONS *
 486              **************************/
 487
 488             r23              = _mm_mul_ps(rsq23,rinv23);
 489
 490             /* Calculate table index by multiplying r with table scale and truncate to integer */
 491             rt               = _mm_mul_ps(r23,vftabscale);
 492             vfitab           = _mm_cvttps_epi32(rt);
 493             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 494             vfitab           = _mm_slli_epi32(vfitab,2);
 495
 496             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 497             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 498             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 499             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 500             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 501             _MM_TRANSPOSE4_PS(Y,F,G,H);
 502             Heps             = _mm_mul_ps(vfeps,H);
 503             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 504             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 505             velec            = _mm_mul_ps(qq23,VV);
 506             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 507             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
 508
 509             /* Update potential sum for this i atom from the interaction with this j atom. */
 510             velecsum         = _mm_add_ps(velecsum,velec);
 511
 512             fscal            = felec;
 513
 514             /* Calculate temporary vectorial force */
 515             tx               = _mm_mul_ps(fscal,dx23);
 516             ty               = _mm_mul_ps(fscal,dy23);
 517             tz               = _mm_mul_ps(fscal,dz23);
 518
 519             /* Update vectorial force */
 520             fix2             = _mm_add_ps(fix2,tx);
 521             fiy2             = _mm_add_ps(fiy2,ty);
 522             fiz2             = _mm_add_ps(fiz2,tz);
 523
 524             fjx3             = _mm_add_ps(fjx3,tx);
 525             fjy3             = _mm_add_ps(fjy3,ty);
 526             fjz3             = _mm_add_ps(fjz3,tz);
 527
 528             /**************************
 529              * CALCULATE INTERACTIONS *
 530              **************************/
 531
 532             r31              = _mm_mul_ps(rsq31,rinv31);
 533
 534             /* Calculate table index by multiplying r with table scale and truncate to integer */
 535             rt               = _mm_mul_ps(r31,vftabscale);
 536             vfitab           = _mm_cvttps_epi32(rt);
 537             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 538             vfitab           = _mm_slli_epi32(vfitab,2);
 539
 540             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 541             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 542             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 543             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 544             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 545             _MM_TRANSPOSE4_PS(Y,F,G,H);
 546             Heps             = _mm_mul_ps(vfeps,H);
 547             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 548             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 549             velec            = _mm_mul_ps(qq31,VV);
 550             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 551             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
 552
 553             /* Update potential sum for this i atom from the interaction with this j atom. */
 554             velecsum         = _mm_add_ps(velecsum,velec);
 555
 556             fscal            = felec;
 557
 558             /* Calculate temporary vectorial force */
 559             tx               = _mm_mul_ps(fscal,dx31);
 560             ty               = _mm_mul_ps(fscal,dy31);
 561             tz               = _mm_mul_ps(fscal,dz31);
 562
 563             /* Update vectorial force */
 564             fix3             = _mm_add_ps(fix3,tx);
 565             fiy3             = _mm_add_ps(fiy3,ty);
 566             fiz3             = _mm_add_ps(fiz3,tz);
 567
 568             fjx1             = _mm_add_ps(fjx1,tx);
 569             fjy1             = _mm_add_ps(fjy1,ty);
 570             fjz1             = _mm_add_ps(fjz1,tz);
 571
 572             /**************************
 573              * CALCULATE INTERACTIONS *
 574              **************************/
 575
 576             r32              = _mm_mul_ps(rsq32,rinv32);
 577
 578             /* Calculate table index by multiplying r with table scale and truncate to integer */
 579             rt               = _mm_mul_ps(r32,vftabscale);
 580             vfitab           = _mm_cvttps_epi32(rt);
 581             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 582             vfitab           = _mm_slli_epi32(vfitab,2);
 583
 584             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 585             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 586             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 587             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 588             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 589             _MM_TRANSPOSE4_PS(Y,F,G,H);
 590             Heps             = _mm_mul_ps(vfeps,H);
 591             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 592             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 593             velec            = _mm_mul_ps(qq32,VV);
 594             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 595             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
 596
 597             /* Update potential sum for this i atom from the interaction with this j atom. */
 598             velecsum         = _mm_add_ps(velecsum,velec);
 599
 600             fscal            = felec;
 601
 602             /* Calculate temporary vectorial force */
 603             tx               = _mm_mul_ps(fscal,dx32);
 604             ty               = _mm_mul_ps(fscal,dy32);
 605             tz               = _mm_mul_ps(fscal,dz32);
 606
 607             /* Update vectorial force */
 608             fix3             = _mm_add_ps(fix3,tx);
 609             fiy3             = _mm_add_ps(fiy3,ty);
 610             fiz3             = _mm_add_ps(fiz3,tz);
 611
 612             fjx2             = _mm_add_ps(fjx2,tx);
 613             fjy2             = _mm_add_ps(fjy2,ty);
 614             fjz2             = _mm_add_ps(fjz2,tz);
 615
 616             /**************************
 617              * CALCULATE INTERACTIONS *
 618              **************************/
 619
 620             r33              = _mm_mul_ps(rsq33,rinv33);
 621
 622             /* Calculate table index by multiplying r with table scale and truncate to integer */
 623             rt               = _mm_mul_ps(r33,vftabscale);
 624             vfitab           = _mm_cvttps_epi32(rt);
 625             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 626             vfitab           = _mm_slli_epi32(vfitab,2);
 627
 628             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 629             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 630             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 631             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 632             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 633             _MM_TRANSPOSE4_PS(Y,F,G,H);
 634             Heps             = _mm_mul_ps(vfeps,H);
 635             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 636             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 637             velec            = _mm_mul_ps(qq33,VV);
 638             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 639             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
 640
 641             /* Update potential sum for this i atom from the interaction with this j atom. */
 642             velecsum         = _mm_add_ps(velecsum,velec);
 643
 644             fscal            = felec;
 645
 646             /* Calculate temporary vectorial force */
 647             tx               = _mm_mul_ps(fscal,dx33);
 648             ty               = _mm_mul_ps(fscal,dy33);
 649             tz               = _mm_mul_ps(fscal,dz33);
 650
 651             /* Update vectorial force */
 652             fix3             = _mm_add_ps(fix3,tx);
 653             fiy3             = _mm_add_ps(fiy3,ty);
 654             fiz3             = _mm_add_ps(fiz3,tz);
 655
 656             fjx3             = _mm_add_ps(fjx3,tx);
 657             fjy3             = _mm_add_ps(fjy3,ty);
 658             fjz3             = _mm_add_ps(fjz3,tz);
 659
 660             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(f+j_coord_offsetA+DIM,f+j_coord_offsetB+DIM,
 661                                                    f+j_coord_offsetC+DIM,f+j_coord_offsetD+DIM,
 662                                                    fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 663
 664             /* Inner loop uses 387 flops */
 665         }
 666
 667         if(jidx<j_index_end)
 668         {
 669
 670             /* Get j neighbor index, and coordinate index */
 671             jnrA             = jjnr[jidx];
 672             jnrB             = jjnr[jidx+1];
 673             jnrC             = jjnr[jidx+2];
 674             jnrD             = jjnr[jidx+3];
 675
 676             /* Sign of each element will be negative for non-real atoms.
