src/gmxlib/nonbonded/nb_kernel_avx_128_fma_double/nb_kernel_ElecRF_VdwCSTab_GeomW3P1_avx_128_fma_double.c

   1 /*
   2  * Note: this file was generated by the Gromacs avx_128_fma_double kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_avx_128_fma_double.h"
  34 #include "kernelutil_x86_avx_128_fma_double.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwCSTab_GeomW3P1_VF_avx_128_fma_double
  38  * Electrostatics interaction: ReactionField
  39  * VdW interaction:            CubicSplineTable
  40  * Geometry:                   Water3-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecRF_VdwCSTab_GeomW3P1_VF_avx_128_fma_double
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB;
  61     int              j_coord_offsetA,j_coord_offsetB;
  62     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  63     real             rcutoff_scalar;
  64     real             *shiftvec,*fshift,*x,*f;
  65     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  66     int              vdwioffset0;
  67     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  68     int              vdwioffset1;
  69     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  70     int              vdwioffset2;
  71     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  72     int              vdwjidx0A,vdwjidx0B;
  73     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  74     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  75     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  76     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  77     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  78     real             *charge;
  79     int              nvdwtype;
  80     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  81     int              *vdwtype;
  82     real             *vdwparam;
  83     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
  84     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
  85     __m128i          vfitab;
  86     __m128i          ifour       = _mm_set1_epi32(4);
  87     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF,twovfeps;
  88     real             *vftab;
  89     __m128d          dummy_mask,cutoff_mask;
  90     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  91     __m128d          one     = _mm_set1_pd(1.0);
  92     __m128d          two     = _mm_set1_pd(2.0);
  93     x                = xx[0];
  94     f                = ff[0];
  95
  96     nri              = nlist->nri;
  97     iinr             = nlist->iinr;
  98     jindex           = nlist->jindex;
  99     jjnr             = nlist->jjnr;
 100     shiftidx         = nlist->shift;
 101     gid              = nlist->gid;
 102     shiftvec         = fr->shift_vec[0];
 103     fshift           = fr->fshift[0];
 104     facel            = _mm_set1_pd(fr->epsfac);
 105     charge           = mdatoms->chargeA;
 106     krf              = _mm_set1_pd(fr->ic->k_rf);
 107     krf2             = _mm_set1_pd(fr->ic->k_rf*2.0);
 108     crf              = _mm_set1_pd(fr->ic->c_rf);
 109     nvdwtype         = fr->ntype;
 110     vdwparam         = fr->nbfp;
 111     vdwtype          = mdatoms->typeA;
 112
 113     vftab            = kernel_data->table_vdw->data;
 114     vftabscale       = _mm_set1_pd(kernel_data->table_vdw->scale);
 115
 116     /* Setup water-specific parameters */
 117     inr              = nlist->iinr[0];
 118     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 119     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 120     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 121     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 122
 123     /* Avoid stupid compiler warnings */
 124     jnrA = jnrB = 0;
 125     j_coord_offsetA = 0;
 126     j_coord_offsetB = 0;
 127
 128     outeriter        = 0;
 129     inneriter        = 0;
 130
 131     /* Start outer loop over neighborlists */
 132     for(iidx=0; iidx<nri; iidx++)
 133     {
 134         /* Load shift vector for this list */
 135         i_shift_offset   = DIM*shiftidx[iidx];
 136
 137         /* Load limits for loop over neighbors */
 138         j_index_start    = jindex[iidx];
 139         j_index_end      = jindex[iidx+1];
 140
 141         /* Get outer coordinate index */
 142         inr              = iinr[iidx];
 143         i_coord_offset   = DIM*inr;
 144
 145         /* Load i particle coords and add shift vector */
 146         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 147                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 148
 149         fix0             = _mm_setzero_pd();
 150         fiy0             = _mm_setzero_pd();
 151         fiz0             = _mm_setzero_pd();
 152         fix1             = _mm_setzero_pd();
 153         fiy1             = _mm_setzero_pd();
 154         fiz1             = _mm_setzero_pd();
 155         fix2             = _mm_setzero_pd();
 156         fiy2             = _mm_setzero_pd();
 157         fiz2             = _mm_setzero_pd();
 158
 159         /* Reset potential sums */
 160         velecsum         = _mm_setzero_pd();
 161         vvdwsum          = _mm_setzero_pd();
 162
 163         /* Start inner kernel loop */
 164         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 165         {
 166
 167             /* Get j neighbor index, and coordinate index */
 168             jnrA             = jjnr[jidx];
 169             jnrB             = jjnr[jidx+1];
 170             j_coord_offsetA  = DIM*jnrA;
 171             j_coord_offsetB  = DIM*jnrB;
 172
 173             /* load j atom coordinates */
 174             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 175                                               &jx0,&jy0,&jz0);
 176
 177             /* Calculate displacement vector */
 178             dx00             = _mm_sub_pd(ix0,jx0);
 179             dy00             = _mm_sub_pd(iy0,jy0);
 180             dz00             = _mm_sub_pd(iz0,jz0);
 181             dx10             = _mm_sub_pd(ix1,jx0);
 182             dy10             = _mm_sub_pd(iy1,jy0);
