src/gmxlib/nonbonded/nb_kernel_avx_128_fma_double/nb_kernel_ElecEwSh_VdwLJSh_GeomW4P1_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_ElecEwSh_VdwLJSh_GeomW4P1_VF_avx_128_fma_double
  38  * Electrostatics interaction: Ewald
  39  * VdW interaction:            LennardJones
  40  * Geometry:                   Water4-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecEwSh_VdwLJSh_GeomW4P1_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              vdwioffset3;
  73     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  74     int              vdwjidx0A,vdwjidx0B;
  75     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  76     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  77     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  78     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  79     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
  80     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  81     real             *charge;
  82     int              nvdwtype;
  83     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  84     int              *vdwtype;
  85     real             *vdwparam;
  86     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
  87     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
  88     __m128i          ewitab;
  89     __m128d          ewtabscale,eweps,twoeweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
  90     real             *ewtab;
  91     __m128d          dummy_mask,cutoff_mask;
  92     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  93     __m128d          one     = _mm_set1_pd(1.0);
  94     __m128d          two     = _mm_set1_pd(2.0);
  95     x                = xx[0];
  96     f                = ff[0];
  97
  98     nri              = nlist->nri;
  99     iinr             = nlist->iinr;
 100     jindex           = nlist->jindex;
 101     jjnr             = nlist->jjnr;
 102     shiftidx         = nlist->shift;
 103     gid              = nlist->gid;
 104     shiftvec         = fr->shift_vec[0];
 105     fshift           = fr->fshift[0];
 106     facel            = _mm_set1_pd(fr->epsfac);
 107     charge           = mdatoms->chargeA;
 108     nvdwtype         = fr->ntype;
 109     vdwparam         = fr->nbfp;
 110     vdwtype          = mdatoms->typeA;
 111
 112     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 113     ewtab            = fr->ic->tabq_coul_FDV0;
 114     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 115     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 116
 117     /* Setup water-specific parameters */
 118     inr              = nlist->iinr[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     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 122     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 123
 124     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 125     rcutoff_scalar   = fr->rcoulomb;
 126     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 127     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 128
 129     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
 130     rvdw             = _mm_set1_pd(fr->rvdw);
 131
 132     /* Avoid stupid compiler warnings */
 133     jnrA = jnrB = 0;
 134     j_coord_offsetA = 0;
 135     j_coord_offsetB = 0;
 136
 137     outeriter        = 0;
 138     inneriter        = 0;
 139
 140     /* Start outer loop over neighborlists */
 141     for(iidx=0; iidx<nri; iidx++)
 142     {
 143         /* Load shift vector for this list */
 144         i_shift_offset   = DIM*shiftidx[iidx];
 145
 146         /* Load limits for loop over neighbors */
 147         j_index_start    = jindex[iidx];
 148         j_index_end      = jindex[iidx+1];
 149
 150         /* Get outer coordinate index */
 151         inr              = iinr[iidx];
 152         i_coord_offset   = DIM*inr;
 153
 154         /* Load i particle coords and add shift vector */
 155         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 156                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 157
 158         fix0             = _mm_setzero_pd();
 159         fiy0             = _mm_setzero_pd();
 160         fiz0             = _mm_setzero_pd();
 161         fix1             = _mm_setzero_pd();
 162         fiy1             = _mm_setzero_pd();
 163         fiz1             = _mm_setzero_pd();
 164         fix2             = _mm_setzero_pd();
 165         fiy2             = _mm_setzero_pd();
 166         fiz2             = _mm_setzero_pd();
 167         fix3             = _mm_setzero_pd();
 168         fiy3             = _mm_setzero_pd();
 169         fiz3             = _mm_setzero_pd();
 170
 171         /* Reset potential sums */
 172         velecsum         = _mm_setzero_pd();
 173         vvdwsum          = _mm_setzero_pd();
 174
 175         /* Start inner kernel loop */
 176         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 177         {
 178
 179             /* Get j neighbor index, and coordinate index */
 180             jnrA             = jjnr[jidx];
 181             jnrB             = jjnr[jidx+1];
 182             j_coord_offsetA  = DIM*jnrA;
 183             j_coord_offsetB  = DIM*jnrB;
 184
 185             /* load j atom coordinates */
 186             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 187                                               &jx0,&jy0,&jz0);
 188
 189             /* Calculate displacement vector */
 190             dx00             = _mm_sub_pd(ix0,jx0);
 191             dy00             = _mm_sub_pd(iy0,jy0);
 192             dz00             = _mm_sub_pd(iz0,jz0);
 193             dx10             = _mm_sub_pd(ix1,jx0);
 194             dy10             = _mm_sub_pd(iy1,jy0);
 195             dz10             = _mm_sub_pd(iz1,jz0);
 196             dx20             = _mm_sub_pd(ix2,jx0);
 197             dy20             = _mm_sub_pd(iy2,jy0);
 198             dz20             = _mm_sub_pd(iz2,jz0);
 199             dx30             = _mm_sub_pd(ix3,jx0);
 200             dy30             = _mm_sub_pd(iy3,jy0);
 201             dz30             = _mm_sub_pd(iz3,jz0);
 202
 203             /* Calculate squared distance and things based on it */
 204             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 205             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 206             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 207             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 208
 209             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 210             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 211             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 212
 213             rinvsq00         = gmx_mm_inv_pd(rsq00);
 214             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 215             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 216             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 217
 218             /* Load parameters for j particles */
 219             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 220             vdwjidx0A        = 2*vdwtype[jnrA+0];
 221             vdwjidx0B        = 2*vdwtype[jnrB+0];
 222
 223             fjx0             = _mm_setzero_pd();
 224             fjy0             = _mm_setzero_pd();
 225             fjz0             = _mm_setzero_pd();
 226
 227             /**************************
 228              * CALCULATE INTERACTIONS *
 229              **************************/
 230
 231             if (gmx_mm_any_lt(rsq00,rcutoff2))
 232             {
 233
 234             /* Compute parameters for interactions between i and j atoms */
 235             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 236                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 237
 238             /* LENNARD-JONES DISPERSION/REPULSION */
 239