 677              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 678              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 679              */
 680             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 681             jnrA       = (jnrA>=0) ? jnrA : 0;
 682             jnrB       = (jnrB>=0) ? jnrB : 0;
 683             jnrC       = (jnrC>=0) ? jnrC : 0;
 684             jnrD       = (jnrD>=0) ? jnrD : 0;
 685
 686             j_coord_offsetA  = DIM*jnrA;
 687             j_coord_offsetB  = DIM*jnrB;
 688             j_coord_offsetC  = DIM*jnrC;
 689             j_coord_offsetD  = DIM*jnrD;
 690
 691             /* load j atom coordinates */
 692             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
 693                                               x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
 694                                               &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
 695
 696             /* Calculate displacement vector */
 697             dx11             = _mm_sub_ps(ix1,jx1);
 698             dy11             = _mm_sub_ps(iy1,jy1);
 699             dz11             = _mm_sub_ps(iz1,jz1);
 700             dx12             = _mm_sub_ps(ix1,jx2);
 701             dy12             = _mm_sub_ps(iy1,jy2);
 702             dz12             = _mm_sub_ps(iz1,jz2);
 703             dx13             = _mm_sub_ps(ix1,jx3);
 704             dy13             = _mm_sub_ps(iy1,jy3);
 705             dz13             = _mm_sub_ps(iz1,jz3);
 706             dx21             = _mm_sub_ps(ix2,jx1);
 707             dy21             = _mm_sub_ps(iy2,jy1);
 708             dz21             = _mm_sub_ps(iz2,jz1);
 709             dx22             = _mm_sub_ps(ix2,jx2);
 710             dy22             = _mm_sub_ps(iy2,jy2);
 711             dz22             = _mm_sub_ps(iz2,jz2);
 712             dx23             = _mm_sub_ps(ix2,jx3);
 713             dy23             = _mm_sub_ps(iy2,jy3);
 714             dz23             = _mm_sub_ps(iz2,jz3);
 715             dx31             = _mm_sub_ps(ix3,jx1);
 716             dy31             = _mm_sub_ps(iy3,jy1);
 717             dz31             = _mm_sub_ps(iz3,jz1);
 718             dx32             = _mm_sub_ps(ix3,jx2);
 719             dy32             = _mm_sub_ps(iy3,jy2);
 720             dz32             = _mm_sub_ps(iz3,jz2);
 721             dx33             = _mm_sub_ps(ix3,jx3);
 722             dy33             = _mm_sub_ps(iy3,jy3);
 723             dz33             = _mm_sub_ps(iz3,jz3);
 724
 725             /* Calculate squared distance and things based on it */
 726             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 727             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 728             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 729             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 730             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 731             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 732             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 733             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 734             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 735
 736             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 737             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 738             rinv13           = gmx_mm_invsqrt_ps(rsq13);
 739             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 740             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 741             rinv23           = gmx_mm_invsqrt_ps(rsq23);
 742             rinv31           = gmx_mm_invsqrt_ps(rsq31);
 743             rinv32           = gmx_mm_invsqrt_ps(rsq32);
 744             rinv33           = gmx_mm_invsqrt_ps(rsq33);
 745
 746             fjx1             = _mm_setzero_ps();
 747             fjy1             = _mm_setzero_ps();
 748             fjz1             = _mm_setzero_ps();
 749             fjx2             = _mm_setzero_ps();
 750             fjy2             = _mm_setzero_ps();
 751             fjz2             = _mm_setzero_ps();
 752             fjx3             = _mm_setzero_ps();
 753             fjy3             = _mm_setzero_ps();
 754             fjz3             = _mm_setzero_ps();
 755
 756             /**************************
 757              * CALCULATE INTERACTIONS *
 758              **************************/
 759
 760             r11              = _mm_mul_ps(rsq11,rinv11);
 761             r11              = _mm_andnot_ps(dummy_mask,r11);
 762
 763             /* Calculate table index by multiplying r with table scale and truncate to integer */
 764             rt               = _mm_mul_ps(r11,vftabscale);
 765             vfitab           = _mm_cvttps_epi32(rt);
 766             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 767             vfitab           = _mm_slli_epi32(vfitab,2);
 768
 769             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 770             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 771             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 772             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 773             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 774             _MM_TRANSPOSE4_PS(Y,F,G,H);
 775             Heps             = _mm_mul_ps(vfeps,H);
 776             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 777             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 778             velec            = _mm_mul_ps(qq11,VV);
 779             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 780             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
 781
 782             /* Update potential sum for this i atom from the interaction with this j atom. */
 783             velec            = _mm_andnot_ps(dummy_mask,velec);
 784             velecsum         = _mm_add_ps(velecsum,velec);
 785
 786             fscal            = felec;
 787
 788             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 789
 790             /* Calculate temporary vectorial force */
 791             tx               = _mm_mul_ps(fscal,dx11);
 792             ty               = _mm_mul_ps(fscal,dy11);
 793             tz               = _mm_mul_ps(fscal,dz11);
 794
 795             /* Update vectorial force */
 796             fix1             = _mm_add_ps(fix1,tx);
 797             fiy1             = _mm_add_ps(fiy1,ty);
 798             fiz1             = _mm_add_ps(fiz1,tz);
 799
 800             fjx1             = _mm_add_ps(fjx1,tx);
 801             fjy1             = _mm_add_ps(fjy1,ty);
 802             fjz1             = _mm_add_ps(fjz1,tz);
 803
 804             /**************************
 805              * CALCULATE INTERACTIONS *
 806              **************************/
 807
 808             r12              = _mm_mul_ps(rsq12,rinv12);
 809             r12              = _mm_andnot_ps(dummy_mask,r12);
 810
 811             /* Calculate table index by multiplying r with table scale and truncate to integer */
 812             rt               = _mm_mul_ps(r12,vftabscale);
 813             vfitab           = _mm_cvttps_epi32(rt);
 814             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 815             vfitab           = _mm_slli_epi32(vfitab,2);
 816
 817             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 818             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 819             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 820             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 821             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 822             _MM_TRANSPOSE4_PS(Y,F,G,H);
 823             Heps             = _mm_mul_ps(vfeps,H);
 824             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 825             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 826             velec            = _mm_mul_ps(qq12,VV);
 827             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 828             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
 829
 830             /* Update potential sum for this i atom from the interaction with this j atom. */
 831             velec            = _mm_andnot_ps(dummy_mask,velec);
 832             velecsum         = _mm_add_ps(velecsum,velec);
 833
 834             fscal            = felec;
 835
 836             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 837
 838             /* Calculate temporary vectorial force */
 839             tx               = _mm_mul_ps(fscal,dx12);
 840             ty               = _mm_mul_ps(fscal,dy12);
 841             tz               = _mm_mul_ps(fscal,dz12);
 842
 843             /* Update vectorial force */
 844             fix1             = _mm_add_ps(fix1,tx);
 845             fiy1             = _mm_add_ps(fiy1,ty);
 846             fiz1             = _mm_add_ps(fiz1,tz);
 847
 848             fjx2             = _mm_add_ps(fjx2,tx);
 849             fjy2             = _mm_add_ps(fjy2,ty);
 850             fjz2             = _mm_add_ps(fjz2,tz);
 851
 852             /**************************
 853              * CALCULATE INTERACTIONS *
 854              **************************/
 855
 856             r13              = _mm_mul_ps(rsq13,rinv13);
 857             r13              = _mm_andnot_ps(dummy_mask,r13);
 858
 859             /* Calculate table index by multiplying r with table scale and truncate to integer */
 860             rt               = _mm_mul_ps(r13,vftabscale);
 861             vfitab           = _mm_cvttps_epi32(rt);
 862             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 863             vfitab           = _mm_slli_epi32(vfitab,2);
 864
 865             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 866             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 867             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 868             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 869             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 870             _MM_TRANSPOSE4_PS(Y,F,G,H);
 871             Heps             = _mm_mul_ps(vfeps,H);
 872             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 873             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 874             velec            = _mm_mul_ps(qq13,VV);
 875             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 876             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
 877
 878             /* Update potential sum for this i atom from the interaction with this j atom. */
 879             velec            = _mm_andnot_ps(dummy_mask,velec);
 880             velecsum         = _mm_add_ps(velecsum,velec);
 881
 882             fscal            = felec;
 883
 884             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 885
 886             /* Calculate temporary vectorial force */
 887             tx               = _mm_mul_ps(fscal,dx13);
 888             ty               = _mm_mul_ps(fscal,dy13);
 889             tz               = _mm_mul_ps(fscal,dz13);
 890
 891             /* Update vectorial force */
 892             fix1             = _mm_add_ps(fix1,tx);
 893             fiy1             = _mm_add_ps(fiy1,ty);
 894             fiz1             = _mm_add_ps(fiz1,tz);
 895
 896             fjx3             = _mm_add_ps(fjx3,tx);
 897             fjy3             = _mm_add_ps(fjy3,ty);
 898             fjz3             = _mm_add_ps(fjz3,tz);
 899
 900             /**************************
 901              * CALCULATE INTERACTIONS *
 902              **************************/
 903
 904             r21              = _mm_mul_ps(rsq21,rinv21);
 905             r21              = _mm_andnot_ps(dummy_mask,r21);
 906
 907             /* Calculate table index by multiplying r with table scale and truncate to integer */
 908             rt               = _mm_mul_ps(r21,vftabscale);
 909             vfitab           = _mm_cvttps_epi32(rt);
 910             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 911             vfitab           = _mm_slli_epi32(vfitab,2);
 912
 913             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 914             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 915             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 916             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 917             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 918             _MM_TRANSPOSE4_PS(Y,F,G,H);
 919             Heps             = _mm_mul_ps(vfeps,H);
 920             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 921             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 922             velec            = _mm_mul_ps(qq21,VV);
 923             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 924             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
 925
 926             /* Update potential sum for this i atom from the interaction with this j atom. */
 927             velec            = _mm_andnot_ps(dummy_mask,velec);
 928             velecsum         = _mm_add_ps(velecsum,velec);
 929
 930             fscal            = felec;
 931
 932             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 933
 934             /* Calculate temporary vectorial force */
 935             tx               = _mm_mul_ps(fscal,dx21);
 936             ty               = _mm_mul_ps(fscal,dy21);
 937             tz               = _mm_mul_ps(fscal,dz21);
 938
 939             /* Update vectorial force */
 940             fix2             = _mm_add_ps(fix2,tx);
 941             fiy2             = _mm_add_ps(fiy2,ty);
 942             fiz2             = _mm_add_ps(fiz2,tz);
 943
 944             fjx1             = _mm_add_ps(fjx1,tx);
 945             fjy1             = _mm_add_ps(fjy1,ty);
 946             fjz1             = _mm_add_ps(fjz1,tz);
 947
 948             /**************************
 949              * CALCULATE INTERACTIONS *
 950              **************************/
