src/gromacs/gmxlib/nonbonded/nb_kernel_avx_128_fma_double/nb_kernel_ElecRF_VdwLJ_GeomW3P1_avx_128_fma_double.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014,2015,2017, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*
  36  * Note: this file was generated by the GROMACS avx_128_fma_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/gmxlib/nrnb.h"
  46
  47 #include "kernelutil_x86_avx_128_fma_double.h"
  48
  49 /*
  50  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwLJ_GeomW3P1_VF_avx_128_fma_double
  51  * Electrostatics interaction: ReactionField
  52  * VdW interaction:            LennardJones
  53  * Geometry:                   Water3-Particle
  54  * Calculate force/pot:        PotentialAndForce
  55  */
  56 void
  57 nb_kernel_ElecRF_VdwLJ_GeomW3P1_VF_avx_128_fma_double
  58                     (t_nblist                    * gmx_restrict       nlist,
  59                      rvec                        * gmx_restrict          xx,
  60                      rvec                        * gmx_restrict          ff,
  61                      struct t_forcerec           * gmx_restrict          fr,
  62                      t_mdatoms                   * gmx_restrict     mdatoms,
  63                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  64                      t_nrnb                      * gmx_restrict        nrnb)
  65 {
  66     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
  67      * just 0 for non-waters.
  68      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
  69      * jnr indices corresponding to data put in the four positions in the SIMD register.
  70      */
  71     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
  72     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
  73     int              jnrA,jnrB;
  74     int              j_coord_offsetA,j_coord_offsetB;
  75     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  76     real             rcutoff_scalar;
  77     real             *shiftvec,*fshift,*x,*f;
  78     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
  79     int              vdwioffset0;
  80     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
  81     int              vdwioffset1;
  82     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  83     int              vdwioffset2;
  84     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  85     int              vdwjidx0A,vdwjidx0B;
  86     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  87     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
  88     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
  89     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
  90     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
  91     real             *charge;
  92     int              nvdwtype;
  93     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
  94     int              *vdwtype;
  95     real             *vdwparam;
  96     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
  97     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
  98     __m128d          dummy_mask,cutoff_mask;
  99     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 100     __m128d          one     = _mm_set1_pd(1.0);
 101     __m128d          two     = _mm_set1_pd(2.0);
 102     x                = xx[0];
 103     f                = ff[0];
 104
 105     nri              = nlist->nri;
 106     iinr             = nlist->iinr;
 107     jindex           = nlist->jindex;
 108     jjnr             = nlist->jjnr;
 109     shiftidx         = nlist->shift;
 110     gid              = nlist->gid;
 111     shiftvec         = fr->shift_vec[0];
 112     fshift           = fr->fshift[0];
 113     facel            = _mm_set1_pd(fr->ic->epsfac);
 114     charge           = mdatoms->chargeA;
 115     krf              = _mm_set1_pd(fr->ic->k_rf);
 116     krf2             = _mm_set1_pd(fr->ic->k_rf*2.0);
 117     crf              = _mm_set1_pd(fr->ic->c_rf);
 118     nvdwtype         = fr->ntype;
 119     vdwparam         = fr->nbfp;
 120     vdwtype          = mdatoms->typeA;
 121
 122     /* Setup water-specific parameters */
 123     inr              = nlist->iinr[0];
 124     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 125     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 126     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 127     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 128
 129     /* Avoid stupid compiler warnings */
 130     jnrA = jnrB = 0;
 131     j_coord_offsetA = 0;
 132     j_coord_offsetB = 0;
 133
 134     outeriter        = 0;
 135     inneriter        = 0;
 136
 137     /* Start outer loop over neighborlists */
 138     for(iidx=0; iidx<nri; iidx++)
 139     {
 140         /* Load shift vector for this list */
 141         i_shift_offset   = DIM*shiftidx[iidx];
 142
 143         /* Load limits for loop over neighbors */
 144         j_index_start    = jindex[iidx];
 145         j_index_end      = jindex[iidx+1];
 146
 147         /* Get outer coordinate index */
 148         inr              = iinr[iidx];
 149         i_coord_offset   = DIM*inr;
 150
 151         /* Load i particle coords and add shift vector */
 152         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 153                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 154
 155         fix0             = _mm_setzero_pd();