 183             dz10             = _mm_sub_pd(iz1,jz0);
 184             dx20             = _mm_sub_pd(ix2,jx0);
 185             dy20             = _mm_sub_pd(iy2,jy0);
 186             dz20             = _mm_sub_pd(iz2,jz0);
 187
 188             /* Calculate squared distance and things based on it */
 189             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 190             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 191             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 192
 193             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 194             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 195             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 196
 197             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 198             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 199             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 200
 201             /* Load parameters for j particles */
 202             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 203             vdwjidx0A        = 2*vdwtype[jnrA+0];
 204             vdwjidx0B        = 2*vdwtype[jnrB+0];
 205
 206             fjx0             = _mm_setzero_pd();
 207             fjy0             = _mm_setzero_pd();
 208             fjz0             = _mm_setzero_pd();
 209
 210             /**************************
 211              * CALCULATE INTERACTIONS *
 212              **************************/
 213
 214             r00              = _mm_mul_pd(rsq00,rinv00);
 215
 216             /* Compute parameters for interactions between i and j atoms */
 217             qq00             = _mm_mul_pd(iq0,jq0);
 218             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 219                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 220
 221             /* Calculate table index by multiplying r with table scale and truncate to integer */
 222             rt               = _mm_mul_pd(r00,vftabscale);
 223             vfitab           = _mm_cvttpd_epi32(rt);
 224 #ifdef __XOP__
 225             vfeps            = _mm_frcz_pd(rt);
 226 #else
 227             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 228 #endif
 229             twovfeps         = _mm_add_pd(vfeps,vfeps);
 230             vfitab           = _mm_slli_epi32(vfitab,3);
 231
 232             /* REACTION-FIELD ELECTROSTATICS */
 233             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_macc_pd(krf,rsq00,rinv00),crf));
 234             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 235
 236             /* CUBIC SPLINE TABLE DISPERSION */
 237             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 238             F                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 239             GMX_MM_TRANSPOSE2_PD(Y,F);
 240             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 241             H                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) +2);
 242             GMX_MM_TRANSPOSE2_PD(G,H);
 243             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 244             VV               = _mm_macc_pd(vfeps,Fp,Y);
 245             vvdw6            = _mm_mul_pd(c6_00,VV);
 246             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 247             fvdw6            = _mm_mul_pd(c6_00,FF);
 248
 249             /* CUBIC SPLINE TABLE REPULSION */
 250             vfitab           = _mm_add_epi32(vfitab,ifour);
 251             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 252             F                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 253             GMX_MM_TRANSPOSE2_PD(Y,F);
 254             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 255             H                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) +2);
 256             GMX_MM_TRANSPOSE2_PD(G,H);
 257             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 258             VV               = _mm_macc_pd(vfeps,Fp,Y);
 259             vvdw12           = _mm_mul_pd(c12_00,VV);
 260             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 261             fvdw12           = _mm_mul_pd(c12_00,FF);
 262             vvdw             = _mm_add_pd(vvdw12,vvdw6);
 263             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 264
 265             /* Update potential sum for this i atom from the interaction with this j atom. */
 266             velecsum         = _mm_add_pd(velecsum,velec);
 267             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 268
 269             fscal            = _mm_add_pd(felec,fvdw);
 270
 271             /* Update vectorial force */
 272             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 273             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 274             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 275
 276             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 277             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 278             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 279
 280             /**************************
 281              * CALCULATE INTERACTIONS *
 282              **************************/
 283
 284             /* Compute parameters for interactions between i and j atoms */
 285             qq10             = _mm_mul_pd(iq1,jq0);
 286
 287             /* REACTION-FIELD ELECTROSTATICS */
 288             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_macc_pd(krf,rsq10,rinv10),crf));
 289             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 290
 291             /* Update potential sum for this i atom from the interaction with this j atom. */
 292             velecsum         = _mm_add_pd(velecsum,velec);
 293
 294             fscal            = felec;
 295
 296             /* Update vectorial force */
 297             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 298             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 299             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 300
