src/gromacs/gmxlib/nonbonded/nb_kernel_sse2_double/nb_kernel_ElecGB_VdwNone_GeomP1P1_sse2_double.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015, 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 sse2_double kernel generator.
  37  */
  38 #include "gmxpre.h"
  39
  40 #include "config.h"
  41
  42 #include <math.h>
  43
  44 #include "../nb_kernel.h"
  45 #include "gromacs/math/vec.h"
  46 #include "gromacs/legacyheaders/nrnb.h"
  47
  48 #include "gromacs/simd/math_x86_sse2_double.h"
  49 #include "kernelutil_x86_sse2_double.h"
  50
  51 /*
  52  * Gromacs nonbonded kernel:   nb_kernel_ElecGB_VdwNone_GeomP1P1_VF_sse2_double
  53  * Electrostatics interaction: GeneralizedBorn
  54  * VdW interaction:            None
  55  * Geometry:                   Particle-Particle
  56  * Calculate force/pot:        PotentialAndForce
  57  */
  58 void
  59 nb_kernel_ElecGB_VdwNone_GeomP1P1_VF_sse2_double
  60                     (t_nblist                    * gmx_restrict       nlist,
  61                      rvec                        * gmx_restrict          xx,
  62                      rvec                        * gmx_restrict          ff,
  63                      t_forcerec                  * gmx_restrict          fr,
  64                      t_mdatoms                   * gmx_restrict     mdatoms,
  65                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  66                      t_nrnb                      * gmx_restrict        nrnb)
  67 {
  68     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  69      * just 0 for non-waters.
  70      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  71      * jnr indices corresponding to data put in the four positions in the SIMD register.
  72      */
  73     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  74     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  75     int              jnrA,jnrB;
  76     int              j_coord_offsetA,j_coord_offsetB;
  77     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  78     real             rcutoff_scalar;
  79     real             *shiftvec,*fshift,*x,*f;
  80     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  81     int              vdwioffset0;
  82     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  83     int              vdwjidx0A,vdwjidx0B;
  84     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  85     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  86     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  87     real             *charge;
  88     __m128i          gbitab;
  89     __m128d          vgb,fgb,vgbsum,dvdasum,gbscale,gbtabscale,isaprod,gbqqfactor,gbinvepsdiff,dvdaj,gbeps,dvdatmp;
  90     __m128d          minushalf = _mm_set1_pd(-0.5);
  91     real             *invsqrta,*dvda,*gbtab;
  92     __m128i          vfitab;
  93     __m128i          ifour       = _mm_set1_epi32(4);
  94     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
  95     real             *vftab;
  96     __m128d          dummy_mask,cutoff_mask;
  97     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
  98     __m128d          one     = _mm_set1_pd(1.0);
  99     __m128d          two     = _mm_set1_pd(2.0);
 100     x                = xx[0];
 101     f                = ff[0];
 102
 103     nri              = nlist->nri;
 104     iinr             = nlist->iinr;
 105     jindex           = nlist->jindex;
 106     jjnr             = nlist->jjnr;
 107     shiftidx         = nlist->shift;
 108     gid              = nlist->gid;
 109     shiftvec         = fr->shift_vec[0];
 110     fshift           = fr->fshift[0];
 111     facel            = _mm_set1_pd(fr->epsfac);
 112     charge           = mdatoms->chargeA;
 113
 114     invsqrta         = fr->invsqrta;
 115     dvda             = fr->dvda;
 116     gbtabscale       = _mm_set1_pd(fr->gbtab.scale);
 117     gbtab            = fr->gbtab.data;
 118     gbinvepsdiff     = _mm_set1_pd((1.0/fr->epsilon_r) - (1.0/fr->gb_epsilon_solvent));
 119
 120     /* Avoid stupid compiler warnings */
 121     jnrA = jnrB = 0;
 122     j_coord_offsetA = 0;
 123     j_coord_offsetB = 0;
 124
 125     outeriter        = 0;
 126     inneriter        = 0;
 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_mm_load_shift_and_1rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 144
 145         fix0             = _mm_setzero_pd();
 146         fiy0             = _mm_setzero_pd();
 147         fiz0             = _mm_setzero_pd();
 148
 149         /* Load parameters for i particles */
 150         iq0              = _mm_mul_pd(facel,_mm_load1_pd(charge+inr+0));
 151         isai0            = _mm_load1_pd(invsqrta+inr+0);
 152
 153         /* Reset potential sums */
 154         velecsum         = _mm_setzero_pd();
 155         vgbsum           = _mm_setzero_pd();
 156         dvdasum          = _mm_setzero_pd();
 157
 158         /* Start inner kernel loop */
 159         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 160         {
 161
 162             /* Get j neighbor index, and coordinate index */
 163             jnrA             = jjnr[jidx];
 164             jnrB             = jjnr[jidx+1];
 165             j_coord_offsetA  = DIM*jnrA;
 166             j_coord_offsetB  = DIM*jnrB;
 167
 168             /* load j atom coordinates */
 169             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 170                                               &jx0,&jy0,&jz0);
 171
 172             /* Calculate displacement vector */
 173             dx00             = _mm_sub_pd(ix0,jx0);
