src/gromacs/gmxlib/nonbonded/nb_kernel_avx_128_fma_single/nb_kernel_ElecCSTab_VdwLJ_GeomW4W4_avx_128_fma_single.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS avx_128_fma_single kernel generator.
  37  */
  38 #include "config.h"
  39
  40 #include <math.h>
  41
  42 #include "../nb_kernel.h"
  43 #include "gromacs/legacyheaders/types/simple.h"
  44 #include "gromacs/math/vec.h"
  45 #include "gromacs/legacyheaders/nrnb.h"
  46
  47 #include "gromacs/simd/math_x86_avx_128_fma_single.h"
  48 #include "kernelutil_x86_avx_128_fma_single.h"
  49
  50 /*
  51  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwLJ_GeomW4W4_VF_avx_128_fma_single
  52  * Electrostatics interaction: CubicSplineTable
  53  * VdW interaction:            LennardJones
  54  * Geometry:                   Water4-Water4
  55  * Calculate force/pot:        PotentialAndForce
  56  */
  57 void
  58 nb_kernel_ElecCSTab_VdwLJ_GeomW4W4_VF_avx_128_fma_single
  59                     (t_nblist                    * gmx_restrict       nlist,
  60                      rvec                        * gmx_restrict          xx,
  61                      rvec                        * gmx_restrict          ff,
  62                      t_forcerec                  * gmx_restrict          fr,
  63                      t_mdatoms                   * gmx_restrict     mdatoms,
  64                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  65                      t_nrnb                      * gmx_restrict        nrnb)
  66 {
  67     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  68      * just 0 for non-waters.
  69      * Suffixes A,B,C,D refer to j loop unrolling done with AVX_128, e.g. for the four different
  70      * jnr indices corresponding to data put in the four positions in the SIMD register.
  71      */
  72     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  73     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  74     int              jnrA,jnrB,jnrC,jnrD;
  75     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  76     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  77     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  78     real             rcutoff_scalar;
  79     real             *shiftvec,*fshift,*x,*f;
  80     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
  81     real             scratch[4*DIM];
  82     __m128           fscal,rcutoff,rcutoff2,jidxall;
  83     int              vdwioffset0;
  84     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  85     int              vdwioffset1;
  86     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  87     int              vdwioffset2;
  88     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  89     int              vdwioffset3;
  90     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  91     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
  92     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  93     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
  94     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
  95     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
  96     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
  97     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
  98     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
  99     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 100     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
 101     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
 102     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
 103     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
 104     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
 105     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
 106     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
 107     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
 108     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
 109     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
 110     real             *charge;
 111     int              nvdwtype;
 112     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 113     int              *vdwtype;
 114     real             *vdwparam;
 115     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
 116     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
 117     __m128i          vfitab;
 118     __m128i          ifour       = _mm_set1_epi32(4);
 119     __m128           rt,vfeps,twovfeps,vftabscale,Y,F,G,H,Fp,VV,FF;
 120     real             *vftab;
 121     __m128           dummy_mask,cutoff_mask;
 122     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
 123     __m128           one     = _mm_set1_ps(1.0);
 124     __m128           two     = _mm_set1_ps(2.0);
 125     x                = xx[0];
 126     f                = ff[0];
 127
 128     nri              = nlist->nri;
 129     iinr             = nlist->iinr;
 130     jindex           = nlist->jindex;
 131     jjnr             = nlist->jjnr;
 132     shiftidx         = nlist->shift;
 133     gid              = nlist->gid;
 134     shiftvec         = fr->shift_vec[0];
 135     fshift           = fr->fshift[0];
 136     facel            = _mm_set1_ps(fr->epsfac);
 137     charge           = mdatoms->chargeA;
 138     nvdwtype         = fr->ntype;
 139     vdwparam         = fr->nbfp;
 140     vdwtype          = mdatoms->typeA;
 141
 142     vftab            = kernel_data->table_elec->data;
 143     vftabscale       = _mm_set1_ps(kernel_data->table_elec->scale);
 144
 145     /* Setup water-specific parameters */
 146     inr              = nlist->iinr[0];
 147     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
 148     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
 149     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
 150     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 151
 152     jq1              = _mm_set1_ps(charge[inr+1]);
 153     jq2              = _mm_set1_ps(charge[inr+2]);
 154     jq3              = _mm_set1_ps(charge[inr+3]);
 155     vdwjidx0A        = 2*vdwtype[inr+0];
 156     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
 157     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
 158     qq11             = _mm_mul_ps(iq1,jq1);
 159     qq12             = _mm_mul_ps(iq1,jq2);
 160     qq13             = _mm_mul_ps(iq1,jq3);
 161     qq21             = _mm_mul_ps(iq2,jq1);
 162     qq22             = _mm_mul_ps(iq2,jq2);
 163     qq23             = _mm_mul_ps(iq2,jq3);
 164     qq31             = _mm_mul_ps(iq3,jq1);
 165     qq32             = _mm_mul_ps(iq3,jq2);
 166     qq33             = _mm_mul_ps(iq3,jq3);
 167
 168     /* Avoid stupid compiler warnings */
 169     jnrA = jnrB = jnrC = jnrD = 0;
 170     j_coord_offsetA = 0;
 171     j_coord_offsetB = 0;
 172     j_coord_offsetC = 0;
 173     j_coord_offsetD = 0;
 174
 175     outeriter        = 0;
 176     inneriter        = 0;
 177
 178     for(iidx=0;iidx<4*DIM;iidx++)
 179     {
 180         scratch[iidx] = 0.0;
 181     }
 182
 183     /* Start outer loop over neighborlists */
 184     for(iidx=0; iidx<nri; iidx++)
 185     {
 186         /* Load shift vector for this list */
 187         i_shift_offset   = DIM*shiftidx[iidx];
 188
 189         /* Load limits for loop over neighbors */
 190         j_index_start    = jindex[iidx];
 191         j_index_end      = jindex[iidx+1];
 192
 193         /* Get outer coordinate index */
 194         inr              = iinr[iidx];
 195         i_coord_offset   = DIM*inr;
 196
 197         /* Load i particle coords and add shift vector */
 198         gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
 199                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 200
 201         fix0             = _mm_setzero_ps();
 202         fiy0             = _mm_setzero_ps();
 203         fiz0             = _mm_setzero_ps();
 204         fix1             = _mm_setzero_ps();
 205         fiy1             = _mm_setzero_ps();
 206         fiz1             = _mm_setzero_ps();
 207         fix2             = _mm_setzero_ps();
 208         fiy2             = _mm_setzero_ps();
 209         fiz2             = _mm_setzero_ps();
 210         fix3             = _mm_setzero_ps();
 211         fiy3             = _mm_setzero_ps();
 212         fiz3             = _mm_setzero_ps();
 213
 214         /* Reset potential sums */
 215         velecsum         = _mm_setzero_ps();
 216         vvdwsum          = _mm_setzero_ps();
 217
 218         /* Start inner kernel loop */
 219         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
 220         {
 221
 222             /* Get j neighbor index, and coordinate index */
 223             jnrA             = jjnr[jidx];
 224             jnrB             = jjnr[jidx+1];
 225             jnrC             = jjnr[jidx+2];
 226             jnrD             = jjnr[jidx+3];
 227             j_coord_offsetA  = DIM*jnrA;
 228             j_coord_offsetB  = DIM*jnrB;
 229             j_coord_offsetC  = DIM*jnrC;
 230             j_coord_offsetD  = DIM*jnrD;
 231
 232             /* load j atom coordinates */
 233             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 234                                               x+j_coord_offsetC,x+j_coord_offsetD,
 235                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 236                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 237
 238             /* Calculate displacement vector */
 239             dx00             = _mm_sub_ps(ix0,jx0);
 240             dy00             = _mm_sub_ps(iy0,jy0);
 241             dz00             = _mm_sub_ps(iz0,jz0);
 242             dx11             = _mm_sub_ps(ix1,jx1);
 243             dy11             = _mm_sub_ps(iy1,jy1);
 244             dz11             = _mm_sub_ps(iz1,jz1);
 245             dx12             = _mm_sub_ps(ix1,jx2);
 246             dy12             = _mm_sub_ps(iy1,jy2);
 247             dz12             = _mm_sub_ps(iz1,jz2);
 248             dx13             = _mm_sub_ps(ix1,jx3);
 249             dy13             = _mm_sub_ps(iy1,jy3);
 250             dz13             = _mm_sub_ps(iz1,jz3);
 251             dx21             = _mm_sub_ps(ix2,jx1);
 252             dy21             = _mm_sub_ps(iy2,jy1);
 253             dz21             = _mm_sub_ps(iz2,jz1);
 254             dx22             = _mm_sub_ps(ix2,jx2);
 255             dy22             = _mm_sub_ps(iy2,jy2);
 256             dz22             = _mm_sub_ps(iz2,jz2);
 257             dx23             = _mm_sub_ps(ix2,jx3);
 258             dy23             = _mm_sub_ps(iy2,jy3);
 259             dz23             = _mm_sub_ps(iz2,jz3);
 260             dx31             = _mm_sub_ps(ix3,jx1);
 261             dy31             = _mm_sub_ps(iy3,jy1);
 262             dz31             = _mm_sub_ps(iz3,jz1);
 263             dx32             = _mm_sub_ps(ix3,jx2);
 264             dy32             = _mm_sub_ps(iy3,jy2);
 265             dz32             = _mm_sub_ps(iz3,jz2);
 266             dx33             = _mm_sub_ps(ix3,jx3);
 267             dy33             = _mm_sub_ps(iy3,jy3);
 268             dz33             = _mm_sub_ps(iz3,jz3);
 269
 270             /* Calculate squared distance and things based on it */
 271             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 272             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 273             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 274             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 275             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 276             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 277             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 278             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 279             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 280             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 281
 282             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 283             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 284             rinv13           = gmx_mm_invsqrt_ps(rsq13);
 285             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 286             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 287             rinv23           = gmx_mm_invsqrt_ps(rsq23);
 288             rinv31           = gmx_mm_invsqrt_ps(rsq31);
 289             rinv32           = gmx_mm_invsqrt_ps(rsq32);
 290             rinv33           = gmx_mm_invsqrt_ps(rsq33);
 291
 292             rinvsq00         = gmx_mm_inv_ps(rsq00);
 293
 294             fjx0             = _mm_setzero_ps();
 295             fjy0             = _mm_setzero_ps();
 296             fjz0             = _mm_setzero_ps();
 297             fjx1             = _mm_setzero_ps();
 298             fjy1             = _mm_setzero_ps();
 299             fjz1             = _mm_setzero_ps();
 300             fjx2             = _mm_setzero_ps();
 301             fjy2             = _mm_setzero_ps();
 302             fjz2             = _mm_setzero_ps();
 303             fjx3             = _mm_setzero_ps();
 304             fjy3             = _mm_setzero_ps();
 305             fjz3             = _mm_setzero_ps();
 306
 307             /**************************
 308              * CALCULATE INTERACTIONS *
 309              **************************/
 310
 311             /* LENNARD-JONES DISPERSION/REPULSION */
 312
 313             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 314             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 315             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 316             vvdw             = _mm_msub_ps(vvdw12,one_twelfth,_mm_mul_ps(vvdw6,one_sixth));
 317             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 318
 319             /* Update potential sum for this i atom from the interaction with this j atom. */
 320             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 321
 322             fscal            = fvdw;
 323
 324              /* Update vectorial force */
 325             fix0             = _mm_macc_ps(dx00,fscal,fix0);
 326             fiy0             = _mm_macc_ps(dy00,fscal,fiy0);
 327             fiz0             = _mm_macc_ps(dz00,fscal,fiz0);
 328
 329             fjx0             = _mm_macc_ps(dx00,fscal,fjx0);
 330             fjy0             = _mm_macc_ps(dy00,fscal,fjy0);
 331             fjz0             = _mm_macc_ps(dz00,fscal,fjz0);
 332
 333             /**************************
 334              * CALCULATE INTERACTIONS *
 335              **************************/
 336
 337             r11              = _mm_mul_ps(rsq11,rinv11);
 338
 339             /* Calculate table index by multiplying r with table scale and truncate to integer */
 340             rt               = _mm_mul_ps(r11,vftabscale);
 341             vfitab           = _mm_cvttps_epi32(rt);
 342 #ifdef __XOP__
 343             vfeps            = _mm_frcz_ps(rt);
 344 #else
 345             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 346 #endif
 347             twovfeps         = _mm_add_ps(vfeps,vfeps);
 348             vfitab           = _mm_slli_epi32(vfitab,2);
 349
 350             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 351             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 352             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 353             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 354             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 355             _MM_TRANSPOSE4_PS(Y,F,G,H);
 356             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 357             VV               = _mm_macc_ps(vfeps,Fp,Y);
 358             velec            = _mm_mul_ps(qq11,VV);
 359             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 360             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
 361
 362             /* Update potential sum for this i atom from the interaction with this j atom. */
 363             velecsum         = _mm_add_ps(velecsum,velec);
 364
 365             fscal            = felec;
 366
 367              /* Update vectorial force */
 368             fix1             = _mm_macc_ps(dx11,fscal,fix1);
 369             fiy1             = _mm_macc_ps(dy11,fscal,fiy1);
 370             fiz1             = _mm_macc_ps(dz11,fscal,fiz1);
 371
 372             fjx1             = _mm_macc_ps(dx11,fscal,fjx1);
 373             fjy1             = _mm_macc_ps(dy11,fscal,fjy1);
 374             fjz1             = _mm_macc_ps(dz11,fscal,fjz1);
 375
 376             /**************************
 377              * CALCULATE INTERACTIONS *
 378              **************************/
 379
 380             r12              = _mm_mul_ps(rsq12,rinv12);
 381
 382             /* Calculate table index by multiplying r with table scale and truncate to integer */
 383             rt               = _mm_mul_ps(r12,vftabscale);
 384             vfitab           = _mm_cvttps_epi32(rt);
 385 #ifdef __XOP__
 386             vfeps            = _mm_frcz_ps(rt);
 387 #else
 388             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 389 #endif