 240             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 241             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 242             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 243             vvdw             = _mm_msub_pd(_mm_nmacc_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6),vvdw12),one_twelfth,
 244                                            _mm_mul_pd(_mm_nmacc_pd( c6_00,sh_vdw_invrcut6,vvdw6),one_sixth));
 245             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 246
 247             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 248
 249             /* Update potential sum for this i atom from the interaction with this j atom. */
 250             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 251             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 252
 253             fscal            = fvdw;
 254
 255             fscal            = _mm_and_pd(fscal,cutoff_mask);
 256
 257             /* Update vectorial force */
 258             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 259             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 260             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 261
 262             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 263             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 264             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 265
 266             }
 267
 268             /**************************
 269              * CALCULATE INTERACTIONS *
 270              **************************/
 271
 272             if (gmx_mm_any_lt(rsq10,rcutoff2))
 273             {
 274
 275             r10              = _mm_mul_pd(rsq10,rinv10);
 276
 277             /* Compute parameters for interactions between i and j atoms */
 278             qq10             = _mm_mul_pd(iq1,jq0);
 279
 280             /* EWALD ELECTROSTATICS */
 281
 282             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 283             ewrt             = _mm_mul_pd(r10,ewtabscale);
 284             ewitab           = _mm_cvttpd_epi32(ewrt);
 285 #ifdef __XOP__
 286             eweps            = _mm_frcz_pd(ewrt);
 287 #else
 288             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 289 #endif
 290             twoeweps         = _mm_add_pd(eweps,eweps);
 291             ewitab           = _mm_slli_epi32(ewitab,2);
 292             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 293             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 294             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 295             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 296             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 297             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 298             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 299             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 300             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 301             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 302
 303             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 304
 305             /* Update potential sum for this i atom from the interaction with this j atom. */
 306             velec            = _mm_and_pd(velec,cutoff_mask);
 307             velecsum         = _mm_add_pd(velecsum,velec);
 308
 309             fscal            = felec;
 310
 311             fscal            = _mm_and_pd(fscal,cutoff_mask);
 312
 313             /* Update vectorial force */
 314             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 315             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 316             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 317
 318             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 319             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 320             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 321
 322             }
 323
 324             /**************************
 325              * CALCULATE INTERACTIONS *
 326              **************************/
 327
 328             if (gmx_mm_any_lt(rsq20,rcutoff2))
 329             {
 330
 331             r20              = _mm_mul_pd(rsq20,rinv20);
 332
 333             /* Compute parameters for interactions between i and j atoms */
 334             qq20             = _mm_mul_pd(iq2,jq0);
 335
 336             /* EWALD ELECTROSTATICS */
 337
 338             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 339             ewrt             = _mm_mul_pd(r20,ewtabscale);
 340             ewitab           = _mm_cvttpd_epi32(ewrt);
 341 #ifdef __XOP__
 342             eweps            = _mm_frcz_pd(ewrt);
 343 #else
 344             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 345 #endif
 346             twoeweps         = _mm_add_pd(eweps,eweps);
 347             ewitab           = _mm_slli_epi32(ewitab,2);
 348             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 349             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 350             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 351             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 352             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 353             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 354             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 355             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 356             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 357             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 358
 359             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 360
 361             /* Update potential sum for this i atom from the interaction with this j atom. */
 362             velec            = _mm_and_pd(velec,cutoff_mask);
 363             velecsum         = _mm_add_pd(velecsum,velec);
 364
 365             fscal            = felec;
 366
 367             fscal            = _mm_and_pd(fscal,cutoff_mask);
 368
 369             /* Update vectorial force */
 370             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 371             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 372             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 373
 374             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 375             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 376             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 377
 378             }
 379
 380             /**************************
 381              * CALCULATE INTERACTIONS *
 382              **************************/
 383
 384             if (gmx_mm_any_lt(rsq30,rcutoff2))
 385             {
 386
 387             r30              = _mm_mul_pd(rsq30,rinv30);
 388
 389             /* Compute parameters for interactions between i and j atoms */
 390             qq30             = _mm_mul_pd(iq3,jq0);
 391
 392             /* EWALD ELECTROSTATICS */
 393
 394             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 395             ewrt             = _mm_mul_pd(r30,ewtabscale);
 396             ewitab           = _mm_cvttpd_epi32(ewrt);
 397 #ifdef __XOP__
 398             eweps            = _mm_frcz_pd(ewrt);
 399 #else
 400             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 401 #endif
 402             twoeweps         = _mm_add_pd(eweps,eweps);
 403             ewitab           = _mm_slli_epi32(ewitab,2);
 404             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 405             ewtabD           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,1) );
 406             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 407             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 408             ewtabFn          = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,1) +2);
