src/gmxlib/nonbonded/nb_kernel_avx_256_single/nb_kernel_ElecNone_VdwLJSh_GeomP1P1_avx_256_single.c

   1 /*
   2  * Note: this file was generated by the Gromacs avx_256_single kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_avx_256_single.h"
  34 #include "kernelutil_x86_avx_256_single.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecNone_VdwLJSh_GeomP1P1_VF_avx_256_single
  38  * Electrostatics interaction: None
  39  * VdW interaction:            LennardJones
  40  * Geometry:                   Particle-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecNone_VdwLJSh_GeomP1P1_VF_avx_256_single
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B,C,D,E,F,G,H refer to j loop unrolling done with AVX, e.g. for the eight different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB,jnrC,jnrD;
  61     int              jnrE,jnrF,jnrG,jnrH;
  62     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
  63     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
  64     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
  65     int              j_coord_offsetE,j_coord_offsetF,j_coord_offsetG,j_coord_offsetH;
  66     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  67     real             rcutoff_scalar;
  68     real             *shiftvec,*fshift,*x,*f;
  69     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD,*fjptrE,*fjptrF,*fjptrG,*fjptrH;
  70     real             scratch[4*DIM];
  71     __m256           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  72     real *           vdwioffsetptr0;
  73     __m256           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  74     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D,vdwjidx0E,vdwjidx0F,vdwjidx0G,vdwjidx0H;
  75     __m256           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  76     __m256           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  77     int              nvdwtype;
  78     __m256           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  79     int              *vdwtype;
  80     real             *vdwparam;
  81     __m256           one_sixth   = _mm256_set1_ps(1.0/6.0);
  82     __m256           one_twelfth = _mm256_set1_ps(1.0/12.0);
  83     __m256           dummy_mask,cutoff_mask;
  84     __m256           signbit = _mm256_castsi256_ps( _mm256_set1_epi32(0x80000000) );
  85     __m256           one     = _mm256_set1_ps(1.0);
  86     __m256           two     = _mm256_set1_ps(2.0);
  87     x                = xx[0];
  88     f                = ff[0];
  89
  90     nri              = nlist->nri;
  91     iinr             = nlist->iinr;
  92     jindex           = nlist->jindex;
  93     jjnr             = nlist->jjnr;
  94     shiftidx         = nlist->shift;
  95     gid              = nlist->gid;
  96     shiftvec         = fr->shift_vec[0];
  97     fshift           = fr->fshift[0];
  98     nvdwtype         = fr->ntype;
  99     vdwparam         = fr->nbfp;
 100     vdwtype          = mdatoms->typeA;
 101
 102     rcutoff_scalar   = fr->rvdw;
 103     rcutoff          = _mm256_set1_ps(rcutoff_scalar);
 104     rcutoff2         = _mm256_mul_ps(rcutoff,rcutoff);
 105
 106     sh_vdw_invrcut6  = _mm256_set1_ps(fr->ic->sh_invrc6);
 107     rvdw             = _mm256_set1_ps(fr->rvdw);
 108
 109     /* Avoid stupid compiler warnings */
 110     jnrA = jnrB = jnrC = jnrD = jnrE = jnrF = jnrG = jnrH = 0;
 111     j_coord_offsetA = 0;
 112     j_coord_offsetB = 0;
 113     j_coord_offsetC = 0;
 114     j_coord_offsetD = 0;
 115     j_coord_offsetE = 0;
 116     j_coord_offsetF = 0;
 117     j_coord_offsetG = 0;
 118     j_coord_offsetH = 0;
 119
 120     outeriter        = 0;
 121     inneriter        = 0;
 122
 123     for(iidx=0;iidx<4*DIM;iidx++)
 124     {
 125         scratch[iidx] = 0.0;
 126     }
 127
 128     /* Start outer loop over neighborlists */
 129     for(iidx=0; iidx<nri; iidx++)
 130     {
 131         /* Load shift vector for this list */
 132         i_shift_offset   = DIM*shiftidx[iidx];
 133
 134         /* Load limits for loop over neighbors */
 135         j_index_start    = jindex[iidx];
 136         j_index_end      = jindex[iidx+1];
 137
 138         /* Get outer coordinate index */
 139         inr              = iinr[iidx];
 140         i_coord_offset   = DIM*inr;
 141
 142         /* Load i particle coords and add shift vector */
 143         gmx_mm256_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 144
 145         fix0             = _mm256_setzero_ps();
 146         fiy0             = _mm256_setzero_ps();
 147         fiz0             = _mm256_setzero_ps();
 148
 149         /* Load parameters for i particles */
 150         vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 151