 390             twovfeps         = _mm_add_ps(vfeps,vfeps);
 391             vfitab           = _mm_slli_epi32(vfitab,2);
 392
 393             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 394             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 395             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 396             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 397             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 398             _MM_TRANSPOSE4_PS(Y,F,G,H);
 399             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 400             VV               = _mm_macc_ps(vfeps,Fp,Y);
 401             velec            = _mm_mul_ps(qq12,VV);
 402             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 403             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
 404
 405             /* Update potential sum for this i atom from the interaction with this j atom. */
 406             velecsum         = _mm_add_ps(velecsum,velec);
 407
 408             fscal            = felec;
 409
 410              /* Update vectorial force */
 411             fix1             = _mm_macc_ps(dx12,fscal,fix1);
 412             fiy1             = _mm_macc_ps(dy12,fscal,fiy1);
 413             fiz1             = _mm_macc_ps(dz12,fscal,fiz1);
 414
 415             fjx2             = _mm_macc_ps(dx12,fscal,fjx2);
 416             fjy2             = _mm_macc_ps(dy12,fscal,fjy2);
 417             fjz2             = _mm_macc_ps(dz12,fscal,fjz2);
 418
 419             /**************************
 420              * CALCULATE INTERACTIONS *
 421              **************************/
 422
 423             r13              = _mm_mul_ps(rsq13,rinv13);
 424
 425             /* Calculate table index by multiplying r with table scale and truncate to integer */
 426             rt               = _mm_mul_ps(r13,vftabscale);
 427             vfitab           = _mm_cvttps_epi32(rt);
 428 #ifdef __XOP__
 429             vfeps            = _mm_frcz_ps(rt);
 430 #else
 431             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 432 #endif
 433             twovfeps         = _mm_add_ps(vfeps,vfeps);
 434             vfitab           = _mm_slli_epi32(vfitab,2);
 435
 436             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 437             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 438             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 439             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 440             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 441             _MM_TRANSPOSE4_PS(Y,F,G,H);
 442             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 443             VV               = _mm_macc_ps(vfeps,Fp,Y);
 444             velec            = _mm_mul_ps(qq13,VV);
 445             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 446             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
 447
 448             /* Update potential sum for this i atom from the interaction with this j atom. */
 449             velecsum         = _mm_add_ps(velecsum,velec);
 450
 451             fscal            = felec;
 452
 453              /* Update vectorial force */
 454             fix1             = _mm_macc_ps(dx13,fscal,fix1);
 455             fiy1             = _mm_macc_ps(dy13,fscal,fiy1);
 456             fiz1             = _mm_macc_ps(dz13,fscal,fiz1);
 457
 458             fjx3             = _mm_macc_ps(dx13,fscal,fjx3);
 459             fjy3             = _mm_macc_ps(dy13,fscal,fjy3);
 460             fjz3             = _mm_macc_ps(dz13,fscal,fjz3);
 461
 462             /**************************
 463              * CALCULATE INTERACTIONS *
 464              **************************/
 465
 466             r21              = _mm_mul_ps(rsq21,rinv21);
 467
 468             /* Calculate table index by multiplying r with table scale and truncate to integer */
 469             rt               = _mm_mul_ps(r21,vftabscale);
 470             vfitab           = _mm_cvttps_epi32(rt);
 471 #ifdef __XOP__
 472             vfeps            = _mm_frcz_ps(rt);
 473 #else
 474             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 475 #endif
 476             twovfeps         = _mm_add_ps(vfeps,vfeps);
 477             vfitab           = _mm_slli_epi32(vfitab,2);
 478
 479             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 480             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 481             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 482             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 483             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 484             _MM_TRANSPOSE4_PS(Y,F,G,H);
 485             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 486             VV               = _mm_macc_ps(vfeps,Fp,Y);
 487             velec            = _mm_mul_ps(qq21,VV);
 488             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 489             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
 490
 491             /* Update potential sum for this i atom from the interaction with this j atom. */
 492             velecsum         = _mm_add_ps(velecsum,velec);
 493
 494             fscal            = felec;
 495
 496              /* Update vectorial force */
 497             fix2             = _mm_macc_ps(dx21,fscal,fix2);
 498             fiy2             = _mm_macc_ps(dy21,fscal,fiy2);
 499             fiz2             = _mm_macc_ps(dz21,fscal,fiz2);
 500
 501             fjx1             = _mm_macc_ps(dx21,fscal,fjx1);
 502             fjy1             = _mm_macc_ps(dy21,fscal,fjy1);
 503             fjz1             = _mm_macc_ps(dz21,fscal,fjz1);
 504
 505             /**************************
 506              * CALCULATE INTERACTIONS *
 507              **************************/
 508
 509             r22              = _mm_mul_ps(rsq22,rinv22);
 510
 511             /* Calculate table index by multiplying r with table scale and truncate to integer */
 512             rt               = _mm_mul_ps(r22,vftabscale);
 513             vfitab           = _mm_cvttps_epi32(rt);
 514 #ifdef __XOP__
 515             vfeps            = _mm_frcz_ps(rt);
 516 #else
 517             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 518 #endif
 519             twovfeps         = _mm_add_ps(vfeps,vfeps);
 520             vfitab           = _mm_slli_epi32(vfitab,2);
 521
 522             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 523             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 524             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 525             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 526             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 527             _MM_TRANSPOSE4_PS(Y,F,G,H);
 528             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 529             VV               = _mm_macc_ps(vfeps,Fp,Y);
 530             velec            = _mm_mul_ps(qq22,VV);
 531             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 532             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
 533
 534             /* Update potential sum for this i atom from the interaction with this j atom. */
 535             velecsum         = _mm_add_ps(velecsum,velec);
 536
 537             fscal            = felec;
 538
 539              /* Update vectorial force */
 540             fix2             = _mm_macc_ps(dx22,fscal,fix2);
 541             fiy2             = _mm_macc_ps(dy22,fscal,fiy2);
 542             fiz2             = _mm_macc_ps(dz22,fscal,fiz2);
 543
 544             fjx2             = _mm_macc_ps(dx22,fscal,fjx2);
 545             fjy2             = _mm_macc_ps(dy22,fscal,fjy2);
 546             fjz2             = _mm_macc_ps(dz22,fscal,fjz2);
 547
 548             /**************************
 549              * CALCULATE INTERACTIONS *
 550              **************************/
 551
 552             r23              = _mm_mul_ps(rsq23,rinv23);
 553
 554             /* Calculate table index by multiplying r with table scale and truncate to integer */
 555             rt               = _mm_mul_ps(r23,vftabscale);
 556             vfitab           = _mm_cvttps_epi32(rt);
 557 #ifdef __XOP__
 558             vfeps            = _mm_frcz_ps(rt);
 559 #else
 560             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 561 #endif
 562             twovfeps         = _mm_add_ps(vfeps,vfeps);
 563             vfitab           = _mm_slli_epi32(vfitab,2);
 564
 565             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 566             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 567             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 568             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 569             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 570             _MM_TRANSPOSE4_PS(Y,F,G,H);
 571             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 572             VV               = _mm_macc_ps(vfeps,Fp,Y);
 573             velec            = _mm_mul_ps(qq23,VV);
 574             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 575             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
 576
 577             /* Update potential sum for this i atom from the interaction with this j atom. */
 578             velecsum         = _mm_add_ps(velecsum,velec);
 579
 580             fscal            = felec;
 581
 582              /* Update vectorial force */
 583             fix2             = _mm_macc_ps(dx23,fscal,fix2);
 584             fiy2             = _mm_macc_ps(dy23,fscal,fiy2);
 585             fiz2             = _mm_macc_ps(dz23,fscal,fiz2);
 586
 587             fjx3             = _mm_macc_ps(dx23,fscal,fjx3);
 588             fjy3             = _mm_macc_ps(dy23,fscal,fjy3);
 589             fjz3             = _mm_macc_ps(dz23,fscal,fjz3);
 590
 591             /**************************
 592              * CALCULATE INTERACTIONS *
 593              **************************/
 594
 595             r31              = _mm_mul_ps(rsq31,rinv31);
 596
 597             /* Calculate table index by multiplying r with table scale and truncate to integer */
 598             rt               = _mm_mul_ps(r31,vftabscale);
 599             vfitab           = _mm_cvttps_epi32(rt);
 600 #ifdef __XOP__
 601             vfeps            = _mm_frcz_ps(rt);
 602 #else
 603             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 604 #endif
 605             twovfeps         = _mm_add_ps(vfeps,vfeps);
 606             vfitab           = _mm_slli_epi32(vfitab,2);
 607
 608             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 609             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 610             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 611             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 612             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 613             _MM_TRANSPOSE4_PS(Y,F,G,H);
 614             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 615             VV               = _mm_macc_ps(vfeps,Fp,Y);
 616             velec            = _mm_mul_ps(qq31,VV);
 617             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 618             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
 619
 620             /* Update potential sum for this i atom from the interaction with this j atom. */
 621             velecsum         = _mm_add_ps(velecsum,velec);
 622
 623             fscal            = felec;
 624
 625              /* Update vectorial force */
 626             fix3             = _mm_macc_ps(dx31,fscal,fix3);
 627             fiy3             = _mm_macc_ps(dy31,fscal,fiy3);
 628             fiz3             = _mm_macc_ps(dz31,fscal,fiz3);
 629
 630             fjx1             = _mm_macc_ps(dx31,fscal,fjx1);
 631             fjy1             = _mm_macc_ps(dy31,fscal,fjy1);
 632             fjz1             = _mm_macc_ps(dz31,fscal,fjz1);
 633
 634             /**************************
 635              * CALCULATE INTERACTIONS *
 636              **************************/
 637
 638             r32              = _mm_mul_ps(rsq32,rinv32);
 639
 640             /* Calculate table index by multiplying r with table scale and truncate to integer */
 641             rt               = _mm_mul_ps(r32,vftabscale);
 642             vfitab           = _mm_cvttps_epi32(rt);
 643 #ifdef __XOP__
 644             vfeps            = _mm_frcz_ps(rt);
 645 #else
 646             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 647 #endif
 648             twovfeps         = _mm_add_ps(vfeps,vfeps);
 649             vfitab           = _mm_slli_epi32(vfitab,2);
 650
 651             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 652             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 653             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 654             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 655             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 656             _MM_TRANSPOSE4_PS(Y,F,G,H);
 657             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 658             VV               = _mm_macc_ps(vfeps,Fp,Y);
 659             velec            = _mm_mul_ps(qq32,VV);
 660             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 661             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
 662
 663             /* Update potential sum for this i atom from the interaction with this j atom. */
 664             velecsum         = _mm_add_ps(velecsum,velec);
 665
 666             fscal            = felec;
 667
 668              /* Update vectorial force */
 669             fix3             = _mm_macc_ps(dx32,fscal,fix3);
 670             fiy3             = _mm_macc_ps(dy32,fscal,fiy3);
 671             fiz3             = _mm_macc_ps(dz32,fscal,fiz3);
 672
 673             fjx2             = _mm_macc_ps(dx32,fscal,fjx2);
 674             fjy2             = _mm_macc_ps(dy32,fscal,fjy2);
 675             fjz2             = _mm_macc_ps(dz32,fscal,fjz2);
 676
 677             /**************************
 678              * CALCULATE INTERACTIONS *
 679              **************************/
 680
 681             r33              = _mm_mul_ps(rsq33,rinv33);
 682
 683             /* Calculate table index by multiplying r with table scale and truncate to integer */
 684             rt               = _mm_mul_ps(r33,vftabscale);
 685             vfitab           = _mm_cvttps_epi32(rt);
 686 #ifdef __XOP__
 687             vfeps            = _mm_frcz_ps(rt);
 688 #else
 689             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 690 #endif
 691             twovfeps         = _mm_add_ps(vfeps,vfeps);
 692             vfitab           = _mm_slli_epi32(vfitab,2);
 693
 694             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 695             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 696             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 697             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 698             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 699             _MM_TRANSPOSE4_PS(Y,F,G,H);
 700             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 701             VV               = _mm_macc_ps(vfeps,Fp,Y);
 702             velec            = _mm_mul_ps(qq33,VV);
 703             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 704             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
 705
 706             /* Update potential sum for this i atom from the interaction with this j atom. */
 707             velecsum         = _mm_add_ps(velecsum,velec);
 708
 709             fscal            = felec;
 710
 711              /* Update vectorial force */
 712             fix3             = _mm_macc_ps(dx33,fscal,fix3);
 713             fiy3             = _mm_macc_ps(dy33,fscal,fiy3);
 714             fiz3             = _mm_macc_ps(dz33,fscal,fiz3);
 715
 716             fjx3             = _mm_macc_ps(dx33,fscal,fjx3);
 717             fjy3             = _mm_macc_ps(dy33,fscal,fjy3);
 718             fjz3             = _mm_macc_ps(dz33,fscal,fjz3);
 719
 720             fjptrA             = f+j_coord_offsetA;
 721             fjptrB             = f+j_coord_offsetB;
 722             fjptrC             = f+j_coord_offsetC;
 723             fjptrD             = f+j_coord_offsetD;
 724
 725             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
 726                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
 727                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
 728
 729             /* Inner loop uses 452 flops */
 730         }
 731
 732         if(jidx<j_index_end)
 733         {
 734
 735             /* Get j neighbor index, and coordinate index */
 736             jnrlistA         = jjnr[jidx];
 737             jnrlistB         = jjnr[jidx+1];
 738             jnrlistC         = jjnr[jidx+2];
 739             jnrlistD         = jjnr[jidx+3];
 740             /* Sign of each element will be negative for non-real atoms.