 174             dy00             = _mm_sub_pd(iy0,jy0);
 175             dz00             = _mm_sub_pd(iz0,jz0);
 176
 177             /* Calculate squared distance and things based on it */
 178             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 179
 180             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 181
 182             /* Load parameters for j particles */
 183             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 184             isaj0            = gmx_mm_load_2real_swizzle_pd(invsqrta+jnrA+0,invsqrta+jnrB+0);
 185
 186             /**************************
 187              * CALCULATE INTERACTIONS *
 188              **************************/
 189
 190             r00              = _mm_mul_pd(rsq00,rinv00);
 191
 192             /* Compute parameters for interactions between i and j atoms */
 193             qq00             = _mm_mul_pd(iq0,jq0);
 194
 195             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 196             isaprod          = _mm_mul_pd(isai0,isaj0);
 197             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 198             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 199
 200             /* Calculate generalized born table index - this is a separate table from the normal one,
 201              * but we use the same procedure by multiplying r with scale and truncating to integer.
 202              */
 203             rt               = _mm_mul_pd(r00,gbscale);
 204             gbitab           = _mm_cvttpd_epi32(rt);
 205             gbeps            = _mm_sub_pd(rt,_mm_cvtepi32_pd(gbitab));
 206             gbitab           = _mm_slli_epi32(gbitab,2);
 207
 208             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 209             F                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) );
 210             GMX_MM_TRANSPOSE2_PD(Y,F);
 211             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 212             H                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) +2);
 213             GMX_MM_TRANSPOSE2_PD(G,H);
 214             Heps             = _mm_mul_pd(gbeps,H);
 215             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 216             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 217             vgb              = _mm_mul_pd(gbqqfactor,VV);
 218
 219             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 220             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 221             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 222             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 223             gmx_mm_increment_2real_swizzle_pd(dvda+jnrA,dvda+jnrB,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 224             velec            = _mm_mul_pd(qq00,rinv00);
 225             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 226
 227             /* Update potential sum for this i atom from the interaction with this j atom. */
 228             velecsum         = _mm_add_pd(velecsum,velec);
 229             vgbsum           = _mm_add_pd(vgbsum,vgb);
 230
 231             fscal            = felec;
 232
 233             /* Calculate temporary vectorial force */
 234             tx               = _mm_mul_pd(fscal,dx00);
 235             ty               = _mm_mul_pd(fscal,dy00);
 236             tz               = _mm_mul_pd(fscal,dz00);
 237
 238             /* Update vectorial force */
 239             fix0             = _mm_add_pd(fix0,tx);
 240             fiy0             = _mm_add_pd(fiy0,ty);
 241             fiz0             = _mm_add_pd(fiz0,tz);
 242
 243             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,tx,ty,tz);
 244
 245             /* Inner loop uses 58 flops */
 246         }
 247
 248         if(jidx<j_index_end)
 249         {
 250
 251             jnrA             = jjnr[jidx];
 252             j_coord_offsetA  = DIM*jnrA;
 253
 254             /* load j atom coordinates */
 255             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 256                                               &jx0,&jy0,&jz0);
 257
 258             /* Calculate displacement vector */
 259             dx00             = _mm_sub_pd(ix0,jx0);
 260             dy00             = _mm_sub_pd(iy0,jy0);
 261             dz00             = _mm_sub_pd(iz0,jz0);
 262
 263             /* Calculate squared distance and things based on it */
 264             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 265
 266             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 267
 268             /* Load parameters for j particles */
 269             jq0              = _mm_load_sd(charge+jnrA+0);
 270             isaj0            = _mm_load_sd(invsqrta+jnrA+0);
 271
 272             /**************************
 273              * CALCULATE INTERACTIONS *
 274              **************************/
 275
 276             r00              = _mm_mul_pd(rsq00,rinv00);
 277
 278             /* Compute parameters for interactions between i and j atoms */
 279             qq00             = _mm_mul_pd(iq0,jq0);
 280
 281             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 282             isaprod          = _mm_mul_pd(isai0,isaj0);
 283             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 284             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 285
 286             /* Calculate generalized born table index - this is a separate table from the normal one,
 287              * but we use the same procedure by multiplying r with scale and truncating to integer.