 156         fiy0             = _mm_setzero_pd();
 157         fiz0             = _mm_setzero_pd();
 158         fix1             = _mm_setzero_pd();
 159         fiy1             = _mm_setzero_pd();
 160         fiz1             = _mm_setzero_pd();
 161         fix2             = _mm_setzero_pd();
 162         fiy2             = _mm_setzero_pd();
 163         fiz2             = _mm_setzero_pd();
 164
 165         /* Reset potential sums */
 166         velecsum         = _mm_setzero_pd();
 167         vvdwsum          = _mm_setzero_pd();
 168
 169         /* Start inner kernel loop */
 170         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 171         {
 172
 173             /* Get j neighbor index, and coordinate index */
 174             jnrA             = jjnr[jidx];
 175             jnrB             = jjnr[jidx+1];
 176             j_coord_offsetA  = DIM*jnrA;
 177             j_coord_offsetB  = DIM*jnrB;
 178
 179             /* load j atom coordinates */
 180             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 181                                               &jx0,&jy0,&jz0);
 182
 183             /* Calculate displacement vector */
 184             dx00             = _mm_sub_pd(ix0,jx0);
 185             dy00             = _mm_sub_pd(iy0,jy0);
 186             dz00             = _mm_sub_pd(iz0,jz0);
 187             dx10             = _mm_sub_pd(ix1,jx0);
 188             dy10             = _mm_sub_pd(iy1,jy0);
 189             dz10             = _mm_sub_pd(iz1,jz0);
 190             dx20             = _mm_sub_pd(ix2,jx0);
 191             dy20             = _mm_sub_pd(iy2,jy0);
 192             dz20             = _mm_sub_pd(iz2,jz0);
 193
 194             /* Calculate squared distance and things based on it */
 195             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 196             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 197             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 198
 199             rinv00           = avx128fma_invsqrt_d(rsq00);
 200             rinv10           = avx128fma_invsqrt_d(rsq10);
 201             rinv20           = avx128fma_invsqrt_d(rsq20);
 202
 203             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 204             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 205             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 206
 207             /* Load parameters for j particles */
 208             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 209             vdwjidx0A        = 2*vdwtype[jnrA+0];
 210             vdwjidx0B        = 2*vdwtype[jnrB+0];
 211
 212             fjx0             = _mm_setzero_pd();
 213             fjy0             = _mm_setzero_pd();
 214             fjz0             = _mm_setzero_pd();
 215
 216             /**************************
 217              * CALCULATE INTERACTIONS *
 218              **************************/
 219
 220             /* Compute parameters for interactions between i and j atoms */
 221             qq00             = _mm_mul_pd(iq0,jq0);
 222             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 223                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 224
 225             /* REACTION-FIELD ELECTROSTATICS */
 226             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_macc_pd(krf,rsq00,rinv00),crf));
 227             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 228
 229             /* LENNARD-JONES DISPERSION/REPULSION */
 230
 231             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 232             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 233             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 234             vvdw             = _mm_msub_pd( vvdw12,one_twelfth, _mm_mul_pd(vvdw6,one_sixth) );
 235             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 236
 237             /* Update potential sum for this i atom from the interaction with this j atom. */
 238             velecsum         = _mm_add_pd(velecsum,velec);
 239             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 240
 241             fscal            = _mm_add_pd(felec,fvdw);
 242
 243             /* Update vectorial force */
 244             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 245             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 246             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 247
 248             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 249             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 250             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 251
 252             /**************************
 253              * CALCULATE INTERACTIONS *
 254              **************************/
 255
 256             /* Compute parameters for interactions between i and j atoms */
 257             qq10             = _mm_mul_pd(iq1,jq0);
 258
 259             /* REACTION-FIELD ELECTROSTATICS */
 260             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_macc_pd(krf,rsq10,rinv10),crf));
 261             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 262
 263             /* Update potential sum for this i atom from the interaction with this j atom. */
 264             velecsum         = _mm_add_pd(velecsum,velec);
 265
 266             fscal            = felec;
