src/gmxlib/nonbonded/nb_kernel_sse4_1_double/nb_kernel_ElecGB_VdwCSTab_GeomP1P1_sse4_1_double.c

   1 /*
   2  * Note: this file was generated by the Gromacs sse4_1_double kernel generator.
   3  *
   4  *                This source code is part of
   5  *
   6  *                 G   R   O   M   A   C   S
   7  *
   8  * Copyright (c) 2001-2012, The GROMACS Development Team
   9  *
  10  * Gromacs is a library for molecular simulation and trajectory analysis,
  11  * written by Erik Lindahl, David van der Spoel, Berk Hess, and others - for
  12  * a full list of developers and information, check out http://www.gromacs.org
  13  *
  14  * This program is free software; you can redistribute it and/or modify it under
  15  * the terms of the GNU Lesser General Public License as published by the Free
  16  * Software Foundation; either version 2 of the License, or (at your option) any
  17  * later version.
  18  *
  19  * To help fund GROMACS development, we humbly ask that you cite
  20  * the papers people have written on it - you can find them on the website.
  21  */
  22 #ifdef HAVE_CONFIG_H
  23 #include <config.h>
  24 #endif
  25
  26 #include <math.h>
  27
  28 #include "../nb_kernel.h"
  29 #include "types/simple.h"
  30 #include "vec.h"
  31 #include "nrnb.h"
  32
  33 #include "gmx_math_x86_sse4_1_double.h"
  34 #include "kernelutil_x86_sse4_1_double.h"
  35
  36 /*
  37  * Gromacs nonbonded kernel:   nb_kernel_ElecGB_VdwCSTab_GeomP1P1_VF_sse4_1_double
  38  * Electrostatics interaction: GeneralizedBorn
  39  * VdW interaction:            CubicSplineTable
  40  * Geometry:                   Particle-Particle
  41  * Calculate force/pot:        PotentialAndForce
  42  */
  43 void
  44 nb_kernel_ElecGB_VdwCSTab_GeomP1P1_VF_sse4_1_double
  45                     (t_nblist * gmx_restrict                nlist,
  46                      rvec * gmx_restrict                    xx,
  47                      rvec * gmx_restrict                    ff,
  48                      t_forcerec * gmx_restrict              fr,
  49                      t_mdatoms * gmx_restrict               mdatoms,
  50                      nb_kernel_data_t * gmx_restrict        kernel_data,
  51                      t_nrnb * gmx_restrict                  nrnb)
  52 {
  53     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  54      * just 0 for non-waters.
  55      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  56      * jnr indices corresponding to data put in the four positions in the SIMD register.
  57      */
  58     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  59     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  60     int              jnrA,jnrB;
  61     int              j_coord_offsetA,j_coord_offsetB;
  62     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  63     real             rcutoff_scalar;
  64     real             *shiftvec,*fshift,*x,*f;
  65     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  66     int              vdwioffset0;
  67     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  68     int              vdwjidx0A,vdwjidx0B;
  69     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  70     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  71     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  72     real             *charge;
  73     __m128i          gbitab;
  74     __m128d          vgb,fgb,vgbsum,dvdasum,gbscale,gbtabscale,isaprod,gbqqfactor,gbinvepsdiff,dvdaj,gbeps,dvdatmp;
  75     __m128d          minushalf = _mm_set1_pd(-0.5);
  76     real             *invsqrta,*dvda,*gbtab;
  77     int              nvdwtype;
  78     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  79     int              *vdwtype;
  80     real             *vdwparam;
  81     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
  82     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
  83     __m128i          vfitab;
  84     __m128i          ifour       = _mm_set1_epi32(4);
  85     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
  86     real             *vftab;
  87     __m128d          dummy_mask,cutoff_mask;
  88     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  89     __m128d          one     = _mm_set1_pd(1.0);
  90     __m128d          two     = _mm_set1_pd(2.0);
  91     x                = xx[0];
  92     f                = ff[0];
  93
  94     nri              = nlist->nri;
  95     iinr             = nlist->iinr;
  96     jindex           = nlist->jindex;
  97     jjnr             = nlist->jjnr;
  98     shiftidx         = nlist->shift;
  99     gid              = nlist->gid;
 100     shiftvec         = fr->shift_vec[0];
 101     fshift           = fr->fshift[0];
 102     facel            = _mm_set1_pd(fr->epsfac);