 741              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 742              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 743              */
 744             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
 745             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 746             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 747             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 748             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 749             j_coord_offsetA  = DIM*jnrA;
 750             j_coord_offsetB  = DIM*jnrB;
 751             j_coord_offsetC  = DIM*jnrC;
 752             j_coord_offsetD  = DIM*jnrD;
 753
 754             /* load j atom coordinates */
 755             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 756                                               x+j_coord_offsetC,x+j_coord_offsetD,
 757                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
 758                                               &jy2,&jz2,&jx3,&jy3,&jz3);
 759
 760             /* Calculate displacement vector */
 761             dx00             = _mm_sub_ps(ix0,jx0);
 762             dy00             = _mm_sub_ps(iy0,jy0);
 763             dz00             = _mm_sub_ps(iz0,jz0);
 764             dx11             = _mm_sub_ps(ix1,jx1);
 765             dy11             = _mm_sub_ps(iy1,jy1);
 766             dz11             = _mm_sub_ps(iz1,jz1);
 767             dx12             = _mm_sub_ps(ix1,jx2);
 768             dy12             = _mm_sub_ps(iy1,jy2);
 769             dz12             = _mm_sub_ps(iz1,jz2);
 770             dx13             = _mm_sub_ps(ix1,jx3);
 771             dy13             = _mm_sub_ps(iy1,jy3);
 772             dz13             = _mm_sub_ps(iz1,jz3);
 773             dx21             = _mm_sub_ps(ix2,jx1);
 774             dy21             = _mm_sub_ps(iy2,jy1);
 775             dz21             = _mm_sub_ps(iz2,jz1);
 776             dx22             = _mm_sub_ps(ix2,jx2);
 777             dy22             = _mm_sub_ps(iy2,jy2);
 778             dz22             = _mm_sub_ps(iz2,jz2);
 779             dx23             = _mm_sub_ps(ix2,jx3);
 780             dy23             = _mm_sub_ps(iy2,jy3);
 781             dz23             = _mm_sub_ps(iz2,jz3);
 782             dx31             = _mm_sub_ps(ix3,jx1);
 783             dy31             = _mm_sub_ps(iy3,jy1);
 784             dz31             = _mm_sub_ps(iz3,jz1);
 785             dx32             = _mm_sub_ps(ix3,jx2);
 786             dy32             = _mm_sub_ps(iy3,jy2);
 787             dz32             = _mm_sub_ps(iz3,jz2);
 788             dx33             = _mm_sub_ps(ix3,jx3);
 789             dy33             = _mm_sub_ps(iy3,jy3);
 790             dz33             = _mm_sub_ps(iz3,jz3);
 791
 792             /* Calculate squared distance and things based on it */
 793             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
 794             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
 795             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
 796             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
 797             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
 798             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
 799             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
 800             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
 801             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
 802             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
 803
 804             rinv11           = gmx_mm_invsqrt_ps(rsq11);
 805             rinv12           = gmx_mm_invsqrt_ps(rsq12);
 806             rinv13           = gmx_mm_invsqrt_ps(rsq13);
 807             rinv21           = gmx_mm_invsqrt_ps(rsq21);
 808             rinv22           = gmx_mm_invsqrt_ps(rsq22);
 809             rinv23           = gmx_mm_invsqrt_ps(rsq23);
 810             rinv31           = gmx_mm_invsqrt_ps(rsq31);
 811             rinv32           = gmx_mm_invsqrt_ps(rsq32);
 812             rinv33           = gmx_mm_invsqrt_ps(rsq33);
 813
 814             rinvsq00         = gmx_mm_inv_ps(rsq00);
 815
 816             fjx0             = _mm_setzero_ps();
 817             fjy0             = _mm_setzero_ps();
 818             fjz0             = _mm_setzero_ps();
 819             fjx1             = _mm_setzero_ps();
 820             fjy1             = _mm_setzero_ps();
 821             fjz1             = _mm_setzero_ps();
 822             fjx2             = _mm_setzero_ps();
 823             fjy2             = _mm_setzero_ps();
 824             fjz2             = _mm_setzero_ps();
 825             fjx3             = _mm_setzero_ps();
 826             fjy3             = _mm_setzero_ps();
 827             fjz3             = _mm_setzero_ps();
 828
 829             /**************************
 830              * CALCULATE INTERACTIONS *
 831              **************************/
 832
 833             /* LENNARD-JONES DISPERSION/REPULSION */
 834
 835             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 836             vvdw6            = _mm_mul_ps(c6_00,rinvsix);
 837             vvdw12           = _mm_mul_ps(c12_00,_mm_mul_ps(rinvsix,rinvsix));
 838             vvdw             = _mm_msub_ps(vvdw12,one_twelfth,_mm_mul_ps(vvdw6,one_sixth));
 839             fvdw             = _mm_mul_ps(_mm_sub_ps(vvdw12,vvdw6),rinvsq00);
 840
 841             /* Update potential sum for this i atom from the interaction with this j atom. */
 842             vvdw             = _mm_andnot_ps(dummy_mask,vvdw);
 843             vvdwsum          = _mm_add_ps(vvdwsum,vvdw);
 844
 845             fscal            = fvdw;
 846
 847             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 848
 849              /* Update vectorial force */
 850             fix0             = _mm_macc_ps(dx00,fscal,fix0);
 851             fiy0             = _mm_macc_ps(dy00,fscal,fiy0);
 852             fiz0             = _mm_macc_ps(dz00,fscal,fiz0);
 853
 854             fjx0             = _mm_macc_ps(dx00,fscal,fjx0);
 855             fjy0             = _mm_macc_ps(dy00,fscal,fjy0);
 856             fjz0             = _mm_macc_ps(dz00,fscal,fjz0);
 857
 858             /**************************
 859              * CALCULATE INTERACTIONS *
 860              **************************/
 861
 862             r11              = _mm_mul_ps(rsq11,rinv11);
 863             r11              = _mm_andnot_ps(dummy_mask,r11);
 864
 865             /* Calculate table index by multiplying r with table scale and truncate to integer */
 866             rt               = _mm_mul_ps(r11,vftabscale);
 867             vfitab           = _mm_cvttps_epi32(rt);
 868 #ifdef __XOP__
 869             vfeps            = _mm_frcz_ps(rt);
 870 #else
 871             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 872 #endif
 873             twovfeps         = _mm_add_ps(vfeps,vfeps);
 874             vfitab           = _mm_slli_epi32(vfitab,2);
 875
 876             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 877             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 878             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 879             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 880             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 881             _MM_TRANSPOSE4_PS(Y,F,G,H);
 882             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 883             VV               = _mm_macc_ps(vfeps,Fp,Y);
 884             velec            = _mm_mul_ps(qq11,VV);
 885             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 886             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
 887
 888             /* Update potential sum for this i atom from the interaction with this j atom. */
 889             velec            = _mm_andnot_ps(dummy_mask,velec);
 890             velecsum         = _mm_add_ps(velecsum,velec);
 891
 892             fscal            = felec;
 893
 894             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 895
 896              /* Update vectorial force */
 897             fix1             = _mm_macc_ps(dx11,fscal,fix1);
 898             fiy1             = _mm_macc_ps(dy11,fscal,fiy1);
 899             fiz1             = _mm_macc_ps(dz11,fscal,fiz1);
 900
 901             fjx1             = _mm_macc_ps(dx11,fscal,fjx1);
 902             fjy1             = _mm_macc_ps(dy11,fscal,fjy1);
 903             fjz1             = _mm_macc_ps(dz11,fscal,fjz1);
 904
 905             /**************************
 906              * CALCULATE INTERACTIONS *
 907              **************************/
 908
 909             r12              = _mm_mul_ps(rsq12,rinv12);
 910             r12              = _mm_andnot_ps(dummy_mask,r12);
 911
 912             /* Calculate table index by multiplying r with table scale and truncate to integer */
 913             rt               = _mm_mul_ps(r12,vftabscale);
 914             vfitab           = _mm_cvttps_epi32(rt);
 915 #ifdef __XOP__
 916             vfeps            = _mm_frcz_ps(rt);
 917 #else
 918             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 919 #endif
 920             twovfeps         = _mm_add_ps(vfeps,vfeps);
 921             vfitab           = _mm_slli_epi32(vfitab,2);
 922
 923             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 924             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 925             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 926             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 927             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 928             _MM_TRANSPOSE4_PS(Y,F,G,H);
 929             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 930             VV               = _mm_macc_ps(vfeps,Fp,Y);
 931             velec            = _mm_mul_ps(qq12,VV);
 932             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 933             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
 934
 935             /* Update potential sum for this i atom from the interaction with this j atom. */
 936             velec            = _mm_andnot_ps(dummy_mask,velec);
 937             velecsum         = _mm_add_ps(velecsum,velec);
 938
 939             fscal            = felec;
 940
 941             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 942
 943              /* Update vectorial force */
 944             fix1             = _mm_macc_ps(dx12,fscal,fix1);
 945             fiy1             = _mm_macc_ps(dy12,fscal,fiy1);
 946             fiz1             = _mm_macc_ps(dz12,fscal,fiz1);
 947
 948             fjx2             = _mm_macc_ps(dx12,fscal,fjx2);
 949             fjy2             = _mm_macc_ps(dy12,fscal,fjy2);
 950             fjz2             = _mm_macc_ps(dz12,fscal,fjz2);
 951
 952             /**************************
 953              * CALCULATE INTERACTIONS *
 954              **************************/
 955
 956             r13              = _mm_mul_ps(rsq13,rinv13);
 957             r13              = _mm_andnot_ps(dummy_mask,r13);
 958
 959             /* Calculate table index by multiplying r with table scale and truncate to integer */
 960             rt               = _mm_mul_ps(r13,vftabscale);
 961             vfitab           = _mm_cvttps_epi32(rt);
 962 #ifdef __XOP__
 963             vfeps            = _mm_frcz_ps(rt);
 964 #else
 965             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
 966 #endif
 967             twovfeps         = _mm_add_ps(vfeps,vfeps);
 968             vfitab           = _mm_slli_epi32(vfitab,2);
 969
 970             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 971             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
 972             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
 973             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
 974             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
 975             _MM_TRANSPOSE4_PS(Y,F,G,H);
 976             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
 977             VV               = _mm_macc_ps(vfeps,Fp,Y);
 978             velec            = _mm_mul_ps(qq13,VV);
 979             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
 980             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
 981
 982             /* Update potential sum for this i atom from the interaction with this j atom. */
 983             velec            = _mm_andnot_ps(dummy_mask,velec);
 984             velecsum         = _mm_add_ps(velecsum,velec);
 985
 986             fscal            = felec;
 987
 988             fscal            = _mm_andnot_ps(dummy_mask,fscal);
 989
 990              /* Update vectorial force */
 991             fix1             = _mm_macc_ps(dx13,fscal,fix1);
 992             fiy1             = _mm_macc_ps(dy13,fscal,fiy1);
 993             fiz1             = _mm_macc_ps(dz13,fscal,fiz1);
 994
 995             fjx3             = _mm_macc_ps(dx13,fscal,fjx3);
 996             fjy3             = _mm_macc_ps(dy13,fscal,fjy3);
 997             fjz3             = _mm_macc_ps(dz13,fscal,fjz3);
 998
 999             /**************************
1000              * CALCULATE INTERACTIONS *
1001              **************************/
1002
1003             r21              = _mm_mul_ps(rsq21,rinv21);
1004             r21              = _mm_andnot_ps(dummy_mask,r21);
1005
1006             /* Calculate table index by multiplying r with table scale and truncate to integer */
1007             rt               = _mm_mul_ps(r21,vftabscale);
1008             vfitab           = _mm_cvttps_epi32(rt);
1009 #ifdef __XOP__
1010             vfeps            = _mm_frcz_ps(rt);
1011 #else
1012             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1013 #endif
1014             twovfeps         = _mm_add_ps(vfeps,vfeps);
1015             vfitab           = _mm_slli_epi32(vfitab,2);
1016
1017             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1018             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1019             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1020             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1021             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1022             _MM_TRANSPOSE4_PS(Y,F,G,H);
1023             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1024             VV               = _mm_macc_ps(vfeps,Fp,Y);
1025             velec            = _mm_mul_ps(qq21,VV);
1026             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1027             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
1028
1029             /* Update potential sum for this i atom from the interaction with this j atom. */