 409             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 410             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 411             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 412             velec            = _mm_mul_pd(qq30,_mm_sub_pd(_mm_sub_pd(rinv30,sh_ewald),velec));
 413             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
 414
 415             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
 416
 417             /* Update potential sum for this i atom from the interaction with this j atom. */
 418             velec            = _mm_and_pd(velec,cutoff_mask);
 419             velecsum         = _mm_add_pd(velecsum,velec);
 420
 421             fscal            = felec;
 422
 423             fscal            = _mm_and_pd(fscal,cutoff_mask);
 424
 425             /* Update vectorial force */
 426             fix3             = _mm_macc_pd(dx30,fscal,fix3);
 427             fiy3             = _mm_macc_pd(dy30,fscal,fiy3);
 428             fiz3             = _mm_macc_pd(dz30,fscal,fiz3);
 429
 430             fjx0             = _mm_macc_pd(dx30,fscal,fjx0);
 431             fjy0             = _mm_macc_pd(dy30,fscal,fjy0);
 432             fjz0             = _mm_macc_pd(dz30,fscal,fjz0);
 433
 434             }
 435
 436             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 437
 438             /* Inner loop uses 194 flops */
 439         }
 440
 441         if(jidx<j_index_end)
 442         {
 443
 444             jnrA             = jjnr[jidx];
 445             j_coord_offsetA  = DIM*jnrA;
 446
 447             /* load j atom coordinates */
 448             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 449                                               &jx0,&jy0,&jz0);
 450
 451             /* Calculate displacement vector */
 452             dx00             = _mm_sub_pd(ix0,jx0);
 453             dy00             = _mm_sub_pd(iy0,jy0);
 454             dz00             = _mm_sub_pd(iz0,jz0);
 455             dx10             = _mm_sub_pd(ix1,jx0);
 456             dy10             = _mm_sub_pd(iy1,jy0);
 457             dz10             = _mm_sub_pd(iz1,jz0);
 458             dx20             = _mm_sub_pd(ix2,jx0);
 459             dy20             = _mm_sub_pd(iy2,jy0);
 460             dz20             = _mm_sub_pd(iz2,jz0);
 461             dx30             = _mm_sub_pd(ix3,jx0);
 462             dy30             = _mm_sub_pd(iy3,jy0);
 463             dz30             = _mm_sub_pd(iz3,jz0);
 464
 465             /* Calculate squared distance and things based on it */
 466             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 467             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 468             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 469             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 470
 471             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 472             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 473             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 474
 475             rinvsq00         = gmx_mm_inv_pd(rsq00);
 476             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 477             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 478             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 479
 480             /* Load parameters for j particles */
 481             jq0              = _mm_load_sd(charge+jnrA+0);
 482             vdwjidx0A        = 2*vdwtype[jnrA+0];
 483
 484             fjx0             = _mm_setzero_pd();
 485             fjy0             = _mm_setzero_pd();
 486             fjz0             = _mm_setzero_pd();
 487
 488             /**************************
 489              * CALCULATE INTERACTIONS *
 490              **************************/
 491
 492             if (gmx_mm_any_lt(rsq00,rcutoff2))
 493             {
 494
 495             /* Compute parameters for interactions between i and j atoms */
 496             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 497
 498             /* LENNARD-JONES DISPERSION/REPULSION */
 499
 500             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 501             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 502             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 503             vvdw             = _mm_msub_pd(_mm_nmacc_pd(c12_00,_mm_mul_pd(sh_vdw_invrcut6,sh_vdw_invrcut6),vvdw12),one_twelfth,
 504                                            _mm_mul_pd(_mm_nmacc_pd( c6_00,sh_vdw_invrcut6,vvdw6),one_sixth));
 505             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 506
 507             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 508
 509             /* Update potential sum for this i atom from the interaction with this j atom. */
 510             vvdw             = _mm_and_pd(vvdw,cutoff_mask);
 511             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 512             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 513
 514             fscal            = fvdw;
 515
 516             fscal            = _mm_and_pd(fscal,cutoff_mask);
 517
 518             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 519
 520             /* Update vectorial force */
 521             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 522             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 523             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 524
 525             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 526             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 527             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 528
 529             }
 530
 531             /**************************
 532              * CALCULATE INTERACTIONS *
 533              **************************/
 534
 535             if (gmx_mm_any_lt(rsq10,rcutoff2))
 536             {
 537
 538             r10              = _mm_mul_pd(rsq10,rinv10);
 539
 540             /* Compute parameters for interactions between i and j atoms */
 541             qq10             = _mm_mul_pd(iq1,jq0);
 542
 543             /* EWALD ELECTROSTATICS */
 544
 545             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 546             ewrt             = _mm_mul_pd(r10,ewtabscale);
 547             ewitab           = _mm_cvttpd_epi32(ewrt);
 548 #ifdef __XOP__
 549             eweps            = _mm_frcz_pd(ewrt);
 550 #else
 551             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 552 #endif
 553             twoeweps         = _mm_add_pd(eweps,eweps);
 554             ewitab           = _mm_slli_epi32(ewitab,2);
 555             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 556             ewtabD           = _mm_setzero_pd();
 557             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 558             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 559             ewtabFn          = _mm_setzero_pd();
 560             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 561             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 562             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 563             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_sub_pd(rinv10,sh_ewald),velec));
 564             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 565
 566             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 567
 568             /* Update potential sum for this i atom from the interaction with this j atom. */
 569             velec            = _mm_and_pd(velec,cutoff_mask);
 570             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 571             velecsum         = _mm_add_pd(velecsum,velec);
 572
 573             fscal            = felec;
 574
 575             fscal            = _mm_and_pd(fscal,cutoff_mask);
 576
 577             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 578
