src/gromacs/gmxlib/nonbonded/nb_kernel_avx_256_single/nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_avx_256_single.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS avx_256_single kernel generator.
  37  */
  38 #ifdef HAVE_CONFIG_H
  39 #include <config.h>
  40 #endif
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "types/simple.h"
  46 #include "vec.h"
  47 #include "nrnb.h"
  48
  49 #include "gromacs/simd/math_x86_avx_256_single.h"
  50 #include "kernelutil_x86_avx_256_single.h"
  51
  52 /*
  53  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_VF_avx_256_single
  54  * Electrostatics interaction: CubicSplineTable
  55  * VdW interaction:            CubicSplineTable
  56  * Geometry:                   Particle-Particle
  57  * Calculate force/pot:        PotentialAndForce
  58  */
  59 void
  60 nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_VF_avx_256_single
  61                     (t_nblist                    * gmx_restrict       nlist,
  62                      rvec                        * gmx_restrict          xx,
  63                      rvec                        * gmx_restrict          ff,
  64                      t_forcerec                  * gmx_restrict          fr,
  65                      t_mdatoms                   * gmx_restrict     mdatoms,
  66                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  67                      t_nrnb                      * gmx_restrict        nrnb)
  68 {
  69     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  70      * just 0 for non-waters.
  71      * Suffixes A,B,C,D,E,F,G,H refer to j loop unrolling done with AVX, e.g. for the eight different
  72      * jnr indices corresponding to data put in the four positions in the SIMD register.
  73      */
  74     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  75     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  76     int              jnrA,jnrB,jnrC,jnrD;
  77     int              jnrE,jnrF,jnrG,jnrH;
  78     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  79     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  80     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  81     int              j_coord_offsetE,j_coord_offsetF,j_coord_offsetG,j_coord_offsetH;
  82     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  83     real             rcutoff_scalar;
  84     real             *shiftvec,*fshift,*x,*f;
  85     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD,*fjptrE,*fjptrF,*fjptrG,*fjptrH;
  86     real             scratch[4*DIM];
  87     __m256           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  88     real *           vdwioffsetptr0;
  89     __m256           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  90     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D,vdwjidx0E,vdwjidx0F,vdwjidx0G,vdwjidx0H;
  91     __m256           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  92     __m256           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  93     __m256           velec,felec,velecsum,facel,crf,krf,krf2;
  94     real             *charge;
  95     int              nvdwtype;
  96     __m256           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  97     int              *vdwtype;
  98     real             *vdwparam;
  99     __m256           one_sixth   = _mm256_set1_ps(1.0/6.0);
 100     __m256           one_twelfth = _mm256_set1_ps(1.0/12.0);
 101     __m256i          vfitab;
 102     __m128i          vfitab_lo,vfitab_hi;
 103     __m128i          ifour       = _mm_set1_epi32(4);
 104     __m256           rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 105     real             *vftab;
 106     __m256           dummy_mask,cutoff_mask;
 107     __m256           signbit = _mm256_castsi256_ps( _mm256_set1_epi32(0x80000000) );
 108     __m256           one     = _mm256_set1_ps(1.0);
 109     __m256           two     = _mm256_set1_ps(2.0);
 110     x                = xx[0];
 111     f                = ff[0];
 112
 113     nri              = nlist->nri;
 114     iinr             = nlist->iinr;
 115     jindex           = nlist->jindex;
 116     jjnr             = nlist->jjnr;
 117     shiftidx         = nlist->shift;
 118     gid              = nlist->gid;
 119     shiftvec         = fr->shift_vec[0];
 120     fshift           = fr->fshift[0];
 121     facel            = _mm256_set1_ps(fr->epsfac);
 122     charge           = mdatoms->chargeA;
 123     nvdwtype         = fr->ntype;
 124     vdwparam         = fr->nbfp;
 125     vdwtype          = mdatoms->typeA;
 126
 127     vftab            = kernel_data->table_elec_vdw->data;
 128     vftabscale       = _mm256_set1_ps(kernel_data->table_elec_vdw->scale);
 129
 130     /* Avoid stupid compiler warnings */
 131     jnrA = jnrB = jnrC = jnrD = jnrE = jnrF = jnrG = jnrH = 0;
 132     j_coord_offsetA = 0;
 133     j_coord_offsetB = 0;
 134     j_coord_offsetC = 0;
 135     j_coord_offsetD = 0;
 136     j_coord_offsetE = 0;
 137     j_coord_offsetF = 0;
 138     j_coord_offsetG = 0;
 139     j_coord_offsetH = 0;
 140
 141     outeriter        = 0;
 142     inneriter        = 0;
 143
 144     for(iidx=0;iidx<4*DIM;iidx++)
 145     {
 146         scratch[iidx] = 0.0;
 147     }
 148
 149     /* Start outer loop over neighborlists */
 150     for(iidx=0; iidx<nri; iidx++)
 151     {
 152         /* Load shift vector for this list */
 153         i_shift_offset   = DIM*shiftidx[iidx];
 154
 155         /* Load limits for loop over neighbors */
 156         j_index_start    = jindex[iidx];
 157         j_index_end      = jindex[iidx+1];
 158
 159         /* Get outer coordinate index */
 160         inr              = iinr[iidx];
 161         i_coord_offset   = DIM*inr;
 162
 163         /* Load i particle coords and add shift vector */
 164         gmx_mm256_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 165
 166         fix0             = _mm256_setzero_ps();
 167         fiy0             = _mm256_setzero_ps();
 168         fiz0             = _mm256_setzero_ps();
 169
 170         /* Load parameters for i particles */
 171         iq0              = _mm256_mul_ps(facel,_mm256_set1_ps(charge[inr+0]));
 172         vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 173
 174         /* Reset potential sums */
 175         velecsum         = _mm256_setzero_ps();
 176         vvdwsum          = _mm256_setzero_ps();
 177
 178         /* Start inner kernel loop */
 179         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+7]>=0; jidx+=8)
 180         {
 181
 182             /* Get j neighbor index, and coordinate index */
 183             jnrA             = jjnr[jidx];
 184             jnrB             = jjnr[jidx+1];
 185             jnrC             = jjnr[jidx+2];
 186             jnrD             = jjnr[jidx+3];
 187             jnrE             = jjnr[jidx+4];
 188             jnrF             = jjnr[jidx+5];
 189             jnrG             = jjnr[jidx+6];
 190             jnrH             = jjnr[jidx+7];
 191             j_coord_offsetA  = DIM*jnrA;
 192             j_coord_offsetB  = DIM*jnrB;
 193             j_coord_offsetC  = DIM*jnrC;
 194             j_coord_offsetD  = DIM*jnrD;
 195             j_coord_offsetE  = DIM*jnrE;
 196             j_coord_offsetF  = DIM*jnrF;
 197             j_coord_offsetG  = DIM*jnrG;