 951
 952             r22              = _mm_mul_ps(rsq22,rinv22);
 953             r22              = _mm_andnot_ps(dummy_mask,r22);
 954
 955             /* Calculate table index by multiplying r with table scale and truncate to integer */
 956             rt               = _mm_mul_ps(r22,vftabscale);
 957             vfitab           = _mm_cvttps_epi32(rt);
 958             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
 959             vfitab           = _mm_slli_epi32(vfitab,2);
 960
 961             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 962             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
 963             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
 964             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
 965             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
 966             _MM_TRANSPOSE4_PS(Y,F,G,H);
 967             Heps             = _mm_mul_ps(vfeps,H);
 968             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
 969             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
 970             velec            = _mm_mul_ps(qq22,VV);
 971             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
 972             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
 973
 974             /* Update potential sum for this i atom from the interaction with this j atom. */
 975             velec            = _mm_andnot_ps(dummy_mask,velec);
 976             velecsum         = _mm_add_ps(velecsum,velec);
 977
 978             fscal            = felec;
 979
 980             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 981
 982             /* Calculate temporary vectorial force */
 983             tx               = _mm_mul_ps(fscal,dx22);
 984             ty               = _mm_mul_ps(fscal,dy22);
 985             tz               = _mm_mul_ps(fscal,dz22);
 986
 987             /* Update vectorial force */
 988             fix2             = _mm_add_ps(fix2,tx);
 989             fiy2             = _mm_add_ps(fiy2,ty);
 990             fiz2             = _mm_add_ps(fiz2,tz);
 991
 992             fjx2             = _mm_add_ps(fjx2,tx);
 993             fjy2             = _mm_add_ps(fjy2,ty);
 994             fjz2             = _mm_add_ps(fjz2,tz);
 995
 996             /**************************
 997              * CALCULATE INTERACTIONS *
 998              **************************/
 999
1000             r23              = _mm_mul_ps(rsq23,rinv23);
1001             r23              = _mm_andnot_ps(dummy_mask,r23);
1002
1003             /* Calculate table index by multiplying r with table scale and truncate to integer */
1004             rt               = _mm_mul_ps(r23,vftabscale);
1005             vfitab           = _mm_cvttps_epi32(rt);
1006             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1007             vfitab           = _mm_slli_epi32(vfitab,2);
1008
1009             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1010             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1011             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1012             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1013             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1014             _MM_TRANSPOSE4_PS(Y,F,G,H);
1015             Heps             = _mm_mul_ps(vfeps,H);
1016             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1017             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1018             velec            = _mm_mul_ps(qq23,VV);
1019             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1020             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1021
1022             /* Update potential sum for this i atom from the interaction with this j atom. */
1023             velec            = _mm_andnot_ps(dummy_mask,velec);
1024             velecsum         = _mm_add_ps(velecsum,velec);
1025
1026             fscal            = felec;
1027
1028             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1029
1030             /* Calculate temporary vectorial force */
1031             tx               = _mm_mul_ps(fscal,dx23);
1032             ty               = _mm_mul_ps(fscal,dy23);
1033             tz               = _mm_mul_ps(fscal,dz23);
1034
1035             /* Update vectorial force */
1036             fix2             = _mm_add_ps(fix2,tx);
1037             fiy2             = _mm_add_ps(fiy2,ty);
1038             fiz2             = _mm_add_ps(fiz2,tz);
1039
1040             fjx3             = _mm_add_ps(fjx3,tx);
1041             fjy3             = _mm_add_ps(fjy3,ty);
1042             fjz3             = _mm_add_ps(fjz3,tz);
1043
1044             /**************************
1045              * CALCULATE INTERACTIONS *
1046              **************************/
1047
1048             r31              = _mm_mul_ps(rsq31,rinv31);
1049             r31              = _mm_andnot_ps(dummy_mask,r31);
1050
1051             /* Calculate table index by multiplying r with table scale and truncate to integer */
1052             rt               = _mm_mul_ps(r31,vftabscale);
1053             vfitab           = _mm_cvttps_epi32(rt);
1054             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1055             vfitab           = _mm_slli_epi32(vfitab,2);
1056
1057             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1058             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1059             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1060             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1061             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1062             _MM_TRANSPOSE4_PS(Y,F,G,H);
1063             Heps             = _mm_mul_ps(vfeps,H);
1064             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1065             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1066             velec            = _mm_mul_ps(qq31,VV);
1067             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1068             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1069
1070             /* Update potential sum for this i atom from the interaction with this j atom. */
1071             velec            = _mm_andnot_ps(dummy_mask,velec);
1072             velecsum         = _mm_add_ps(velecsum,velec);
1073
1074             fscal            = felec;
1075
1076             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1077
1078             /* Calculate temporary vectorial force */
1079             tx               = _mm_mul_ps(fscal,dx31);
1080             ty               = _mm_mul_ps(fscal,dy31);
1081             tz               = _mm_mul_ps(fscal,dz31);
1082
1083             /* Update vectorial force */
1084             fix3             = _mm_add_ps(fix3,tx);
1085             fiy3             = _mm_add_ps(fiy3,ty);
1086             fiz3             = _mm_add_ps(fiz3,tz);
1087
1088             fjx1             = _mm_add_ps(fjx1,tx);
1089             fjy1             = _mm_add_ps(fjy1,ty);
1090             fjz1             = _mm_add_ps(fjz1,tz);
1091
1092             /**************************
1093              * CALCULATE INTERACTIONS *
1094              **************************/
1095
1096             r32              = _mm_mul_ps(rsq32,rinv32);
1097             r32              = _mm_andnot_ps(dummy_mask,r32);
1098
1099             /* Calculate table index by multiplying r with table scale and truncate to integer */
1100             rt               = _mm_mul_ps(r32,vftabscale);
1101             vfitab           = _mm_cvttps_epi32(rt);
1102             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1103             vfitab           = _mm_slli_epi32(vfitab,2);
1104
1105             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1106             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1107             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1108             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1109             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1110             _MM_TRANSPOSE4_PS(Y,F,G,H);
1111             Heps             = _mm_mul_ps(vfeps,H);
1112             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1113             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1114             velec            = _mm_mul_ps(qq32,VV);
1115             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1116             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1117
1118             /* Update potential sum for this i atom from the interaction with this j atom. */
1119             velec            = _mm_andnot_ps(dummy_mask,velec);
1120             velecsum         = _mm_add_ps(velecsum,velec);
1121
1122             fscal            = felec;
1123
1124             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1125
1126             /* Calculate temporary vectorial force */
1127             tx               = _mm_mul_ps(fscal,dx32);
1128             ty               = _mm_mul_ps(fscal,dy32);
1129             tz               = _mm_mul_ps(fscal,dz32);
1130
1131             /* Update vectorial force */
1132             fix3             = _mm_add_ps(fix3,tx);
1133             fiy3             = _mm_add_ps(fiy3,ty);
1134             fiz3             = _mm_add_ps(fiz3,tz);
1135
1136             fjx2             = _mm_add_ps(fjx2,tx);
1137             fjy2             = _mm_add_ps(fjy2,ty);
1138             fjz2             = _mm_add_ps(fjz2,tz);
1139
1140             /**************************
1141              * CALCULATE INTERACTIONS *
1142              **************************/
1143
1144             r33              = _mm_mul_ps(rsq33,rinv33);
1145             r33              = _mm_andnot_ps(dummy_mask,r33);
1146
1147             /* Calculate table index by multiplying r with table scale and truncate to integer */
1148             rt               = _mm_mul_ps(r33,vftabscale);
1149             vfitab           = _mm_cvttps_epi32(rt);
1150             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1151             vfitab           = _mm_slli_epi32(vfitab,2);
1152
1153             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1154             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1155             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1156             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1157             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1158             _MM_TRANSPOSE4_PS(Y,F,G,H);
1159             Heps             = _mm_mul_ps(vfeps,H);
1160             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1161             VV               = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp));
1162             velec            = _mm_mul_ps(qq33,VV);
1163             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1164             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1165
1166             /* Update potential sum for this i atom from the interaction with this j atom. */
1167             velec            = _mm_andnot_ps(dummy_mask,velec);
1168             velecsum         = _mm_add_ps(velecsum,velec);
1169
1170             fscal            = felec;
1171
1172             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1173
1174             /* Calculate temporary vectorial force */
1175             tx               = _mm_mul_ps(fscal,dx33);
1176             ty               = _mm_mul_ps(fscal,dy33);
1177             tz               = _mm_mul_ps(fscal,dz33);
1178
1179             /* Update vectorial force */
1180             fix3             = _mm_add_ps(fix3,tx);
1181             fiy3             = _mm_add_ps(fiy3,ty);
1182             fiz3             = _mm_add_ps(fiz3,tz);
1183
1184             fjx3             = _mm_add_ps(fjx3,tx);
1185             fjy3             = _mm_add_ps(fjy3,ty);
1186             fjz3             = _mm_add_ps(fjz3,tz);
1187
1188             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(f+j_coord_offsetA+DIM,f+j_coord_offsetB+DIM,
1189                                                    f+j_coord_offsetC+DIM,f+j_coord_offsetD+DIM,
1190                                                    fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1191
1192             /* Inner loop uses 396 flops */
1193         }
1194
1195         /* End of innermost loop */
1196
1197         gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1198                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
1199
1200         ggid                        = gid[iidx];
1201         /* Update potential energies */
1202         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1203
1204         /* Increment number of inner iterations */
1205         inneriter                  += j_index_end - j_index_start;
1206
1207         /* Outer loop uses 28 flops */
1208     }
1209
1210     /* Increment number of outer iterations */
1211     outeriter        += nri;
1212
1213     /* Update outer/inner flops */
1214
1215     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_VF,outeriter*28 + inneriter*396);
1216 }
1217 /*
1218  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwNone_GeomW4W4_F_sse2_single
1219  * Electrostatics interaction: CubicSplineTable
1220  * VdW interaction:            None
1221  * Geometry:                   Water4-Water4
1222  * Calculate force/pot:        Force
1223  */
1224 void
1225 nb_kernel_ElecCSTab_VdwNone_GeomW4W4_F_sse2_single
1226                     (t_nblist * gmx_restrict                nlist,
1227                      rvec * gmx_restrict                    xx,
1228                      rvec * gmx_restrict                    ff,
1229                      t_forcerec * gmx_restrict              fr,
1230                      t_mdatoms * gmx_restrict               mdatoms,
1231                      nb_kernel_data_t * gmx_restrict        kernel_data,
1232                      t_nrnb * gmx_restrict                  nrnb)
1233 {
1234     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1235      * just 0 for non-waters.