 152         /* Reset potential sums */
 153         vvdwsum          = _mm256_setzero_ps();
 154
 155         /* Start inner kernel loop */
 156         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+7]>=0; jidx+=8)
 157         {
 158
 159             /* Get j neighbor index, and coordinate index */
 160             jnrA             = jjnr[jidx];
 161             jnrB             = jjnr[jidx+1];
 162             jnrC             = jjnr[jidx+2];
 163             jnrD             = jjnr[jidx+3];
 164             jnrE             = jjnr[jidx+4];
 165             jnrF             = jjnr[jidx+5];
 166             jnrG             = jjnr[jidx+6];
 167             jnrH             = jjnr[jidx+7];
 168             j_coord_offsetA  = DIM*jnrA;
 169             j_coord_offsetB  = DIM*jnrB;
 170             j_coord_offsetC  = DIM*jnrC;
 171             j_coord_offsetD  = DIM*jnrD;
 172             j_coord_offsetE  = DIM*jnrE;
 173             j_coord_offsetF  = DIM*jnrF;
 174             j_coord_offsetG  = DIM*jnrG;
 175             j_coord_offsetH  = DIM*jnrH;
 176
 177             /* load j atom coordinates */
 178             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 179                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 180                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 181                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 182                                                  &jx0,&jy0,&jz0);
 183
 184             /* Calculate displacement vector */
 185             dx00             = _mm256_sub_ps(ix0,jx0);
 186             dy00             = _mm256_sub_ps(iy0,jy0);
 187             dz00             = _mm256_sub_ps(iz0,jz0);
 188
 189             /* Calculate squared distance and things based on it */
 190             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 191
 192             rinvsq00         = gmx_mm256_inv_ps(rsq00);
 193
 194             /* Load parameters for j particles */
 195             vdwjidx0A        = 2*vdwtype[jnrA+0];
 196             vdwjidx0B        = 2*vdwtype[jnrB+0];
 197             vdwjidx0C        = 2*vdwtype[jnrC+0];
 198             vdwjidx0D        = 2*vdwtype[jnrD+0];
 199             vdwjidx0E        = 2*vdwtype[jnrE+0];
 200             vdwjidx0F        = 2*vdwtype[jnrF+0];
 201             vdwjidx0G        = 2*vdwtype[jnrG+0];
 202             vdwjidx0H        = 2*vdwtype[jnrH+0];
 203
 204             /**************************
 205              * CALCULATE INTERACTIONS *
 206              **************************/
 207
 208             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 209             {
 210
 211             /* Compute parameters for interactions between i and j atoms */
 212             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 213                                             vdwioffsetptr0+vdwjidx0B,
 214                                             vdwioffsetptr0+vdwjidx0C,
 215                                             vdwioffsetptr0+vdwjidx0D,
 216                                             vdwioffsetptr0+vdwjidx0E,
 217                                             vdwioffsetptr0+vdwjidx0F,
 218                                             vdwioffsetptr0+vdwjidx0G,
 219                                             vdwioffsetptr0+vdwjidx0H,
 220                                             &c6_00,&c12_00);
 221
 222             /* LENNARD-JONES DISPERSION/REPULSION */
 223
 224             rinvsix          = _mm256_mul_ps(_mm256_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 225             vvdw6            = _mm256_mul_ps(c6_00,rinvsix);
 226             vvdw12           = _mm256_mul_ps(c12_00,_mm256_mul_ps(rinvsix,rinvsix));
 227             vvdw             = _mm256_sub_ps(_mm256_mul_ps( _mm256_sub_ps(vvdw12 , _mm256_mul_ps(c12_00,_mm256_mul_ps(sh_vdw_invrcut6,sh_vdw_invrcut6))), one_twelfth) ,
 228                                           _mm256_mul_ps( _mm256_sub_ps(vvdw6,_mm256_mul_ps(c6_00,sh_vdw_invrcut6)),one_sixth));
 229             fvdw             = _mm256_mul_ps(_mm256_sub_ps(vvdw12,vvdw6),rinvsq00);
 230
 231             cutoff_mask      = _mm256_cmp_ps(rsq00,rcutoff2,_CMP_LT_OQ);
 232
 233             /* Update potential sum for this i atom from the interaction with this j atom. */
 234             vvdw             = _mm256_and_ps(vvdw,cutoff_mask);
 235             vvdwsum          = _mm256_add_ps(vvdwsum,vvdw);
 236
 237             fscal            = fvdw;
 238
 239             fscal            = _mm256_and_ps(fscal,cutoff_mask);
 240
 241             /* Calculate temporary vectorial force */
 242             tx               = _mm256_mul_ps(fscal,dx00);
 243             ty               = _mm256_mul_ps(fscal,dy00);