1030             velec            = _mm_andnot_ps(dummy_mask,velec);
1031             velecsum         = _mm_add_ps(velecsum,velec);
1032
1033             fscal            = felec;
1034
1035             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1036
1037              /* Update vectorial force */
1038             fix2             = _mm_macc_ps(dx21,fscal,fix2);
1039             fiy2             = _mm_macc_ps(dy21,fscal,fiy2);
1040             fiz2             = _mm_macc_ps(dz21,fscal,fiz2);
1041
1042             fjx1             = _mm_macc_ps(dx21,fscal,fjx1);
1043             fjy1             = _mm_macc_ps(dy21,fscal,fjy1);
1044             fjz1             = _mm_macc_ps(dz21,fscal,fjz1);
1045
1046             /**************************
1047              * CALCULATE INTERACTIONS *
1048              **************************/
1049
1050             r22              = _mm_mul_ps(rsq22,rinv22);
1051             r22              = _mm_andnot_ps(dummy_mask,r22);
1052
1053             /* Calculate table index by multiplying r with table scale and truncate to integer */
1054             rt               = _mm_mul_ps(r22,vftabscale);
1055             vfitab           = _mm_cvttps_epi32(rt);
1056 #ifdef __XOP__
1057             vfeps            = _mm_frcz_ps(rt);
1058 #else
1059             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1060 #endif
1061             twovfeps         = _mm_add_ps(vfeps,vfeps);
1062             vfitab           = _mm_slli_epi32(vfitab,2);
1063
1064             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1065             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1066             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1067             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1068             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1069             _MM_TRANSPOSE4_PS(Y,F,G,H);
1070             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1071             VV               = _mm_macc_ps(vfeps,Fp,Y);
1072             velec            = _mm_mul_ps(qq22,VV);
1073             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1074             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
1075
1076             /* Update potential sum for this i atom from the interaction with this j atom. */
1077             velec            = _mm_andnot_ps(dummy_mask,velec);
1078             velecsum         = _mm_add_ps(velecsum,velec);
1079
1080             fscal            = felec;
1081
1082             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1083
1084              /* Update vectorial force */
1085             fix2             = _mm_macc_ps(dx22,fscal,fix2);
1086             fiy2             = _mm_macc_ps(dy22,fscal,fiy2);
1087             fiz2             = _mm_macc_ps(dz22,fscal,fiz2);
1088
1089             fjx2             = _mm_macc_ps(dx22,fscal,fjx2);
1090             fjy2             = _mm_macc_ps(dy22,fscal,fjy2);
1091             fjz2             = _mm_macc_ps(dz22,fscal,fjz2);
1092
1093             /**************************
1094              * CALCULATE INTERACTIONS *
1095              **************************/
1096
1097             r23              = _mm_mul_ps(rsq23,rinv23);
1098             r23              = _mm_andnot_ps(dummy_mask,r23);
1099
1100             /* Calculate table index by multiplying r with table scale and truncate to integer */
1101             rt               = _mm_mul_ps(r23,vftabscale);
1102             vfitab           = _mm_cvttps_epi32(rt);
1103 #ifdef __XOP__
1104             vfeps            = _mm_frcz_ps(rt);
1105 #else
1106             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1107 #endif
1108             twovfeps         = _mm_add_ps(vfeps,vfeps);
1109             vfitab           = _mm_slli_epi32(vfitab,2);
1110
1111             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1112             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1113             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1114             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1115             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1116             _MM_TRANSPOSE4_PS(Y,F,G,H);
1117             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1118             VV               = _mm_macc_ps(vfeps,Fp,Y);
1119             velec            = _mm_mul_ps(qq23,VV);
1120             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1121             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1122
1123             /* Update potential sum for this i atom from the interaction with this j atom. */
1124             velec            = _mm_andnot_ps(dummy_mask,velec);
1125             velecsum         = _mm_add_ps(velecsum,velec);
1126
1127             fscal            = felec;
1128
1129             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1130
1131              /* Update vectorial force */
1132             fix2             = _mm_macc_ps(dx23,fscal,fix2);
1133             fiy2             = _mm_macc_ps(dy23,fscal,fiy2);
1134             fiz2             = _mm_macc_ps(dz23,fscal,fiz2);
1135
1136             fjx3             = _mm_macc_ps(dx23,fscal,fjx3);
1137             fjy3             = _mm_macc_ps(dy23,fscal,fjy3);
1138             fjz3             = _mm_macc_ps(dz23,fscal,fjz3);
1139
1140             /**************************
1141              * CALCULATE INTERACTIONS *
1142              **************************/
1143
1144             r31              = _mm_mul_ps(rsq31,rinv31);
1145             r31              = _mm_andnot_ps(dummy_mask,r31);
1146
1147             /* Calculate table index by multiplying r with table scale and truncate to integer */
1148             rt               = _mm_mul_ps(r31,vftabscale);
1149             vfitab           = _mm_cvttps_epi32(rt);
1150 #ifdef __XOP__
1151             vfeps            = _mm_frcz_ps(rt);
1152 #else
1153             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1154 #endif
1155             twovfeps         = _mm_add_ps(vfeps,vfeps);
1156             vfitab           = _mm_slli_epi32(vfitab,2);
1157
1158             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1159             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1160             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1161             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1162             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1163             _MM_TRANSPOSE4_PS(Y,F,G,H);
1164             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1165             VV               = _mm_macc_ps(vfeps,Fp,Y);
1166             velec            = _mm_mul_ps(qq31,VV);
1167             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1168             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1169
1170             /* Update potential sum for this i atom from the interaction with this j atom. */
1171             velec            = _mm_andnot_ps(dummy_mask,velec);
1172             velecsum         = _mm_add_ps(velecsum,velec);
1173
1174             fscal            = felec;
1175
1176             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1177
1178              /* Update vectorial force */
1179             fix3             = _mm_macc_ps(dx31,fscal,fix3);
1180             fiy3             = _mm_macc_ps(dy31,fscal,fiy3);
1181             fiz3             = _mm_macc_ps(dz31,fscal,fiz3);
1182
1183             fjx1             = _mm_macc_ps(dx31,fscal,fjx1);
1184             fjy1             = _mm_macc_ps(dy31,fscal,fjy1);
1185             fjz1             = _mm_macc_ps(dz31,fscal,fjz1);
1186
1187             /**************************
1188              * CALCULATE INTERACTIONS *
1189              **************************/
1190
1191             r32              = _mm_mul_ps(rsq32,rinv32);
1192             r32              = _mm_andnot_ps(dummy_mask,r32);
1193
1194             /* Calculate table index by multiplying r with table scale and truncate to integer */
1195             rt               = _mm_mul_ps(r32,vftabscale);
1196             vfitab           = _mm_cvttps_epi32(rt);
1197 #ifdef __XOP__
1198             vfeps            = _mm_frcz_ps(rt);
1199 #else
1200             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1201 #endif
1202             twovfeps         = _mm_add_ps(vfeps,vfeps);
1203             vfitab           = _mm_slli_epi32(vfitab,2);
1204
1205             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1206             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1207             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1208             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1209             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1210             _MM_TRANSPOSE4_PS(Y,F,G,H);
1211             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1212             VV               = _mm_macc_ps(vfeps,Fp,Y);
1213             velec            = _mm_mul_ps(qq32,VV);
1214             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1215             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1216
1217             /* Update potential sum for this i atom from the interaction with this j atom. */
1218             velec            = _mm_andnot_ps(dummy_mask,velec);
1219             velecsum         = _mm_add_ps(velecsum,velec);
1220
1221             fscal            = felec;
1222
1223             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1224
1225              /* Update vectorial force */
1226             fix3             = _mm_macc_ps(dx32,fscal,fix3);
1227             fiy3             = _mm_macc_ps(dy32,fscal,fiy3);
1228             fiz3             = _mm_macc_ps(dz32,fscal,fiz3);
1229
1230             fjx2             = _mm_macc_ps(dx32,fscal,fjx2);
1231             fjy2             = _mm_macc_ps(dy32,fscal,fjy2);
1232             fjz2             = _mm_macc_ps(dz32,fscal,fjz2);
1233
1234             /**************************
1235              * CALCULATE INTERACTIONS *
1236              **************************/
1237
1238             r33              = _mm_mul_ps(rsq33,rinv33);
1239             r33              = _mm_andnot_ps(dummy_mask,r33);
1240
1241             /* Calculate table index by multiplying r with table scale and truncate to integer */
1242             rt               = _mm_mul_ps(r33,vftabscale);
1243             vfitab           = _mm_cvttps_epi32(rt);
1244 #ifdef __XOP__
1245             vfeps            = _mm_frcz_ps(rt);
1246 #else
1247             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1248 #endif
1249             twovfeps         = _mm_add_ps(vfeps,vfeps);
1250             vfitab           = _mm_slli_epi32(vfitab,2);
1251
1252             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1253             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1254             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1255             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1256             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1257             _MM_TRANSPOSE4_PS(Y,F,G,H);
1258             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1259             VV               = _mm_macc_ps(vfeps,Fp,Y);
1260             velec            = _mm_mul_ps(qq33,VV);
1261             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1262             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1263
1264             /* Update potential sum for this i atom from the interaction with this j atom. */
1265             velec            = _mm_andnot_ps(dummy_mask,velec);
1266             velecsum         = _mm_add_ps(velecsum,velec);
1267
1268             fscal            = felec;
1269
1270             fscal            = _mm_andnot_ps(dummy_mask,fscal);
1271
1272              /* Update vectorial force */
1273             fix3             = _mm_macc_ps(dx33,fscal,fix3);
1274             fiy3             = _mm_macc_ps(dy33,fscal,fiy3);
1275             fiz3             = _mm_macc_ps(dz33,fscal,fiz3);
1276
1277             fjx3             = _mm_macc_ps(dx33,fscal,fjx3);
1278             fjy3             = _mm_macc_ps(dy33,fscal,fjy3);
1279             fjz3             = _mm_macc_ps(dz33,fscal,fjz3);
1280
1281             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
1282             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
1283             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
1284             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
1285
1286             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
1287                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1288                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1289
1290             /* Inner loop uses 461 flops */
1291         }
1292
1293         /* End of innermost loop */
1294
1295         gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1296                                               f+i_coord_offset,fshift+i_shift_offset);
1297
1298         ggid                        = gid[iidx];
1299         /* Update potential energies */
1300         gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
1301         gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
1302
1303         /* Increment number of inner iterations */
1304         inneriter                  += j_index_end - j_index_start;
1305
1306         /* Outer loop uses 26 flops */
1307     }
1308
1309     /* Increment number of outer iterations */
1310     outeriter        += nri;
1311
1312     /* Update outer/inner flops */
1313
1314     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_VF,outeriter*26 + inneriter*461);
1315 }
1316 /*
1317  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwLJ_GeomW4W4_F_avx_128_fma_single
1318  * Electrostatics interaction: CubicSplineTable
1319  * VdW interaction:            LennardJones
1320  * Geometry:                   Water4-Water4
1321  * Calculate force/pot:        Force
1322  */
1323 void
1324 nb_kernel_ElecCSTab_VdwLJ_GeomW4W4_F_avx_128_fma_single
1325                     (t_nblist                    * gmx_restrict       nlist,
1326                      rvec                        * gmx_restrict          xx,
1327                      rvec                        * gmx_restrict          ff,
1328                      t_forcerec                  * gmx_restrict          fr,
1329                      t_mdatoms                   * gmx_restrict     mdatoms,
1330                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
1331                      t_nrnb                      * gmx_restrict        nrnb)
1332 {
1333     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
1334      * just 0 for non-waters.