 288              */
 289             rt               = _mm_mul_pd(r00,gbscale);
 290             gbitab           = _mm_cvttpd_epi32(rt);
 291             gbeps            = _mm_sub_pd(rt,_mm_cvtepi32_pd(gbitab));
 292             gbitab           = _mm_slli_epi32(gbitab,2);
 293
 294             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 295             F                = _mm_setzero_pd();
 296             GMX_MM_TRANSPOSE2_PD(Y,F);
 297             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 298             H                = _mm_setzero_pd();
 299             GMX_MM_TRANSPOSE2_PD(G,H);
 300             Heps             = _mm_mul_pd(gbeps,H);
 301             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 302             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 303             vgb              = _mm_mul_pd(gbqqfactor,VV);
 304
 305             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 306             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 307             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 308             dvdatmp          = _mm_unpacklo_pd(dvdatmp,_mm_setzero_pd());
 309             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 310             gmx_mm_increment_1real_pd(dvda+jnrA,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 311             velec            = _mm_mul_pd(qq00,rinv00);
 312             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 313
 314             /* Update potential sum for this i atom from the interaction with this j atom. */
 315             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 316             velecsum         = _mm_add_pd(velecsum,velec);
 317             vgb              = _mm_unpacklo_pd(vgb,_mm_setzero_pd());
 318             vgbsum           = _mm_add_pd(vgbsum,vgb);
 319
 320             fscal            = felec;
 321
 322             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 323
 324             /* Calculate temporary vectorial force */
 325             tx               = _mm_mul_pd(fscal,dx00);
 326             ty               = _mm_mul_pd(fscal,dy00);
 327             tz               = _mm_mul_pd(fscal,dz00);
 328
 329             /* Update vectorial force */
 330             fix0             = _mm_add_pd(fix0,tx);
 331             fiy0             = _mm_add_pd(fiy0,ty);
 332             fiz0             = _mm_add_pd(fiz0,tz);
 333
 334             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,tx,ty,tz);
 335
 336             /* Inner loop uses 58 flops */
 337         }
 338
 339         /* End of innermost loop */
 340
 341         gmx_mm_update_iforce_1atom_swizzle_pd(fix0,fiy0,fiz0,
 342                                               f+i_coord_offset,fshift+i_shift_offset);
 343
 344         ggid                        = gid[iidx];
 345         /* Update potential energies */
 346         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 347         gmx_mm_update_1pot_pd(vgbsum,kernel_data->energygrp_polarization+ggid);
 348         dvdasum = _mm_mul_pd(dvdasum, _mm_mul_pd(isai0,isai0));
 349         gmx_mm_update_1pot_pd(dvdasum,dvda+inr);
 350
 351         /* Increment number of inner iterations */
 352         inneriter                  += j_index_end - j_index_start;
 353
 354         /* Outer loop uses 9 flops */
 355     }
 356
 357     /* Increment number of outer iterations */
 358     outeriter        += nri;
 359
 360     /* Update outer/inner flops */
 361
 362     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VF,outeriter*9 + inneriter*58);
 363 }
 364 /*
 365  * Gromacs nonbonded kernel:   nb_kernel_ElecGB_VdwNone_GeomP1P1_F_sse2_double
 366  * Electrostatics interaction: GeneralizedBorn
 367  * VdW interaction:            None
 368  * Geometry:                   Particle-Particle
 369  * Calculate force/pot:        Force
 370  */
 371 void
 372 nb_kernel_ElecGB_VdwNone_GeomP1P1_F_sse2_double
 373                     (t_nblist                    * gmx_restrict       nlist,
 374                      rvec                        * gmx_restrict          xx,
 375                      rvec                        * gmx_restrict          ff,
 376                      t_forcerec                  * gmx_restrict          fr,
 377                      t_mdatoms                   * gmx_restrict     mdatoms,
 378                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 379                      t_nrnb                      * gmx_restrict        nrnb)
 380 {
 381     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 382      * just 0 for non-waters.