 198             j_coord_offsetH  = DIM*jnrH;
 199
 200             /* load j atom coordinates */
 201             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 202                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 203                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 204                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 205                                                  &jx0,&jy0,&jz0);
 206
 207             /* Calculate displacement vector */
 208             dx00             = _mm256_sub_ps(ix0,jx0);
 209             dy00             = _mm256_sub_ps(iy0,jy0);
 210             dz00             = _mm256_sub_ps(iz0,jz0);
 211
 212             /* Calculate squared distance and things based on it */
 213             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 214
 215             rinv00           = gmx_mm256_invsqrt_ps(rsq00);
 216
 217             /* Load parameters for j particles */
 218             jq0              = gmx_mm256_load_8real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 219                                                                  charge+jnrC+0,charge+jnrD+0,
 220                                                                  charge+jnrE+0,charge+jnrF+0,
 221                                                                  charge+jnrG+0,charge+jnrH+0);
 222             vdwjidx0A        = 2*vdwtype[jnrA+0];
 223             vdwjidx0B        = 2*vdwtype[jnrB+0];
 224             vdwjidx0C        = 2*vdwtype[jnrC+0];
 225             vdwjidx0D        = 2*vdwtype[jnrD+0];
 226             vdwjidx0E        = 2*vdwtype[jnrE+0];
 227             vdwjidx0F        = 2*vdwtype[jnrF+0];
 228             vdwjidx0G        = 2*vdwtype[jnrG+0];
 229             vdwjidx0H        = 2*vdwtype[jnrH+0];
 230
 231             /**************************
 232              * CALCULATE INTERACTIONS *
 233              **************************/
 234
 235             r00              = _mm256_mul_ps(rsq00,rinv00);
 236
 237             /* Compute parameters for interactions between i and j atoms */
 238             qq00             = _mm256_mul_ps(iq0,jq0);
 239             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 240                                             vdwioffsetptr0+vdwjidx0B,
 241                                             vdwioffsetptr0+vdwjidx0C,
 242                                             vdwioffsetptr0+vdwjidx0D,
 243                                             vdwioffsetptr0+vdwjidx0E,
 244                                             vdwioffsetptr0+vdwjidx0F,
 245                                             vdwioffsetptr0+vdwjidx0G,
 246                                             vdwioffsetptr0+vdwjidx0H,
 247                                             &c6_00,&c12_00);
 248
 249             /* Calculate table index by multiplying r with table scale and truncate to integer */
 250             rt               = _mm256_mul_ps(r00,vftabscale);
 251             vfitab           = _mm256_cvttps_epi32(rt);
 252             vfeps            = _mm256_sub_ps(rt,_mm256_round_ps(rt, _MM_FROUND_FLOOR));
 253             /*         AVX1 does not support 256-bit integer operations, so now we go to 128-bit mode... */
 254             vfitab_lo        = _mm256_extractf128_si256(vfitab,0x0);
 255             vfitab_hi        = _mm256_extractf128_si256(vfitab,0x1);
 256             vfitab_lo        = _mm_slli_epi32(_mm_add_epi32(vfitab_lo,_mm_slli_epi32(vfitab_lo,1)),2);
 257             vfitab_hi        = _mm_slli_epi32(_mm_add_epi32(vfitab_hi,_mm_slli_epi32(vfitab_hi,1)),2);
 258
 259             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 260             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 261                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 262             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 263                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 264             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 265                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 266             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 267                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 268             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 269             Heps             = _mm256_mul_ps(vfeps,H);
 270             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 271             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 272             velec            = _mm256_mul_ps(qq00,VV);
 273             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 274             felec            = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_mul_ps(qq00,FF),_mm256_mul_ps(vftabscale,rinv00)));
 275
 276             /* CUBIC SPLINE TABLE DISPERSION */
 277             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 278             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 279             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 280                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 281             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 282                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 283             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 284                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 285             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 286                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 287             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 288             Heps             = _mm256_mul_ps(vfeps,H);
 289             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 290             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 291             vvdw6            = _mm256_mul_ps(c6_00,VV);
 292             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 293             fvdw6            = _mm256_mul_ps(c6_00,FF);
 294
 295             /* CUBIC SPLINE TABLE REPULSION */
 296             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 297             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 298             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 299                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 300             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 301                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 302             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 303                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 304             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 305                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 306             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 307             Heps             = _mm256_mul_ps(vfeps,H);
 308             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 309             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 310             vvdw12           = _mm256_mul_ps(c12_00,VV);
 311             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 312             fvdw12           = _mm256_mul_ps(c12_00,FF);
 313             vvdw             = _mm256_add_ps(vvdw12,vvdw6);