1335      * Suffixes A,B,C,D refer to j loop unrolling done with AVX_128, e.g. for the four different
1336      * jnr indices corresponding to data put in the four positions in the SIMD register.
1337      */
1338     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
1339     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
1340     int              jnrA,jnrB,jnrC,jnrD;
1341     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
1342     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
1343     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
1344     real             rcutoff_scalar;
1345     real             *shiftvec,*fshift,*x,*f;
1346     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD;
1347     real             scratch[4*DIM];
1348     __m128           fscal,rcutoff,rcutoff2,jidxall;
1349     int              vdwioffset0;
1350     __m128           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
1351     int              vdwioffset1;
1352     __m128           ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
1353     int              vdwioffset2;
1354     __m128           ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
1355     int              vdwioffset3;
1356     __m128           ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
1357     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D;
1358     __m128           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
1359     int              vdwjidx1A,vdwjidx1B,vdwjidx1C,vdwjidx1D;
1360     __m128           jx1,jy1,jz1,fjx1,fjy1,fjz1,jq1,isaj1;
1361     int              vdwjidx2A,vdwjidx2B,vdwjidx2C,vdwjidx2D;
1362     __m128           jx2,jy2,jz2,fjx2,fjy2,fjz2,jq2,isaj2;
1363     int              vdwjidx3A,vdwjidx3B,vdwjidx3C,vdwjidx3D;
1364     __m128           jx3,jy3,jz3,fjx3,fjy3,fjz3,jq3,isaj3;
1365     __m128           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
1366     __m128           dx11,dy11,dz11,rsq11,rinv11,rinvsq11,r11,qq11,c6_11,c12_11;
1367     __m128           dx12,dy12,dz12,rsq12,rinv12,rinvsq12,r12,qq12,c6_12,c12_12;
1368     __m128           dx13,dy13,dz13,rsq13,rinv13,rinvsq13,r13,qq13,c6_13,c12_13;
1369     __m128           dx21,dy21,dz21,rsq21,rinv21,rinvsq21,r21,qq21,c6_21,c12_21;
1370     __m128           dx22,dy22,dz22,rsq22,rinv22,rinvsq22,r22,qq22,c6_22,c12_22;
1371     __m128           dx23,dy23,dz23,rsq23,rinv23,rinvsq23,r23,qq23,c6_23,c12_23;
1372     __m128           dx31,dy31,dz31,rsq31,rinv31,rinvsq31,r31,qq31,c6_31,c12_31;
1373     __m128           dx32,dy32,dz32,rsq32,rinv32,rinvsq32,r32,qq32,c6_32,c12_32;
1374     __m128           dx33,dy33,dz33,rsq33,rinv33,rinvsq33,r33,qq33,c6_33,c12_33;
1375     __m128           velec,felec,velecsum,facel,crf,krf,krf2;
1376     real             *charge;
1377     int              nvdwtype;
1378     __m128           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
1379     int              *vdwtype;
1380     real             *vdwparam;
1381     __m128           one_sixth   = _mm_set1_ps(1.0/6.0);
1382     __m128           one_twelfth = _mm_set1_ps(1.0/12.0);
1383     __m128i          vfitab;
1384     __m128i          ifour       = _mm_set1_epi32(4);
1385     __m128           rt,vfeps,twovfeps,vftabscale,Y,F,G,H,Fp,VV,FF;
1386     real             *vftab;
1387     __m128           dummy_mask,cutoff_mask;
1388     __m128           signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) );
1389     __m128           one     = _mm_set1_ps(1.0);
1390     __m128           two     = _mm_set1_ps(2.0);
1391     x                = xx[0];
1392     f                = ff[0];
1393
1394     nri              = nlist->nri;
1395     iinr             = nlist->iinr;
1396     jindex           = nlist->jindex;
1397     jjnr             = nlist->jjnr;
1398     shiftidx         = nlist->shift;
1399     gid              = nlist->gid;
1400     shiftvec         = fr->shift_vec[0];
1401     fshift           = fr->fshift[0];
1402     facel            = _mm_set1_ps(fr->epsfac);
1403     charge           = mdatoms->chargeA;
1404     nvdwtype         = fr->ntype;
1405     vdwparam         = fr->nbfp;
1406     vdwtype          = mdatoms->typeA;
1407
1408     vftab            = kernel_data->table_elec->data;
1409     vftabscale       = _mm_set1_ps(kernel_data->table_elec->scale);
1410
1411     /* Setup water-specific parameters */
1412     inr              = nlist->iinr[0];
1413     iq1              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1]));
1414     iq2              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2]));
1415     iq3              = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3]));
1416     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
1417
1418     jq1              = _mm_set1_ps(charge[inr+1]);
1419     jq2              = _mm_set1_ps(charge[inr+2]);
1420     jq3              = _mm_set1_ps(charge[inr+3]);
1421     vdwjidx0A        = 2*vdwtype[inr+0];
1422     c6_00            = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A]);
1423     c12_00           = _mm_set1_ps(vdwparam[vdwioffset0+vdwjidx0A+1]);
1424     qq11             = _mm_mul_ps(iq1,jq1);
1425     qq12             = _mm_mul_ps(iq1,jq2);
1426     qq13             = _mm_mul_ps(iq1,jq3);
1427     qq21             = _mm_mul_ps(iq2,jq1);
1428     qq22             = _mm_mul_ps(iq2,jq2);
1429     qq23             = _mm_mul_ps(iq2,jq3);
1430     qq31             = _mm_mul_ps(iq3,jq1);
1431     qq32             = _mm_mul_ps(iq3,jq2);
1432     qq33             = _mm_mul_ps(iq3,jq3);
1433
1434     /* Avoid stupid compiler warnings */
1435     jnrA = jnrB = jnrC = jnrD = 0;
1436     j_coord_offsetA = 0;
1437     j_coord_offsetB = 0;
1438     j_coord_offsetC = 0;
1439     j_coord_offsetD = 0;
1440
1441     outeriter        = 0;
1442     inneriter        = 0;
1443
1444     for(iidx=0;iidx<4*DIM;iidx++)
1445     {
1446         scratch[iidx] = 0.0;
1447     }
1448
1449     /* Start outer loop over neighborlists */
1450     for(iidx=0; iidx<nri; iidx++)
1451     {
1452         /* Load shift vector for this list */
1453         i_shift_offset   = DIM*shiftidx[iidx];
1454
1455         /* Load limits for loop over neighbors */
1456         j_index_start    = jindex[iidx];
1457         j_index_end      = jindex[iidx+1];
1458
1459         /* Get outer coordinate index */
1460         inr              = iinr[iidx];
1461         i_coord_offset   = DIM*inr;
1462
1463         /* Load i particle coords and add shift vector */
1464         gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,
1465                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
1466
1467         fix0             = _mm_setzero_ps();
1468         fiy0             = _mm_setzero_ps();
1469         fiz0             = _mm_setzero_ps();
1470         fix1             = _mm_setzero_ps();
1471         fiy1             = _mm_setzero_ps();
1472         fiz1             = _mm_setzero_ps();
1473         fix2             = _mm_setzero_ps();
1474         fiy2             = _mm_setzero_ps();
1475         fiz2             = _mm_setzero_ps();
1476         fix3             = _mm_setzero_ps();
1477         fiy3             = _mm_setzero_ps();
1478         fiz3             = _mm_setzero_ps();
1479
1480         /* Start inner kernel loop */
1481         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4)
1482         {
1483
1484             /* Get j neighbor index, and coordinate index */
1485             jnrA             = jjnr[jidx];
1486             jnrB             = jjnr[jidx+1];
1487             jnrC             = jjnr[jidx+2];
1488             jnrD             = jjnr[jidx+3];
1489             j_coord_offsetA  = DIM*jnrA;
1490             j_coord_offsetB  = DIM*jnrB;
1491             j_coord_offsetC  = DIM*jnrC;
1492             j_coord_offsetD  = DIM*jnrD;
1493
1494             /* load j atom coordinates */
1495             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1496                                               x+j_coord_offsetC,x+j_coord_offsetD,
1497                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1498                                               &jy2,&jz2,&jx3,&jy3,&jz3);
1499
1500             /* Calculate displacement vector */
1501             dx00             = _mm_sub_ps(ix0,jx0);
1502             dy00             = _mm_sub_ps(iy0,jy0);
1503             dz00             = _mm_sub_ps(iz0,jz0);
1504             dx11             = _mm_sub_ps(ix1,jx1);
1505             dy11             = _mm_sub_ps(iy1,jy1);
1506             dz11             = _mm_sub_ps(iz1,jz1);
1507             dx12             = _mm_sub_ps(ix1,jx2);
1508             dy12             = _mm_sub_ps(iy1,jy2);
1509             dz12             = _mm_sub_ps(iz1,jz2);
1510             dx13             = _mm_sub_ps(ix1,jx3);
1511             dy13             = _mm_sub_ps(iy1,jy3);
1512             dz13             = _mm_sub_ps(iz1,jz3);
1513             dx21             = _mm_sub_ps(ix2,jx1);
1514             dy21             = _mm_sub_ps(iy2,jy1);
1515             dz21             = _mm_sub_ps(iz2,jz1);
1516             dx22             = _mm_sub_ps(ix2,jx2);
1517             dy22             = _mm_sub_ps(iy2,jy2);
1518             dz22             = _mm_sub_ps(iz2,jz2);
1519             dx23             = _mm_sub_ps(ix2,jx3);
1520             dy23             = _mm_sub_ps(iy2,jy3);
1521             dz23             = _mm_sub_ps(iz2,jz3);
1522             dx31             = _mm_sub_ps(ix3,jx1);
1523             dy31             = _mm_sub_ps(iy3,jy1);
1524             dz31             = _mm_sub_ps(iz3,jz1);
1525             dx32             = _mm_sub_ps(ix3,jx2);
1526             dy32             = _mm_sub_ps(iy3,jy2);
1527             dz32             = _mm_sub_ps(iz3,jz2);
1528             dx33             = _mm_sub_ps(ix3,jx3);
1529             dy33             = _mm_sub_ps(iy3,jy3);
1530             dz33             = _mm_sub_ps(iz3,jz3);
1531
1532             /* Calculate squared distance and things based on it */
1533             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
1534             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
1535             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
1536             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
1537             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
1538             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
1539             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
1540             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
1541             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
1542             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
1543
1544             rinv11           = gmx_mm_invsqrt_ps(rsq11);
1545             rinv12           = gmx_mm_invsqrt_ps(rsq12);
1546             rinv13           = gmx_mm_invsqrt_ps(rsq13);
1547             rinv21           = gmx_mm_invsqrt_ps(rsq21);
1548             rinv22           = gmx_mm_invsqrt_ps(rsq22);
1549             rinv23           = gmx_mm_invsqrt_ps(rsq23);
1550             rinv31           = gmx_mm_invsqrt_ps(rsq31);
1551             rinv32           = gmx_mm_invsqrt_ps(rsq32);
1552             rinv33           = gmx_mm_invsqrt_ps(rsq33);
1553
1554             rinvsq00         = gmx_mm_inv_ps(rsq00);
1555
1556             fjx0             = _mm_setzero_ps();
1557             fjy0             = _mm_setzero_ps();
1558             fjz0             = _mm_setzero_ps();
1559             fjx1             = _mm_setzero_ps();
1560             fjy1             = _mm_setzero_ps();
1561             fjz1             = _mm_setzero_ps();
1562             fjx2             = _mm_setzero_ps();
1563             fjy2             = _mm_setzero_ps();
1564             fjz2             = _mm_setzero_ps();
1565             fjx3             = _mm_setzero_ps();
1566             fjy3             = _mm_setzero_ps();
1567             fjz3             = _mm_setzero_ps();
1568
1569             /**************************
1570              * CALCULATE INTERACTIONS *
1571              **************************/
1572
1573             /* LENNARD-JONES DISPERSION/REPULSION */
1574
1575             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
1576             fvdw             = _mm_mul_ps(_mm_msub_ps(c12_00,rinvsix,c6_00),_mm_mul_ps(rinvsix,rinvsq00));
1577
1578             fscal            = fvdw;
1579
1580              /* Update vectorial force */
1581             fix0             = _mm_macc_ps(dx00,fscal,fix0);
1582             fiy0             = _mm_macc_ps(dy00,fscal,fiy0);
1583             fiz0             = _mm_macc_ps(dz00,fscal,fiz0);
1584
1585             fjx0             = _mm_macc_ps(dx00,fscal,fjx0);
1586             fjy0             = _mm_macc_ps(dy00,fscal,fjy0);
1587             fjz0             = _mm_macc_ps(dz00,fscal,fjz0);
1588
1589             /**************************
1590              * CALCULATE INTERACTIONS *
1591              **************************/
1592
1593             r11              = _mm_mul_ps(rsq11,rinv11);
1594
1595             /* Calculate table index by multiplying r with table scale and truncate to integer */
1596             rt               = _mm_mul_ps(r11,vftabscale);
1597             vfitab           = _mm_cvttps_epi32(rt);
1598 #ifdef __XOP__
1599             vfeps            = _mm_frcz_ps(rt);
1600 #else
1601             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1602 #endif
1603             twovfeps         = _mm_add_ps(vfeps,vfeps);
1604             vfitab           = _mm_slli_epi32(vfitab,2);
1605
1606             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1607             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1608             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1609             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1610             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1611             _MM_TRANSPOSE4_PS(Y,F,G,H);
1612             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1613             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1614             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
1615
1616             fscal            = felec;
1617
1618              /* Update vectorial force */
1619             fix1             = _mm_macc_ps(dx11,fscal,fix1);
1620             fiy1             = _mm_macc_ps(dy11,fscal,fiy1);
1621             fiz1             = _mm_macc_ps(dz11,fscal,fiz1);
1622
1623             fjx1             = _mm_macc_ps(dx11,fscal,fjx1);
1624             fjy1             = _mm_macc_ps(dy11,fscal,fjy1);
1625             fjz1             = _mm_macc_ps(dz11,fscal,fjz1);
1626
1627             /**************************
1628              * CALCULATE INTERACTIONS *
1629              **************************/
1630
1631             r12              = _mm_mul_ps(rsq12,rinv12);
1632
1633             /* Calculate table index by multiplying r with table scale and truncate to integer */
1634             rt               = _mm_mul_ps(r12,vftabscale);
1635             vfitab           = _mm_cvttps_epi32(rt);