 103     charge           = mdatoms->chargeA;
 104     nvdwtype         = fr->ntype;
 105     vdwparam         = fr->nbfp;
 106     vdwtype          = mdatoms->typeA;
 107
 108     vftab            = kernel_data->table_vdw->data;
 109     vftabscale       = _mm_set1_pd(kernel_data->table_vdw->scale);
 110
 111     invsqrta         = fr->invsqrta;
 112     dvda             = fr->dvda;
 113     gbtabscale       = _mm_set1_pd(fr->gbtab.scale);
 114     gbtab            = fr->gbtab.data;
 115     gbinvepsdiff     = _mm_set1_pd((1.0/fr->epsilon_r) - (1.0/fr->gb_epsilon_solvent));
 116
 117     /* Avoid stupid compiler warnings */
 118     jnrA = jnrB = 0;
 119     j_coord_offsetA = 0;
 120     j_coord_offsetB = 0;
 121
 122     outeriter        = 0;
 123     inneriter        = 0;
 124
 125     /* Start outer loop over neighborlists */
 126     for(iidx=0; iidx<nri; iidx++)
 127     {
 128         /* Load shift vector for this list */
 129         i_shift_offset   = DIM*shiftidx[iidx];
 130
 131         /* Load limits for loop over neighbors */
 132         j_index_start    = jindex[iidx];
 133         j_index_end      = jindex[iidx+1];
 134
 135         /* Get outer coordinate index */
 136         inr              = iinr[iidx];
 137         i_coord_offset   = DIM*inr;
 138
 139         /* Load i particle coords and add shift vector */
 140         gmx_mm_load_shift_and_1rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 141
 142         fix0             = _mm_setzero_pd();
 143         fiy0             = _mm_setzero_pd();
 144         fiz0             = _mm_setzero_pd();
 145
 146         /* Load parameters for i particles */
 147         iq0              = _mm_mul_pd(facel,_mm_load1_pd(charge+inr+0));
 148         isai0            = _mm_load1_pd(invsqrta+inr+0);
 149         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 150
 151         /* Reset potential sums */
 152         velecsum         = _mm_setzero_pd();
 153         vgbsum           = _mm_setzero_pd();
 154         vvdwsum          = _mm_setzero_pd();
 155         dvdasum          = _mm_setzero_pd();
 156
 157         /* Start inner kernel loop */
 158         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 159         {
 160
 161             /* Get j neighbor index, and coordinate index */
 162             jnrA             = jjnr[jidx];
 163             jnrB             = jjnr[jidx+1];
 164             j_coord_offsetA  = DIM*jnrA;
 165             j_coord_offsetB  = DIM*jnrB;
 166
 167             /* load j atom coordinates */
 168             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 169                                               &jx0,&jy0,&jz0);
 170
 171             /* Calculate displacement vector */
 172             dx00             = _mm_sub_pd(ix0,jx0);
 173             dy00             = _mm_sub_pd(iy0,jy0);
 174             dz00             = _mm_sub_pd(iz0,jz0);
 175
 176             /* Calculate squared distance and things based on it */
 177             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 178
 179             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 180
 181             /* Load parameters for j particles */
 182             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 183             isaj0            = gmx_mm_load_2real_swizzle_pd(invsqrta+jnrA+0,invsqrta+jnrB+0);
 184             vdwjidx0A        = 2*vdwtype[jnrA+0];
 185             vdwjidx0B        = 2*vdwtype[jnrB+0];
 186
 187             /**************************
 188              * CALCULATE INTERACTIONS *
 189              **************************/
 190
 191             r00              = _mm_mul_pd(rsq00,rinv00);
 192
 193             /* Compute parameters for interactions between i and j atoms */
 194             qq00             = _mm_mul_pd(iq0,jq0);
 195             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 196                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 197
 198             /* Calculate table index by multiplying r with table scale and truncate to integer */
 199             rt               = _mm_mul_pd(r00,vftabscale);
 200             vfitab           = _mm_cvttpd_epi32(rt);
 201             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 202             vfitab           = _mm_slli_epi32(vfitab,3);
 203
 204             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 205             isaprod          = _mm_mul_pd(isai0,isaj0);
 206             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 207             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 208
 209             /* Calculate generalized born table index - this is a separate table from the normal one,
 210              * but we use the same procedure by multiplying r with scale and truncating to integer.