1236      * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different
1237      * jnr indices corresponding to data put in the four positions in the SIMD register.
1238      */
1239     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1240     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1241     int              jnrA,jnrB,jnrC,jnrD;
1242     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1243     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1244     real             shX,shY,shZ,rcutoff_scalar;
1245     real             *shiftvec,*fshift,*x,*f;
1246     __m128           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
1247     int              vdwioffset1;
1248     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1249     int              vdwioffset2;
1250     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1251     int              vdwioffset3;
1252     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1253     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1254     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1255     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1256     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1257     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1258     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1259     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1260     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1261     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1262     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1263     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1264     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1265     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1266     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1267     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1268     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
1269     real             *charge;
1270     __m128i          vfitab;
1271     __m128i          ifour       = _mm_set1_epi32(4);
1272     __m128           rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
1273     real             *vftab;
1274     __m128           dummy_mask,cutoff_mask;
1275     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1276     __m128           one     = _mm_set1_ps(1.0);
1277     __m128           two     = _mm_set1_ps(2.0);
1278     x                = xx[0];
1279     f                = ff[0];
1280
1281     nri              = nlist->nri;
1282     iinr             = nlist->iinr;
1283     jindex           = nlist->jindex;
1284     jjnr             = nlist->jjnr;
1285     shiftidx         = nlist->shift;
1286     gid              = nlist->gid;
1287     shiftvec         = fr->shift_vec[0];
1288     fshift           = fr->fshift[0];
1289     facel            = _mm_set1_ps(fr->epsfac);
1290     charge           = mdatoms->chargeA;
1291
1292     vftab            = kernel_data->table_elec->data;
1293     vftabscale       = _mm_set1_ps(kernel_data->table_elec->scale);
1294
1295     /* Setup water-specific parameters */
1296     inr              = nlist->iinr[0];
1297     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1298     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1299     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
1300
1301     jq1              = _mm_set1_ps(charge[inr+1]);
1302     jq2              = _mm_set1_ps(charge[inr+2]);
1303     jq3              = _mm_set1_ps(charge[inr+3]);
1304     qq11             = _mm_mul_ps(iq1,jq1);
1305     qq12             = _mm_mul_ps(iq1,jq2);
1306     qq13             = _mm_mul_ps(iq1,jq3);
1307     qq21             = _mm_mul_ps(iq2,jq1);
1308     qq22             = _mm_mul_ps(iq2,jq2);
1309     qq23             = _mm_mul_ps(iq2,jq3);
1310     qq31             = _mm_mul_ps(iq3,jq1);
1311     qq32             = _mm_mul_ps(iq3,jq2);
1312     qq33             = _mm_mul_ps(iq3,jq3);
1313
1314     /* Avoid stupid compiler warnings */
1315     jnrA = jnrB = jnrC = jnrD = 0;
1316     j_coord_offsetA = 0;
1317     j_coord_offsetB = 0;
1318     j_coord_offsetC = 0;
1319     j_coord_offsetD = 0;
1320
1321     outeriter        = 0;
1322     inneriter        = 0;
1323
1324     /* Start outer loop over neighborlists */
1325     for(iidx=0; iidx<nri; iidx++)
1326     {
1327         /* Load shift vector for this list */
1328         i_shift_offset   = DIM*shiftidx[iidx];
1329         shX              = shiftvec[i_shift_offset+XX];
1330         shY              = shiftvec[i_shift_offset+YY];
1331         shZ              = shiftvec[i_shift_offset+ZZ];
1332
1333         /* Load limits for loop over neighbors */
1334         j_index_start    = jindex[iidx];
1335         j_index_end      = jindex[iidx+1];
1336
1337         /* Get outer coordinate index */
1338         inr              = iinr[iidx];
1339         i_coord_offset   = DIM*inr;
1340
1341         /* Load i particle coords and add shift vector */
1342         ix1              = _mm_set1_ps(shX + x[i_coord_offset+DIM*1+XX]);
1343         iy1              = _mm_set1_ps(shY + x[i_coord_offset+DIM*1+YY]);
1344         iz1              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*1+ZZ]);
1345         ix2              = _mm_set1_ps(shX + x[i_coord_offset+DIM*2+XX]);
1346         iy2              = _mm_set1_ps(shY + x[i_coord_offset+DIM*2+YY]);
1347         iz2              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*2+ZZ]);
1348         ix3              = _mm_set1_ps(shX + x[i_coord_offset+DIM*3+XX]);
1349         iy3              = _mm_set1_ps(shY + x[i_coord_offset+DIM*3+YY]);
1350         iz3              = _mm_set1_ps(shZ + x[i_coord_offset+DIM*3+ZZ]);
1351
1352         fix1             = _mm_setzero_ps();
1353         fiy1             = _mm_setzero_ps();
1354         fiz1             = _mm_setzero_ps();
1355         fix2             = _mm_setzero_ps();
1356         fiy2             = _mm_setzero_ps();
1357         fiz2             = _mm_setzero_ps();
1358         fix3             = _mm_setzero_ps();
1359         fiy3             = _mm_setzero_ps();
1360         fiz3             = _mm_setzero_ps();
1361
1362         /* Start inner kernel loop */
1363         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1364         {
1365
1366             /* Get j neighbor index, and coordinate index */
1367             jnrA             = jjnr[jidx];
1368             jnrB             = jjnr[jidx+1];
1369             jnrC             = jjnr[jidx+2];
1370             jnrD             = jjnr[jidx+3];
1371
1372             j_coord_offsetA  = DIM*jnrA;
1373             j_coord_offsetB  = DIM*jnrB;
1374             j_coord_offsetC  = DIM*jnrC;
1375             j_coord_offsetD  = DIM*jnrD;
1376
1377             /* load j atom coordinates */
1378             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1379                                               x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1380                                               &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1381
1382             /* Calculate displacement vector */
1383             dx11             = _mm_sub_ps(ix1,jx1);
1384             dy11             = _mm_sub_ps(iy1,jy1);
1385             dz11             = _mm_sub_ps(iz1,jz1);
1386             dx12             = _mm_sub_ps(ix1,jx2);
1387             dy12             = _mm_sub_ps(iy1,jy2);
1388             dz12             = _mm_sub_ps(iz1,jz2);
1389             dx13             = _mm_sub_ps(ix1,jx3);
1390             dy13             = _mm_sub_ps(iy1,jy3);
1391             dz13             = _mm_sub_ps(iz1,jz3);
1392             dx21             = _mm_sub_ps(ix2,jx1);
1393             dy21             = _mm_sub_ps(iy2,jy1);
1394             dz21             = _mm_sub_ps(iz2,jz1);
1395             dx22             = _mm_sub_ps(ix2,jx2);
1396             dy22             = _mm_sub_ps(iy2,jy2);
1397             dz22             = _mm_sub_ps(iz2,jz2);
1398             dx23             = _mm_sub_ps(ix2,jx3);
1399             dy23             = _mm_sub_ps(iy2,jy3);
1400             dz23             = _mm_sub_ps(iz2,jz3);
1401             dx31             = _mm_sub_ps(ix3,jx1);
1402             dy31             = _mm_sub_ps(iy3,jy1);
1403             dz31             = _mm_sub_ps(iz3,jz1);
1404             dx32             = _mm_sub_ps(ix3,jx2);
1405             dy32             = _mm_sub_ps(iy3,jy2);
1406             dz32             = _mm_sub_ps(iz3,jz2);
1407             dx33             = _mm_sub_ps(ix3,jx3);
1408             dy33             = _mm_sub_ps(iy3,jy3);
1409             dz33             = _mm_sub_ps(iz3,jz3);
1410
1411             /* Calculate squared distance and things based on it */
1412             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1413             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1414             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1415             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1416             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1417             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1418             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1419             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1420             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1421
1422             rinv11           = gmx_mm_invsqrt_ps(rsq11);
1423             rinv12           = gmx_mm_invsqrt_ps(rsq12);