1636 #ifdef __XOP__
1637             vfeps            = _mm_frcz_ps(rt);
1638 #else
1639             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1640 #endif
1641             twovfeps         = _mm_add_ps(vfeps,vfeps);
1642             vfitab           = _mm_slli_epi32(vfitab,2);
1643
1644             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1645             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1646             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1647             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1648             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1649             _MM_TRANSPOSE4_PS(Y,F,G,H);
1650             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1651             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1652             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
1653
1654             fscal            = felec;
1655
1656              /* Update vectorial force */
1657             fix1             = _mm_macc_ps(dx12,fscal,fix1);
1658             fiy1             = _mm_macc_ps(dy12,fscal,fiy1);
1659             fiz1             = _mm_macc_ps(dz12,fscal,fiz1);
1660
1661             fjx2             = _mm_macc_ps(dx12,fscal,fjx2);
1662             fjy2             = _mm_macc_ps(dy12,fscal,fjy2);
1663             fjz2             = _mm_macc_ps(dz12,fscal,fjz2);
1664
1665             /**************************
1666              * CALCULATE INTERACTIONS *
1667              **************************/
1668
1669             r13              = _mm_mul_ps(rsq13,rinv13);
1670
1671             /* Calculate table index by multiplying r with table scale and truncate to integer */
1672             rt               = _mm_mul_ps(r13,vftabscale);
1673             vfitab           = _mm_cvttps_epi32(rt);
1674 #ifdef __XOP__
1675             vfeps            = _mm_frcz_ps(rt);
1676 #else
1677             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1678 #endif
1679             twovfeps         = _mm_add_ps(vfeps,vfeps);
1680             vfitab           = _mm_slli_epi32(vfitab,2);
1681
1682             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1683             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1684             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1685             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1686             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1687             _MM_TRANSPOSE4_PS(Y,F,G,H);
1688             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1689             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1690             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
1691
1692             fscal            = felec;
1693
1694              /* Update vectorial force */
1695             fix1             = _mm_macc_ps(dx13,fscal,fix1);
1696             fiy1             = _mm_macc_ps(dy13,fscal,fiy1);
1697             fiz1             = _mm_macc_ps(dz13,fscal,fiz1);
1698
1699             fjx3             = _mm_macc_ps(dx13,fscal,fjx3);
1700             fjy3             = _mm_macc_ps(dy13,fscal,fjy3);
1701             fjz3             = _mm_macc_ps(dz13,fscal,fjz3);
1702
1703             /**************************
1704              * CALCULATE INTERACTIONS *
1705              **************************/
1706
1707             r21              = _mm_mul_ps(rsq21,rinv21);
1708
1709             /* Calculate table index by multiplying r with table scale and truncate to integer */
1710             rt               = _mm_mul_ps(r21,vftabscale);
1711             vfitab           = _mm_cvttps_epi32(rt);
1712 #ifdef __XOP__
1713             vfeps            = _mm_frcz_ps(rt);
1714 #else
1715             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1716 #endif
1717             twovfeps         = _mm_add_ps(vfeps,vfeps);
1718             vfitab           = _mm_slli_epi32(vfitab,2);
1719
1720             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1721             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1722             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1723             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1724             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1725             _MM_TRANSPOSE4_PS(Y,F,G,H);
1726             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1727             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1728             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
1729
1730             fscal            = felec;
1731
1732              /* Update vectorial force */
1733             fix2             = _mm_macc_ps(dx21,fscal,fix2);
1734             fiy2             = _mm_macc_ps(dy21,fscal,fiy2);
1735             fiz2             = _mm_macc_ps(dz21,fscal,fiz2);
1736
1737             fjx1             = _mm_macc_ps(dx21,fscal,fjx1);
1738             fjy1             = _mm_macc_ps(dy21,fscal,fjy1);
1739             fjz1             = _mm_macc_ps(dz21,fscal,fjz1);
1740
1741             /**************************
1742              * CALCULATE INTERACTIONS *
1743              **************************/
1744
1745             r22              = _mm_mul_ps(rsq22,rinv22);
1746
1747             /* Calculate table index by multiplying r with table scale and truncate to integer */
1748             rt               = _mm_mul_ps(r22,vftabscale);
1749             vfitab           = _mm_cvttps_epi32(rt);
1750 #ifdef __XOP__
1751             vfeps            = _mm_frcz_ps(rt);
1752 #else
1753             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1754 #endif
1755             twovfeps         = _mm_add_ps(vfeps,vfeps);
1756             vfitab           = _mm_slli_epi32(vfitab,2);
1757
1758             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1759             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1760             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1761             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1762             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1763             _MM_TRANSPOSE4_PS(Y,F,G,H);
1764             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1765             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1766             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
1767
1768             fscal            = felec;
1769
1770              /* Update vectorial force */
1771             fix2             = _mm_macc_ps(dx22,fscal,fix2);
1772             fiy2             = _mm_macc_ps(dy22,fscal,fiy2);
1773             fiz2             = _mm_macc_ps(dz22,fscal,fiz2);
1774
1775             fjx2             = _mm_macc_ps(dx22,fscal,fjx2);
1776             fjy2             = _mm_macc_ps(dy22,fscal,fjy2);
1777             fjz2             = _mm_macc_ps(dz22,fscal,fjz2);
1778
1779             /**************************
1780              * CALCULATE INTERACTIONS *
1781              **************************/
1782
1783             r23              = _mm_mul_ps(rsq23,rinv23);
1784
1785             /* Calculate table index by multiplying r with table scale and truncate to integer */
1786             rt               = _mm_mul_ps(r23,vftabscale);
1787             vfitab           = _mm_cvttps_epi32(rt);
1788 #ifdef __XOP__
1789             vfeps            = _mm_frcz_ps(rt);
1790 #else
1791             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1792 #endif
1793             twovfeps         = _mm_add_ps(vfeps,vfeps);
1794             vfitab           = _mm_slli_epi32(vfitab,2);
1795
1796             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1797             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1798             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1799             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1800             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1801             _MM_TRANSPOSE4_PS(Y,F,G,H);
1802             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1803             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1804             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
1805
1806             fscal            = felec;
1807
1808              /* Update vectorial force */
1809             fix2             = _mm_macc_ps(dx23,fscal,fix2);
1810             fiy2             = _mm_macc_ps(dy23,fscal,fiy2);
1811             fiz2             = _mm_macc_ps(dz23,fscal,fiz2);
1812
1813             fjx3             = _mm_macc_ps(dx23,fscal,fjx3);
1814             fjy3             = _mm_macc_ps(dy23,fscal,fjy3);
1815             fjz3             = _mm_macc_ps(dz23,fscal,fjz3);
1816
1817             /**************************
1818              * CALCULATE INTERACTIONS *
1819              **************************/
1820
1821             r31              = _mm_mul_ps(rsq31,rinv31);
1822
1823             /* Calculate table index by multiplying r with table scale and truncate to integer */
1824             rt               = _mm_mul_ps(r31,vftabscale);
1825             vfitab           = _mm_cvttps_epi32(rt);
1826 #ifdef __XOP__
1827             vfeps            = _mm_frcz_ps(rt);
1828 #else
1829             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1830 #endif
1831             twovfeps         = _mm_add_ps(vfeps,vfeps);
1832             vfitab           = _mm_slli_epi32(vfitab,2);
1833
1834             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1835             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1836             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1837             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1838             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1839             _MM_TRANSPOSE4_PS(Y,F,G,H);
1840             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1841             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1842             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
1843
1844             fscal            = felec;
1845
1846              /* Update vectorial force */
1847             fix3             = _mm_macc_ps(dx31,fscal,fix3);
1848             fiy3             = _mm_macc_ps(dy31,fscal,fiy3);
1849             fiz3             = _mm_macc_ps(dz31,fscal,fiz3);
1850
1851             fjx1             = _mm_macc_ps(dx31,fscal,fjx1);
1852             fjy1             = _mm_macc_ps(dy31,fscal,fjy1);
1853             fjz1             = _mm_macc_ps(dz31,fscal,fjz1);
1854
1855             /**************************
1856              * CALCULATE INTERACTIONS *
1857              **************************/
1858
1859             r32              = _mm_mul_ps(rsq32,rinv32);
1860
1861             /* Calculate table index by multiplying r with table scale and truncate to integer */
1862             rt               = _mm_mul_ps(r32,vftabscale);
1863             vfitab           = _mm_cvttps_epi32(rt);
1864 #ifdef __XOP__
1865             vfeps            = _mm_frcz_ps(rt);
1866 #else
1867             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1868 #endif
1869             twovfeps         = _mm_add_ps(vfeps,vfeps);
1870             vfitab           = _mm_slli_epi32(vfitab,2);
1871
1872             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1873             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1874             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1875             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1876             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1877             _MM_TRANSPOSE4_PS(Y,F,G,H);
1878             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1879             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1880             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
1881
1882             fscal            = felec;
1883
1884              /* Update vectorial force */
1885             fix3             = _mm_macc_ps(dx32,fscal,fix3);
1886             fiy3             = _mm_macc_ps(dy32,fscal,fiy3);
1887             fiz3             = _mm_macc_ps(dz32,fscal,fiz3);
1888
1889             fjx2             = _mm_macc_ps(dx32,fscal,fjx2);
1890             fjy2             = _mm_macc_ps(dy32,fscal,fjy2);
1891             fjz2             = _mm_macc_ps(dz32,fscal,fjz2);
1892
1893             /**************************
1894              * CALCULATE INTERACTIONS *
1895              **************************/
1896
1897             r33              = _mm_mul_ps(rsq33,rinv33);
1898
1899             /* Calculate table index by multiplying r with table scale and truncate to integer */
1900             rt               = _mm_mul_ps(r33,vftabscale);
1901             vfitab           = _mm_cvttps_epi32(rt);
1902 #ifdef __XOP__
1903             vfeps            = _mm_frcz_ps(rt);
1904 #else
1905             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
1906 #endif
1907             twovfeps         = _mm_add_ps(vfeps,vfeps);
1908             vfitab           = _mm_slli_epi32(vfitab,2);
1909
1910             /* CUBIC SPLINE TABLE ELECTROSTATICS */
1911             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
1912             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
1913             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
1914             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
1915             _MM_TRANSPOSE4_PS(Y,F,G,H);
1916             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
1917             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
1918             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
1919
1920             fscal            = felec;
1921
1922              /* Update vectorial force */
1923             fix3             = _mm_macc_ps(dx33,fscal,fix3);
1924             fiy3             = _mm_macc_ps(dy33,fscal,fiy3);
1925             fiz3             = _mm_macc_ps(dz33,fscal,fiz3);
1926
1927             fjx3             = _mm_macc_ps(dx33,fscal,fjx3);
1928             fjy3             = _mm_macc_ps(dy33,fscal,fjy3);
1929             fjz3             = _mm_macc_ps(dz33,fscal,fjz3);
1930
1931             fjptrA             = f+j_coord_offsetA;
1932             fjptrB             = f+j_coord_offsetB;
1933             fjptrC             = f+j_coord_offsetC;
1934             fjptrD             = f+j_coord_offsetD;
1935
1936             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
1937                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
1938                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
1939
1940             /* Inner loop uses 411 flops */
1941         }
1942
1943         if(jidx<j_index_end)
1944         {
1945
1946             /* Get j neighbor index, and coordinate index */
1947             jnrlistA         = jjnr[jidx];
1948             jnrlistB         = jjnr[jidx+1];
1949             jnrlistC         = jjnr[jidx+2];
1950             jnrlistD         = jjnr[jidx+3];
1951             /* Sign of each element will be negative for non-real atoms.