 383      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 384      * jnr indices corresponding to data put in the four positions in the SIMD register.
 385      */
 386     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 387     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 388     int              jnrA,jnrB;
 389     int              j_coord_offsetA,j_coord_offsetB;
 390     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 391     real             rcutoff_scalar;
 392     real             *shiftvec,*fshift,*x,*f;
 393     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 394     int              vdwioffset0;
 395     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 396     int              vdwjidx0A,vdwjidx0B;
 397     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 398     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 399     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 400     real             *charge;
 401     __m128i          gbitab;
 402     __m128d          vgb,fgb,vgbsum,dvdasum,gbscale,gbtabscale,isaprod,gbqqfactor,gbinvepsdiff,dvdaj,gbeps,dvdatmp;
 403     __m128d          minushalf = _mm_set1_pd(-0.5);
 404     real             *invsqrta,*dvda,*gbtab;
 405     __m128i          vfitab;
 406     __m128i          ifour       = _mm_set1_epi32(4);
 407     __m128d          rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF;
 408     real             *vftab;
 409     __m128d          dummy_mask,cutoff_mask;
 410     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 411     __m128d          one     = _mm_set1_pd(1.0);
 412     __m128d          two     = _mm_set1_pd(2.0);
 413     x                = xx[0];
 414     f                = ff[0];
 415
 416     nri              = nlist->nri;
 417     iinr             = nlist->iinr;
 418     jindex           = nlist->jindex;
 419     jjnr             = nlist->jjnr;
 420     shiftidx         = nlist->shift;
 421     gid              = nlist->gid;
 422     shiftvec         = fr->shift_vec[0];
 423     fshift           = fr->fshift[0];
 424     facel            = _mm_set1_pd(fr->epsfac);
 425     charge           = mdatoms->chargeA;
 426
 427     invsqrta         = fr->invsqrta;
 428     dvda             = fr->dvda;
 429     gbtabscale       = _mm_set1_pd(fr->gbtab.scale);
 430     gbtab            = fr->gbtab.data;
 431     gbinvepsdiff     = _mm_set1_pd((1.0/fr->epsilon_r) - (1.0/fr->gb_epsilon_solvent));
 432
 433     /* Avoid stupid compiler warnings */
 434     jnrA = jnrB = 0;
 435     j_coord_offsetA = 0;
 436     j_coord_offsetB = 0;
 437
 438     outeriter        = 0;
 439     inneriter        = 0;
 440
 441     /* Start outer loop over neighborlists */
 442     for(iidx=0; iidx<nri; iidx++)
 443     {
 444         /* Load shift vector for this list */
 445         i_shift_offset   = DIM*shiftidx[iidx];
 446
 447         /* Load limits for loop over neighbors */
 448         j_index_start    = jindex[iidx];
 449         j_index_end      = jindex[iidx+1];
 450
 451         /* Get outer coordinate index */
 452         inr              = iinr[iidx];
 453         i_coord_offset   = DIM*inr;
 454
 455         /* Load i particle coords and add shift vector */
 456         gmx_mm_load_shift_and_1rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,&ix0,&iy0,&iz0);
 457
 458         fix0             = _mm_setzero_pd();
 459         fiy0             = _mm_setzero_pd();
 460         fiz0             = _mm_setzero_pd();
 461
 462         /* Load parameters for i particles */
 463         iq0              = _mm_mul_pd(facel,_mm_load1_pd(charge+inr+0));
 464         isai0            = _mm_load1_pd(invsqrta+inr+0);
 465
 466         dvdasum          = _mm_setzero_pd();
 467
 468         /* Start inner kernel loop */
 469         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 470         {
 471
 472             /* Get j neighbor index, and coordinate index */
 473             jnrA             = jjnr[jidx];
 474             jnrB             = jjnr[jidx+1];
 475             j_coord_offsetA  = DIM*jnrA;
 476             j_coord_offsetB  = DIM*jnrB;
 477
 478             /* load j atom coordinates */
 479             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 480                                               &jx0,&jy0,&jz0);
 481
 482             /* Calculate displacement vector */
 483             dx00             = _mm_sub_pd(ix0,jx0);
 484             dy00             = _mm_sub_pd(iy0,jy0);
 485             dz00             = _mm_sub_pd(iz0,jz0);
 486
 487             /* Calculate squared distance and things based on it */
 488             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 489
 490             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 491
 492             /* Load parameters for j particles */
 493             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 494             isaj0            = gmx_mm_load_2real_swizzle_pd(invsqrta+jnrA+0,invsqrta+jnrB+0);