1424             rinv13           = gmx_mm_invsqrt_ps(rsq13);
1425             rinv21           = gmx_mm_invsqrt_ps(rsq21);
1426             rinv22           = gmx_mm_invsqrt_ps(rsq22);
1427             rinv23           = gmx_mm_invsqrt_ps(rsq23);
1428             rinv31           = gmx_mm_invsqrt_ps(rsq31);
1429             rinv32           = gmx_mm_invsqrt_ps(rsq32);
1430             rinv33           = gmx_mm_invsqrt_ps(rsq33);
1431
1432             fjx1             = _mm_setzero_ps();
1433             fjy1             = _mm_setzero_ps();
1434             fjz1             = _mm_setzero_ps();
1435             fjx2             = _mm_setzero_ps();
1436             fjy2             = _mm_setzero_ps();
1437             fjz2             = _mm_setzero_ps();
1438             fjx3             = _mm_setzero_ps();
1439             fjy3             = _mm_setzero_ps();
1440             fjz3             = _mm_setzero_ps();
1441
1442             /**************************
1443              * CALCULATE INTERACTIONS *
1444              **************************/
1445
1446             r11              = _mm_mul_ps(rsq11,rinv11);
1447
1448             /* Calculate table index by multiplying r with table scale and truncate to integer */
1449             rt               = _mm_mul_ps(r11,vftabscale);
1450             vfitab           = _mm_cvttps_epi32(rt);
1451             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1452             vfitab           = _mm_slli_epi32(vfitab,2);
1453
1454             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1455             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1456             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1457             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1458             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1459             _MM_TRANSPOSE4_PS(Y,F,G,H);
1460             Heps             = _mm_mul_ps(vfeps,H);
1461             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1462             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1463             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
1464
1465             fscal            = felec;
1466
1467             /* Calculate temporary vectorial force */
1468             tx               = _mm_mul_ps(fscal,dx11);
1469             ty               = _mm_mul_ps(fscal,dy11);
1470             tz               = _mm_mul_ps(fscal,dz11);
1471
1472             /* Update vectorial force */
1473             fix1             = _mm_add_ps(fix1,tx);
1474             fiy1             = _mm_add_ps(fiy1,ty);
1475             fiz1             = _mm_add_ps(fiz1,tz);
1476
1477             fjx1             = _mm_add_ps(fjx1,tx);
1478             fjy1             = _mm_add_ps(fjy1,ty);
1479             fjz1             = _mm_add_ps(fjz1,tz);
1480
1481             /**************************
1482              * CALCULATE INTERACTIONS *
1483              **************************/
1484
1485             r12              = _mm_mul_ps(rsq12,rinv12);
1486
1487             /* Calculate table index by multiplying r with table scale and truncate to integer */
1488             rt               = _mm_mul_ps(r12,vftabscale);
1489             vfitab           = _mm_cvttps_epi32(rt);
1490             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1491             vfitab           = _mm_slli_epi32(vfitab,2);
1492
1493             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1494             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1495             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1496             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1497             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1498             _MM_TRANSPOSE4_PS(Y,F,G,H);
1499             Heps             = _mm_mul_ps(vfeps,H);
1500             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1501             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1502             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
1503
1504             fscal            = felec;
1505
1506             /* Calculate temporary vectorial force */
1507             tx               = _mm_mul_ps(fscal,dx12);
1508             ty               = _mm_mul_ps(fscal,dy12);
1509             tz               = _mm_mul_ps(fscal,dz12);
1510
1511             /* Update vectorial force */
1512             fix1             = _mm_add_ps(fix1,tx);
1513             fiy1             = _mm_add_ps(fiy1,ty);
1514             fiz1             = _mm_add_ps(fiz1,tz);
1515
1516             fjx2             = _mm_add_ps(fjx2,tx);
1517             fjy2             = _mm_add_ps(fjy2,ty);
1518             fjz2             = _mm_add_ps(fjz2,tz);
1519
1520             /**************************
1521              * CALCULATE INTERACTIONS *
1522              **************************/
1523
1524             r13              = _mm_mul_ps(rsq13,rinv13);
1525
1526             /* Calculate table index by multiplying r with table scale and truncate to integer */
1527             rt               = _mm_mul_ps(r13,vftabscale);
1528             vfitab           = _mm_cvttps_epi32(rt);
1529             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1530             vfitab           = _mm_slli_epi32(vfitab,2);
1531
1532             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1533             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1534             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1535             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1536             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1537             _MM_TRANSPOSE4_PS(Y,F,G,H);
1538             Heps             = _mm_mul_ps(vfeps,H);
1539             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1540             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1541             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
1542
1543             fscal            = felec;
1544
1545             /* Calculate temporary vectorial force */
1546             tx               = _mm_mul_ps(fscal,dx13);
1547             ty               = _mm_mul_ps(fscal,dy13);
1548             tz               = _mm_mul_ps(fscal,dz13);
1549
1550             /* Update vectorial force */
1551             fix1             = _mm_add_ps(fix1,tx);
1552             fiy1             = _mm_add_ps(fiy1,ty);
1553             fiz1             = _mm_add_ps(fiz1,tz);
1554
1555             fjx3             = _mm_add_ps(fjx3,tx);
1556             fjy3             = _mm_add_ps(fjy3,ty);
1557             fjz3             = _mm_add_ps(fjz3,tz);
1558
1559             /**************************
1560              * CALCULATE INTERACTIONS *
1561              **************************/
1562
1563             r21              = _mm_mul_ps(rsq21,rinv21);
1564
1565             /* Calculate table index by multiplying r with table scale and truncate to integer */
1566             rt               = _mm_mul_ps(r21,vftabscale);
1567             vfitab           = _mm_cvttps_epi32(rt);
1568             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1569             vfitab           = _mm_slli_epi32(vfitab,2);
1570
1571             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1572             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1573             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1574             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1575             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1576             _MM_TRANSPOSE4_PS(Y,F,G,H);
1577             Heps             = _mm_mul_ps(vfeps,H);
1578             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1579             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1580             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
1581
1582             fscal            = felec;
1583
1584             /* Calculate temporary vectorial force */
1585             tx               = _mm_mul_ps(fscal,dx21);
1586             ty               = _mm_mul_ps(fscal,dy21);
1587             tz               = _mm_mul_ps(fscal,dz21);
1588
1589             /* Update vectorial force */
1590             fix2             = _mm_add_ps(fix2,tx);
1591             fiy2             = _mm_add_ps(fiy2,ty);
1592             fiz2             = _mm_add_ps(fiz2,tz);
1593
1594             fjx1             = _mm_add_ps(fjx1,tx);
1595             fjy1             = _mm_add_ps(fjy1,ty);
1596             fjz1             = _mm_add_ps(fjz1,tz);
1597
1598             /**************************
1599              * CALCULATE INTERACTIONS *
1600              **************************/
1601
1602             r22              = _mm_mul_ps(rsq22,rinv22);
1603
1604             /* Calculate table index by multiplying r with table scale and truncate to integer */
1605             rt               = _mm_mul_ps(r22,vftabscale);
1606             vfitab           = _mm_cvttps_epi32(rt);
1607             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1608             vfitab           = _mm_slli_epi32(vfitab,2);
1609
1610             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1611             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1612             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1613             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1614             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1615             _MM_TRANSPOSE4_PS(Y,F,G,H);