 314             fvdw             = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_add_ps(fvdw6,fvdw12),_mm256_mul_ps(vftabscale,rinv00)));
 315
 316             /* Update potential sum for this i atom from the interaction with this j atom. */
 317             velecsum         = _mm256_add_ps(velecsum,velec);
 318             vvdwsum          = _mm256_add_ps(vvdwsum,vvdw);
 319
 320             fscal            = _mm256_add_ps(felec,fvdw);
 321
 322             /* Calculate temporary vectorial force */
 323             tx               = _mm256_mul_ps(fscal,dx00);
 324             ty               = _mm256_mul_ps(fscal,dy00);
 325             tz               = _mm256_mul_ps(fscal,dz00);
 326
 327             /* Update vectorial force */
 328             fix0             = _mm256_add_ps(fix0,tx);
 329             fiy0             = _mm256_add_ps(fiy0,ty);
 330             fiz0             = _mm256_add_ps(fiz0,tz);
 331
 332             fjptrA             = f+j_coord_offsetA;
 333             fjptrB             = f+j_coord_offsetB;
 334             fjptrC             = f+j_coord_offsetC;
 335             fjptrD             = f+j_coord_offsetD;
 336             fjptrE             = f+j_coord_offsetE;
 337             fjptrF             = f+j_coord_offsetF;
 338             fjptrG             = f+j_coord_offsetG;
 339             fjptrH             = f+j_coord_offsetH;
 340             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 341
 342             /* Inner loop uses 73 flops */
 343         }
 344
 345         if(jidx<j_index_end)
 346         {
 347
 348             /* Get j neighbor index, and coordinate index */
 349             jnrlistA         = jjnr[jidx];
 350             jnrlistB         = jjnr[jidx+1];
 351             jnrlistC         = jjnr[jidx+2];
 352             jnrlistD         = jjnr[jidx+3];
 353             jnrlistE         = jjnr[jidx+4];
 354             jnrlistF         = jjnr[jidx+5];
 355             jnrlistG         = jjnr[jidx+6];
 356             jnrlistH         = jjnr[jidx+7];
 357             /* Sign of each element will be negative for non-real atoms.
 358              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 359              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 360              */
 361             dummy_mask = gmx_mm256_set_m128(gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx+4)),_mm_setzero_si128())),
 362                                             gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())));
 363
 364             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 365             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 366             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 367             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 368             jnrE       = (jnrlistE>=0) ? jnrlistE : 0;
 369             jnrF       = (jnrlistF>=0) ? jnrlistF : 0;
 370             jnrG       = (jnrlistG>=0) ? jnrlistG : 0;
 371             jnrH       = (jnrlistH>=0) ? jnrlistH : 0;
 372             j_coord_offsetA  = DIM*jnrA;
 373             j_coord_offsetB  = DIM*jnrB;
 374             j_coord_offsetC  = DIM*jnrC;
 375             j_coord_offsetD  = DIM*jnrD;
 376             j_coord_offsetE  = DIM*jnrE;
 377             j_coord_offsetF  = DIM*jnrF;
 378             j_coord_offsetG  = DIM*jnrG;
 379             j_coord_offsetH  = DIM*jnrH;
 380
 381             /* load j atom coordinates */
 382             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 383                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 384                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 385                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 386                                                  &jx0,&jy0,&jz0);
 387
 388             /* Calculate displacement vector */
 389             dx00             = _mm256_sub_ps(ix0,jx0);
 390             dy00             = _mm256_sub_ps(iy0,jy0);
 391             dz00             = _mm256_sub_ps(iz0,jz0);
 392
 393             /* Calculate squared distance and things based on it */
 394             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 395
 396             rinv00           = gmx_mm256_invsqrt_ps(rsq00);
 397
 398             /* Load parameters for j particles */
 399             jq0              = gmx_mm256_load_8real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 400                                                                  charge+jnrC+0,charge+jnrD+0,
 401                                                                  charge+jnrE+0,charge+jnrF+0,
 402                                                                  charge+jnrG+0,charge+jnrH+0);
 403             vdwjidx0A        = 2*vdwtype[jnrA+0];
 404             vdwjidx0B        = 2*vdwtype[jnrB+0];
 405             vdwjidx0C        = 2*vdwtype[jnrC+0];
 406             vdwjidx0D        = 2*vdwtype[jnrD+0];
 407             vdwjidx0E        = 2*vdwtype[jnrE+0];
 408             vdwjidx0F        = 2*vdwtype[jnrF+0];
 409             vdwjidx0G        = 2*vdwtype[jnrG+0];
 410             vdwjidx0H        = 2*vdwtype[jnrH+0];
 411
 412             /**************************
 413              * CALCULATE INTERACTIONS *
 414              **************************/
 415
 416             r00              = _mm256_mul_ps(rsq00,rinv00);
 417             r00              = _mm256_andnot_ps(dummy_mask,r00);
 418
 419             /* Compute parameters for interactions between i and j atoms */
 420             qq00             = _mm256_mul_ps(iq0,jq0);
 421             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 422                                             vdwioffsetptr0+vdwjidx0B,
 423                                             vdwioffsetptr0+vdwjidx0C,
 424                                             vdwioffsetptr0+vdwjidx0D,
 425                                             vdwioffsetptr0+vdwjidx0E,
 426                                             vdwioffsetptr0+vdwjidx0F,
 427                                             vdwioffsetptr0+vdwjidx0G,
 428                                             vdwioffsetptr0+vdwjidx0H,
 429                                             &c6_00,&c12_00);
 430
 431             /* Calculate table index by multiplying r with table scale and truncate to integer */
 432             rt               = _mm256_mul_ps(r00,vftabscale);
 433             vfitab           = _mm256_cvttps_epi32(rt);
 434             vfeps            = _mm256_sub_ps(rt,_mm256_round_ps(rt, _MM_FROUND_FLOOR));
 435             /*         AVX1 does not support 256-bit integer operations, so now we go to 128-bit mode... */
 436             vfitab_lo        = _mm256_extractf128_si256(vfitab,0x0);
 437             vfitab_hi        = _mm256_extractf128_si256(vfitab,0x1);
 438             vfitab_lo        = _mm_slli_epi32(_mm_add_epi32(vfitab_lo,_mm_slli_epi32(vfitab_lo,1)),2);
 439             vfitab_hi        = _mm_slli_epi32(_mm_add_epi32(vfitab_hi,_mm_slli_epi32(vfitab_hi,1)),2);
 440