 495
 496             /**************************
 497              * CALCULATE INTERACTIONS *
 498              **************************/
 499
 500             r00              = _mm_mul_pd(rsq00,rinv00);
 501
 502             /* Compute parameters for interactions between i and j atoms */
 503             qq00             = _mm_mul_pd(iq0,jq0);
 504
 505             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 506             isaprod          = _mm_mul_pd(isai0,isaj0);
 507             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 508             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 509
 510             /* Calculate generalized born table index - this is a separate table from the normal one,
 511              * but we use the same procedure by multiplying r with scale and truncating to integer.
 512              */
 513             rt               = _mm_mul_pd(r00,gbscale);
 514             gbitab           = _mm_cvttpd_epi32(rt);
 515             gbeps            = _mm_sub_pd(rt,_mm_cvtepi32_pd(gbitab));
 516             gbitab           = _mm_slli_epi32(gbitab,2);
 517
 518             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 519             F                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) );
 520             GMX_MM_TRANSPOSE2_PD(Y,F);
 521             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 522             H                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,1) +2);
 523             GMX_MM_TRANSPOSE2_PD(G,H);
 524             Heps             = _mm_mul_pd(gbeps,H);
 525             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 526             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 527             vgb              = _mm_mul_pd(gbqqfactor,VV);
 528
 529             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 530             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 531             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 532             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 533             gmx_mm_increment_2real_swizzle_pd(dvda+jnrA,dvda+jnrB,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 534             velec            = _mm_mul_pd(qq00,rinv00);
 535             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 536
 537             fscal            = felec;
 538
 539             /* Calculate temporary vectorial force */
 540             tx               = _mm_mul_pd(fscal,dx00);
 541             ty               = _mm_mul_pd(fscal,dy00);
 542             tz               = _mm_mul_pd(fscal,dz00);
 543
 544             /* Update vectorial force */
 545             fix0             = _mm_add_pd(fix0,tx);
 546             fiy0             = _mm_add_pd(fiy0,ty);
 547             fiz0             = _mm_add_pd(fiz0,tz);
 548
 549             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,tx,ty,tz);
 550
 551             /* Inner loop uses 56 flops */
 552         }
 553
 554         if(jidx<j_index_end)
 555         {
 556
 557             jnrA             = jjnr[jidx];
 558             j_coord_offsetA  = DIM*jnrA;
 559
 560             /* load j atom coordinates */
 561             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 562                                               &jx0,&jy0,&jz0);
 563
 564             /* Calculate displacement vector */
 565             dx00             = _mm_sub_pd(ix0,jx0);
 566             dy00             = _mm_sub_pd(iy0,jy0);
 567             dz00             = _mm_sub_pd(iz0,jz0);
 568
 569             /* Calculate squared distance and things based on it */
 570             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 571
 572             rinv00           = gmx_mm_invsqrt_pd(rsq00);
 573
 574             /* Load parameters for j particles */
 575             jq0              = _mm_load_sd(charge+jnrA+0);
 576             isaj0            = _mm_load_sd(invsqrta+jnrA+0);
 577
 578             /**************************
 579              * CALCULATE INTERACTIONS *
 580              **************************/
 581
 582             r00              = _mm_mul_pd(rsq00,rinv00);
 583
 584             /* Compute parameters for interactions between i and j atoms */
 585             qq00             = _mm_mul_pd(iq0,jq0);
 586
 587             /* GENERALIZED BORN AND COULOMB ELECTROSTATICS */
 588             isaprod          = _mm_mul_pd(isai0,isaj0);
 589             gbqqfactor       = _mm_xor_pd(signbit,_mm_mul_pd(qq00,_mm_mul_pd(isaprod,gbinvepsdiff)));
 590             gbscale          = _mm_mul_pd(isaprod,gbtabscale);
 591
 592             /* Calculate generalized born table index - this is a separate table from the normal one,
 593              * but we use the same procedure by multiplying r with scale and truncating to integer.