 267
 268             /* Update vectorial force */
 269             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 270             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 271             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 272
 273             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 274             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 275             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 276
 277             /**************************
 278              * CALCULATE INTERACTIONS *
 279              **************************/
 280
 281             /* Compute parameters for interactions between i and j atoms */
 282             qq20             = _mm_mul_pd(iq2,jq0);
 283
 284             /* REACTION-FIELD ELECTROSTATICS */
 285             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_macc_pd(krf,rsq20,rinv20),crf));
 286             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 287
 288             /* Update potential sum for this i atom from the interaction with this j atom. */
 289             velecsum         = _mm_add_pd(velecsum,velec);
 290
 291             fscal            = felec;
 292
 293             /* Update vectorial force */
 294             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 295             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 296             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 297
 298             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 299             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 300             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 301
 302             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 303
 304             /* Inner loop uses 120 flops */
 305         }
 306
 307         if(jidx<j_index_end)
 308         {
 309
 310             jnrA             = jjnr[jidx];
 311             j_coord_offsetA  = DIM*jnrA;
 312
 313             /* load j atom coordinates */
 314             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 315                                               &jx0,&jy0,&jz0);
 316
 317             /* Calculate displacement vector */
 318             dx00             = _mm_sub_pd(ix0,jx0);
 319             dy00             = _mm_sub_pd(iy0,jy0);
 320             dz00             = _mm_sub_pd(iz0,jz0);
 321             dx10             = _mm_sub_pd(ix1,jx0);
 322             dy10             = _mm_sub_pd(iy1,jy0);
 323             dz10             = _mm_sub_pd(iz1,jz0);
 324             dx20             = _mm_sub_pd(ix2,jx0);
 325             dy20             = _mm_sub_pd(iy2,jy0);
 326             dz20             = _mm_sub_pd(iz2,jz0);
 327
 328             /* Calculate squared distance and things based on it */
 329             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 330             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 331             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 332
 333             rinv00           = avx128fma_invsqrt_d(rsq00);
 334             rinv10           = avx128fma_invsqrt_d(rsq10);
 335             rinv20           = avx128fma_invsqrt_d(rsq20);
 336
 337             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 338             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 339             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 340
 341             /* Load parameters for j particles */
 342             jq0              = _mm_load_sd(charge+jnrA+0);
 343             vdwjidx0A        = 2*vdwtype[jnrA+0];
 344
 345             fjx0             = _mm_setzero_pd();
 346             fjy0             = _mm_setzero_pd();
 347             fjz0             = _mm_setzero_pd();
 348
 349             /**************************
 350              * CALCULATE INTERACTIONS *
 351              **************************/
 352
 353             /* Compute parameters for interactions between i and j atoms */
 354             qq00             = _mm_mul_pd(iq0,jq0);
 355             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 356
 357             /* REACTION-FIELD ELECTROSTATICS */
 358             velec            = _mm_mul_pd(qq00,_mm_sub_pd(_mm_macc_pd(krf,rsq00,rinv00),crf));
 359             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 360
 361             /* LENNARD-JONES DISPERSION/REPULSION */
 362
 363             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 364             vvdw6            = _mm_mul_pd(c6_00,rinvsix);
 365             vvdw12           = _mm_mul_pd(c12_00,_mm_mul_pd(rinvsix,rinvsix));
 366             vvdw             = _mm_msub_pd( vvdw12,one_twelfth, _mm_mul_pd(vvdw6,one_sixth) );
 367             fvdw             = _mm_mul_pd(_mm_sub_pd(vvdw12,vvdw6),rinvsq00);
 368
 369             /* Update potential sum for this i atom from the interaction with this j atom. */
 370             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 371             velecsum         = _mm_add_pd(velecsum,velec);
 372             vvdw             = _mm_unpacklo_pd(vvdw,_mm_setzero_pd());
 373             vvdwsum          = _mm_add_pd(vvdwsum,vvdw);
 374
 375             fscal            = _mm_add_pd(felec,fvdw);
 376
 377             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 378
 379             /* Update vectorial force */