 441             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 442             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 443                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 444             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 445                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 446             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 447                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 448             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 449                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 450             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 451             Heps             = _mm256_mul_ps(vfeps,H);
 452             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 453             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 454             velec            = _mm256_mul_ps(qq00,VV);
 455             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 456             felec            = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_mul_ps(qq00,FF),_mm256_mul_ps(vftabscale,rinv00)));
 457
 458             /* CUBIC SPLINE TABLE DISPERSION */
 459             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 460             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 461             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 462                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 463             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 464                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 465             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 466                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 467             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 468                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 469             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 470             Heps             = _mm256_mul_ps(vfeps,H);
 471             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 472             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 473             vvdw6            = _mm256_mul_ps(c6_00,VV);
 474             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 475             fvdw6            = _mm256_mul_ps(c6_00,FF);
 476
 477             /* CUBIC SPLINE TABLE REPULSION */
 478             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 479             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 480             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 481                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 482             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 483                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 484             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 485                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 486             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 487                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 488             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 489             Heps             = _mm256_mul_ps(vfeps,H);
 490             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 491             VV               = _mm256_add_ps(Y,_mm256_mul_ps(vfeps,Fp));
 492             vvdw12           = _mm256_mul_ps(c12_00,VV);
 493             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 494             fvdw12           = _mm256_mul_ps(c12_00,FF);
 495             vvdw             = _mm256_add_ps(vvdw12,vvdw6);
 496             fvdw             = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_add_ps(fvdw6,fvdw12),_mm256_mul_ps(vftabscale,rinv00)));
 497
 498             /* Update potential sum for this i atom from the interaction with this j atom. */
 499             velec            = _mm256_andnot_ps(dummy_mask,velec);
 500             velecsum         = _mm256_add_ps(velecsum,velec);
 501             vvdw             = _mm256_andnot_ps(dummy_mask,vvdw);
 502             vvdwsum          = _mm256_add_ps(vvdwsum,vvdw);
 503
 504             fscal            = _mm256_add_ps(felec,fvdw);
 505
 506             fscal            = _mm256_andnot_ps(dummy_mask,fscal);
 507
 508             /* Calculate temporary vectorial force */
 509             tx               = _mm256_mul_ps(fscal,dx00);
 510             ty               = _mm256_mul_ps(fscal,dy00);
 511             tz               = _mm256_mul_ps(fscal,dz00);
 512
 513             /* Update vectorial force */
 514             fix0             = _mm256_add_ps(fix0,tx);
 515             fiy0             = _mm256_add_ps(fiy0,ty);
 516             fiz0             = _mm256_add_ps(fiz0,tz);
 517
 518             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 519             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 520             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 521             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 522             fjptrE             = (jnrlistE>=0) ? f+j_coord_offsetE : scratch;
 523             fjptrF             = (jnrlistF>=0) ? f+j_coord_offsetF : scratch;
 524             fjptrG             = (jnrlistG>=0) ? f+j_coord_offsetG : scratch;
 525             fjptrH             = (jnrlistH>=0) ? f+j_coord_offsetH : scratch;
 526             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 527
 528             /* Inner loop uses 74 flops */
 529         }
 530
 531         /* End of innermost loop */
 532
 533         gmx_mm256_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
 534                                                  f+i_coord_offset,fshift+i_shift_offset);
 535
 536         ggid                        = gid[iidx];
 537         /* Update potential energies */
 538         gmx_mm256_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid);
 539         gmx_mm256_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
 540
 541         /* Increment number of inner iterations */
 542         inneriter                  += j_index_end - j_index_start;
 543
 544         /* Outer loop uses 9 flops */
 545     }
 546
 547     /* Increment number of outer iterations */
 548     outeriter        += nri;
 549
 550     /* Update outer/inner flops */
 551
 552     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*9 + inneriter*74);
 553 }
 554 /*
 555  * Gromacs nonbonded kernel:   nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_F_avx_256_single
 556  * Electrostatics interaction: CubicSplineTable
 557  * VdW interaction:            CubicSplineTable
 558  * Geometry:                   Particle-Particle
 559  * Calculate force/pot:        Force
 560  */
 561 void
 562 nb_kernel_ElecCSTab_VdwCSTab_GeomP1P1_F_avx_256_single
 563                     (t_nblist                    * gmx_restrict       nlist,
 564                      rvec                        * gmx_restrict          xx,
 565                      rvec                        * gmx_restrict          ff,
 566                      t_forcerec                  * gmx_restrict          fr,
 567                      t_mdatoms                   * gmx_restrict     mdatoms,
 568                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 569                      t_nrnb                      * gmx_restrict        nrnb)
 570 {
 571     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 572      * just 0 for non-waters.