 301             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 302             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 303             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 304
 305             /**************************
 306              * CALCULATE INTERACTIONS *
 307              **************************/
 308
 309             /* Compute parameters for interactions between i and j atoms */
 310             qq20             = _mm_mul_pd(iq2,jq0);
 311
 312             /* REACTION-FIELD ELECTROSTATICS */
 313             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_macc_pd(krf,rsq20,rinv20),crf));
 314             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 315
 316             /* Update potential sum for this i atom from the interaction with this j atom. */
 317             velecsum         = _mm_add_pd(velecsum,velec);
 318
 319             fscal            = felec;
 320
 321             /* Update vectorial force */
 322             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 323             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 324             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 325
 326             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 327             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 328             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 329
 330             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 331
 332             /* Inner loop uses 143 flops */
 333         }
 334
 335         if(jidx<j_index_end)
 336         {
 337
 338             jnrA             = jjnr[jidx];
 339             j_coord_offsetA  = DIM*jnrA;
 340
 341             /* load j atom coordinates */
 342             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 343                                               &jx0,&jy0,&jz0);
 344
 345             /* Calculate displacement vector */
 346             dx00             = _mm_sub_pd(ix0,jx0);
 347             dy00             = _mm_sub_pd(iy0,jy0);
 348             dz00             = _mm_sub_pd(iz0,jz0);
 349             dx10             = _mm_sub_pd(ix1,jx0);
 350             dy10             = _mm_sub_pd(iy1,jy0);
 351             dz10             = _mm_sub_pd(iz1,jz0);
 352             dx20             = _mm_sub_pd(ix2,jx0);
 353             dy20             = _mm_sub_pd(iy2,jy0);
 354             dz20             = _mm_sub_pd(iz2,jz0);
 355
 356             /* Calculate squared distance and things based on it */
 357             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 358             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 359             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 360
 361             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 362             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 363             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 364
 365             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 366             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 367             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 368
 369             /* Load parameters for j particles */
 370             jq0              = _mm_load_sd(charge+jnrA+0);
 371             vdwjidx0A        = 2*vdwtype[jnrA+0];
 372
 373             fjx0             = _mm_setzero_pd();
 374             fjy0             = _mm_setzero_pd();
 375             fjz0             = _mm_setzero_pd();
 376
 377             /**************************
 378              * CALCULATE INTERACTIONS *
 379              **************************/
 380
 381             r00              = _mm_mul_pd(rsq00,rinv00);
 382
 383             /* Compute parameters for interactions between i and j atoms */
 384             qq00             = _mm_mul_pd(iq0,jq0);
 385             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 386
 387             /* Calculate table index by multiplying r with table scale and truncate to integer */
 388             rt               = _mm_mul_pd(r00,vftabscale);
 389             vfitab           = _mm_cvttpd_epi32(rt);
 390 #ifdef __XOP__
 391             vfeps            = _mm_frcz_pd(rt);
 392 #else
 393             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 394 #endif
 395             twovfeps         = _mm_add_pd(vfeps,vfeps);
 396             vfitab           = _mm_slli_epi32(vfitab,3);
 397
 398             /* REACTION-FIELD ELECTROSTATICS */
 399             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_macc_pd(krf,rsq00,rinv00),crf));
 400             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 401
 402             /* CUBIC SPLINE TABLE DISPERSION */
 403             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 404             F                = _mm_setzero_pd();
 405             GMX_MM_TRANSPOSE2_PD(Y,F);
 406             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 407             H                = _mm_setzero_pd();
 408             GMX_MM_TRANSPOSE2_PD(G,H);
 409             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 410             VV               = _mm_macc_pd(vfeps,Fp,Y);
 411             vvdw6            = _mm_mul_pd(c6_00,VV);
 412             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 413             fvdw6            = _mm_mul_pd(c6_00,FF);
 414
 415             /* CUBIC SPLINE TABLE REPULSION */
 416             vfitab           = _mm_add_epi32(vfitab,ifour);
 417             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 418             F                = _mm_setzero_pd();
 419             GMX_MM_TRANSPOSE2_PD(Y,F);
 420             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 421             H                = _mm_setzero_pd();
 422             GMX_MM_TRANSPOSE2_PD(G,H);
 423             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 424             VV               = _mm_macc_pd(vfeps,Fp,Y);
 425             vvdw12           = _mm_mul_pd(c12_00,VV);