 579             /* Update vectorial force */
 580             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 581             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 582             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 583
 584             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 585             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 586             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 587
 588             }
 589
 590             /**************************
 591              * CALCULATE INTERACTIONS *
 592              **************************/
 593
 594             if (gmx_mm_any_lt(rsq20,rcutoff2))
 595             {
 596
 597             r20              = _mm_mul_pd(rsq20,rinv20);
 598
 599             /* Compute parameters for interactions between i and j atoms */
 600             qq20             = _mm_mul_pd(iq2,jq0);
 601
 602             /* EWALD ELECTROSTATICS */
 603
 604             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 605             ewrt             = _mm_mul_pd(r20,ewtabscale);
 606             ewitab           = _mm_cvttpd_epi32(ewrt);
 607 #ifdef __XOP__
 608             eweps            = _mm_frcz_pd(ewrt);
 609 #else
 610             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 611 #endif
 612             twoeweps         = _mm_add_pd(eweps,eweps);
 613             ewitab           = _mm_slli_epi32(ewitab,2);
 614             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 615             ewtabD           = _mm_setzero_pd();
 616             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 617             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 618             ewtabFn          = _mm_setzero_pd();
 619             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 620             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 621             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 622             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_sub_pd(rinv20,sh_ewald),velec));
 623             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
 624
 625             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
 626
 627             /* Update potential sum for this i atom from the interaction with this j atom. */
 628             velec            = _mm_and_pd(velec,cutoff_mask);
 629             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 630             velecsum         = _mm_add_pd(velecsum,velec);
 631
 632             fscal            = felec;
 633
 634             fscal            = _mm_and_pd(fscal,cutoff_mask);
 635
 636             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 637
 638             /* Update vectorial force */
 639             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 640             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 641             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 642
 643             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 644             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 645             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 646
 647             }
 648
 649             /**************************
 650              * CALCULATE INTERACTIONS *
 651              **************************/
 652
 653             if (gmx_mm_any_lt(rsq30,rcutoff2))
 654             {
 655
 656             r30              = _mm_mul_pd(rsq30,rinv30);
 657
 658             /* Compute parameters for interactions between i and j atoms */
 659             qq30             = _mm_mul_pd(iq3,jq0);
 660
 661             /* EWALD ELECTROSTATICS */
 662
 663             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 664             ewrt             = _mm_mul_pd(r30,ewtabscale);
 665             ewitab           = _mm_cvttpd_epi32(ewrt);
 666 #ifdef __XOP__
 667             eweps            = _mm_frcz_pd(ewrt);
 668 #else
 669             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 670 #endif
 671             twoeweps         = _mm_add_pd(eweps,eweps);
 672             ewitab           = _mm_slli_epi32(ewitab,2);
 673             ewtabF           = _mm_load_pd( ewtab + _mm_extract_epi32(ewitab,0) );
 674             ewtabD           = _mm_setzero_pd();
 675             GMX_MM_TRANSPOSE2_PD(ewtabF,ewtabD);
 676             ewtabV           = _mm_load_sd( ewtab + _mm_extract_epi32(ewitab,0) +2);
 677             ewtabFn          = _mm_setzero_pd();
 678             GMX_MM_TRANSPOSE2_PD(ewtabV,ewtabFn);
 679             felec            = _mm_macc_pd(eweps,ewtabD,ewtabF);
 680             velec            = _mm_nmacc_pd(_mm_mul_pd(ewtabhalfspace,eweps) ,_mm_add_pd(ewtabF,felec), ewtabV);
 681             velec            = _mm_mul_pd(qq30,_mm_sub_pd(_mm_sub_pd(rinv30,sh_ewald),velec));
 682             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
 683
 684             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
 685
 686             /* Update potential sum for this i atom from the interaction with this j atom. */
 687             velec            = _mm_and_pd(velec,cutoff_mask);
 688             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 689             velecsum         = _mm_add_pd(velecsum,velec);
 690
 691             fscal            = felec;
 692
 693             fscal            = _mm_and_pd(fscal,cutoff_mask);
 694
 695             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 696
 697             /* Update vectorial force */
 698             fix3             = _mm_macc_pd(dx30,fscal,fix3);
 699             fiy3             = _mm_macc_pd(dy30,fscal,fiy3);
 700             fiz3             = _mm_macc_pd(dz30,fscal,fiz3);
 701
 702             fjx0             = _mm_macc_pd(dx30,fscal,fjx0);
 703             fjy0             = _mm_macc_pd(dy30,fscal,fjy0);
 704             fjz0             = _mm_macc_pd(dz30,fscal,fjz0);
 705
 706             }
 707
 708             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 709
 710             /* Inner loop uses 194 flops */
 711         }
 712
 713         /* End of innermost loop */
 714
 715         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
 716                                               f+i_coord_offset,fshift+i_shift_offset);
 717
 718         ggid                        = gid[iidx];
 719         /* Update potential energies */
 720         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 721         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
 722
 723         /* Increment number of inner iterations */
 724         inneriter                  += j_index_end - j_index_start;
 725
 726         /* Outer loop uses 26 flops */
 727     }
 728
 729     /* Increment number of outer iterations */
 730     outeriter        += nri;
 731
 732     /* Update outer/inner flops */
 733
 734     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*26 + inneriter*194);
 735 }
 736 /*
 737  * Gromacs nonbonded kernel:   nb_kernel_ElecEwSh_VdwLJSh_GeomW4P1_F_avx_128_fma_double
 738  * Electrostatics interaction: Ewald
 739  * VdW interaction:            LennardJones
 740  * Geometry:                   Water4-Particle
 741  * Calculate force/pot:        Force
 742  */
 743 void
 744 nb_kernel_ElecEwSh_VdwLJSh_GeomW4P1_F_avx_128_fma_double
 745                     (t_nblist * gmx_restrict                nlist,
 746                      rvec * gmx_restrict                    xx,
 747                      rvec * gmx_restrict                    ff,
 748                      t_forcerec * gmx_restrict              fr,
 749                      t_mdatoms * gmx_restrict               mdatoms,
 750                      nb_kernel_data_t * gmx_restrict        kernel_data,
 751                      t_nrnb * gmx_restrict                  nrnb)
 752 {
 753     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 754      * just 0 for non-waters.