1952              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
1953              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
1954              */
1955             dummy_mask = gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128()));
1956             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
1957             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
1958             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
1959             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
1960             j_coord_offsetA  = DIM*jnrA;
1961             j_coord_offsetB  = DIM*jnrB;
1962             j_coord_offsetC  = DIM*jnrC;
1963             j_coord_offsetD  = DIM*jnrD;
1964
1965             /* load j atom coordinates */
1966             gmx_mm_load_4rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
1967                                               x+j_coord_offsetC,x+j_coord_offsetD,
1968                                               &jx0,&jy0,&jz0,&jx1,&jy1,&jz1,&jx2,
1969                                               &jy2,&jz2,&jx3,&jy3,&jz3);
1970
1971             /* Calculate displacement vector */
1972             dx00             = _mm_sub_ps(ix0,jx0);
1973             dy00             = _mm_sub_ps(iy0,jy0);
1974             dz00             = _mm_sub_ps(iz0,jz0);
1975             dx11             = _mm_sub_ps(ix1,jx1);
1976             dy11             = _mm_sub_ps(iy1,jy1);
1977             dz11             = _mm_sub_ps(iz1,jz1);
1978             dx12             = _mm_sub_ps(ix1,jx2);
1979             dy12             = _mm_sub_ps(iy1,jy2);
1980             dz12             = _mm_sub_ps(iz1,jz2);
1981             dx13             = _mm_sub_ps(ix1,jx3);
1982             dy13             = _mm_sub_ps(iy1,jy3);
1983             dz13             = _mm_sub_ps(iz1,jz3);
1984             dx21             = _mm_sub_ps(ix2,jx1);
1985             dy21             = _mm_sub_ps(iy2,jy1);
1986             dz21             = _mm_sub_ps(iz2,jz1);
1987             dx22             = _mm_sub_ps(ix2,jx2);
1988             dy22             = _mm_sub_ps(iy2,jy2);
1989             dz22             = _mm_sub_ps(iz2,jz2);
1990             dx23             = _mm_sub_ps(ix2,jx3);
1991             dy23             = _mm_sub_ps(iy2,jy3);
1992             dz23             = _mm_sub_ps(iz2,jz3);
1993             dx31             = _mm_sub_ps(ix3,jx1);
1994             dy31             = _mm_sub_ps(iy3,jy1);
1995             dz31             = _mm_sub_ps(iz3,jz1);
1996             dx32             = _mm_sub_ps(ix3,jx2);
1997             dy32             = _mm_sub_ps(iy3,jy2);
1998             dz32             = _mm_sub_ps(iz3,jz2);
1999             dx33             = _mm_sub_ps(ix3,jx3);
2000             dy33             = _mm_sub_ps(iy3,jy3);
2001             dz33             = _mm_sub_ps(iz3,jz3);
2002
2003             /* Calculate squared distance and things based on it */
2004             rsq00            = gmx_mm_calc_rsq_ps(dx00,dy00,dz00);
2005             rsq11            = gmx_mm_calc_rsq_ps(dx11,dy11,dz11);
2006             rsq12            = gmx_mm_calc_rsq_ps(dx12,dy12,dz12);
2007             rsq13            = gmx_mm_calc_rsq_ps(dx13,dy13,dz13);
2008             rsq21            = gmx_mm_calc_rsq_ps(dx21,dy21,dz21);
2009             rsq22            = gmx_mm_calc_rsq_ps(dx22,dy22,dz22);
2010             rsq23            = gmx_mm_calc_rsq_ps(dx23,dy23,dz23);
2011             rsq31            = gmx_mm_calc_rsq_ps(dx31,dy31,dz31);
2012             rsq32            = gmx_mm_calc_rsq_ps(dx32,dy32,dz32);
2013             rsq33            = gmx_mm_calc_rsq_ps(dx33,dy33,dz33);
2014
2015             rinv11           = gmx_mm_invsqrt_ps(rsq11);
2016             rinv12           = gmx_mm_invsqrt_ps(rsq12);
2017             rinv13           = gmx_mm_invsqrt_ps(rsq13);
2018             rinv21           = gmx_mm_invsqrt_ps(rsq21);
2019             rinv22           = gmx_mm_invsqrt_ps(rsq22);
2020             rinv23           = gmx_mm_invsqrt_ps(rsq23);
2021             rinv31           = gmx_mm_invsqrt_ps(rsq31);
2022             rinv32           = gmx_mm_invsqrt_ps(rsq32);
2023             rinv33           = gmx_mm_invsqrt_ps(rsq33);
2024
2025             rinvsq00         = gmx_mm_inv_ps(rsq00);
2026
2027             fjx0             = _mm_setzero_ps();
2028             fjy0             = _mm_setzero_ps();
2029             fjz0             = _mm_setzero_ps();
2030             fjx1             = _mm_setzero_ps();
2031             fjy1             = _mm_setzero_ps();
2032             fjz1             = _mm_setzero_ps();
2033             fjx2             = _mm_setzero_ps();
2034             fjy2             = _mm_setzero_ps();
2035             fjz2             = _mm_setzero_ps();
2036             fjx3             = _mm_setzero_ps();
2037             fjy3             = _mm_setzero_ps();
2038             fjz3             = _mm_setzero_ps();
2039
2040             /**************************
2041              * CALCULATE INTERACTIONS *
2042              **************************/
2043
2044             /* LENNARD-JONES DISPERSION/REPULSION */
2045
2046             rinvsix          = _mm_mul_ps(_mm_mul_ps(rinvsq00,rinvsq00),rinvsq00);
2047             fvdw             = _mm_mul_ps(_mm_msub_ps(c12_00,rinvsix,c6_00),_mm_mul_ps(rinvsix,rinvsq00));
2048
2049             fscal            = fvdw;
2050
2051             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2052
2053              /* Update vectorial force */
2054             fix0             = _mm_macc_ps(dx00,fscal,fix0);
2055             fiy0             = _mm_macc_ps(dy00,fscal,fiy0);
2056             fiz0             = _mm_macc_ps(dz00,fscal,fiz0);
2057
2058             fjx0             = _mm_macc_ps(dx00,fscal,fjx0);
2059             fjy0             = _mm_macc_ps(dy00,fscal,fjy0);
2060             fjz0             = _mm_macc_ps(dz00,fscal,fjz0);
2061
2062             /**************************
2063              * CALCULATE INTERACTIONS *
2064              **************************/
2065
2066             r11              = _mm_mul_ps(rsq11,rinv11);
2067             r11              = _mm_andnot_ps(dummy_mask,r11);
2068
2069             /* Calculate table index by multiplying r with table scale and truncate to integer */
2070             rt               = _mm_mul_ps(r11,vftabscale);
2071             vfitab           = _mm_cvttps_epi32(rt);
2072 #ifdef __XOP__
2073             vfeps            = _mm_frcz_ps(rt);
2074 #else
2075             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2076 #endif
2077             twovfeps         = _mm_add_ps(vfeps,vfeps);
2078             vfitab           = _mm_slli_epi32(vfitab,2);
2079
2080             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2081             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2082             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2083             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2084             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2085             _MM_TRANSPOSE4_PS(Y,F,G,H);
2086             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2087             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2088             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq11,FF),_mm_mul_ps(vftabscale,rinv11)));
2089
2090             fscal            = felec;
2091
2092             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2093
2094              /* Update vectorial force */
2095             fix1             = _mm_macc_ps(dx11,fscal,fix1);
2096             fiy1             = _mm_macc_ps(dy11,fscal,fiy1);
2097             fiz1             = _mm_macc_ps(dz11,fscal,fiz1);
2098
2099             fjx1             = _mm_macc_ps(dx11,fscal,fjx1);
2100             fjy1             = _mm_macc_ps(dy11,fscal,fjy1);
2101             fjz1             = _mm_macc_ps(dz11,fscal,fjz1);
2102
2103             /**************************
2104              * CALCULATE INTERACTIONS *
2105              **************************/
2106
2107             r12              = _mm_mul_ps(rsq12,rinv12);
2108             r12              = _mm_andnot_ps(dummy_mask,r12);
2109
2110             /* Calculate table index by multiplying r with table scale and truncate to integer */
2111             rt               = _mm_mul_ps(r12,vftabscale);
2112             vfitab           = _mm_cvttps_epi32(rt);
2113 #ifdef __XOP__
2114             vfeps            = _mm_frcz_ps(rt);
2115 #else
2116             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2117 #endif
2118             twovfeps         = _mm_add_ps(vfeps,vfeps);
2119             vfitab           = _mm_slli_epi32(vfitab,2);
2120
2121             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2122             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2123             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2124             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2125             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2126             _MM_TRANSPOSE4_PS(Y,F,G,H);
2127             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2128             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2129             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq12,FF),_mm_mul_ps(vftabscale,rinv12)));
2130
2131             fscal            = felec;
2132
2133             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2134
2135              /* Update vectorial force */
2136             fix1             = _mm_macc_ps(dx12,fscal,fix1);
2137             fiy1             = _mm_macc_ps(dy12,fscal,fiy1);
2138             fiz1             = _mm_macc_ps(dz12,fscal,fiz1);
2139
2140             fjx2             = _mm_macc_ps(dx12,fscal,fjx2);
2141             fjy2             = _mm_macc_ps(dy12,fscal,fjy2);
2142             fjz2             = _mm_macc_ps(dz12,fscal,fjz2);
2143
2144             /**************************
2145              * CALCULATE INTERACTIONS *
2146              **************************/
2147
2148             r13              = _mm_mul_ps(rsq13,rinv13);
2149             r13              = _mm_andnot_ps(dummy_mask,r13);
2150
2151             /* Calculate table index by multiplying r with table scale and truncate to integer */
2152             rt               = _mm_mul_ps(r13,vftabscale);
2153             vfitab           = _mm_cvttps_epi32(rt);
2154 #ifdef __XOP__
2155             vfeps            = _mm_frcz_ps(rt);
2156 #else
2157             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2158 #endif
2159             twovfeps         = _mm_add_ps(vfeps,vfeps);
2160             vfitab           = _mm_slli_epi32(vfitab,2);
2161
2162             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2163             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2164             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2165             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2166             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2167             _MM_TRANSPOSE4_PS(Y,F,G,H);
2168             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2169             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2170             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq13,FF),_mm_mul_ps(vftabscale,rinv13)));
2171
2172             fscal            = felec;
2173
2174             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2175
2176              /* Update vectorial force */
2177             fix1             = _mm_macc_ps(dx13,fscal,fix1);
2178             fiy1             = _mm_macc_ps(dy13,fscal,fiy1);
2179             fiz1             = _mm_macc_ps(dz13,fscal,fiz1);
2180
2181             fjx3             = _mm_macc_ps(dx13,fscal,fjx3);
2182             fjy3             = _mm_macc_ps(dy13,fscal,fjy3);
2183             fjz3             = _mm_macc_ps(dz13,fscal,fjz3);
2184
2185             /**************************
2186              * CALCULATE INTERACTIONS *
2187              **************************/
2188
2189             r21              = _mm_mul_ps(rsq21,rinv21);
2190             r21              = _mm_andnot_ps(dummy_mask,r21);
2191
2192             /* Calculate table index by multiplying r with table scale and truncate to integer */
2193             rt               = _mm_mul_ps(r21,vftabscale);
2194             vfitab           = _mm_cvttps_epi32(rt);
2195 #ifdef __XOP__
2196             vfeps            = _mm_frcz_ps(rt);
2197 #else
2198             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2199 #endif
2200             twovfeps         = _mm_add_ps(vfeps,vfeps);
2201             vfitab           = _mm_slli_epi32(vfitab,2);