1616             Heps             = _mm_mul_ps(vfeps,H);
1617             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1618             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1619             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
1620
1621             fscal            = felec;
1622
1623             /* Calculate temporary vectorial force */
1624             tx               = _mm_mul_ps(fscal,dx22);
1625             ty               = _mm_mul_ps(fscal,dy22);
1626             tz               = _mm_mul_ps(fscal,dz22);
1627
1628             /* Update vectorial force */
1629             fix2             = _mm_add_ps(fix2,tx);
1630             fiy2             = _mm_add_ps(fiy2,ty);
1631             fiz2             = _mm_add_ps(fiz2,tz);
1632
1633             fjx2             = _mm_add_ps(fjx2,tx);
1634             fjy2             = _mm_add_ps(fjy2,ty);
1635             fjz2             = _mm_add_ps(fjz2,tz);
1636
1637             /**************************
1638              * CALCULATE INTERACTIONS *
1639              **************************/
1640
1641             r23              = _mm_mul_ps(rsq23,rinv23);
1642
1643             /* Calculate table index by multiplying r with table scale and truncate to integer */
1644             rt               = _mm_mul_ps(r23,vftabscale);
1645             vfitab           = _mm_cvttps_epi32(rt);
1646             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1647             vfitab           = _mm_slli_epi32(vfitab,2);
1648
1649             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1650             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1651             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1652             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1653             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1654             _MM_TRANSPOSE4_PS(Y,F,G,H);
1655             Heps             = _mm_mul_ps(vfeps,H);
1656             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1657             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1658             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1659
1660             fscal            = felec;
1661
1662             /* Calculate temporary vectorial force */
1663             tx               = _mm_mul_ps(fscal,dx23);
1664             ty               = _mm_mul_ps(fscal,dy23);
1665             tz               = _mm_mul_ps(fscal,dz23);
1666
1667             /* Update vectorial force */
1668             fix2             = _mm_add_ps(fix2,tx);
1669             fiy2             = _mm_add_ps(fiy2,ty);
1670             fiz2             = _mm_add_ps(fiz2,tz);
1671
1672             fjx3             = _mm_add_ps(fjx3,tx);
1673             fjy3             = _mm_add_ps(fjy3,ty);
1674             fjz3             = _mm_add_ps(fjz3,tz);
1675
1676             /**************************
1677              * CALCULATE INTERACTIONS *
1678              **************************/
1679
1680             r31              = _mm_mul_ps(rsq31,rinv31);
1681
1682             /* Calculate table index by multiplying r with table scale and truncate to integer */
1683             rt               = _mm_mul_ps(r31,vftabscale);
1684             vfitab           = _mm_cvttps_epi32(rt);
1685             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1686             vfitab           = _mm_slli_epi32(vfitab,2);
1687
1688             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1689             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1690             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1691             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1692             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1693             _MM_TRANSPOSE4_PS(Y,F,G,H);
1694             Heps             = _mm_mul_ps(vfeps,H);
1695             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1696             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1697             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1698
1699             fscal            = felec;
1700
1701             /* Calculate temporary vectorial force */
1702             tx               = _mm_mul_ps(fscal,dx31);
1703             ty               = _mm_mul_ps(fscal,dy31);
1704             tz               = _mm_mul_ps(fscal,dz31);
1705
1706             /* Update vectorial force */
1707             fix3             = _mm_add_ps(fix3,tx);
1708             fiy3             = _mm_add_ps(fiy3,ty);
1709             fiz3             = _mm_add_ps(fiz3,tz);
1710
1711             fjx1             = _mm_add_ps(fjx1,tx);
1712             fjy1             = _mm_add_ps(fjy1,ty);
1713             fjz1             = _mm_add_ps(fjz1,tz);
1714
1715             /**************************
1716              * CALCULATE INTERACTIONS *
1717              **************************/
1718
1719             r32              = _mm_mul_ps(rsq32,rinv32);
1720
1721             /* Calculate table index by multiplying r with table scale and truncate to integer */
1722             rt               = _mm_mul_ps(r32,vftabscale);
1723             vfitab           = _mm_cvttps_epi32(rt);
1724             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1725             vfitab           = _mm_slli_epi32(vfitab,2);
1726
1727             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1728             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1729             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1730             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1731             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1732             _MM_TRANSPOSE4_PS(Y,F,G,H);
1733             Heps             = _mm_mul_ps(vfeps,H);
1734             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1735             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1736             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1737
1738             fscal            = felec;
1739
1740             /* Calculate temporary vectorial force */
1741             tx               = _mm_mul_ps(fscal,dx32);
1742             ty               = _mm_mul_ps(fscal,dy32);
1743             tz               = _mm_mul_ps(fscal,dz32);
1744
1745             /* Update vectorial force */
1746             fix3             = _mm_add_ps(fix3,tx);
1747             fiy3             = _mm_add_ps(fiy3,ty);
1748             fiz3             = _mm_add_ps(fiz3,tz);
1749
1750             fjx2             = _mm_add_ps(fjx2,tx);
1751             fjy2             = _mm_add_ps(fjy2,ty);
1752             fjz2             = _mm_add_ps(fjz2,tz);
1753
1754             /**************************
1755              * CALCULATE INTERACTIONS *
1756              **************************/
1757
1758             r33              = _mm_mul_ps(rsq33,rinv33);
1759
1760             /* Calculate table index by multiplying r with table scale and truncate to integer */
1761             rt               = _mm_mul_ps(r33,vftabscale);
1762             vfitab           = _mm_cvttps_epi32(rt);
1763             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1764             vfitab           = _mm_slli_epi32(vfitab,2);
1765
1766             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1767             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1768             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1769             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1770             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1771             _MM_TRANSPOSE4_PS(Y,F,G,H);
1772             Heps             = _mm_mul_ps(vfeps,H);
1773             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1774             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1775             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1776
1777             fscal            = felec;
1778
1779             /* Calculate temporary vectorial force */
1780             tx               = _mm_mul_ps(fscal,dx33);
1781             ty               = _mm_mul_ps(fscal,dy33);
1782             tz               = _mm_mul_ps(fscal,dz33);
1783
1784             /* Update vectorial force */
1785             fix3             = _mm_add_ps(fix3,tx);
1786             fiy3             = _mm_add_ps(fiy3,ty);
1787             fiz3             = _mm_add_ps(fiz3,tz);
1788
1789             fjx3             = _mm_add_ps(fjx3,tx);
1790             fjy3             = _mm_add_ps(fjy3,ty);
1791             fjz3             = _mm_add_ps(fjz3,tz);
1792
1793             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(f+j_coord_offsetA+DIM,f+j_coord_offsetB+DIM,
1794                                                    f+j_coord_offsetC+DIM,f+j_coord_offsetD+DIM,
1795                                                    fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1796
1797             /* Inner loop uses 351 flops */
1798         }
1799
1800         if(jidx<j_index_end)
1801         {
1802
1803             /* Get j neighbor index, and coordinate index */
1804             jnrA             = jjnr[jidx];
1805             jnrB             = jjnr[jidx+1];
1806             jnrC             = jjnr[jidx+2];
1807             jnrD             = jjnr[jidx+3];
1808
1809             /* Sign of each element will be negative for non-real atoms.