 755      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 756      * jnr indices corresponding to data put in the four positions in the SIMD register.
 757      */
 758     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 759     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 760     int              jnrA,jnrB;
 761     int              j_coord_offsetA,j_coord_offsetB;
 762     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 763     real             rcutoff_scalar;
 764     real             *shiftvec,*fshift,*x,*f;
 765     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 766     int              vdwioffset0;
 767     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 768     int              vdwioffset1;
 769     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 770     int              vdwioffset2;
 771     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 772     int              vdwioffset3;
 773     __m128d          ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 774     int              vdwjidx0A,vdwjidx0B;
 775     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 776     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 777     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 778     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 779     __m128d          dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30;
 780     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 781     real             *charge;
 782     int              nvdwtype;
 783     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 784     int              *vdwtype;
 785     real             *vdwparam;
 786     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 787     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 788     __m128i          ewitab;
 789     __m128d          ewtabscale,eweps,twoeweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV;
 790     real             *ewtab;
 791     __m128d          dummy_mask,cutoff_mask;
 792     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 793     __m128d          one     = _mm_set1_pd(1.0);
 794     __m128d          two     = _mm_set1_pd(2.0);
 795     x                = xx[0];
 796     f                = ff[0];
 797
 798     nri              = nlist->nri;
 799     iinr             = nlist->iinr;
 800     jindex           = nlist->jindex;
 801     jjnr             = nlist->jjnr;
 802     shiftidx         = nlist->shift;
 803     gid              = nlist->gid;
 804     shiftvec         = fr->shift_vec[0];
 805     fshift           = fr->fshift[0];
 806     facel            = _mm_set1_pd(fr->epsfac);
 807     charge           = mdatoms->chargeA;
 808     nvdwtype         = fr->ntype;
 809     vdwparam         = fr->nbfp;
 810     vdwtype          = mdatoms->typeA;
 811
 812     sh_ewald         = _mm_set1_pd(fr->ic->sh_ewald);
 813     ewtab            = fr->ic->tabq_coul_F;
 814     ewtabscale       = _mm_set1_pd(fr->ic->tabq_scale);
 815     ewtabhalfspace   = _mm_set1_pd(0.5/fr->ic->tabq_scale);
 816
 817     /* Setup water-specific parameters */
 818     inr              = nlist->iinr[0];
 819     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 820     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 821     iq3              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+3]));
 822     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 823
 824     /* When we use explicit cutoffs the value must be identical for elec and VdW, so use elec as an arbitrary choice */
 825     rcutoff_scalar   = fr->rcoulomb;
 826     rcutoff          = _mm_set1_pd(rcutoff_scalar);
 827     rcutoff2         = _mm_mul_pd(rcutoff,rcutoff);
 828
 829     sh_vdw_invrcut6  = _mm_set1_pd(fr->ic->sh_invrc6);
 830     rvdw             = _mm_set1_pd(fr->rvdw);
 831
 832     /* Avoid stupid compiler warnings */
 833     jnrA = jnrB = 0;
 834     j_coord_offsetA = 0;
 835     j_coord_offsetB = 0;
 836
 837     outeriter        = 0;
 838     inneriter        = 0;
 839
 840     /* Start outer loop over neighborlists */
 841     for(iidx=0; iidx<nri; iidx++)
 842     {
 843         /* Load shift vector for this list */
 844         i_shift_offset   = DIM*shiftidx[iidx];
 845
 846         /* Load limits for loop over neighbors */
 847         j_index_start    = jindex[iidx];
 848         j_index_end      = jindex[iidx+1];
 849
 850         /* Get outer coordinate index */
 851         inr              = iinr[iidx];
 852         i_coord_offset   = DIM*inr;
 853
 854         /* Load i particle coords and add shift vector */
 855         gmx_mm_load_shift_and_4rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 856                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3);
 857
 858         fix0             = _mm_setzero_pd();
 859         fiy0             = _mm_setzero_pd();
 860         fiz0             = _mm_setzero_pd();
 861         fix1             = _mm_setzero_pd();
 862         fiy1             = _mm_setzero_pd();
 863         fiz1             = _mm_setzero_pd();
 864         fix2             = _mm_setzero_pd();
 865         fiy2             = _mm_setzero_pd();
 866         fiz2             = _mm_setzero_pd();
 867         fix3             = _mm_setzero_pd();
 868         fiy3             = _mm_setzero_pd();
 869         fiz3             = _mm_setzero_pd();
 870
 871         /* Start inner kernel loop */
 872         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 873         {
 874
 875             /* Get j neighbor index, and coordinate index */
 876             jnrA             = jjnr[jidx];
 877             jnrB             = jjnr[jidx+1];
 878             j_coord_offsetA  = DIM*jnrA;
 879             j_coord_offsetB  = DIM*jnrB;
 880
 881             /* load j atom coordinates */
 882             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 883                                               &jx0,&jy0,&jz0);
 884
 885             /* Calculate displacement vector */
 886             dx00             = _mm_sub_pd(ix0,jx0);
 887             dy00             = _mm_sub_pd(iy0,jy0);
 888             dz00             = _mm_sub_pd(iz0,jz0);
 889             dx10             = _mm_sub_pd(ix1,jx0);
 890             dy10             = _mm_sub_pd(iy1,jy0);
 891             dz10             = _mm_sub_pd(iz1,jz0);