 573      * Suffixes A,B,C,D,E,F,G,H refer to j loop unrolling done with AVX, e.g. for the eight different
 574      * jnr indices corresponding to data put in the four positions in the SIMD register.
 575      */
 576     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 577     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 578     int              jnrA,jnrB,jnrC,jnrD;
 579     int              jnrE,jnrF,jnrG,jnrH;
 580     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
 581     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
 582     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
 583     int              j_coord_offsetE,j_coord_offsetF,j_coord_offsetG,j_coord_offsetH;
 584     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 585     real             rcutoff_scalar;
 586     real             *shiftvec,*fshift,*x,*f;
 587     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD,*fjptrE,*fjptrF,*fjptrG,*fjptrH;
 588     real             scratch[4*DIM];
 589     __m256           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 590     real *           vdwioffsetptr0;
 591     __m256           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 592     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D,vdwjidx0E,vdwjidx0F,vdwjidx0G,vdwjidx0H;
 593     __m256           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 594     __m256           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 595     __m256           velec,felec,velecsum,facel,crf,krf,krf2;
 596     real             *charge;
 597     int              nvdwtype;
 598     __m256           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 599     int              *vdwtype;
 600     real             *vdwparam;
 601     __m256           one_sixth   = _mm256_set1_ps(1.0/6.0);
 602     __m256           one_twelfth = _mm256_set1_ps(1.0/12.0);
 603     __m256i          vfitab;
 604     __m128i          vfitab_lo,vfitab_hi;
 605     __m128i          ifour       = _mm_set1_epi32(4);
 606     __m256           rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 607     real             *vftab;
 608     __m256           dummy_mask,cutoff_mask;
 609     __m256           signbit = _mm256_castsi256_ps( _mm256_set1_epi32(0x80000000) );
 610     __m256           one     = _mm256_set1_ps(1.0);
 611     __m256           two     = _mm256_set1_ps(2.0);
 612     x                = xx[0];
 613     f                = ff[0];
 614
 615     nri              = nlist->nri;
 616     iinr             = nlist->iinr;
 617     jindex           = nlist->jindex;
 618     jjnr             = nlist->jjnr;
 619     shiftidx         = nlist->shift;
 620     gid              = nlist->gid;
 621     shiftvec         = fr->shift_vec[0];
 622     fshift           = fr->fshift[0];
 623     facel            = _mm256_set1_ps(fr->epsfac);
 624     charge           = mdatoms->chargeA;
 625     nvdwtype         = fr->ntype;
 626     vdwparam         = fr->nbfp;
 627     vdwtype          = mdatoms->typeA;
 628
 629     vftab            = kernel_data->table_elec_vdw->data;
 630     vftabscale       = _mm256_set1_ps(kernel_data->table_elec_vdw->scale);
 631
 632     /* Avoid stupid compiler warnings */
 633     jnrA = jnrB = jnrC = jnrD = jnrE = jnrF = jnrG = jnrH = 0;
 634     j_coord_offsetA = 0;
 635     j_coord_offsetB = 0;
 636     j_coord_offsetC = 0;
 637     j_coord_offsetD = 0;
 638     j_coord_offsetE = 0;
 639     j_coord_offsetF = 0;
 640     j_coord_offsetG = 0;
 641     j_coord_offsetH = 0;
 642
 643     outeriter        = 0;
 644     inneriter        = 0;
 645
 646     for(iidx=0;iidx<4*DIM;iidx++)
 647     {
 648         scratch[iidx] = 0.0;
 649     }
 650
 651     /* Start outer loop over neighborlists */
 652     for(iidx=0; iidx<nri; iidx++)
 653     {
 654         /* Load shift vector for this list */
 655         i_shift_offset   = DIM*shiftidx[iidx];
 656
 657         /* Load limits for loop over neighbors */
 658         j_index_start    = jindex[iidx];
 659         j_index_end      = jindex[iidx+1];
 660
 661         /* Get outer coordinate index */
 662         inr              = iinr[iidx];
 663         i_coord_offset   = DIM*inr;
 664
 665         /* Load i particle coords and add shift vector */
 666         gmx_mm256_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 667
 668         fix0             = _mm256_setzero_ps();
 669         fiy0             = _mm256_setzero_ps();
 670         fiz0             = _mm256_setzero_ps();
 671
 672         /* Load parameters for i particles */
 673         iq0              = _mm256_mul_ps(facel,_mm256_set1_ps(charge[inr+0]));
 674         vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 675
 676         /* Start inner kernel loop */
 677         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+7]>=0; jidx+=8)
 678         {
 679
 680             /* Get j neighbor index, and coordinate index */
 681             jnrA             = jjnr[jidx];
 682             jnrB             = jjnr[jidx+1];
 683             jnrC             = jjnr[jidx+2];
 684             jnrD             = jjnr[jidx+3];
 685             jnrE             = jjnr[jidx+4];
 686             jnrF             = jjnr[jidx+5];
 687             jnrG             = jjnr[jidx+6];
 688             jnrH             = jjnr[jidx+7];
 689             j_coord_offsetA  = DIM*jnrA;
 690             j_coord_offsetB  = DIM*jnrB;
 691             j_coord_offsetC  = DIM*jnrC;
 692             j_coord_offsetD  = DIM*jnrD;
 693             j_coord_offsetE  = DIM*jnrE;
 694             j_coord_offsetF  = DIM*jnrF;
 695             j_coord_offsetG  = DIM*jnrG;
 696             j_coord_offsetH  = DIM*jnrH;
 697
 698             /* load j atom coordinates */
 699             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 700                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 701                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 702                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 703                                                  &jx0,&jy0,&jz0);
 704
 705             /* Calculate displacement vector */
 706             dx00             = _mm256_sub_ps(ix0,jx0);