 426             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 427             fvdw12           = _mm_mul_pd(c12_00,FF);
 428             vvdw             = _mm_add_pd(vvdw12,vvdw6);
 429             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 430
 431             /* Update potential sum for this i atom from the interaction with this j atom. */
 432             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 433             velecsum         = _mm_add_pd(velecsum,velec);
 434             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 435             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 436
 437             fscal            = _mm_add_pd(felec,fvdw);
 438
 439             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 440
 441             /* Update vectorial force */
 442             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 443             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 444             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 445
 446             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 447             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 448             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 449
 450             /**************************
 451              * CALCULATE INTERACTIONS *
 452              **************************/
 453
 454             /* Compute parameters for interactions between i and j atoms */
 455             qq10             = _mm_mul_pd(iq1,jq0);
 456
 457             /* REACTION-FIELD ELECTROSTATICS */
 458             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_macc_pd(krf,rsq10,rinv10),crf));
 459             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 460
 461             /* Update potential sum for this i atom from the interaction with this j atom. */
 462             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 463             velecsum         = _mm_add_pd(velecsum,velec);
 464
 465             fscal            = felec;
 466
 467             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 468
 469             /* Update vectorial force */
 470             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 471             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 472             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 473
 474             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 475             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 476             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 477
 478             /**************************
 479              * CALCULATE INTERACTIONS *
 480              **************************/
 481
 482             /* Compute parameters for interactions between i and j atoms */
 483             qq20             = _mm_mul_pd(iq2,jq0);
 484
 485             /* REACTION-FIELD ELECTROSTATICS */
 486             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_macc_pd(krf,rsq20,rinv20),crf));
 487             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 488
 489             /* Update potential sum for this i atom from the interaction with this j atom. */
 490             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 491             velecsum         = _mm_add_pd(velecsum,velec);
 492
 493             fscal            = felec;
 494
 495             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 496
 497             /* Update vectorial force */
 498             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 499             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 500             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 501
 502             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 503             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 504             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 505
 506             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 507
 508             /* Inner loop uses 143 flops */
 509         }
 510
 511         /* End of innermost loop */
 512
 513         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
 514                                               f+i_coord_offset,fshift+i_shift_offset);
 515
 516         ggid                        = gid[iidx];
 517         /* Update potential energies */
 518         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 519         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
 520
 521         /* Increment number of inner iterations */
 522         inneriter                  += j_index_end - j_index_start;
 523
 524         /* Outer loop uses 20 flops */
 525     }
 526
 527     /* Increment number of outer iterations */
 528     outeriter        += nri;
 529
 530     /* Update outer/inner flops */
 531
 532     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*20 + inneriter*143);
 533 }
 534 /*
 535  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwCSTab_GeomW3P1_F_avx_128_fma_double
 536  * Electrostatics interaction: ReactionField
 537  * VdW interaction:            CubicSplineTable
 538  * Geometry:                   Water3-Particle
 539  * Calculate force/pot:        Force
 540  */
 541 void
 542 nb_kernel_ElecRF_VdwCSTab_GeomW3P1_F_avx_128_fma_double
 543                     (t_nblist * gmx_restrict                nlist,
 544                      rvec * gmx_restrict                    xx,
 545                      rvec * gmx_restrict                    ff,
 546                      t_forcerec * gmx_restrict              fr,
 547                      t_mdatoms * gmx_restrict               mdatoms,
 548                      nb_kernel_data_t * gmx_restrict        kernel_data,
 549                      t_nrnb * gmx_restrict                  nrnb)
 550 {
 551     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 552      * just 0 for non-waters.
 553      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 554      * jnr indices corresponding to data put in the four positions in the SIMD register.