 892             dx20             = _mm_sub_pd(ix2,jx0);
 893             dy20             = _mm_sub_pd(iy2,jy0);
 894             dz20             = _mm_sub_pd(iz2,jz0);
 895             dx30             = _mm_sub_pd(ix3,jx0);
 896             dy30             = _mm_sub_pd(iy3,jy0);
 897             dz30             = _mm_sub_pd(iz3,jz0);
 898
 899             /* Calculate squared distance and things based on it */
 900             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 901             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 902             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 903             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
 904
 905             rinv10           = gmx_mm_invsqrt_pd(rsq10);
 906             rinv20           = gmx_mm_invsqrt_pd(rsq20);
 907             rinv30           = gmx_mm_invsqrt_pd(rsq30);
 908
 909             rinvsq00         = gmx_mm_inv_pd(rsq00);
 910             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 911             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 912             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
 913
 914             /* Load parameters for j particles */
 915             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 916             vdwjidx0A        = 2*vdwtype[jnrA+0];
 917             vdwjidx0B        = 2*vdwtype[jnrB+0];
 918
 919             fjx0             = _mm_setzero_pd();
 920             fjy0             = _mm_setzero_pd();
 921             fjz0             = _mm_setzero_pd();
 922
 923             /**************************
 924              * CALCULATE INTERACTIONS *
 925              **************************/
 926
 927             if (gmx_mm_any_lt(rsq00,rcutoff2))
 928             {
 929
 930             /* Compute parameters for interactions between i and j atoms */
 931             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 932                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 933
 934             /* LENNARD-JONES DISPERSION/REPULSION */
 935
 936             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 937             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
 938
 939             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
 940
 941             fscal            = fvdw;
 942
 943             fscal            = _mm_and_pd(fscal,cutoff_mask);
 944
 945             /* Update vectorial force */
 946             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 947             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 948             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 949
 950             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 951             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 952             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 953
 954             }
 955
 956             /**************************
 957              * CALCULATE INTERACTIONS *
 958              **************************/
 959
 960             if (gmx_mm_any_lt(rsq10,rcutoff2))
 961             {
 962
 963             r10              = _mm_mul_pd(rsq10,rinv10);
 964
 965             /* Compute parameters for interactions between i and j atoms */
 966             qq10             = _mm_mul_pd(iq1,jq0);
 967
 968             /* EWALD ELECTROSTATICS */
 969
 970             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
 971             ewrt             = _mm_mul_pd(r10,ewtabscale);
 972             ewitab           = _mm_cvttpd_epi32(ewrt);
 973 #ifdef __XOP__
 974             eweps            = _mm_frcz_pd(ewrt);
 975 #else
 976             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
 977 #endif
 978             twoeweps         = _mm_add_pd(eweps,eweps);
 979             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
 980                                          &ewtabF,&ewtabFn);
 981             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
 982             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
 983
 984             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
 985
 986             fscal            = felec;
 987
 988             fscal            = _mm_and_pd(fscal,cutoff_mask);
 989
 990             /* Update vectorial force */
 991             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 992             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 993             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 994
 995             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 996             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 997             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 998
 999             }
1000
1001             /**************************
1002              * CALCULATE INTERACTIONS *
1003              **************************/
1004
1005             if (gmx_mm_any_lt(rsq20,rcutoff2))
1006             {
1007
1008             r20              = _mm_mul_pd(rsq20,rinv20);
1009
1010             /* Compute parameters for interactions between i and j atoms */
1011             qq20             = _mm_mul_pd(iq2,jq0);
1012
1013             /* EWALD ELECTROSTATICS */
1014
1015             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1016             ewrt             = _mm_mul_pd(r20,ewtabscale);
1017             ewitab           = _mm_cvttpd_epi32(ewrt);
1018 #ifdef __XOP__
1019             eweps            = _mm_frcz_pd(ewrt);
1020 #else
1021             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1022 #endif
1023             twoeweps         = _mm_add_pd(eweps,eweps);
1024             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1025                                          &ewtabF,&ewtabFn);
1026             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1027             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1028
1029             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
1030
1031             fscal            = felec;
1032
1033             fscal            = _mm_and_pd(fscal,cutoff_mask);
1034
1035             /* Update vectorial force */
1036             fix2             = _mm_macc_pd(dx20,fscal,fix2);