 244             tz               = _mm256_mul_ps(fscal,dz00);
 245
 246             /* Update vectorial force */
 247             fix0             = _mm256_add_ps(fix0,tx);
 248             fiy0             = _mm256_add_ps(fiy0,ty);
 249             fiz0             = _mm256_add_ps(fiz0,tz);
 250
 251             fjptrA             = f+j_coord_offsetA;
 252             fjptrB             = f+j_coord_offsetB;
 253             fjptrC             = f+j_coord_offsetC;
 254             fjptrD             = f+j_coord_offsetD;
 255             fjptrE             = f+j_coord_offsetE;
 256             fjptrF             = f+j_coord_offsetF;
 257             fjptrG             = f+j_coord_offsetG;
 258             fjptrH             = f+j_coord_offsetH;
 259             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 260
 261             }
 262
 263             /* Inner loop uses 41 flops */
 264         }
 265
 266         if(jidx<j_index_end)
 267         {
 268
 269             /* Get j neighbor index, and coordinate index */
 270             jnrlistA         = jjnr[jidx];
 271             jnrlistB         = jjnr[jidx+1];
 272             jnrlistC         = jjnr[jidx+2];
 273             jnrlistD         = jjnr[jidx+3];
 274             jnrlistE         = jjnr[jidx+4];
 275             jnrlistF         = jjnr[jidx+5];
 276             jnrlistG         = jjnr[jidx+6];
 277             jnrlistH         = jjnr[jidx+7];
 278             /* Sign of each element will be negative for non-real atoms.
 279              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 280              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 281              */
 282             dummy_mask = gmx_mm256_set_m128(gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx+4)),_mm_setzero_si128())),
 283                                             gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())));
 284
 285             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 286             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 287             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 288             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 289             jnrE       = (jnrlistE>=0) ? jnrlistE : 0;
 290             jnrF       = (jnrlistF>=0) ? jnrlistF : 0;
 291             jnrG       = (jnrlistG>=0) ? jnrlistG : 0;
 292             jnrH       = (jnrlistH>=0) ? jnrlistH : 0;
 293             j_coord_offsetA  = DIM*jnrA;
 294             j_coord_offsetB  = DIM*jnrB;
 295             j_coord_offsetC  = DIM*jnrC;
 296             j_coord_offsetD  = DIM*jnrD;
 297             j_coord_offsetE  = DIM*jnrE;
 298             j_coord_offsetF  = DIM*jnrF;
 299             j_coord_offsetG  = DIM*jnrG;
 300             j_coord_offsetH  = DIM*jnrH;
 301
 302             /* load j atom coordinates */
 303             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 304                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 305                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 306                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 307                                                  &jx0,&jy0,&jz0);
 308
 309             /* Calculate displacement vector */
 310             dx00             = _mm256_sub_ps(ix0,jx0);
 311             dy00             = _mm256_sub_ps(iy0,jy0);
 312             dz00             = _mm256_sub_ps(iz0,jz0);
 313
 314             /* Calculate squared distance and things based on it */
 315             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 316
 317             rinvsq00         = gmx_mm256_inv_ps(rsq00);
 318
 319             /* Load parameters for j particles */
 320             vdwjidx0A        = 2*vdwtype[jnrA+0];
 321             vdwjidx0B        = 2*vdwtype[jnrB+0];
 322             vdwjidx0C        = 2*vdwtype[jnrC+0];
 323             vdwjidx0D        = 2*vdwtype[jnrD+0];
 324             vdwjidx0E        = 2*vdwtype[jnrE+0];
 325             vdwjidx0F        = 2*vdwtype[jnrF+0];
 326             vdwjidx0G        = 2*vdwtype[jnrG+0];
 327             vdwjidx0H        = 2*vdwtype[jnrH+0];
 328
 329             /**************************
 330              * CALCULATE INTERACTIONS *
 331              **************************/
 332
 333             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 334             {
 335
 336             /* Compute parameters for interactions between i and j atoms */
 337             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 338                                             vdwioffsetptr0+vdwjidx0B,
 339                                             vdwioffsetptr0+vdwjidx0C,
 340                                             vdwioffsetptr0+vdwjidx0D,