2202
2203             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2204             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2205             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2206             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2207             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2208             _MM_TRANSPOSE4_PS(Y,F,G,H);
2209             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2210             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2211             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq21,FF),_mm_mul_ps(vftabscale,rinv21)));
2212
2213             fscal            = felec;
2214
2215             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2216
2217              /* Update vectorial force */
2218             fix2             = _mm_macc_ps(dx21,fscal,fix2);
2219             fiy2             = _mm_macc_ps(dy21,fscal,fiy2);
2220             fiz2             = _mm_macc_ps(dz21,fscal,fiz2);
2221
2222             fjx1             = _mm_macc_ps(dx21,fscal,fjx1);
2223             fjy1             = _mm_macc_ps(dy21,fscal,fjy1);
2224             fjz1             = _mm_macc_ps(dz21,fscal,fjz1);
2225
2226             /**************************
2227              * CALCULATE INTERACTIONS *
2228              **************************/
2229
2230             r22              = _mm_mul_ps(rsq22,rinv22);
2231             r22              = _mm_andnot_ps(dummy_mask,r22);
2232
2233             /* Calculate table index by multiplying r with table scale and truncate to integer */
2234             rt               = _mm_mul_ps(r22,vftabscale);
2235             vfitab           = _mm_cvttps_epi32(rt);
2236 #ifdef __XOP__
2237             vfeps            = _mm_frcz_ps(rt);
2238 #else
2239             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2240 #endif
2241             twovfeps         = _mm_add_ps(vfeps,vfeps);
2242             vfitab           = _mm_slli_epi32(vfitab,2);
2243
2244             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2245             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2246             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2247             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2248             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2249             _MM_TRANSPOSE4_PS(Y,F,G,H);
2250             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2251             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2252             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq22,FF),_mm_mul_ps(vftabscale,rinv22)));
2253
2254             fscal            = felec;
2255
2256             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2257
2258              /* Update vectorial force */
2259             fix2             = _mm_macc_ps(dx22,fscal,fix2);
2260             fiy2             = _mm_macc_ps(dy22,fscal,fiy2);
2261             fiz2             = _mm_macc_ps(dz22,fscal,fiz2);
2262
2263             fjx2             = _mm_macc_ps(dx22,fscal,fjx2);
2264             fjy2             = _mm_macc_ps(dy22,fscal,fjy2);
2265             fjz2             = _mm_macc_ps(dz22,fscal,fjz2);
2266
2267             /**************************
2268              * CALCULATE INTERACTIONS *
2269              **************************/
2270
2271             r23              = _mm_mul_ps(rsq23,rinv23);
2272             r23              = _mm_andnot_ps(dummy_mask,r23);
2273
2274             /* Calculate table index by multiplying r with table scale and truncate to integer */
2275             rt               = _mm_mul_ps(r23,vftabscale);
2276             vfitab           = _mm_cvttps_epi32(rt);
2277 #ifdef __XOP__
2278             vfeps            = _mm_frcz_ps(rt);
2279 #else
2280             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2281 #endif
2282             twovfeps         = _mm_add_ps(vfeps,vfeps);
2283             vfitab           = _mm_slli_epi32(vfitab,2);
2284
2285             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2286             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2287             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2288             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2289             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2290             _MM_TRANSPOSE4_PS(Y,F,G,H);
2291             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2292             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2293             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq23,FF),_mm_mul_ps(vftabscale,rinv23)));
2294
2295             fscal            = felec;
2296
2297             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2298
2299              /* Update vectorial force */
2300             fix2             = _mm_macc_ps(dx23,fscal,fix2);
2301             fiy2             = _mm_macc_ps(dy23,fscal,fiy2);
2302             fiz2             = _mm_macc_ps(dz23,fscal,fiz2);
2303
2304             fjx3             = _mm_macc_ps(dx23,fscal,fjx3);
2305             fjy3             = _mm_macc_ps(dy23,fscal,fjy3);
2306             fjz3             = _mm_macc_ps(dz23,fscal,fjz3);
2307
2308             /**************************
2309              * CALCULATE INTERACTIONS *
2310              **************************/
2311
2312             r31              = _mm_mul_ps(rsq31,rinv31);
2313             r31              = _mm_andnot_ps(dummy_mask,r31);
2314
2315             /* Calculate table index by multiplying r with table scale and truncate to integer */
2316             rt               = _mm_mul_ps(r31,vftabscale);
2317             vfitab           = _mm_cvttps_epi32(rt);
2318 #ifdef __XOP__
2319             vfeps            = _mm_frcz_ps(rt);
2320 #else
2321             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2322 #endif
2323             twovfeps         = _mm_add_ps(vfeps,vfeps);
2324             vfitab           = _mm_slli_epi32(vfitab,2);
2325
2326             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2327             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2328             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2329             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2330             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2331             _MM_TRANSPOSE4_PS(Y,F,G,H);
2332             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2333             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2334             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq31,FF),_mm_mul_ps(vftabscale,rinv31)));
2335
2336             fscal            = felec;
2337
2338             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2339
2340              /* Update vectorial force */
2341             fix3             = _mm_macc_ps(dx31,fscal,fix3);
2342             fiy3             = _mm_macc_ps(dy31,fscal,fiy3);
2343             fiz3             = _mm_macc_ps(dz31,fscal,fiz3);
2344
2345             fjx1             = _mm_macc_ps(dx31,fscal,fjx1);
2346             fjy1             = _mm_macc_ps(dy31,fscal,fjy1);
2347             fjz1             = _mm_macc_ps(dz31,fscal,fjz1);
2348
2349             /**************************
2350              * CALCULATE INTERACTIONS *
2351              **************************/
2352
2353             r32              = _mm_mul_ps(rsq32,rinv32);
2354             r32              = _mm_andnot_ps(dummy_mask,r32);
2355
2356             /* Calculate table index by multiplying r with table scale and truncate to integer */
2357             rt               = _mm_mul_ps(r32,vftabscale);
2358             vfitab           = _mm_cvttps_epi32(rt);
2359 #ifdef __XOP__
2360             vfeps            = _mm_frcz_ps(rt);
2361 #else
2362             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2363 #endif
2364             twovfeps         = _mm_add_ps(vfeps,vfeps);
2365             vfitab           = _mm_slli_epi32(vfitab,2);
2366
2367             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2368             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2369             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2370             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2371             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2372             _MM_TRANSPOSE4_PS(Y,F,G,H);
2373             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2374             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2375             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq32,FF),_mm_mul_ps(vftabscale,rinv32)));
2376
2377             fscal            = felec;
2378
2379             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2380
2381              /* Update vectorial force */
2382             fix3             = _mm_macc_ps(dx32,fscal,fix3);
2383             fiy3             = _mm_macc_ps(dy32,fscal,fiy3);
2384             fiz3             = _mm_macc_ps(dz32,fscal,fiz3);
2385
2386             fjx2             = _mm_macc_ps(dx32,fscal,fjx2);
2387             fjy2             = _mm_macc_ps(dy32,fscal,fjy2);
2388             fjz2             = _mm_macc_ps(dz32,fscal,fjz2);
2389
2390             /**************************
2391              * CALCULATE INTERACTIONS *
2392              **************************/
2393
2394             r33              = _mm_mul_ps(rsq33,rinv33);
2395             r33              = _mm_andnot_ps(dummy_mask,r33);
2396
2397             /* Calculate table index by multiplying r with table scale and truncate to integer */
2398             rt               = _mm_mul_ps(r33,vftabscale);
2399             vfitab           = _mm_cvttps_epi32(rt);
2400 #ifdef __XOP__
2401             vfeps            = _mm_frcz_ps(rt);
2402 #else
2403             vfeps            = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR));
2404 #endif
2405             twovfeps         = _mm_add_ps(vfeps,vfeps);
2406             vfitab           = _mm_slli_epi32(vfitab,2);
2407
2408             /* CUBIC SPLINE TABLE ELECTROSTATICS */
2409             Y                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,0) );
2410             F                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,1) );
2411             G                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,2) );
2412             H                = _mm_load_ps( vftab + _mm_extract_epi32(vfitab,3) );
2413             _MM_TRANSPOSE4_PS(Y,F,G,H);
2414             Fp               = _mm_macc_ps(vfeps,_mm_macc_ps(H,vfeps,G),F);
2415             FF               = _mm_macc_ps(vfeps,_mm_macc_ps(twovfeps,H,G),Fp);
2416             felec            = _mm_xor_ps(signbit,_mm_mul_ps(_mm_mul_ps(qq33,FF),_mm_mul_ps(vftabscale,rinv33)));
2417
2418             fscal            = felec;
2419
2420             fscal            = _mm_andnot_ps(dummy_mask,fscal);
2421
2422              /* Update vectorial force */
2423             fix3             = _mm_macc_ps(dx33,fscal,fix3);
2424             fiy3             = _mm_macc_ps(dy33,fscal,fiy3);
2425             fiz3             = _mm_macc_ps(dz33,fscal,fiz3);
2426
2427             fjx3             = _mm_macc_ps(dx33,fscal,fjx3);
2428             fjy3             = _mm_macc_ps(dy33,fscal,fjy3);
2429             fjz3             = _mm_macc_ps(dz33,fscal,fjz3);
2430
2431             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
2432             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
2433             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
2434             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
2435
2436             gmx_mm_decrement_4rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,
2437                                                    fjx0,fjy0,fjz0,fjx1,fjy1,fjz1,
2438                                                    fjx2,fjy2,fjz2,fjx3,fjy3,fjz3);
2439
2440             /* Inner loop uses 420 flops */
2441         }
2442
2443         /* End of innermost loop */
2444
2445         gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
2446                                               f+i_coord_offset,fshift+i_shift_offset);
2447
2448         /* Increment number of inner iterations */
2449         inneriter                  += j_index_end - j_index_start;
2450
2451         /* Outer loop uses 24 flops */
2452     }
2453
2454     /* Increment number of outer iterations */
2455     outeriter        += nri;
2456
2457     /* Update outer/inner flops */
2458
2459     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4W4_F,outeriter*24 + inneriter*420);
2460 }