1037             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
1038             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
1039
1040             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
1041             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
1042             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
1043
1044             }
1045
1046             /**************************
1047              * CALCULATE INTERACTIONS *
1048              **************************/
1049
1050             if (gmx_mm_any_lt(rsq30,rcutoff2))
1051             {
1052
1053             r30              = _mm_mul_pd(rsq30,rinv30);
1054
1055             /* Compute parameters for interactions between i and j atoms */
1056             qq30             = _mm_mul_pd(iq3,jq0);
1057
1058             /* EWALD ELECTROSTATICS */
1059
1060             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1061             ewrt             = _mm_mul_pd(r30,ewtabscale);
1062             ewitab           = _mm_cvttpd_epi32(ewrt);
1063 #ifdef __XOP__
1064             eweps            = _mm_frcz_pd(ewrt);
1065 #else
1066             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1067 #endif
1068             twoeweps         = _mm_add_pd(eweps,eweps);
1069             gmx_mm_load_2pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),ewtab+_mm_extract_epi32(ewitab,1),
1070                                          &ewtabF,&ewtabFn);
1071             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1072             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
1073
1074             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
1075
1076             fscal            = felec;
1077
1078             fscal            = _mm_and_pd(fscal,cutoff_mask);
1079
1080             /* Update vectorial force */
1081             fix3             = _mm_macc_pd(dx30,fscal,fix3);
1082             fiy3             = _mm_macc_pd(dy30,fscal,fiy3);
1083             fiz3             = _mm_macc_pd(dz30,fscal,fiz3);
1084
1085             fjx0             = _mm_macc_pd(dx30,fscal,fjx0);
1086             fjy0             = _mm_macc_pd(dy30,fscal,fjy0);
1087             fjz0             = _mm_macc_pd(dz30,fscal,fjz0);
1088
1089             }
1090
1091             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
1092
1093             /* Inner loop uses 162 flops */
1094         }
1095
1096         if(jidx<j_index_end)
1097         {
1098
1099             jnrA             = jjnr[jidx];
1100             j_coord_offsetA  = DIM*jnrA;
1101
1102             /* load j atom coordinates */
1103             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
1104                                               &jx0,&jy0,&jz0);
1105
1106             /* Calculate displacement vector */
1107             dx00             = _mm_sub_pd(ix0,jx0);
1108             dy00             = _mm_sub_pd(iy0,jy0);
1109             dz00             = _mm_sub_pd(iz0,jz0);
1110             dx10             = _mm_sub_pd(ix1,jx0);
1111             dy10             = _mm_sub_pd(iy1,jy0);
1112             dz10             = _mm_sub_pd(iz1,jz0);
1113             dx20             = _mm_sub_pd(ix2,jx0);
1114             dy20             = _mm_sub_pd(iy2,jy0);
1115             dz20             = _mm_sub_pd(iz2,jz0);
1116             dx30             = _mm_sub_pd(ix3,jx0);
1117             dy30             = _mm_sub_pd(iy3,jy0);
1118             dz30             = _mm_sub_pd(iz3,jz0);
1119
1120             /* Calculate squared distance and things based on it */
1121             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
1122             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
1123             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
1124             rsq30            = gmx_mm_calc_rsq_pd(dx30,dy30,dz30);
1125
1126             rinv10           = gmx_mm_invsqrt_pd(rsq10);
1127             rinv20           = gmx_mm_invsqrt_pd(rsq20);
1128             rinv30           = gmx_mm_invsqrt_pd(rsq30);
1129
1130             rinvsq00         = gmx_mm_inv_pd(rsq00);
1131             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
1132             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
1133             rinvsq30         = _mm_mul_pd(rinv30,rinv30);
1134
1135             /* Load parameters for j particles */
1136             jq0              = _mm_load_sd(charge+jnrA+0);
1137             vdwjidx0A        = 2*vdwtype[jnrA+0];
1138
1139             fjx0             = _mm_setzero_pd();
1140             fjy0             = _mm_setzero_pd();
1141             fjz0             = _mm_setzero_pd();
1142
1143             /**************************
1144              * CALCULATE INTERACTIONS *
1145              **************************/
1146
1147             if (gmx_mm_any_lt(rsq00,rcutoff2))
1148             {
1149
1150             /* Compute parameters for interactions between i and j atoms */
1151             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
1152
1153             /* LENNARD-JONES DISPERSION/REPULSION */
1154
1155             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
1156             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
1157
1158             cutoff_mask      = _mm_cmplt_pd(rsq00,rcutoff2);
1159
1160             fscal            = fvdw;
1161
1162             fscal            = _mm_and_pd(fscal,cutoff_mask);
1163
1164             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1165
1166             /* Update vectorial force */
1167             fix0             = _mm_macc_pd(dx00,fscal,fix0);
1168             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
1169             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
1170
1171             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
1172             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
1173             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
1174
1175             }
1176
1177             /**************************
1178              * CALCULATE INTERACTIONS *
1179              **************************/
1180
1181             if (gmx_mm_any_lt(rsq10,rcutoff2))
1182             {
1183
1184             r10              = _mm_mul_pd(rsq10,rinv10);
1185