 341                                             vdwioffsetptr0+vdwjidx0E,
 342                                             vdwioffsetptr0+vdwjidx0F,
 343                                             vdwioffsetptr0+vdwjidx0G,
 344                                             vdwioffsetptr0+vdwjidx0H,
 345                                             &c6_00,&c12_00);
 346
 347             /* LENNARD-JONES DISPERSION/REPULSION */
 348
 349             rinvsix          = _mm256_mul_ps(_mm256_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 350             vvdw6            = _mm256_mul_ps(c6_00,rinvsix);
 351             vvdw12           = _mm256_mul_ps(c12_00,_mm256_mul_ps(rinvsix,rinvsix));
 352             vvdw             = _mm256_sub_ps(_mm256_mul_ps( _mm256_sub_ps(vvdw12 , _mm256_mul_ps(c12_00,_mm256_mul_ps(sh_vdw_invrcut6,sh_vdw_invrcut6))), one_twelfth) ,
 353                                           _mm256_mul_ps( _mm256_sub_ps(vvdw6,_mm256_mul_ps(c6_00,sh_vdw_invrcut6)),one_sixth));
 354             fvdw             = _mm256_mul_ps(_mm256_sub_ps(vvdw12,vvdw6),rinvsq00);
 355
 356             cutoff_mask      = _mm256_cmp_ps(rsq00,rcutoff2,_CMP_LT_OQ);
 357
 358             /* Update potential sum for this i atom from the interaction with this j atom. */
 359             vvdw             = _mm256_and_ps(vvdw,cutoff_mask);
 360             vvdw             = _mm256_andnot_ps(dummy_mask,vvdw);
 361             vvdwsum          = _mm256_add_ps(vvdwsum,vvdw);
 362
 363             fscal            = fvdw;
 364
 365             fscal            = _mm256_and_ps(fscal,cutoff_mask);
 366
 367             fscal            = _mm256_andnot_ps(dummy_mask,fscal);
 368
 369             /* Calculate temporary vectorial force */
 370             tx               = _mm256_mul_ps(fscal,dx00);
 371             ty               = _mm256_mul_ps(fscal,dy00);
 372             tz               = _mm256_mul_ps(fscal,dz00);
 373
 374             /* Update vectorial force */
 375             fix0             = _mm256_add_ps(fix0,tx);
 376             fiy0             = _mm256_add_ps(fiy0,ty);
 377             fiz0             = _mm256_add_ps(fiz0,tz);
 378
 379             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 380             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 381             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 382             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 383             fjptrE             = (jnrlistE>=0) ? f+j_coord_offsetE : scratch;
 384             fjptrF             = (jnrlistF>=0) ? f+j_coord_offsetF : scratch;
 385             fjptrG             = (jnrlistG>=0) ? f+j_coord_offsetG : scratch;
 386             fjptrH             = (jnrlistH>=0) ? f+j_coord_offsetH : scratch;
 387             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 388
 389             }
 390
 391             /* Inner loop uses 41 flops */
 392         }
 393
 394         /* End of innermost loop */
 395
 396         gmx_mm256_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
 397                                                  f+i_coord_offset,fshift+i_shift_offset);
 398
 399         ggid                        = gid[iidx];
 400         /* Update potential energies */
 401         gmx_mm256_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid);
 402
 403         /* Increment number of inner iterations */
 404         inneriter                  += j_index_end - j_index_start;
 405
 406         /* Outer loop uses 7 flops */
 407     }
 408
 409     /* Increment number of outer iterations */
 410     outeriter        += nri;
 411
 412     /* Update outer/inner flops */
 413
 414     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_VF,outeriter*7 + inneriter*41);
 415 }
 416 /*
 417  * Gromacs nonbonded kernel:   nb_kernel_ElecNone_VdwLJSh_GeomP1P1_F_avx_256_single
 418  * Electrostatics interaction: None
 419  * VdW interaction:            LennardJones
 420  * Geometry:                   Particle-Particle
 421  * Calculate force/pot:        Force
 422  */
 423 void
 424 nb_kernel_ElecNone_VdwLJSh_GeomP1P1_F_avx_256_single
 425                     (t_nblist * gmx_restrict                nlist,
 426                      rvec * gmx_restrict                    xx,
 427                      rvec * gmx_restrict                    ff,
 428                      t_forcerec * gmx_restrict              fr,
 429                      t_mdatoms * gmx_restrict               mdatoms,
 430                      nb_kernel_data_t * gmx_restrict        kernel_data,
 431                      t_nrnb * gmx_restrict                  nrnb)
 432 {
 433     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 434      * just 0 for non-waters.