 594              */
 595             rt               = _mm_mul_pd(r00,gbscale);
 596             gbitab           = _mm_cvttpd_epi32(rt);
 597             gbeps            = _mm_sub_pd(rt,_mm_cvtepi32_pd(gbitab));
 598             gbitab           = _mm_slli_epi32(gbitab,2);
 599
 600             Y                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) );
 601             F                = _mm_setzero_pd();
 602             GMX_MM_TRANSPOSE2_PD(Y,F);
 603             G                = _mm_load_pd( gbtab + gmx_mm_extract_epi32(gbitab,0) +2);
 604             H                = _mm_setzero_pd();
 605             GMX_MM_TRANSPOSE2_PD(G,H);
 606             Heps             = _mm_mul_pd(gbeps,H);
 607             Fp               = _mm_add_pd(F,_mm_mul_pd(gbeps,_mm_add_pd(G,Heps)));
 608             VV               = _mm_add_pd(Y,_mm_mul_pd(gbeps,Fp));
 609             vgb              = _mm_mul_pd(gbqqfactor,VV);
 610
 611             FF               = _mm_add_pd(Fp,_mm_mul_pd(gbeps,_mm_add_pd(G,_mm_add_pd(Heps,Heps))));
 612             fgb              = _mm_mul_pd(gbqqfactor,_mm_mul_pd(FF,gbscale));
 613             dvdatmp          = _mm_mul_pd(minushalf,_mm_add_pd(vgb,_mm_mul_pd(fgb,r00)));
 614             dvdatmp          = _mm_unpacklo_pd(dvdatmp,_mm_setzero_pd());
 615             dvdasum          = _mm_add_pd(dvdasum,dvdatmp);
 616             gmx_mm_increment_1real_pd(dvda+jnrA,_mm_mul_pd(dvdatmp,_mm_mul_pd(isaj0,isaj0)));
 617             velec            = _mm_mul_pd(qq00,rinv00);
 618             felec            = _mm_mul_pd(_mm_sub_pd(_mm_mul_pd(velec,rinv00),fgb),rinv00);
 619
 620             fscal            = felec;
 621
 622             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 623
 624             /* Calculate temporary vectorial force */
 625             tx               = _mm_mul_pd(fscal,dx00);
 626             ty               = _mm_mul_pd(fscal,dy00);
 627             tz               = _mm_mul_pd(fscal,dz00);
 628
 629             /* Update vectorial force */
 630             fix0             = _mm_add_pd(fix0,tx);
 631             fiy0             = _mm_add_pd(fiy0,ty);
 632             fiz0             = _mm_add_pd(fiz0,tz);
 633
 634             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,tx,ty,tz);
 635
 636             /* Inner loop uses 56 flops */
 637         }
 638
 639         /* End of innermost loop */
 640
 641         gmx_mm_update_iforce_1atom_swizzle_pd(fix0,fiy0,fiz0,
 642                                               f+i_coord_offset,fshift+i_shift_offset);
 643
 644         dvdasum = _mm_mul_pd(dvdasum, _mm_mul_pd(isai0,isai0));
 645         gmx_mm_update_1pot_pd(dvdasum,dvda+inr);
 646
 647         /* Increment number of inner iterations */
 648         inneriter                  += j_index_end - j_index_start;
 649
 650         /* Outer loop uses 7 flops */
 651     }
 652
 653     /* Increment number of outer iterations */
 654     outeriter        += nri;
 655
 656     /* Update outer/inner flops */
 657
 658     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_F,outeriter*7 + inneriter*56);
 659 }