 380             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 381             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 382             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 383
 384             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 385             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 386             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 387
 388             /**************************
 389              * CALCULATE INTERACTIONS *
 390              **************************/
 391
 392             /* Compute parameters for interactions between i and j atoms */
 393             qq10             = _mm_mul_pd(iq1,jq0);
 394
 395             /* REACTION-FIELD ELECTROSTATICS */
 396             velec            = _mm_mul_pd(qq10,_mm_sub_pd(_mm_macc_pd(krf,rsq10,rinv10),crf));
 397             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 398
 399             /* Update potential sum for this i atom from the interaction with this j atom. */
 400             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 401             velecsum         = _mm_add_pd(velecsum,velec);
 402
 403             fscal            = felec;
 404
 405             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 406
 407             /* Update vectorial force */
 408             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 409             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 410             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 411
 412             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 413             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 414             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 415
 416             /**************************
 417              * CALCULATE INTERACTIONS *
 418              **************************/
 419
 420             /* Compute parameters for interactions between i and j atoms */
 421             qq20             = _mm_mul_pd(iq2,jq0);
 422
 423             /* REACTION-FIELD ELECTROSTATICS */
 424             velec            = _mm_mul_pd(qq20,_mm_sub_pd(_mm_macc_pd(krf,rsq20,rinv20),crf));
 425             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 426
 427             /* Update potential sum for this i atom from the interaction with this j atom. */
 428             velec            = _mm_unpacklo_pd(velec,_mm_setzero_pd());
 429             velecsum         = _mm_add_pd(velecsum,velec);
 430
 431             fscal            = felec;
 432
 433             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 434
 435             /* Update vectorial force */
 436             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 437             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 438             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 439
 440             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 441             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 442             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 443
 444             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 445
 446             /* Inner loop uses 120 flops */
 447         }
 448
 449         /* End of innermost loop */
 450
 451         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
 452                                               f+i_coord_offset,fshift+i_shift_offset);
 453
 454         ggid                        = gid[iidx];
 455         /* Update potential energies */
 456         gmx_mm_update_1pot_pd(velecsum,kernel_data->energygrp_elec+ggid);
 457         gmx_mm_update_1pot_pd(vvdwsum,kernel_data->energygrp_vdw+ggid);
 458
 459         /* Increment number of inner iterations */
 460         inneriter                  += j_index_end - j_index_start;
 461
 462         /* Outer loop uses 20 flops */
 463     }
 464
 465     /* Increment number of outer iterations */
 466     outeriter        += nri;
 467
 468     /* Update outer/inner flops */
 469
 470     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_VF,outeriter*20 + inneriter*120);
 471 }
 472 /*
 473  * Gromacs nonbonded kernel:   nb_kernel_ElecRF_VdwLJ_GeomW3P1_F_avx_128_fma_double
 474  * Electrostatics interaction: ReactionField
 475  * VdW interaction:            LennardJones
 476  * Geometry:                   Water3-Particle
 477  * Calculate force/pot:        Force
 478  */
 479 void
 480 nb_kernel_ElecRF_VdwLJ_GeomW3P1_F_avx_128_fma_double
 481                     (t_nblist                    * gmx_restrict       nlist,
 482                      rvec                        * gmx_restrict          xx,
 483                      rvec                        * gmx_restrict          ff,
 484                      struct t_forcerec           * gmx_restrict          fr,
 485                      t_mdatoms                   * gmx_restrict     mdatoms,
 486                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 487                      t_nrnb                      * gmx_restrict        nrnb)
 488 {
 489     /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or
 490      * just 0 for non-waters.
 491      * Suffixes A,B refer to j loop unrolling done with SSE double precision, e.g. for the two different
 492      * jnr indices corresponding to data put in the four positions in the SIMD register.