 707             dy00             = _mm256_sub_ps(iy0,jy0);
 708             dz00             = _mm256_sub_ps(iz0,jz0);
 709
 710             /* Calculate squared distance and things based on it */
 711             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 712
 713             rinv00           = gmx_mm256_invsqrt_ps(rsq00);
 714
 715             /* Load parameters for j particles */
 716             jq0              = gmx_mm256_load_8real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 717                                                                  charge+jnrC+0,charge+jnrD+0,
 718                                                                  charge+jnrE+0,charge+jnrF+0,
 719                                                                  charge+jnrG+0,charge+jnrH+0);
 720             vdwjidx0A        = 2*vdwtype[jnrA+0];
 721             vdwjidx0B        = 2*vdwtype[jnrB+0];
 722             vdwjidx0C        = 2*vdwtype[jnrC+0];
 723             vdwjidx0D        = 2*vdwtype[jnrD+0];
 724             vdwjidx0E        = 2*vdwtype[jnrE+0];
 725             vdwjidx0F        = 2*vdwtype[jnrF+0];
 726             vdwjidx0G        = 2*vdwtype[jnrG+0];
 727             vdwjidx0H        = 2*vdwtype[jnrH+0];
 728
 729             /**************************
 730              * CALCULATE INTERACTIONS *
 731              **************************/
 732
 733             r00              = _mm256_mul_ps(rsq00,rinv00);
 734
 735             /* Compute parameters for interactions between i and j atoms */
 736             qq00             = _mm256_mul_ps(iq0,jq0);
 737             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 738                                             vdwioffsetptr0+vdwjidx0B,
 739                                             vdwioffsetptr0+vdwjidx0C,
 740                                             vdwioffsetptr0+vdwjidx0D,
 741                                             vdwioffsetptr0+vdwjidx0E,
 742                                             vdwioffsetptr0+vdwjidx0F,
 743                                             vdwioffsetptr0+vdwjidx0G,
 744                                             vdwioffsetptr0+vdwjidx0H,
 745                                             &c6_00,&c12_00);
 746
 747             /* Calculate table index by multiplying r with table scale and truncate to integer */
 748             rt               = _mm256_mul_ps(r00,vftabscale);
 749             vfitab           = _mm256_cvttps_epi32(rt);
 750             vfeps            = _mm256_sub_ps(rt,_mm256_round_ps(rt, _MM_FROUND_FLOOR));
 751             /*         AVX1 does not support 256-bit integer operations, so now we go to 128-bit mode... */
 752             vfitab_lo        = _mm256_extractf128_si256(vfitab,0x0);
 753             vfitab_hi        = _mm256_extractf128_si256(vfitab,0x1);
 754             vfitab_lo        = _mm_slli_epi32(_mm_add_epi32(vfitab_lo,_mm_slli_epi32(vfitab_lo,1)),2);
 755             vfitab_hi        = _mm_slli_epi32(_mm_add_epi32(vfitab_hi,_mm_slli_epi32(vfitab_hi,1)),2);
 756
 757             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 758             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 759                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 760             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 761                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 762             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 763                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 764             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 765                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 766             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 767             Heps             = _mm256_mul_ps(vfeps,H);
 768             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 769             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 770             felec            = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_mul_ps(qq00,FF),_mm256_mul_ps(vftabscale,rinv00)));
 771
 772             /* CUBIC SPLINE TABLE DISPERSION */
 773             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 774             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 775             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 776                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 777             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 778                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 779             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 780                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 781             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 782                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 783             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 784             Heps             = _mm256_mul_ps(vfeps,H);
 785             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 786             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 787             fvdw6            = _mm256_mul_ps(c6_00,FF);
 788
 789             /* CUBIC SPLINE TABLE REPULSION */
 790             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 791             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 792             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 793                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 794             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 795                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 796             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 797                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 798             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 799                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 800             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 801             Heps             = _mm256_mul_ps(vfeps,H);
 802             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 803             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 804             fvdw12           = _mm256_mul_ps(c12_00,FF);