 211              */
 212             rt               = _mm_mul_pd(r00,gbscale);
 213             gbitab           = _mm_cvttpd_epi32(rt);
 214             gbeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 215             gbitab           = _mm_slli_epi32(gbitab,2);
 216
 217             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 218             F                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) );
 219             GMX_MM_TRANSPOSE2_PD(Y,F);
 220             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 221             H                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) +2);
 222             GMX_MM_TRANSPOSE2_PD(G,H);
 223             Heps             = _mm_mul_pd(gbeps,H);
 224             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 225             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 226             vgb              = _mm_mul_pd(gbqqfactor,VV);
 227
 228             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 229             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 230             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 231             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 232             gmx_mm_increment_2real_swizzle_pd(dvda+jnrA,dvda+jnrB,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 233             velec            = _mm_mul_pd(qq00,rinv00);
 234             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 235
 236             /* CUBIC SPLINE TABLE DISPERSION */
 237             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 238             F                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
 239             GMX_MM_TRANSPOSE2_PD(Y,F);
 240             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 241             H                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
 242             GMX_MM_TRANSPOSE2_PD(G,H);
 243             Heps             = _mm_mul_pd(vfeps,H);
 244             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 245             VV               = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
 246             vvdw6            = _mm_mul_pd(c6_00,VV);
 247             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 248             fvdw6            = _mm_mul_pd(c6_00,FF);
 249
 250             /* CUBIC SPLINE TABLE REPULSION */
 251             vfitab           = _mm_add_epi32(vfitab,ifour);
 252             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 253             F                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
 254             GMX_MM_TRANSPOSE2_PD(Y,F);
 255             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 256             H                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
 257             GMX_MM_TRANSPOSE2_PD(G,H);
 258             Heps             = _mm_mul_pd(vfeps,H);
 259             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 260             VV               = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
 261             vvdw12           = _mm_mul_pd(c12_00,VV);
 262             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 263             fvdw12           = _mm_mul_pd(c12_00,FF);
 264             vvdw             = _mm_add_pd(vvdw12,vvdw6);
 265             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 266
 267             /* Update potential sum for this i atom from the interaction with this j atom. */
 268             velecsum         = _mm_add_pd(velecsum,velec);
 269             vgbsum           = _mm_add_pd(vgbsum,vgb);
 270             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 271
 272             fscal            = _mm_add_pd(felec,fvdw);
 273
 274             /* Calculate temporary vectorial force */
 275             tx               = _mm_mul_pd(fscal,dx00);
 276             ty               = _mm_mul_pd(fscal,dy00);
 277             tz               = _mm_mul_pd(fscal,dz00);
 278
 279             /* Update vectorial force */
 280             fix0             = _mm_add_pd(fix0,tx);
 281             fiy0             = _mm_add_pd(fiy0,ty);
 282             fiz0             = _mm_add_pd(fiz0,tz);
 283
 284             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,tx,ty,tz);
 285
 286             /* Inner loop uses 92 flops */
 287         }
 288
 289         if(jidx<j_index_end)
 290         {
 291
 292             jnrA             = jjnr[jidx];
 293             j_coord_offsetA  = DIM*jnrA;
 294
 295             /* load j atom coordinates */
 296             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 297                                               &jx0,&jy0,&jz0);
 298
 299             /* Calculate displacement vector */
 300             dx00             = _mm_sub_pd(ix0,jx0);
 301             dy00             = _mm_sub_pd(iy0,jy0);
 302             dz00             = _mm_sub_pd(iz0,jz0);
 303
 304             /* Calculate squared distance and things based on it */
 305             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 306
 307             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 308
 309             /* Load parameters for j particles */
 310             jq0              = _mm_load_sd(charge+jnrA+0);
 311             isaj0            = _mm_load_sd(invsqrta+jnrA+0);
 312             vdwjidx0A        = 2*vdwtype[jnrA+0];
 313
 314             /**************************
 315              * CALCULATE INTERACTIONS *
 316              **************************/
 317
 318             r00              = _mm_mul_pd(rsq00,rinv00);
 319
 320             /* Compute parameters for interactions between i and j atoms */
 321             qq00             = _mm_mul_pd(iq0,jq0);
 322             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 323
 324             /* Calculate table index by multiplying r with table scale and truncate to integer */
 325             rt               = _mm_mul_pd(r00,vftabscale);
 326             vfitab           = _mm_cvttpd_epi32(rt);
 327             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 328             vfitab           = _mm_slli_epi32(vfitab,3);
 329
 330             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 331             isaprod          = _mm_mul_pd(isai0,isaj0);
 332             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 333             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 334
 335             /* Calculate generalized born table index - this is a separate table from the normal one,
 336              * but we use the same procedure by multiplying r with scale and truncating to integer.