 435      * Suffixes A,B,C,D,E,F,G,H refer to j loop unrolling done with AVX, e.g. for the eight different
 436      * jnr indices corresponding to data put in the four positions in the SIMD register.
 437      */
 438     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 439     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 440     int              jnrA,jnrB,jnrC,jnrD;
 441     int              jnrE,jnrF,jnrG,jnrH;
 442     int              jnrlistA,jnrlistB,jnrlistC,jnrlistD;
 443     int              jnrlistE,jnrlistF,jnrlistG,jnrlistH;
 444     int              j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD;
 445     int              j_coord_offsetE,j_coord_offsetF,j_coord_offsetG,j_coord_offsetH;
 446     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 447     real             rcutoff_scalar;
 448     real             *shiftvec,*fshift,*x,*f;
 449     real             *fjptrA,*fjptrB,*fjptrC,*fjptrD,*fjptrE,*fjptrF,*fjptrG,*fjptrH;
 450     real             scratch[4*DIM];
 451     __m256           tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 452     real *           vdwioffsetptr0;
 453     __m256           ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 454     int              vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D,vdwjidx0E,vdwjidx0F,vdwjidx0G,vdwjidx0H;
 455     __m256           jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 456     __m256           dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 457     int              nvdwtype;
 458     __m256           rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 459     int              *vdwtype;
 460     real             *vdwparam;
 461     __m256           one_sixth   = _mm256_set1_ps(1.0/6.0);
 462     __m256           one_twelfth = _mm256_set1_ps(1.0/12.0);
 463     __m256           dummy_mask,cutoff_mask;
 464     __m256           signbit = _mm256_castsi256_ps( _mm256_set1_epi32(0x80000000) );
 465     __m256           one     = _mm256_set1_ps(1.0);
 466     __m256           two     = _mm256_set1_ps(2.0);
 467     x                = xx[0];
 468     f                = ff[0];
 469
 470     nri              = nlist->nri;
 471     iinr             = nlist->iinr;
 472     jindex           = nlist->jindex;
 473     jjnr             = nlist->jjnr;
 474     shiftidx         = nlist->shift;
 475     gid              = nlist->gid;
 476     shiftvec         = fr->shift_vec[0];
 477     fshift           = fr->fshift[0];
 478     nvdwtype         = fr->ntype;
 479     vdwparam         = fr->nbfp;
 480     vdwtype          = mdatoms->typeA;
 481
 482     rcutoff_scalar   = fr->rvdw;
 483     rcutoff          = _mm256_set1_ps(rcutoff_scalar);
 484     rcutoff2         = _mm256_mul_ps(rcutoff,rcutoff);
 485
 486     sh_vdw_invrcut6  = _mm256_set1_ps(fr->ic->sh_invrc6);
 487     rvdw             = _mm256_set1_ps(fr->rvdw);
 488
 489     /* Avoid stupid compiler warnings */
 490     jnrA = jnrB = jnrC = jnrD = jnrE = jnrF = jnrG = jnrH = 0;
 491     j_coord_offsetA = 0;
 492     j_coord_offsetB = 0;
 493     j_coord_offsetC = 0;
 494     j_coord_offsetD = 0;
 495     j_coord_offsetE = 0;
 496     j_coord_offsetF = 0;
 497     j_coord_offsetG = 0;
 498     j_coord_offsetH = 0;
 499
 500     outeriter        = 0;
 501     inneriter        = 0;
 502
 503     for(iidx=0;iidx<4*DIM;iidx++)
 504     {
 505         scratch[iidx] = 0.0;
 506     }
 507
 508     /* Start outer loop over neighborlists */
 509     for(iidx=0; iidx<nri; iidx++)
 510     {
 511         /* Load shift vector for this list */
 512         i_shift_offset   = DIM*shiftidx[iidx];
 513
 514         /* Load limits for loop over neighbors */
 515         j_index_start    = jindex[iidx];
 516         j_index_end      = jindex[iidx+1];
 517
 518         /* Get outer coordinate index */
 519         inr              = iinr[iidx];
 520         i_coord_offset   = DIM*inr;
 521
 522         /* Load i particle coords and add shift vector */
 523         gmx_mm256_load_shift_and_1rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 524