1810              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1811              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
1812              */
1813             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1814             jnrA       = (jnrA>=0) ? jnrA : 0;
1815             jnrB       = (jnrB>=0) ? jnrB : 0;
1816             jnrC       = (jnrC>=0) ? jnrC : 0;
1817             jnrD       = (jnrD>=0) ? jnrD : 0;
1818
1819             j_coord_offsetA  = DIM*jnrA;
1820             j_coord_offsetB  = DIM*jnrB;
1821             j_coord_offsetC  = DIM*jnrC;
1822             j_coord_offsetD  = DIM*jnrD;
1823
1824             /* load j atom coordinates */
1825             gmx_mm_load_3rvec_4ptr_swizzle_ps(x+j_coord_offsetA+DIM,x+j_coord_offsetB+DIM,
1826                                               x+j_coord_offsetC+DIM,x+j_coord_offsetD+DIM,
1827                                               &jx1,&jy1,&jz1,&jx2,&jy2,&jz2,&jx3,&jy3,&jz3);
1828
1829             /* Calculate displacement vector */
1830             dx11             = _mm_sub_ps(ix1,jx1);
1831             dy11             = _mm_sub_ps(iy1,jy1);
1832             dz11             = _mm_sub_ps(iz1,jz1);
1833             dx12             = _mm_sub_ps(ix1,jx2);
1834             dy12             = _mm_sub_ps(iy1,jy2);
1835             dz12             = _mm_sub_ps(iz1,jz2);
1836             dx13             = _mm_sub_ps(ix1,jx3);
1837             dy13             = _mm_sub_ps(iy1,jy3);
1838             dz13             = _mm_sub_ps(iz1,jz3);
1839             dx21             = _mm_sub_ps(ix2,jx1);
1840             dy21             = _mm_sub_ps(iy2,jy1);
1841             dz21             = _mm_sub_ps(iz2,jz1);
1842             dx22             = _mm_sub_ps(ix2,jx2);
1843             dy22             = _mm_sub_ps(iy2,jy2);
1844             dz22             = _mm_sub_ps(iz2,jz2);
1845             dx23             = _mm_sub_ps(ix2,jx3);
1846             dy23             = _mm_sub_ps(iy2,jy3);
1847             dz23             = _mm_sub_ps(iz2,jz3);
1848             dx31             = _mm_sub_ps(ix3,jx1);
1849             dy31             = _mm_sub_ps(iy3,jy1);
1850             dz31             = _mm_sub_ps(iz3,jz1);
1851             dx32             = _mm_sub_ps(ix3,jx2);
1852             dy32             = _mm_sub_ps(iy3,jy2);
1853             dz32             = _mm_sub_ps(iz3,jz2);
1854             dx33             = _mm_sub_ps(ix3,jx3);
1855             dy33             = _mm_sub_ps(iy3,jy3);
1856             dz33             = _mm_sub_ps(iz3,jz3);
1857
1858             /* Calculate squared distance and things based on it */
1859             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1860             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1861             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1862             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1863             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1864             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1865             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1866             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1867             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1868
1869             rinv11           = gmx_mm_invsqrt_ps(rsq11);
1870             rinv12           = gmx_mm_invsqrt_ps(rsq12);
1871             rinv13           = gmx_mm_invsqrt_ps(rsq13);
1872             rinv21           = gmx_mm_invsqrt_ps(rsq21);
1873             rinv22           = gmx_mm_invsqrt_ps(rsq22);
1874             rinv23           = gmx_mm_invsqrt_ps(rsq23);
1875             rinv31           = gmx_mm_invsqrt_ps(rsq31);
1876             rinv32           = gmx_mm_invsqrt_ps(rsq32);
1877             rinv33           = gmx_mm_invsqrt_ps(rsq33);
1878
1879             fjx1             = _mm_setzero_ps();
1880             fjy1             = _mm_setzero_ps();
1881             fjz1             = _mm_setzero_ps();
1882             fjx2             = _mm_setzero_ps();
1883             fjy2             = _mm_setzero_ps();
1884             fjz2             = _mm_setzero_ps();
1885             fjx3             = _mm_setzero_ps();
1886             fjy3             = _mm_setzero_ps();
1887             fjz3             = _mm_setzero_ps();
1888
1889             /**************************
1890              * CALCULATE INTERACTIONS *
1891              **************************/
1892
1893             r11              = _mm_mul_ps(rsq11,rinv11);
1894             r11              = _mm_andnot_ps(dummy_mask,r11);
1895
1896             /* Calculate table index by multiplying r with table scale and truncate to integer */
1897             rt               = _mm_mul_ps(r11,vftabscale);
1898             vfitab           = _mm_cvttps_epi32(rt);
1899             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1900             vfitab           = _mm_slli_epi32(vfitab,2);
1901
1902             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1903             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1904             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1905             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1906             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1907             _MM_TRANSPOSE4_PS(Y,F,G,H);
1908             Heps             = _mm_mul_ps(vfeps,H);
1909             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1910             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1911             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
1912
1913             fscal            = felec;
1914
1915             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1916
1917             /* Calculate temporary vectorial force */
1918             tx               = _mm_mul_ps(fscal,dx11);
1919             ty               = _mm_mul_ps(fscal,dy11);
1920             tz               = _mm_mul_ps(fscal,dz11);
1921
1922             /* Update vectorial force */
1923             fix1             = _mm_add_ps(fix1,tx);
1924             fiy1             = _mm_add_ps(fiy1,ty);
1925             fiz1             = _mm_add_ps(fiz1,tz);
1926
1927             fjx1             = _mm_add_ps(fjx1,tx);
1928             fjy1             = _mm_add_ps(fjy1,ty);
1929             fjz1             = _mm_add_ps(fjz1,tz);
1930
1931             /**************************
1932              * CALCULATE INTERACTIONS *
1933              **************************/
1934
1935             r12              = _mm_mul_ps(rsq12,rinv12);
1936             r12              = _mm_andnot_ps(dummy_mask,r12);
1937
1938             /* Calculate table index by multiplying r with table scale and truncate to integer */
1939             rt               = _mm_mul_ps(r12,vftabscale);
1940             vfitab           = _mm_cvttps_epi32(rt);
1941             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1942             vfitab           = _mm_slli_epi32(vfitab,2);
1943
1944             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1945             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1946             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1947             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1948             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1949             _MM_TRANSPOSE4_PS(Y,F,G,H);
1950             Heps             = _mm_mul_ps(vfeps,H);
1951             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1952             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1953             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
1954
1955             fscal            = felec;
1956
1957             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1958
1959             /* Calculate temporary vectorial force */
1960             tx               = _mm_mul_ps(fscal,dx12);
1961             ty               = _mm_mul_ps(fscal,dy12);
1962             tz               = _mm_mul_ps(fscal,dz12);
1963
1964             /* Update vectorial force */
1965             fix1             = _mm_add_ps(fix1,tx);
1966             fiy1             = _mm_add_ps(fiy1,ty);
1967             fiz1             = _mm_add_ps(fiz1,tz);
1968
1969             fjx2             = _mm_add_ps(fjx2,tx);
1970             fjy2             = _mm_add_ps(fjy2,ty);
1971             fjz2             = _mm_add_ps(fjz2,tz);
1972
1973             /**************************
1974              * CALCULATE INTERACTIONS *
1975              **************************/
1976
1977             r13              = _mm_mul_ps(rsq13,rinv13);
1978             r13              = _mm_andnot_ps(dummy_mask,r13);
1979
1980             /* Calculate table index by multiplying r with table scale and truncate to integer */
1981             rt               = _mm_mul_ps(r13,vftabscale);
1982             vfitab           = _mm_cvttps_epi32(rt);
1983             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
1984             vfitab           = _mm_slli_epi32(vfitab,2);
1985
1986             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1987             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
1988             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
1989             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
1990             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
1991             _MM_TRANSPOSE4_PS(Y,F,G,H);
1992             Heps             = _mm_mul_ps(vfeps,H);
1993             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
1994             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
1995             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
1996
1997             fscal            = felec;
1998
1999             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2000
2001             /* Calculate temporary vectorial force */
2002             tx               = _mm_mul_ps(fscal,dx13);
2003             ty               = _mm_mul_ps(fscal,dy13);
2004             tz               = _mm_mul_ps(fscal,dz13);
2005
2006             /* Update vectorial force */
2007             fix1             = _mm_add_ps(fix1,tx);
2008             fiy1             = _mm_add_ps(fiy1,ty);
2009             fiz1             = _mm_add_ps(fiz1,tz);
2010
2011             fjx3             = _mm_add_ps(fjx3,tx);
2012             fjy3             = _mm_add_ps(fjy3,ty);
2013             fjz3             = _mm_add_ps(fjz3,tz);
2014
2015             /**************************
2016              * CALCULATE INTERACTIONS *
2017              **************************/
2018
2019             r21              = _mm_mul_ps(rsq21,rinv21);
2020             r21              = _mm_andnot_ps(dummy_mask,r21);
2021
2022             /* Calculate table index by multiplying r with table scale and truncate to integer */
2023             rt               = _mm_mul_ps(r21,vftabscale);
2024             vfitab           = _mm_cvttps_epi32(rt);
2025             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2026             vfitab           = _mm_slli_epi32(vfitab,2);
2027
2028             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2029             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2030             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2031             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2032             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2033             _MM_TRANSPOSE4_PS(Y,F,G,H);
2034             Heps             = _mm_mul_ps(vfeps,H);
2035             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2036             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2037             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
2038
2039             fscal            = felec;
2040
2041             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2042
2043             /* Calculate temporary vectorial force */
2044             tx               = _mm_mul_ps(fscal,dx21);