 337              */
 338             rt               = _mm_mul_pd(r00,gbscale);
 339             gbitab           = _mm_cvttpd_epi32(rt);
 340             gbeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 341             gbitab           = _mm_slli_epi32(gbitab,2);
 342
 343             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 344             F                = _mm_setzero_pd();
 345             GMX_MM_TRANSPOSE2_PD(Y,F);
 346             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 347             H                = _mm_setzero_pd();
 348             GMX_MM_TRANSPOSE2_PD(G,H);
 349             Heps             = _mm_mul_pd(gbeps,H);
 350             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 351             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 352             vgb              = _mm_mul_pd(gbqqfactor,VV);
 353
 354             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 355             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 356             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 357             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 358             gmx_mm_increment_1real_pd(dvda+jnrA,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 359             velec            = _mm_mul_pd(qq00,rinv00);
 360             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 361
 362             /* CUBIC SPLINE TABLE DISPERSION */
 363             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 364             F                = _mm_setzero_pd();
 365             GMX_MM_TRANSPOSE2_PD(Y,F);
 366             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 367             H                = _mm_setzero_pd();
 368             GMX_MM_TRANSPOSE2_PD(G,H);
 369             Heps             = _mm_mul_pd(vfeps,H);
 370             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 371             VV               = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
 372             vvdw6            = _mm_mul_pd(c6_00,VV);
 373             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 374             fvdw6            = _mm_mul_pd(c6_00,FF);
 375
 376             /* CUBIC SPLINE TABLE REPULSION */
 377             vfitab           = _mm_add_epi32(vfitab,ifour);
 378             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 379             F                = _mm_setzero_pd();
 380             GMX_MM_TRANSPOSE2_PD(Y,F);
 381             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 382             H                = _mm_setzero_pd();
 383             GMX_MM_TRANSPOSE2_PD(G,H);
 384             Heps             = _mm_mul_pd(vfeps,H);
 385             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 386             VV               = _mm_add_pd(Y,_mm_mul_pd(vfeps,Fp));
 387             vvdw12           = _mm_mul_pd(c12_00,VV);
 388             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 389             fvdw12           = _mm_mul_pd(c12_00,FF);
 390             vvdw             = _mm_add_pd(vvdw12,vvdw6);
 391             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 392
 393             /* Update potential sum for this i atom from the interaction with this j atom. */
 394             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 395             velecsum         = _mm_add_pd(velecsum,velec);
 396             vgb              = _mm_unpacklo_pd(vgb,_mm_setzero_pd());
 397             vgbsum           = _mm_add_pd(vgbsum,vgb);
 398             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 399             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 400
 401             fscal            = _mm_add_pd(felec,fvdw);
 402
 403             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 404
 405             /* Calculate temporary vectorial force */
 406             tx               = _mm_mul_pd(fscal,dx00);
 407             ty               = _mm_mul_pd(fscal,dy00);
 408             tz               = _mm_mul_pd(fscal,dz00);
 409
 410             /* Update vectorial force */
 411             fix0             = _mm_add_pd(fix0,tx);
 412             fiy0             = _mm_add_pd(fiy0,ty);
 413             fiz0             = _mm_add_pd(fiz0,tz);
 414
 415             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,tx,ty,tz);
 416
 417             /* Inner loop uses 92 flops */
 418         }
 419
 420         /* End of innermost loop */
 421