 525         fix0             = _mm256_setzero_ps();
 526         fiy0             = _mm256_setzero_ps();
 527         fiz0             = _mm256_setzero_ps();
 528
 529         /* Load parameters for i particles */
 530         vdwioffsetptr0   = vdwparam+2*nvdwtype*vdwtype[inr+0];
 531
 532         /* Start inner kernel loop */
 533         for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+7]>=0; jidx+=8)
 534         {
 535
 536             /* Get j neighbor index, and coordinate index */
 537             jnrA             = jjnr[jidx];
 538             jnrB             = jjnr[jidx+1];
 539             jnrC             = jjnr[jidx+2];
 540             jnrD             = jjnr[jidx+3];
 541             jnrE             = jjnr[jidx+4];
 542             jnrF             = jjnr[jidx+5];
 543             jnrG             = jjnr[jidx+6];
 544             jnrH             = jjnr[jidx+7];
 545             j_coord_offsetA  = DIM*jnrA;
 546             j_coord_offsetB  = DIM*jnrB;
 547             j_coord_offsetC  = DIM*jnrC;
 548             j_coord_offsetD  = DIM*jnrD;
 549             j_coord_offsetE  = DIM*jnrE;
 550             j_coord_offsetF  = DIM*jnrF;
 551             j_coord_offsetG  = DIM*jnrG;
 552             j_coord_offsetH  = DIM*jnrH;
 553
 554             /* load j atom coordinates */
 555             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 556                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 557                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 558                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 559                                                  &jx0,&jy0,&jz0);
 560
 561             /* Calculate displacement vector */
 562             dx00             = _mm256_sub_ps(ix0,jx0);
 563             dy00             = _mm256_sub_ps(iy0,jy0);
 564             dz00             = _mm256_sub_ps(iz0,jz0);
 565
 566             /* Calculate squared distance and things based on it */
 567             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 568
 569             rinvsq00         = gmx_mm256_inv_ps(rsq00);
 570
 571             /* Load parameters for j particles */
 572             vdwjidx0A        = 2*vdwtype[jnrA+0];
 573             vdwjidx0B        = 2*vdwtype[jnrB+0];
 574             vdwjidx0C        = 2*vdwtype[jnrC+0];
 575             vdwjidx0D        = 2*vdwtype[jnrD+0];
 576             vdwjidx0E        = 2*vdwtype[jnrE+0];
 577             vdwjidx0F        = 2*vdwtype[jnrF+0];
 578             vdwjidx0G        = 2*vdwtype[jnrG+0];
 579             vdwjidx0H        = 2*vdwtype[jnrH+0];
 580
 581             /**************************
 582              * CALCULATE INTERACTIONS *
 583              **************************/
 584
 585             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 586             {
 587
 588             /* Compute parameters for interactions between i and j atoms */
 589             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 590                                             vdwioffsetptr0+vdwjidx0B,
 591                                             vdwioffsetptr0+vdwjidx0C,
 592                                             vdwioffsetptr0+vdwjidx0D,
 593                                             vdwioffsetptr0+vdwjidx0E,
 594                                             vdwioffsetptr0+vdwjidx0F,
 595                                             vdwioffsetptr0+vdwjidx0G,
 596                                             vdwioffsetptr0+vdwjidx0H,
 597                                             &c6_00,&c12_00);
 598
 599             /* LENNARD-JONES DISPERSION/REPULSION */
 600
 601             rinvsix          = _mm256_mul_ps(_mm256_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 602             fvdw             = _mm256_mul_ps(_mm256_sub_ps(_mm256_mul_ps(c12_00,rinvsix),c6_00),_mm256_mul_ps(rinvsix,rinvsq00));
 603
 604             cutoff_mask      = _mm256_cmp_ps(rsq00,rcutoff2,_CMP_LT_OQ);
 605