 493      */
 494     int              i_shift_offset,i_coord_offset,outeriter,inneriter;
 495     int              j_index_start,j_index_end,jidx,nri,inr,ggid,iidx;
 496     int              jnrA,jnrB;
 497     int              j_coord_offsetA,j_coord_offsetB;
 498     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 499     real             rcutoff_scalar;
 500     real             *shiftvec,*fshift,*x,*f;
 501     __m128d          tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall;
 502     int              vdwioffset0;
 503     __m128d          ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0;
 504     int              vdwioffset1;
 505     __m128d          ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 506     int              vdwioffset2;
 507     __m128d          ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 508     int              vdwjidx0A,vdwjidx0B;
 509     __m128d          jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 510     __m128d          dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00;
 511     __m128d          dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10;
 512     __m128d          dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20;
 513     __m128d          velec,felec,velecsum,facel,crf,krf,krf2;
 514     real             *charge;
 515     int              nvdwtype;
 516     __m128d          rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6;
 517     int              *vdwtype;
 518     real             *vdwparam;
 519     __m128d          one_sixth   = _mm_set1_pd(1.0/6.0);
 520     __m128d          one_twelfth = _mm_set1_pd(1.0/12.0);
 521     __m128d          dummy_mask,cutoff_mask;
 522     __m128d          signbit   = gmx_mm_castsi128_pd( _mm_set_epi32(0x80000000,0x00000000,0x80000000,0x00000000) );
 523     __m128d          one     = _mm_set1_pd(1.0);
 524     __m128d          two     = _mm_set1_pd(2.0);
 525     x                = xx[0];
 526     f                = ff[0];
 527
 528     nri              = nlist->nri;
 529     iinr             = nlist->iinr;
 530     jindex           = nlist->jindex;
 531     jjnr             = nlist->jjnr;
 532     shiftidx         = nlist->shift;
 533     gid              = nlist->gid;
 534     shiftvec         = fr->shift_vec[0];
 535     fshift           = fr->fshift[0];
 536     facel            = _mm_set1_pd(fr->ic->epsfac);
 537     charge           = mdatoms->chargeA;
 538     krf              = _mm_set1_pd(fr->ic->k_rf);
 539     krf2             = _mm_set1_pd(fr->ic->k_rf*2.0);
 540     crf              = _mm_set1_pd(fr->ic->c_rf);
 541     nvdwtype         = fr->ntype;
 542     vdwparam         = fr->nbfp;
 543     vdwtype          = mdatoms->typeA;
 544
 545     /* Setup water-specific parameters */
 546     inr              = nlist->iinr[0];
 547     iq0              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+0]));
 548     iq1              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+1]));
 549     iq2              = _mm_mul_pd(facel,_mm_set1_pd(charge[inr+2]));
 550     vdwioffset0      = 2*nvdwtype*vdwtype[inr+0];
 551
 552     /* Avoid stupid compiler warnings */
 553     jnrA = jnrB = 0;
 554     j_coord_offsetA = 0;
 555     j_coord_offsetB = 0;
 556
 557     outeriter        = 0;
 558     inneriter        = 0;
 559
 560     /* Start outer loop over neighborlists */
 561     for(iidx=0; iidx<nri; iidx++)
 562     {
 563         /* Load shift vector for this list */
 564         i_shift_offset   = DIM*shiftidx[iidx];
 565
 566         /* Load limits for loop over neighbors */
 567         j_index_start    = jindex[iidx];
 568         j_index_end      = jindex[iidx+1];
 569
 570         /* Get outer coordinate index */
 571         inr              = iinr[iidx];
 572         i_coord_offset   = DIM*inr;
 573
 574         /* Load i particle coords and add shift vector */
 575         gmx_mm_load_shift_and_3rvec_broadcast_pd(shiftvec+i_shift_offset,x+i_coord_offset,
 576                                                  &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2);
 577
 578         fix0             = _mm_setzero_pd();
 579         fiy0             = _mm_setzero_pd();
 580         fiz0             = _mm_setzero_pd();
 581         fix1             = _mm_setzero_pd();
 582         fiy1             = _mm_setzero_pd();
 583         fiz1             = _mm_setzero_pd();
 584         fix2             = _mm_setzero_pd();
 585         fiy2             = _mm_setzero_pd();
 586         fiz2             = _mm_setzero_pd();
 587
 588         /* Start inner kernel loop */
 589         for(jidx=j_index_start; jidx<j_index_end-1; jidx+=2)
 590         {
 591
 592             /* Get j neighbor index, and coordinate index */
 593             jnrA             = jjnr[jidx];