 805             fvdw             = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_add_ps(fvdw6,fvdw12),_mm256_mul_ps(vftabscale,rinv00)));
 806
 807             fscal            = _mm256_add_ps(felec,fvdw);
 808
 809             /* Calculate temporary vectorial force */
 810             tx               = _mm256_mul_ps(fscal,dx00);
 811             ty               = _mm256_mul_ps(fscal,dy00);
 812             tz               = _mm256_mul_ps(fscal,dz00);
 813
 814             /* Update vectorial force */
 815             fix0             = _mm256_add_ps(fix0,tx);
 816             fiy0             = _mm256_add_ps(fiy0,ty);
 817             fiz0             = _mm256_add_ps(fiz0,tz);
 818
 819             fjptrA             = f+j_coord_offsetA;
 820             fjptrB             = f+j_coord_offsetB;
 821             fjptrC             = f+j_coord_offsetC;
 822             fjptrD             = f+j_coord_offsetD;
 823             fjptrE             = f+j_coord_offsetE;
 824             fjptrF             = f+j_coord_offsetF;
 825             fjptrG             = f+j_coord_offsetG;
 826             fjptrH             = f+j_coord_offsetH;
 827             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 828
 829             /* Inner loop uses 61 flops */
 830         }
 831
 832         if(jidx<j_index_end)
 833         {
 834
 835             /* Get j neighbor index, and coordinate index */
 836             jnrlistA         = jjnr[jidx];
 837             jnrlistB         = jjnr[jidx+1];
 838             jnrlistC         = jjnr[jidx+2];
 839             jnrlistD         = jjnr[jidx+3];
 840             jnrlistE         = jjnr[jidx+4];
 841             jnrlistF         = jjnr[jidx+5];
 842             jnrlistG         = jjnr[jidx+6];
 843             jnrlistH         = jjnr[jidx+7];
 844             /* Sign of each element will be negative for non-real atoms.
 845              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 846              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 847              */
 848             dummy_mask = gmx_mm256_set_m128(gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx+4)),_mm_setzero_si128())),
 849                                             gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())));
 850
 851             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 852             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 853             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 854             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 855             jnrE       = (jnrlistE>=0) ? jnrlistE : 0;
 856             jnrF       = (jnrlistF>=0) ? jnrlistF : 0;
 857             jnrG       = (jnrlistG>=0) ? jnrlistG : 0;
 858             jnrH       = (jnrlistH>=0) ? jnrlistH : 0;
 859             j_coord_offsetA  = DIM*jnrA;
 860             j_coord_offsetB  = DIM*jnrB;
 861             j_coord_offsetC  = DIM*jnrC;
 862             j_coord_offsetD  = DIM*jnrD;
 863             j_coord_offsetE  = DIM*jnrE;
 864             j_coord_offsetF  = DIM*jnrF;
 865             j_coord_offsetG  = DIM*jnrG;
 866             j_coord_offsetH  = DIM*jnrH;
 867
 868             /* load j atom coordinates */
 869             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 870                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 871                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 872                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 873                                                  &jx0,&jy0,&jz0);
 874
 875             /* Calculate displacement vector */
 876             dx00             = _mm256_sub_ps(ix0,jx0);
 877             dy00             = _mm256_sub_ps(iy0,jy0);
 878             dz00             = _mm256_sub_ps(iz0,jz0);
 879
 880             /* Calculate squared distance and things based on it */
 881             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 882
 883             rinv00           = gmx_mm256_invsqrt_ps(rsq00);
 884
 885             /* Load parameters for j particles */
 886             jq0              = gmx_mm256_load_8real_swizzle_ps(charge+jnrA+0,charge+jnrB+0,
 887                                                                  charge+jnrC+0,charge+jnrD+0,
 888                                                                  charge+jnrE+0,charge+jnrF+0,
 889                                                                  charge+jnrG+0,charge+jnrH+0);
 890             vdwjidx0A        = 2*vdwtype[jnrA+0];
 891             vdwjidx0B        = 2*vdwtype[jnrB+0];
 892             vdwjidx0C        = 2*vdwtype[jnrC+0];
 893             vdwjidx0D        = 2*vdwtype[jnrD+0];
 894             vdwjidx0E        = 2*vdwtype[jnrE+0];
 895             vdwjidx0F        = 2*vdwtype[jnrF+0];
 896             vdwjidx0G        = 2*vdwtype[jnrG+0];
 897             vdwjidx0H        = 2*vdwtype[jnrH+0];
 898
 899             /**************************
 900              * CALCULATE INTERACTIONS *
 901              **************************/
 902
 903             r00              = _mm256_mul_ps(rsq00,rinv00);
 904             r00              = _mm256_andnot_ps(dummy_mask,r00);
 905
 906             /* Compute parameters for interactions between i and j atoms */
 907             qq00             = _mm256_mul_ps(iq0,jq0);
 908             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 909                                             vdwioffsetptr0+vdwjidx0B,
 910                                             vdwioffsetptr0+vdwjidx0C,
 911                                             vdwioffsetptr0+vdwjidx0D,
 912                                             vdwioffsetptr0+vdwjidx0E,
 913                                             vdwioffsetptr0+vdwjidx0F,
 914                                             vdwioffsetptr0+vdwjidx0G,
 915                                             vdwioffsetptr0+vdwjidx0H,
 916                                             &c6_00,&c12_00);
 917
 918             /* Calculate table index by multiplying r with table scale and truncate to integer */
 919             rt               = _mm256_mul_ps(r00,vftabscale);