 606             fscal            = fvdw;
 607
 608             fscal            = _mm256_and_ps(fscal,cutoff_mask);
 609
 610             /* Calculate temporary vectorial force */
 611             tx               = _mm256_mul_ps(fscal,dx00);
 612             ty               = _mm256_mul_ps(fscal,dy00);
 613             tz               = _mm256_mul_ps(fscal,dz00);
 614
 615             /* Update vectorial force */
 616             fix0             = _mm256_add_ps(fix0,tx);
 617             fiy0             = _mm256_add_ps(fiy0,ty);
 618             fiz0             = _mm256_add_ps(fiz0,tz);
 619
 620             fjptrA             = f+j_coord_offsetA;
 621             fjptrB             = f+j_coord_offsetB;
 622             fjptrC             = f+j_coord_offsetC;
 623             fjptrD             = f+j_coord_offsetD;
 624             fjptrE             = f+j_coord_offsetE;
 625             fjptrF             = f+j_coord_offsetF;
 626             fjptrG             = f+j_coord_offsetG;
 627             fjptrH             = f+j_coord_offsetH;
 628             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 629
 630             }
 631
 632             /* Inner loop uses 30 flops */
 633         }
 634
 635         if(jidx<j_index_end)
 636         {
 637
 638             /* Get j neighbor index, and coordinate index */
 639             jnrlistA         = jjnr[jidx];
 640             jnrlistB         = jjnr[jidx+1];
 641             jnrlistC         = jjnr[jidx+2];
 642             jnrlistD         = jjnr[jidx+3];
 643             jnrlistE         = jjnr[jidx+4];
 644             jnrlistF         = jjnr[jidx+5];
 645             jnrlistG         = jjnr[jidx+6];
 646             jnrlistH         = jjnr[jidx+7];
 647             /* Sign of each element will be negative for non-real atoms.
 648              * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones,
 649              * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries.
 650              */
 651             dummy_mask = gmx_mm256_set_m128(gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx+4)),_mm_setzero_si128())),
 652                                             gmx_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())));
 653
 654             jnrA       = (jnrlistA>=0) ? jnrlistA : 0;
 655             jnrB       = (jnrlistB>=0) ? jnrlistB : 0;
 656             jnrC       = (jnrlistC>=0) ? jnrlistC : 0;
 657             jnrD       = (jnrlistD>=0) ? jnrlistD : 0;
 658             jnrE       = (jnrlistE>=0) ? jnrlistE : 0;
 659             jnrF       = (jnrlistF>=0) ? jnrlistF : 0;
 660             jnrG       = (jnrlistG>=0) ? jnrlistG : 0;
 661             jnrH       = (jnrlistH>=0) ? jnrlistH : 0;
 662             j_coord_offsetA  = DIM*jnrA;
 663             j_coord_offsetB  = DIM*jnrB;
 664             j_coord_offsetC  = DIM*jnrC;
 665             j_coord_offsetD  = DIM*jnrD;
 666             j_coord_offsetE  = DIM*jnrE;
 667             j_coord_offsetF  = DIM*jnrF;
 668             j_coord_offsetG  = DIM*jnrG;
 669             j_coord_offsetH  = DIM*jnrH;
 670
 671             /* load j atom coordinates */
 672             gmx_mm256_load_1rvec_8ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB,
 673                                                  x+j_coord_offsetC,x+j_coord_offsetD,
 674                                                  x+j_coord_offsetE,x+j_coord_offsetF,
 675                                                  x+j_coord_offsetG,x+j_coord_offsetH,
 676                                                  &jx0,&jy0,&jz0);
 677
 678             /* Calculate displacement vector */
 679             dx00             = _mm256_sub_ps(ix0,jx0);
 680             dy00             = _mm256_sub_ps(iy0,jy0);
 681             dz00             = _mm256_sub_ps(iz0,jz0);
 682
 683             /* Calculate squared distance and things based on it */
 684             rsq00            = gmx_mm256_calc_rsq_ps(dx00,dy00,dz00);
 685
 686             rinvsq00         = gmx_mm256_inv_ps(rsq00);
 687