 594             jnrB             = jjnr[jidx+1];
 595             j_coord_offsetA  = DIM*jnrA;
 596             j_coord_offsetB  = DIM*jnrB;
 597
 598             /* load j atom coordinates */
 599             gmx_mm_load_1rvec_2ptr_swizzle_pd(x+j_coord_offsetA,x+j_coord_offsetB,
 600                                               &jx0,&jy0,&jz0);
 601
 602             /* Calculate displacement vector */
 603             dx00             = _mm_sub_pd(ix0,jx0);
 604             dy00             = _mm_sub_pd(iy0,jy0);
 605             dz00             = _mm_sub_pd(iz0,jz0);
 606             dx10             = _mm_sub_pd(ix1,jx0);
 607             dy10             = _mm_sub_pd(iy1,jy0);
 608             dz10             = _mm_sub_pd(iz1,jz0);
 609             dx20             = _mm_sub_pd(ix2,jx0);
 610             dy20             = _mm_sub_pd(iy2,jy0);
 611             dz20             = _mm_sub_pd(iz2,jz0);
 612
 613             /* Calculate squared distance and things based on it */
 614             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 615             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 616             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 617
 618             rinv00           = avx128fma_invsqrt_d(rsq00);
 619             rinv10           = avx128fma_invsqrt_d(rsq10);
 620             rinv20           = avx128fma_invsqrt_d(rsq20);
 621
 622             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 623             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 624             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 625
 626             /* Load parameters for j particles */
 627             jq0              = gmx_mm_load_2real_swizzle_pd(charge+jnrA+0,charge+jnrB+0);
 628             vdwjidx0A        = 2*vdwtype[jnrA+0];
 629             vdwjidx0B        = 2*vdwtype[jnrB+0];
 630
 631             fjx0             = _mm_setzero_pd();
 632             fjy0             = _mm_setzero_pd();
 633             fjz0             = _mm_setzero_pd();
 634
 635             /**************************
 636              * CALCULATE INTERACTIONS *
 637              **************************/
 638
 639             /* Compute parameters for interactions between i and j atoms */
 640             qq00             = _mm_mul_pd(iq0,jq0);
 641             gmx_mm_load_2pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,
 642                                          vdwparam+vdwioffset0+vdwjidx0B,&c6_00,&c12_00);
 643
 644             /* REACTION-FIELD ELECTROSTATICS */
 645             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 646
 647             /* LENNARD-JONES DISPERSION/REPULSION */
 648
 649             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 650             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
 651
 652             fscal            = _mm_add_pd(felec,fvdw);
 653
 654             /* Update vectorial force */
 655             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 656             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 657             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 658
 659             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 660             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 661             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 662
 663             /**************************
 664              * CALCULATE INTERACTIONS *
 665              **************************/
 666
 667             /* Compute parameters for interactions between i and j atoms */
 668             qq10             = _mm_mul_pd(iq1,jq0);
 669
 670             /* REACTION-FIELD ELECTROSTATICS */
 671             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 672
 673             fscal            = felec;
 674
 675             /* Update vectorial force */
 676             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 677             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 678             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 679
 680             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 681             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 682             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 683
 684             /**************************
 685              * CALCULATE INTERACTIONS *
 686              **************************/
 687
 688             /* Compute parameters for interactions between i and j atoms */
 689             qq20             = _mm_mul_pd(iq2,jq0);
 690
 691             /* REACTION-FIELD ELECTROSTATICS */
 692             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 693
 694             fscal            = felec;
 695