 422         gmx_mm_update_iforce_1atom_swizzle_pd(fix0,fiy0,fiz0,
 423                                               f+i_coord_offset,fshift+i_shift_offset);
 424
 425         ggid                        = gid[iidx];
 426         /* Update potential energies */
 427         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 428         gmx_mm_update_1pot_pd(vgbsum,kernel_data->energygrp_polarization+ggid);
 429         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
 430         dvdasum = _mm_mul_pd(dvdasum, _mm_mul_pd(isai0,isai0));
 431         gmx_mm_update_1pot_pd(dvdasum,dvda+inr);
 432
 433         /* Increment number of inner iterations */
 434         inneriter                  += j_index_end - j_index_start;
 435
 436         /* Outer loop uses 10 flops */
 437     }
 438
 439     /* Increment number of outer iterations */
 440     outeriter        += nri;
 441
 442     /* Update outer/inner flops */
 443
 444     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_VF,outeriter*10 + inneriter*92);
 445 }
 446 /*
 447  * Gromacs nonbonded kernel:   nb_kernel_ElecGB_VdwCSTab_GeomP1P1_F_sse4_1_double
 448  * Electrostatics interaction: GeneralizedBorn
 449  * VdW interaction:            CubicSplineTable
 450  * Geometry:                   Particle-Particle
 451  * Calculate force/pot:        Force
 452  */
 453 void
 454 nb_kernel_ElecGB_VdwCSTab_GeomP1P1_F_sse4_1_double
 455                     (t_nblist * gmx_restrict                nlist,
 456                      rvec * gmx_restrict                    xx,
 457                      rvec * gmx_restrict                    ff,
 458                      t_forcerec * gmx_restrict              fr,
 459                      t_mdatoms * gmx_restrict               mdatoms,
 460                      nb_kernel_data_t * gmx_restrict        kernel_data,
 461                      t_nrnb * gmx_restrict                  nrnb)
 462 {
 463     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 464      * just 0 for non-waters.
 465      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 466      * jnr indices corresponding to data put in the four positions in the SIMD register.
 467      */
 468     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 469     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 470     int              jnrA,jnrB;
 471     int              j_coord_offsetA,j_coord_offsetB;
 472     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 473     real             rcutoff_scalar;
 474     real             *shiftvec,*fshift,*x,*f;
 475     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 476     int              vdwioffset0;
 477     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 478     int              vdwjidx0A,vdwjidx0B;
 479     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 480     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 481     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 482     real             *charge;
 483     __m128i          gbitab;
 484     __m128d          vgb,fgb,vgbsum,dvdasum,gbscale,gbtabscale,isaprod,gbqqfactor,gbinvepsdiff,dvdaj,gbeps,dvdatmp;
 485     __m128d          minushalf = _mm_set1_pd(-0.5);
 486     real             *invsqrta,*dvda,*gbtab;
 487     int              nvdwtype;
 488     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 489     int              *vdwtype;
 490     real             *vdwparam;
 491     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 492     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 493     __m128i          vfitab;
 494     __m128i          ifour       = _mm_set1_epi32(4);
 495     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 496     real             *vftab;
 497     __m128d          dummy_mask,cutoff_mask;
 498     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 499     __m128d          one     = _mm_set1_pd(1.0);
 500     __m128d          two     = _mm_set1_pd(2.0);
 501     x                = xx[0];
 502     f                = ff[0];
 503
 504     nri              = nlist->nri;
 505     iinr             = nlist->iinr;
 506     jindex           = nlist->jindex;
 507     jjnr             = nlist->jjnr;
 508     shiftidx         = nlist->shift;
 509     gid              = nlist->gid;
 510     shiftvec         = fr->shift_vec[0];
 511     fshift           = fr->fshift[0];
 512     facel            = _mm_set1_pd(fr->epsfac);
 513     charge           = mdatoms->chargeA;
 514     nvdwtype         = fr->ntype;
 515     vdwparam         = fr->nbfp;