2045             ty               = _mm_mul_ps(fscal,dy21);
2046             tz               = _mm_mul_ps(fscal,dz21);
2047
2048             /* Update vectorial force */
2049             fix2             = _mm_add_ps(fix2,tx);
2050             fiy2             = _mm_add_ps(fiy2,ty);
2051             fiz2             = _mm_add_ps(fiz2,tz);
2052
2053             fjx1             = _mm_add_ps(fjx1,tx);
2054             fjy1             = _mm_add_ps(fjy1,ty);
2055             fjz1             = _mm_add_ps(fjz1,tz);
2056
2057             /**************************
2058              * CALCULATE INTERACTIONS *
2059              **************************/
2060
2061             r22              = _mm_mul_ps(rsq22,rinv22);
2062             r22              = _mm_andnot_ps(dummy_mask,r22);
2063
2064             /* Calculate table index by multiplying r with table scale and truncate to integer */
2065             rt               = _mm_mul_ps(r22,vftabscale);
2066             vfitab           = _mm_cvttps_epi32(rt);
2067             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2068             vfitab           = _mm_slli_epi32(vfitab,2);
2069
2070             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2071             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2072             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2073             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2074             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2075             _MM_TRANSPOSE4_PS(Y,F,G,H);
2076             Heps             = _mm_mul_ps(vfeps,H);
2077             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2078             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2079             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
2080
2081             fscal            = felec;
2082
2083             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2084
2085             /* Calculate temporary vectorial force */
2086             tx               = _mm_mul_ps(fscal,dx22);
2087             ty               = _mm_mul_ps(fscal,dy22);
2088             tz               = _mm_mul_ps(fscal,dz22);
2089
2090             /* Update vectorial force */
2091             fix2             = _mm_add_ps(fix2,tx);
2092             fiy2             = _mm_add_ps(fiy2,ty);
2093             fiz2             = _mm_add_ps(fiz2,tz);
2094
2095             fjx2             = _mm_add_ps(fjx2,tx);
2096             fjy2             = _mm_add_ps(fjy2,ty);
2097             fjz2             = _mm_add_ps(fjz2,tz);
2098
2099             /**************************
2100              * CALCULATE INTERACTIONS *
2101              **************************/
2102
2103             r23              = _mm_mul_ps(rsq23,rinv23);
2104             r23              = _mm_andnot_ps(dummy_mask,r23);
2105
2106             /* Calculate table index by multiplying r with table scale and truncate to integer */
2107             rt               = _mm_mul_ps(r23,vftabscale);
2108             vfitab           = _mm_cvttps_epi32(rt);
2109             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2110             vfitab           = _mm_slli_epi32(vfitab,2);
2111
2112             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2113             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2114             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2115             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2116             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2117             _MM_TRANSPOSE4_PS(Y,F,G,H);
2118             Heps             = _mm_mul_ps(vfeps,H);
2119             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2120             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2121             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
2122
2123             fscal            = felec;
2124
2125             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2126
2127             /* Calculate temporary vectorial force */
2128             tx               = _mm_mul_ps(fscal,dx23);
2129             ty               = _mm_mul_ps(fscal,dy23);
2130             tz               = _mm_mul_ps(fscal,dz23);
2131
2132             /* Update vectorial force */
2133             fix2             = _mm_add_ps(fix2,tx);
2134             fiy2             = _mm_add_ps(fiy2,ty);
2135             fiz2             = _mm_add_ps(fiz2,tz);
2136
2137             fjx3             = _mm_add_ps(fjx3,tx);
2138             fjy3             = _mm_add_ps(fjy3,ty);
2139             fjz3             = _mm_add_ps(fjz3,tz);
2140
2141             /**************************
2142              * CALCULATE INTERACTIONS *
2143              **************************/
2144
2145             r31              = _mm_mul_ps(rsq31,rinv31);
2146             r31              = _mm_andnot_ps(dummy_mask,r31);
2147
2148             /* Calculate table index by multiplying r with table scale and truncate to integer */
2149             rt               = _mm_mul_ps(r31,vftabscale);
2150             vfitab           = _mm_cvttps_epi32(rt);
2151             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2152             vfitab           = _mm_slli_epi32(vfitab,2);
2153
2154             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2155             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2156             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2157             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2158             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2159             _MM_TRANSPOSE4_PS(Y,F,G,H);
2160             Heps             = _mm_mul_ps(vfeps,H);
2161             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2162             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2163             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
2164
2165             fscal            = felec;
2166
2167             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2168
2169             /* Calculate temporary vectorial force */
2170             tx               = _mm_mul_ps(fscal,dx31);
2171             ty               = _mm_mul_ps(fscal,dy31);
2172             tz               = _mm_mul_ps(fscal,dz31);
2173
2174             /* Update vectorial force */
2175             fix3             = _mm_add_ps(fix3,tx);
2176             fiy3             = _mm_add_ps(fiy3,ty);
2177             fiz3             = _mm_add_ps(fiz3,tz);
2178
2179             fjx1             = _mm_add_ps(fjx1,tx);
2180             fjy1             = _mm_add_ps(fjy1,ty);
2181             fjz1             = _mm_add_ps(fjz1,tz);
2182
2183             /**************************
2184              * CALCULATE INTERACTIONS *
2185              **************************/
2186
2187             r32              = _mm_mul_ps(rsq32,rinv32);
2188             r32              = _mm_andnot_ps(dummy_mask,r32);
2189
2190             /* Calculate table index by multiplying r with table scale and truncate to integer */
2191             rt               = _mm_mul_ps(r32,vftabscale);
2192             vfitab           = _mm_cvttps_epi32(rt);
2193             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2194             vfitab           = _mm_slli_epi32(vfitab,2);
2195
2196             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2197             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2198             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2199             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2200             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2201             _MM_TRANSPOSE4_PS(Y,F,G,H);
2202             Heps             = _mm_mul_ps(vfeps,H);
2203             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2204             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2205             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
2206
2207             fscal            = felec;
2208
2209             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2210
2211             /* Calculate temporary vectorial force */
2212             tx               = _mm_mul_ps(fscal,dx32);
2213             ty               = _mm_mul_ps(fscal,dy32);
2214             tz               = _mm_mul_ps(fscal,dz32);
2215
2216             /* Update vectorial force */
2217             fix3             = _mm_add_ps(fix3,tx);
2218             fiy3             = _mm_add_ps(fiy3,ty);
2219             fiz3             = _mm_add_ps(fiz3,tz);
2220
2221             fjx2             = _mm_add_ps(fjx2,tx);
2222             fjy2             = _mm_add_ps(fjy2,ty);
2223             fjz2             = _mm_add_ps(fjz2,tz);
2224
2225             /**************************
2226              * CALCULATE INTERACTIONS *
2227              **************************/
2228
2229             r33              = _mm_mul_ps(rsq33,rinv33);
2230             r33              = _mm_andnot_ps(dummy_mask,r33);
2231
2232             /* Calculate table index by multiplying r with table scale and truncate to integer */
2233             rt               = _mm_mul_ps(r33,vftabscale);
2234             vfitab           = _mm_cvttps_epi32(rt);
2235             vfeps            = _mm_sub_ps(rt,_mm_cvtepi32_ps(vfitab));
2236             vfitab           = _mm_slli_epi32(vfitab,2);
2237
2238             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2239             Y                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0) );
2240             F                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1) );
2241             G                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2) );
2242             H                = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3) );
2243             _MM_TRANSPOSE4_PS(Y,F,G,H);
2244             Heps             = _mm_mul_ps(vfeps,H);
2245             Fp               = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps)));
2246             FF               = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps))));
2247             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
2248
2249             fscal            = felec;
2250
2251             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2252
2253             /* Calculate temporary vectorial force */
2254             tx               = _mm_mul_ps(fscal,dx33);
2255             ty               = _mm_mul_ps(fscal,dy33);
2256             tz               = _mm_mul_ps(fscal,dz33);
2257
2258             /* Update vectorial force */
2259             fix3             = _mm_add_ps(fix3,tx);
2260             fiy3             = _mm_add_ps(fiy3,ty);
2261             fiz3             = _mm_add_ps(fiz3,tz);
2262
2263             fjx3             = _mm_add_ps(fjx3,tx);
2264             fjy3             = _mm_add_ps(fjy3,ty);
2265             fjz3             = _mm_add_ps(fjz3,tz);
2266
2267             gmx_mm_decrement_3rvec_4ptr_swizzle_ps(f+j_coord_offsetA+DIM,f+j_coord_offsetB+DIM,
2268                                                    f+j_coord_offsetC+DIM,f+j_coord_offsetD+DIM,
2269                                                    fjx1,fjy1,fjz1,fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2270
2271             /* Inner loop uses 360 flops */
2272         }
2273
2274         /* End of innermost loop */
2275
2276         gmx_mm_update_iforce_3atom_swizzle_ps(fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2277                                               f+i_coord_offset+DIM,fshift+i_shift_offset);
2278
2279         /* Increment number of inner iterations */
2280         inneriter                  += j_index_end - j_index_start;
2281
2282         /* Outer loop uses 27 flops */
2283     }
2284
2285     /* Increment number of outer iterations */
2286     outeriter        += nri;
2287
2288     /* Update outer/inner flops */
2289
2290     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4W4_F,outeriter*27 + inneriter*360);
2291 }