 555      */
 556     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 557     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 558     int              jnrA,jnrB;
 559     int              j_coord_offsetA,j_coord_offsetB;
 560     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 561     real             rcutoff_scalar;
 562     real             *shiftvec,*fshift,*x,*f;
 563     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 564     int              vdwioffset0;
 565     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 566     int              vdwioffset1;
 567     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 568     int              vdwioffset2;
 569     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 570     int              vdwjidx0A,vdwjidx0B;
 571     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 572     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 573     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 574     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 575     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 576     real             *charge;
 577     int              nvdwtype;
 578     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 579     int              *vdwtype;
 580     real             *vdwparam;
 581     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 582     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 583     __m128i          vfitab;
 584     __m128i          ifour       = _mm_set1_epi32(4);
 585     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF,twovfeps;
 586     real             *vftab;
 587     __m128d          dummy_mask,cutoff_mask;
 588     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 589     __m128d          one     = _mm_set1_pd(1.0);
 590     __m128d          two     = _mm_set1_pd(2.0);
 591     x                = xx[0];
 592     f                = ff[0];
 593
 594     nri              = nlist->nri;
 595     iinr             = nlist->iinr;
 596     jindex           = nlist->jindex;
 597     jjnr             = nlist->jjnr;
 598     shiftidx         = nlist->shift;
 599     gid              = nlist->gid;
 600     shiftvec         = fr->shift_vec[0];
 601     fshift           = fr->fshift[0];
 602     facel            = _mm_set1_pd(fr->epsfac);
 603     charge           = mdatoms->chargeA;
 604     krf              = _mm_set1_pd(fr->ic->k_rf);
 605     krf2             = _mm_set1_pd(fr->ic->k_rf*2.0);
 606     crf              = _mm_set1_pd(fr->ic->c_rf);
 607     nvdwtype         = fr->ntype;
 608     vdwparam         = fr->nbfp;
 609     vdwtype          = mdatoms->typeA;
 610
 611     vftab            = kernel_data->table_vdw->data;
 612     vftabscale       = _mm_set1_pd(kernel_data->table_vdw->scale);
 613
 614     /* Setup water-specific parameters */
 615     inr              = nlist->iinr[0];
 616     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 617     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 618     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 619     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 620
 621     /* Avoid stupid compiler warnings */
 622     jnrA = jnrB = 0;
 623     j_coord_offsetA = 0;
 624     j_coord_offsetB = 0;
 625
 626     outeriter        = 0;
 627     inneriter        = 0;
 628
 629     /* Start outer loop over neighborlists */
 630     for(iidx=0; iidx<nri; iidx++)
 631     {
 632         /* Load shift vector for this list */
 633         i_shift_offset   = DIM*shiftidx[iidx];
 634
 635         /* Load limits for loop over neighbors */
 636         j_index_start    = jindex[iidx];
 637         j_index_end      = jindex[iidx+1];
 638
 639         /* Get outer coordinate index */
 640         inr              = iinr[iidx];
 641         i_coord_offset   = DIM*inr;
 642
 643         /* Load i particle coords and add shift vector */
 644         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 645                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 646
 647         fix0             = _mm_setzero_pd();
 648         fiy0             = _mm_setzero_pd();
 649         fiz0             = _mm_setzero_pd();
 650         fix1             = _mm_setzero_pd();
 651         fiy1             = _mm_setzero_pd();
 652         fiz1             = _mm_setzero_pd();
 653         fix2             = _mm_setzero_pd();
 654         fiy2             = _mm_setzero_pd();
 655         fiz2             = _mm_setzero_pd();
 656
 657         /* Start inner kernel loop */
 658         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 659         {
 660
 661             /* Get j neighbor index, and coordinate index */
 662             jnrA             = jjnr[jidx];
 663             jnrB             = jjnr[jidx+1];
 664             j_coord_offsetA  = DIM*jnrA;
 665             j_coord_offsetB  = DIM*jnrB;
 666
 667             /* load j atom coordinates */
 668             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 669                                               &jx0,&jy0,&jz0);
 670
 671             /* Calculate displacement vector */
 672             dx00             = _mm_sub_pd(ix0,jx0);
 673             dy00             = _mm_sub_pd(iy0,jy0);
 674             dz00             = _mm_sub_pd(iz0,jz0);
 675             dx10             = _mm_sub_pd(ix1,jx0);
 676             dy10             = _mm_sub_pd(iy1,jy0);
 677             dz10             = _mm_sub_pd(iz1,jz0);
 678             dx20             = _mm_sub_pd(ix2,jx0);
 679             dy20             = _mm_sub_pd(iy2,jy0);
 680             dz20             = _mm_sub_pd(iz2,jz0);
 681
 682             /* Calculate squared distance and things based on it */
 683             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 684             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 685             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 686
 687             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 688             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 689             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 690
 691             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 692             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 693             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 694
 695             /* Load parameters for j particles */
 696             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 697             vdwjidx0A        = 2*vdwtype[jnrA+0];
 698             vdwjidx0B        = 2*vdwtype[jnrB+0];