 688             /* Load parameters for j particles */
 689             vdwjidx0A        = 2*vdwtype[jnrA+0];
 690             vdwjidx0B        = 2*vdwtype[jnrB+0];
 691             vdwjidx0C        = 2*vdwtype[jnrC+0];
 692             vdwjidx0D        = 2*vdwtype[jnrD+0];
 693             vdwjidx0E        = 2*vdwtype[jnrE+0];
 694             vdwjidx0F        = 2*vdwtype[jnrF+0];
 695             vdwjidx0G        = 2*vdwtype[jnrG+0];
 696             vdwjidx0H        = 2*vdwtype[jnrH+0];
 697
 698             /**************************
 699              * CALCULATE INTERACTIONS *
 700              **************************/
 701
 702             if (gmx_mm256_any_lt(rsq00,rcutoff2))
 703             {
 704
 705             /* Compute parameters for interactions between i and j atoms */
 706             gmx_mm256_load_8pair_swizzle_ps(vdwioffsetptr0+vdwjidx0A,
 707                                             vdwioffsetptr0+vdwjidx0B,
 708                                             vdwioffsetptr0+vdwjidx0C,
 709                                             vdwioffsetptr0+vdwjidx0D,
 710                                             vdwioffsetptr0+vdwjidx0E,
 711                                             vdwioffsetptr0+vdwjidx0F,
 712                                             vdwioffsetptr0+vdwjidx0G,
 713                                             vdwioffsetptr0+vdwjidx0H,
 714                                             &c6_00,&c12_00);
 715
 716             /* LENNARD-JONES DISPERSION/REPULSION */
 717
 718             rinvsix          = _mm256_mul_ps(_mm256_mul_ps(rinvsq00,rinvsq00),rinvsq00);
 719             fvdw             = _mm256_mul_ps(_mm256_sub_ps(_mm256_mul_ps(c12_00,rinvsix),c6_00),_mm256_mul_ps(rinvsix,rinvsq00));
 720
 721             cutoff_mask      = _mm256_cmp_ps(rsq00,rcutoff2,_CMP_LT_OQ);
 722
 723             fscal            = fvdw;
 724
 725             fscal            = _mm256_and_ps(fscal,cutoff_mask);
 726
 727             fscal            = _mm256_andnot_ps(dummy_mask,fscal);
 728
 729             /* Calculate temporary vectorial force */
 730             tx               = _mm256_mul_ps(fscal,dx00);
 731             ty               = _mm256_mul_ps(fscal,dy00);
 732             tz               = _mm256_mul_ps(fscal,dz00);
 733
 734             /* Update vectorial force */
 735             fix0             = _mm256_add_ps(fix0,tx);
 736             fiy0             = _mm256_add_ps(fiy0,ty);
 737             fiz0             = _mm256_add_ps(fiz0,tz);
 738
 739             fjptrA             = (jnrlistA>=0) ? f+j_coord_offsetA : scratch;
 740             fjptrB             = (jnrlistB>=0) ? f+j_coord_offsetB : scratch;
 741             fjptrC             = (jnrlistC>=0) ? f+j_coord_offsetC : scratch;
 742             fjptrD             = (jnrlistD>=0) ? f+j_coord_offsetD : scratch;
 743             fjptrE             = (jnrlistE>=0) ? f+j_coord_offsetE : scratch;
 744             fjptrF             = (jnrlistF>=0) ? f+j_coord_offsetF : scratch;
 745             fjptrG             = (jnrlistG>=0) ? f+j_coord_offsetG : scratch;
 746             fjptrH             = (jnrlistH>=0) ? f+j_coord_offsetH : scratch;
 747             gmx_mm256_decrement_1rvec_8ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjptrE,fjptrF,fjptrG,fjptrH,tx,ty,tz);
 748
 749             }
 750
 751             /* Inner loop uses 30 flops */
 752         }
 753
 754         /* End of innermost loop */
 755
 756         gmx_mm256_update_iforce_1atom_swizzle_ps(fix0,fiy0,fiz0,
 757                                                  f+i_coord_offset,fshift+i_shift_offset);
 758
 759         /* Increment number of inner iterations */
 760         inneriter                  += j_index_end - j_index_start;
 761
 762         /* Outer loop uses 6 flops */
 763     }
 764
 765     /* Increment number of outer iterations */
 766     outeriter        += nri;
 767
 768     /* Update outer/inner flops */
 769
 770     inc_nrnb(nrnb,eNR_NBKERNEL_VDW_F,outeriter*6 + inneriter*30);
 771 }