 920             vfitab           = _mm256_cvttps_epi32(rt);
 921             vfeps            = _mm256_sub_ps(rt,_mm256_round_ps(rt, _MM_FROUND_FLOOR));
 922             /*         AVX1 does not support 256-bit integer operations, so now we go to 128-bit mode... */
 923             vfitab_lo        = _mm256_extractf128_si256(vfitab,0x0);
 924             vfitab_hi        = _mm256_extractf128_si256(vfitab,0x1);
 925             vfitab_lo        = _mm_slli_epi32(_mm_add_epi32(vfitab_lo,_mm_slli_epi32(vfitab_lo,1)),2);
 926             vfitab_hi        = _mm_slli_epi32(_mm_add_epi32(vfitab_hi,_mm_slli_epi32(vfitab_hi,1)),2);
 927
 928             /* CUBIC SPLINE TABLE ELECTROSTATICS */
 929             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 930                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 931             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 932                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 933             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 934                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 935             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 936                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 937             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 938             Heps             = _mm256_mul_ps(vfeps,H);
 939             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 940             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 941             felec            = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_mul_ps(qq00,FF),_mm256_mul_ps(vftabscale,rinv00)));
 942
 943             /* CUBIC SPLINE TABLE DISPERSION */
 944             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 945             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 946             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 947                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 948             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 949                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 950             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 951                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 952             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 953                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 954             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 955             Heps             = _mm256_mul_ps(vfeps,H);
 956             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 957             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 958             fvdw6            = _mm256_mul_ps(c6_00,FF);
 959
 960             /* CUBIC SPLINE TABLE REPULSION */
 961             vfitab_lo        = _mm_add_epi32(vfitab_lo,ifour);
 962             vfitab_hi        = _mm_add_epi32(vfitab_hi,ifour);
 963             Y                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,0)),
 964                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,0)));
 965             F                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,1)),
 966                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,1)));
 967             G                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,2)),
 968                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,2)));
 969             H                = gmx_mm256_set_m128(_mm_load_ps(vftab + _mm_extract_epi32(vfitab_hi,3)),
 970                                                   _mm_load_ps(vftab + _mm_extract_epi32(vfitab_lo,3)));
 971             GMX_MM256_HALFTRANSPOSE4_PS(Y,F,G,H);
 972             Heps             = _mm256_mul_ps(vfeps,H);
 973             Fp               = _mm256_add_ps(F,_mm256_mul_ps(vfeps,_mm256_add_ps(G,Heps)));
 974             FF               = _mm256_add_ps(Fp,_mm256_mul_ps(vfeps,_mm256_add_ps(G,_mm256_add_ps(Heps,Heps))));
 975             fvdw12           = _mm256_mul_ps(c12_00,FF);
 976             fvdw             = _mm256_xor_ps(signbit,_mm256_mul_ps(_mm256_add_ps(fvdw6,fvdw12),_mm256_mul_ps(vftabscale,rinv00)));
 977
 978             fscal            = _mm256_add_ps(felec,fvdw);
 979
 980             fscal            = _mm256_andnot_ps(dummy_mask,fscal);
 981
 982             /* Calculate temporary vectorial force */
 983             tx               = _mm256_mul_ps(fscal,dx00);
 984             ty               = _mm256_mul_ps(fscal,dy00);
 985             tz               = _mm256_mul_ps(fscal,dz00);
 986
 987             /* Update vectorial force */
 988             fix0             = _mm256_add_ps(fix0,tx);
 989             fiy0             = _mm256_add_ps(fiy0,ty);
 990             fiz0             = _mm256_add_ps(fiz0,tz);
 991
 992             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 993             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 994             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 995             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 996             fjptrE             = (jnrlistE>=0) ? f+j_coord_offsetE : scratch;
 997             fjptrF             = (jnrlistF>=0) ? f+j_coord_offsetF : scratch;
 998             fjptrG             = (jnrlistG>=0) ? f+j_coord_offsetG : scratch;
 999             fjptrH             = (jnrlistH>=0) ? f+j_coord_offsetH : scratch;
1000             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
1001
1002             /* Inner loop uses 62 flops */
1003         }
1004
1005         /* End of innermost loop */
1006
1007         gmx_mm256_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
1008                                                  f+i_coord_offset,fshift+i_shift_offset);
1009
1010         /* Increment number of inner iterations */
1011         inneriter                  += j_index_end - j_index_start;
1012
1013         /* Outer loop uses 7 flops */
1014     }
1015
1016     /* Increment number of outer iterations */
1017     outeriter        += nri;
1018
1019     /* Update outer/inner flops */
1020
1021     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*7 + inneriter*62);
1022 }