 516     vdwtype          = mdatoms->typeA;
 517
 518     vftab            = kernel_data->table_vdw->data;
 519     vftabscale       = _mm_set1_pd(kernel_data->table_vdw->scale);
 520
 521     invsqrta         = fr->invsqrta;
 522     dvda             = fr->dvda;
 523     gbtabscale       = _mm_set1_pd(fr->gbtab.scale);
 524     gbtab            = fr->gbtab.data;
 525     gbinvepsdiff     = _mm_set1_pd((1.0/fr->epsilon_r) - (1.0/fr->gb_epsilon_solvent));
 526
 527     /* Avoid stupid compiler warnings */
 528     jnrA = jnrB = 0;
 529     j_coord_offsetA = 0;
 530     j_coord_offsetB = 0;
 531
 532     outeriter        = 0;
 533     inneriter        = 0;
 534
 535     /* Start outer loop over neighborlists */
 536     for(iidx=0; iidx<nri; iidx++)
 537     {
 538         /* Load shift vector for this list */
 539         i_shift_offset   = DIM*shiftidx[iidx];
 540
 541         /* Load limits for loop over neighbors */
 542         j_index_start    = jindex[iidx];
 543         j_index_end      = jindex[iidx+1];
 544
 545         /* Get outer coordinate index */
 546         inr              = iinr[iidx];
 547         i_coord_offset   = DIM*inr;
 548
 549         /* Load i particle coords and add shift vector */
 550         gmx_mm_load_shift_and_1rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 551
 552         fix0             = _mm_setzero_pd();
 553         fiy0             = _mm_setzero_pd();
 554         fiz0             = _mm_setzero_pd();
 555
 556         /* Load parameters for i particles */
 557         iq0              = _mm_mul_pd(facel,_mm_load1_pd(charge+inr+0));
 558         isai0            = _mm_load1_pd(invsqrta+inr+0);
 559         vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 560
 561         dvdasum          = _mm_setzero_pd();
 562
 563         /* Start inner kernel loop */
 564         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 565         {
 566
 567             /* Get j neighbor index, and coordinate index */
 568             jnrA             = jjnr[jidx];
 569             jnrB             = jjnr[jidx+1];
 570             j_coord_offsetA  = DIM*jnrA;
 571             j_coord_offsetB  = DIM*jnrB;
 572
 573             /* load j atom coordinates */
 574             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 575                                               &jx0,&jy0,&jz0);
 576
 577             /* Calculate displacement vector */
 578             dx00             = _mm_sub_pd(ix0,jx0);
 579             dy00             = _mm_sub_pd(iy0,jy0);
 580             dz00             = _mm_sub_pd(iz0,jz0);
 581
 582             /* Calculate squared distance and things based on it */
 583             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 584
 585             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 586
 587             /* Load parameters for j particles */
 588             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 589             isaj0            = gmx_mm_load_2real_swizzle_pd(invsqrta+jnrA+0,invsqrta+jnrB+0);
 590             vdwjidx0A        = 2*vdwtype[jnrA+0];
 591             vdwjidx0B        = 2*vdwtype[jnrB+0];
 592
 593             /**************************
 594              * CALCULATE INTERACTIONS *
 595              **************************/
 596
 597             r00              = _mm_mul_pd(rsq00,rinv00);
 598
 599             /* Compute parameters for interactions between i and j atoms */
 600             qq00             = _mm_mul_pd(iq0,jq0);
 601             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 602                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 603
 604             /* Calculate table index by multiplying r with table scale and truncate to integer */
 605             rt               = _mm_mul_pd(r00,vftabscale);
 606             vfitab           = _mm_cvttpd_epi32(rt);
 607             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 608             vfitab           = _mm_slli_epi32(vfitab,3);
 609
 610             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 611             isaprod          = _mm_mul_pd(isai0,isaj0);
 612             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 613             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 614
 615             /* Calculate generalized born table index - this is a separate table from the normal one,
 616              * but we use the same procedure by multiplying r with scale and truncating to integer.