 699
 700             fjx0             = _mm_setzero_pd();
 701             fjy0             = _mm_setzero_pd();
 702             fjz0             = _mm_setzero_pd();
 703
 704             /**************************
 705              * CALCULATE INTERACTIONS *
 706              **************************/
 707
 708             r00              = _mm_mul_pd(rsq00,rinv00);
 709
 710             /* Compute parameters for interactions between i and j atoms */
 711             qq00             = _mm_mul_pd(iq0,jq0);
 712             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 713                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 714
 715             /* Calculate table index by multiplying r with table scale and truncate to integer */
 716             rt               = _mm_mul_pd(r00,vftabscale);
 717             vfitab           = _mm_cvttpd_epi32(rt);
 718 #ifdef __XOP__
 719             vfeps            = _mm_frcz_pd(rt);
 720 #else
 721             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 722 #endif
 723             twovfeps         = _mm_add_pd(vfeps,vfeps);
 724             vfitab           = _mm_slli_epi32(vfitab,3);
 725
 726             /* REACTION-FIELD ELECTROSTATICS */
 727             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 728
 729             /* CUBIC SPLINE TABLE DISPERSION */
 730             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 731             F                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 732             GMX_MM_TRANSPOSE2_PD(Y,F);
 733             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 734             H                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) +2);
 735             GMX_MM_TRANSPOSE2_PD(G,H);
 736             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 737             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 738             fvdw6            = _mm_mul_pd(c6_00,FF);
 739
 740             /* CUBIC SPLINE TABLE REPULSION */
 741             vfitab           = _mm_add_epi32(vfitab,ifour);
 742             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 743             F                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) );
 744             GMX_MM_TRANSPOSE2_PD(Y,F);
 745             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 746             H                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,1) +2);
 747             GMX_MM_TRANSPOSE2_PD(G,H);
 748             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 749             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 750             fvdw12           = _mm_mul_pd(c12_00,FF);
 751             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 752
 753             fscal            = _mm_add_pd(felec,fvdw);
 754
 755             /* Update vectorial force */
 756             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 757             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 758             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 759
 760             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 761             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 762             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 763
 764             /**************************
 765              * CALCULATE INTERACTIONS *
 766              **************************/
 767
 768             /* Compute parameters for interactions between i and j atoms */
 769             qq10             = _mm_mul_pd(iq1,jq0);
 770
 771             /* REACTION-FIELD ELECTROSTATICS */
 772             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 773
 774             fscal            = felec;
 775
 776             /* Update vectorial force */
 777             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 778             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 779             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 780
 781             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 782             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 783             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 784
 785             /**************************
 786              * CALCULATE INTERACTIONS *
 787              **************************/
 788
 789             /* Compute parameters for interactions between i and j atoms */
 790             qq20             = _mm_mul_pd(iq2,jq0);
 791
 792             /* REACTION-FIELD ELECTROSTATICS */
 793             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 794
 795             fscal            = felec;
 796
 797             /* Update vectorial force */
 798             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 799             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 800             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 801
 802             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 803             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 804             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 805
 806             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 807
 808             /* Inner loop uses 120 flops */
 809         }
 810
 811         if(jidx<j_index_end)
 812         {
 813
 814             jnrA             = jjnr[jidx];
 815             j_coord_offsetA  = DIM*jnrA;
 816
 817             /* load j atom coordinates */
 818             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 819                                               &jx0,&jy0,&jz0);
 820
 821             /* Calculate displacement vector */
 822             dx00             = _mm_sub_pd(ix0,jx0);
 823             dy00             = _mm_sub_pd(iy0,jy0);
 824             dz00             = _mm_sub_pd(iz0,jz0);
 825             dx10             = _mm_sub_pd(ix1,jx0);
 826             dy10             = _mm_sub_pd(iy1,jy0);
 827             dz10             = _mm_sub_pd(iz1,jz0);
 828             dx20             = _mm_sub_pd(ix2,jx0);
 829             dy20             = _mm_sub_pd(iy2,jy0);
 830             dz20             = _mm_sub_pd(iz2,jz0);
 831
 832             /* Calculate squared distance and things based on it */
 833             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 834             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 835             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 836