1186             /* Compute parameters for interactions between i and j atoms */
1187             qq10             = _mm_mul_pd(iq1,jq0);
1188
1189             /* EWALD ELECTROSTATICS */
1190
1191             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1192             ewrt             = _mm_mul_pd(r10,ewtabscale);
1193             ewitab           = _mm_cvttpd_epi32(ewrt);
1194 #ifdef __XOP__
1195             eweps            = _mm_frcz_pd(ewrt);
1196 #else
1197             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1198 #endif
1199             twoeweps         = _mm_add_pd(eweps,eweps);
1200             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1201             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1202             felec            = _mm_mul_pd(_mm_mul_pd(qq10,rinv10),_mm_sub_pd(rinvsq10,felec));
1203
1204             cutoff_mask      = _mm_cmplt_pd(rsq10,rcutoff2);
1205
1206             fscal            = felec;
1207
1208             fscal            = _mm_and_pd(fscal,cutoff_mask);
1209
1210             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1211
1212             /* Update vectorial force */
1213             fix1             = _mm_macc_pd(dx10,fscal,fix1);
1214             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
1215             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
1216
1217             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
1218             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
1219             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
1220
1221             }
1222
1223             /**************************
1224              * CALCULATE INTERACTIONS *
1225              **************************/
1226
1227             if (gmx_mm_any_lt(rsq20,rcutoff2))
1228             {
1229
1230             r20              = _mm_mul_pd(rsq20,rinv20);
1231
1232             /* Compute parameters for interactions between i and j atoms */
1233             qq20             = _mm_mul_pd(iq2,jq0);
1234
1235             /* EWALD ELECTROSTATICS */
1236
1237             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1238             ewrt             = _mm_mul_pd(r20,ewtabscale);
1239             ewitab           = _mm_cvttpd_epi32(ewrt);
1240 #ifdef __XOP__
1241             eweps            = _mm_frcz_pd(ewrt);
1242 #else
1243             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1244 #endif
1245             twoeweps         = _mm_add_pd(eweps,eweps);
1246             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1247             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1248             felec            = _mm_mul_pd(_mm_mul_pd(qq20,rinv20),_mm_sub_pd(rinvsq20,felec));
1249
1250             cutoff_mask      = _mm_cmplt_pd(rsq20,rcutoff2);
1251
1252             fscal            = felec;
1253
1254             fscal            = _mm_and_pd(fscal,cutoff_mask);
1255
1256             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1257
1258             /* Update vectorial force */
1259             fix2             = _mm_macc_pd(dx20,fscal,fix2);
1260             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
1261             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
1262
1263             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
1264             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
1265             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
1266
1267             }
1268
1269             /**************************
1270              * CALCULATE INTERACTIONS *
1271              **************************/
1272
1273             if (gmx_mm_any_lt(rsq30,rcutoff2))
1274             {
1275
1276             r30              = _mm_mul_pd(rsq30,rinv30);
1277
1278             /* Compute parameters for interactions between i and j atoms */
1279             qq30             = _mm_mul_pd(iq3,jq0);
1280
1281             /* EWALD ELECTROSTATICS */
1282
1283             /* Calculate Ewald table index by multiplying r with scale and truncate to integer */
1284             ewrt             = _mm_mul_pd(r30,ewtabscale);
1285             ewitab           = _mm_cvttpd_epi32(ewrt);
1286 #ifdef __XOP__
1287             eweps            = _mm_frcz_pd(ewrt);
1288 #else
1289             eweps            = _mm_sub_pd(ewrt,_mm_round_pd(ewrt, _MM_FROUND_FLOOR));
1290 #endif
1291             twoeweps         = _mm_add_pd(eweps,eweps);
1292             gmx_mm_load_1pair_swizzle_pd(ewtab+_mm_extract_epi32(ewitab,0),&ewtabF,&ewtabFn);
1293             felec            = _mm_macc_pd(eweps,ewtabFn,_mm_mul_pd( _mm_sub_pd(one,eweps),ewtabF));
1294             felec            = _mm_mul_pd(_mm_mul_pd(qq30,rinv30),_mm_sub_pd(rinvsq30,felec));
1295
1296             cutoff_mask      = _mm_cmplt_pd(rsq30,rcutoff2);
1297
1298             fscal            = felec;
1299
1300             fscal            = _mm_and_pd(fscal,cutoff_mask);
1301
1302             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
1303
1304             /* Update vectorial force */
1305             fix3             = _mm_macc_pd(dx30,fscal,fix3);
1306             fiy3             = _mm_macc_pd(dy30,fscal,fiy3);
1307             fiz3             = _mm_macc_pd(dz30,fscal,fiz3);
1308
1309             fjx0             = _mm_macc_pd(dx30,fscal,fjx0);
1310             fjy0             = _mm_macc_pd(dy30,fscal,fjy0);
1311             fjz0             = _mm_macc_pd(dz30,fscal,fjz0);
1312
1313             }
1314
1315             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
1316
1317             /* Inner loop uses 162 flops */
1318         }
1319
1320         /* End of innermost loop */
1321
1322         gmx_mm_update_iforce_4atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3,
1323                                               f+i_coord_offset,fshift+i_shift_offset);
1324
1325         /* Increment number of inner iterations */
1326         inneriter                  += j_index_end - j_index_start;
1327
1328         /* Outer loop uses 24 flops */
1329     }
1330
1331     /* Increment number of outer iterations */
1332     outeriter        += nri;
1333
1334     /* Update outer/inner flops */
1335
1336     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*24 + inneriter*162);
1337 }