 617              */
 618             rt               = _mm_mul_pd(r00,gbscale);
 619             gbitab           = _mm_cvttpd_epi32(rt);
 620             gbeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 621             gbitab           = _mm_slli_epi32(gbitab,2);
 622
 623             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 624             F                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) );
 625             GMX_MM_TRANSPOSE2_PD(Y,F);
 626             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 627             H                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) +2);
 628             GMX_MM_TRANSPOSE2_PD(G,H);
 629             Heps             = _mm_mul_pd(gbeps,H);
 630             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 631             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 632             vgb              = _mm_mul_pd(gbqqfactor,VV);
 633
 634             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 635             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 636             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 637             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 638             gmx_mm_increment_2real_swizzle_pd(dvda+jnrA,dvda+jnrB,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 639             velec            = _mm_mul_pd(qq00,rinv00);
 640             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 641
 642             /* CUBIC SPLINE TABLE DISPERSION */
 643             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 644             F                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
 645             GMX_MM_TRANSPOSE2_PD(Y,F);
 646             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 647             H                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
 648             GMX_MM_TRANSPOSE2_PD(G,H);
 649             Heps             = _mm_mul_pd(vfeps,H);
 650             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 651             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 652             fvdw6            = _mm_mul_pd(c6_00,FF);
 653
 654             /* CUBIC SPLINE TABLE REPULSION */
 655             vfitab           = _mm_add_epi32(vfitab,ifour);
 656             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 657             F                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) );
 658             GMX_MM_TRANSPOSE2_PD(Y,F);
 659             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 660             H                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,1) +2);
 661             GMX_MM_TRANSPOSE2_PD(G,H);
 662             Heps             = _mm_mul_pd(vfeps,H);
 663             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 664             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 665             fvdw12           = _mm_mul_pd(c12_00,FF);
 666             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 667
 668             fscal            = _mm_add_pd(felec,fvdw);
 669
 670             /* Calculate temporary vectorial force */
 671             tx               = _mm_mul_pd(fscal,dx00);
 672             ty               = _mm_mul_pd(fscal,dy00);
 673             tz               = _mm_mul_pd(fscal,dz00);
 674
 675             /* Update vectorial force */
 676             fix0             = _mm_add_pd(fix0,tx);
 677             fiy0             = _mm_add_pd(fiy0,ty);
 678             fiz0             = _mm_add_pd(fiz0,tz);
 679
 680             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,tx,ty,tz);
 681
 682             /* Inner loop uses 82 flops */
 683         }
 684
 685         if(jidx<j_index_end)
 686         {
 687
 688             jnrA             = jjnr[jidx];
 689             j_coord_offsetA  = DIM*jnrA;
 690
 691             /* load j atom coordinates */
 692             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 693                                               &jx0,&jy0,&jz0);
 694
 695             /* Calculate displacement vector */
 696             dx00             = _mm_sub_pd(ix0,jx0);
 697             dy00             = _mm_sub_pd(iy0,jy0);
 698             dz00             = _mm_sub_pd(iz0,jz0);
 699
 700             /* Calculate squared distance and things based on it */
 701             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 702
 703             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 704
 705             /* Load parameters for j particles */
 706             jq0              = _mm_load_sd(charge+jnrA+0);
 707             isaj0            = _mm_load_sd(invsqrta+jnrA+0);
 708             vdwjidx0A        = 2*vdwtype[jnrA+0];
 709
 710             /**************************
 711              * CALCULATE INTERACTIONS *
 712              **************************/
 713
 714             r00              = _mm_mul_pd(rsq00,rinv00);
 715
 716             /* Compute parameters for interactions between i and j atoms */
 717             qq00             = _mm_mul_pd(iq0,jq0);
 718             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 719
 720             /* Calculate table index by multiplying r with table scale and truncate to integer */
 721             rt               = _mm_mul_pd(r00,vftabscale);
 722             vfitab           = _mm_cvttpd_epi32(rt);
 723             vfeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 724             vfitab           = _mm_slli_epi32(vfitab,3);
 725
 726             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 727             isaprod          = _mm_mul_pd(isai0,isaj0);
 728             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 729             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 730
 731             /* Calculate generalized born table index - this is a separate table from the normal one,
 732              * but we use the same procedure by multiplying r with scale and truncating to integer.