 837             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 838             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 839             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 840
 841             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 842             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 843             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 844
 845             /* Load parameters for j particles */
 846             jq0              = _mm_load_sd(charge+jnrA+0);
 847             vdwjidx0A        = 2*vdwtype[jnrA+0];
 848
 849             fjx0             = _mm_setzero_pd();
 850             fjy0             = _mm_setzero_pd();
 851             fjz0             = _mm_setzero_pd();
 852
 853             /**************************
 854              * CALCULATE INTERACTIONS *
 855              **************************/
 856
 857             r00              = _mm_mul_pd(rsq00,rinv00);
 858
 859             /* Compute parameters for interactions between i and j atoms */
 860             qq00             = _mm_mul_pd(iq0,jq0);
 861             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 862
 863             /* Calculate table index by multiplying r with table scale and truncate to integer */
 864             rt               = _mm_mul_pd(r00,vftabscale);
 865             vfitab           = _mm_cvttpd_epi32(rt);
 866 #ifdef __XOP__
 867             vfeps            = _mm_frcz_pd(rt);
 868 #else
 869             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 870 #endif
 871             twovfeps         = _mm_add_pd(vfeps,vfeps);
 872             vfitab           = _mm_slli_epi32(vfitab,3);
 873
 874             /* REACTION-FIELD ELECTROSTATICS */
 875             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 876
 877             /* CUBIC SPLINE TABLE DISPERSION */
 878             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 879             F                = _mm_setzero_pd();
 880             GMX_MM_TRANSPOSE2_PD(Y,F);
 881             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 882             H                = _mm_setzero_pd();
 883             GMX_MM_TRANSPOSE2_PD(G,H);
 884             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 885             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 886             fvdw6            = _mm_mul_pd(c6_00,FF);
 887
 888             /* CUBIC SPLINE TABLE REPULSION */
 889             vfitab           = _mm_add_epi32(vfitab,ifour);
 890             Y                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) );
 891             F                = _mm_setzero_pd();
 892             GMX_MM_TRANSPOSE2_PD(Y,F);
 893             G                = _mm_load_pd( vftab + _mm_extract_epi32(vfitab,0) +2);
 894             H                = _mm_setzero_pd();
 895             GMX_MM_TRANSPOSE2_PD(G,H);
 896             Fp               = _mm_macc_pd(vfeps,_mm_macc_pd(H,vfeps,G),F);
 897             FF               = _mm_macc_pd(vfeps,_mm_macc_pd(twovfeps,H,G),Fp);
 898             fvdw12           = _mm_mul_pd(c12_00,FF);
 899             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 900
 901             fscal            = _mm_add_pd(felec,fvdw);
 902
 903             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 904
 905             /* Update vectorial force */
 906             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 907             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 908             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 909
 910             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 911             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 912             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 913
 914             /**************************
 915              * CALCULATE INTERACTIONS *
 916              **************************/
 917
 918             /* Compute parameters for interactions between i and j atoms */
 919             qq10             = _mm_mul_pd(iq1,jq0);
 920
 921             /* REACTION-FIELD ELECTROSTATICS */
 922             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 923
 924             fscal            = felec;
 925
 926             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 927
 928             /* Update vectorial force */
 929             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 930             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 931             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 932
 933             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 934             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 935             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 936
 937             /**************************
 938              * CALCULATE INTERACTIONS *
 939              **************************/
 940
 941             /* Compute parameters for interactions between i and j atoms */
 942             qq20             = _mm_mul_pd(iq2,jq0);
 943
 944             /* REACTION-FIELD ELECTROSTATICS */
 945             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 946
 947             fscal            = felec;
 948
 949             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 950
 951             /* Update vectorial force */
 952             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 953             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 954             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 955
 956             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 957             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 958             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 959
 960             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 961
 962             /* Inner loop uses 120 flops */
 963         }
 964
 965         /* End of innermost loop */
 966
 967         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
 968                                               f+i_coord_offset,fshift+i_shift_offset);
 969
 970         /* Increment number of inner iterations */
 971         inneriter                  += j_index_end - j_index_start;
 972
 973         /* Outer loop uses 18 flops */
 974     }
 975
 976     /* Increment number of outer iterations */
 977     outeriter        += nri;
 978
 979     /* Update outer/inner flops */
 980
 981     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*18 + inneriter*120);
 982 }