 733              */
 734             rt               = _mm_mul_pd(r00,gbscale);
 735             gbitab           = _mm_cvttpd_epi32(rt);
 736             gbeps            = _mm_sub_pd(rt,_mm_round_pd(rt, _MM_FROUND_FLOOR));
 737             gbitab           = _mm_slli_epi32(gbitab,2);
 738
 739             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 740             F                = _mm_setzero_pd();
 741             GMX_MM_TRANSPOSE2_PD(Y,F);
 742             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 743             H                = _mm_setzero_pd();
 744             GMX_MM_TRANSPOSE2_PD(G,H);
 745             Heps             = _mm_mul_pd(gbeps,H);
 746             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 747             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 748             vgb              = _mm_mul_pd(gbqqfactor,VV);
 749
 750             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 751             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 752             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 753             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 754             gmx_mm_increment_1real_pd(dvda+jnrA,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 755             velec            = _mm_mul_pd(qq00,rinv00);
 756             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 757
 758             /* CUBIC SPLINE TABLE DISPERSION */
 759             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 760             F                = _mm_setzero_pd();
 761             GMX_MM_TRANSPOSE2_PD(Y,F);
 762             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 763             H                = _mm_setzero_pd();
 764             GMX_MM_TRANSPOSE2_PD(G,H);
 765             Heps             = _mm_mul_pd(vfeps,H);
 766             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 767             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 768             fvdw6            = _mm_mul_pd(c6_00,FF);
 769
 770             /* CUBIC SPLINE TABLE REPULSION */
 771             vfitab           = _mm_add_epi32(vfitab,ifour);
 772             Y                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) );
 773             F                = _mm_setzero_pd();
 774             GMX_MM_TRANSPOSE2_PD(Y,F);
 775             G                = _mm_load_pd( vftab + gmx_mm_extract_epi32(vfitab,0) +2);
 776             H                = _mm_setzero_pd();
 777             GMX_MM_TRANSPOSE2_PD(G,H);
 778             Heps             = _mm_mul_pd(vfeps,H);
 779             Fp               = _mm_add_pd(F,_mm_mul_pd(vfeps,_mm_add_pd(G,Heps)));
 780             FF               = _mm_add_pd(Fp,_mm_mul_pd(vfeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 781             fvdw12           = _mm_mul_pd(c12_00,FF);
 782             fvdw             = _mm_xor_pd(signbit,_mm_mul_pd(_mm_add_pd(fvdw6,fvdw12),_mm_mul_pd(vftabscale,rinv00)));
 783
 784             fscal            = _mm_add_pd(felec,fvdw);
 785
 786             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 787
 788             /* Calculate temporary vectorial force */
 789             tx               = _mm_mul_pd(fscal,dx00);
 790             ty               = _mm_mul_pd(fscal,dy00);
 791             tz               = _mm_mul_pd(fscal,dz00);
 792
 793             /* Update vectorial force */
 794             fix0             = _mm_add_pd(fix0,tx);
 795             fiy0             = _mm_add_pd(fiy0,ty);
 796             fiz0             = _mm_add_pd(fiz0,tz);
 797
 798             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,tx,ty,tz);
 799
 800             /* Inner loop uses 82 flops */
 801         }
 802
 803         /* End of innermost loop */
 804
 805         gmx_mm_update_iforce_1atom_swizzle_pd(fix0,fiy0,fiz0,
 806                                               f+i_coord_offset,fshift+i_shift_offset);
 807
 808         dvdasum = _mm_mul_pd(dvdasum, _mm_mul_pd(isai0,isai0));
 809         gmx_mm_update_1pot_pd(dvdasum,dvda+inr);
 810
 811         /* Increment number of inner iterations */
 812         inneriter                  += j_index_end - j_index_start;
 813
 814         /* Outer loop uses 7 flops */
 815     }
 816
 817     /* Increment number of outer iterations */
 818     outeriter        += nri;
 819
 820     /* Update outer/inner flops */
 821
 822     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_F,outeriter*7 + inneriter*82);
 823 }