 696             /* Update vectorial force */
 697             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 698             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 699             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 700
 701             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 702             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 703             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 704
 705             gmx_mm_decrement_1rvec_2ptr_swizzle_pd(f+j_coord_offsetA,f+j_coord_offsetB,fjx0,fjy0,fjz0);
 706
 707             /* Inner loop uses 100 flops */
 708         }
 709
 710         if(jidx<j_index_end)
 711         {
 712
 713             jnrA             = jjnr[jidx];
 714             j_coord_offsetA  = DIM*jnrA;
 715
 716             /* load j atom coordinates */
 717             gmx_mm_load_1rvec_1ptr_swizzle_pd(x+j_coord_offsetA,
 718                                               &jx0,&jy0,&jz0);
 719
 720             /* Calculate displacement vector */
 721             dx00             = _mm_sub_pd(ix0,jx0);
 722             dy00             = _mm_sub_pd(iy0,jy0);
 723             dz00             = _mm_sub_pd(iz0,jz0);
 724             dx10             = _mm_sub_pd(ix1,jx0);
 725             dy10             = _mm_sub_pd(iy1,jy0);
 726             dz10             = _mm_sub_pd(iz1,jz0);
 727             dx20             = _mm_sub_pd(ix2,jx0);
 728             dy20             = _mm_sub_pd(iy2,jy0);
 729             dz20             = _mm_sub_pd(iz2,jz0);
 730
 731             /* Calculate squared distance and things based on it */
 732             rsq00            = gmx_mm_calc_rsq_pd(dx00,dy00,dz00);
 733             rsq10            = gmx_mm_calc_rsq_pd(dx10,dy10,dz10);
 734             rsq20            = gmx_mm_calc_rsq_pd(dx20,dy20,dz20);
 735
 736             rinv00           = avx128fma_invsqrt_d(rsq00);
 737             rinv10           = avx128fma_invsqrt_d(rsq10);
 738             rinv20           = avx128fma_invsqrt_d(rsq20);
 739
 740             rinvsq00         = _mm_mul_pd(rinv00,rinv00);
 741             rinvsq10         = _mm_mul_pd(rinv10,rinv10);
 742             rinvsq20         = _mm_mul_pd(rinv20,rinv20);
 743
 744             /* Load parameters for j particles */
 745             jq0              = _mm_load_sd(charge+jnrA+0);
 746             vdwjidx0A        = 2*vdwtype[jnrA+0];
 747
 748             fjx0             = _mm_setzero_pd();
 749             fjy0             = _mm_setzero_pd();
 750             fjz0             = _mm_setzero_pd();
 751
 752             /**************************
 753              * CALCULATE INTERACTIONS *
 754              **************************/
 755
 756             /* Compute parameters for interactions between i and j atoms */
 757             qq00             = _mm_mul_pd(iq0,jq0);
 758             gmx_mm_load_1pair_swizzle_pd(vdwparam+vdwioffset0+vdwjidx0A,&c6_00,&c12_00);
 759
 760             /* REACTION-FIELD ELECTROSTATICS */
 761             felec            = _mm_mul_pd(qq00,_mm_msub_pd(rinv00,rinvsq00,krf2));
 762
 763             /* LENNARD-JONES DISPERSION/REPULSION */
 764
 765             rinvsix          = _mm_mul_pd(_mm_mul_pd(rinvsq00,rinvsq00),rinvsq00);
 766             fvdw             = _mm_mul_pd(_mm_msub_pd(c12_00,rinvsix,c6_00),_mm_mul_pd(rinvsix,rinvsq00));
 767
 768             fscal            = _mm_add_pd(felec,fvdw);
 769
 770             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 771
 772             /* Update vectorial force */
 773             fix0             = _mm_macc_pd(dx00,fscal,fix0);
 774             fiy0             = _mm_macc_pd(dy00,fscal,fiy0);
 775             fiz0             = _mm_macc_pd(dz00,fscal,fiz0);
 776
 777             fjx0             = _mm_macc_pd(dx00,fscal,fjx0);
 778             fjy0             = _mm_macc_pd(dy00,fscal,fjy0);
 779             fjz0             = _mm_macc_pd(dz00,fscal,fjz0);
 780
 781             /**************************
 782              * CALCULATE INTERACTIONS *
 783              **************************/
 784
 785             /* Compute parameters for interactions between i and j atoms */
 786             qq10             = _mm_mul_pd(iq1,jq0);
 787
 788             /* REACTION-FIELD ELECTROSTATICS */
 789             felec            = _mm_mul_pd(qq10,_mm_msub_pd(rinv10,rinvsq10,krf2));
 790
 791             fscal            = felec;
 792
 793             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 794
 795             /* Update vectorial force */
 796             fix1             = _mm_macc_pd(dx10,fscal,fix1);
 797             fiy1             = _mm_macc_pd(dy10,fscal,fiy1);
 798             fiz1             = _mm_macc_pd(dz10,fscal,fiz1);
 799
 800             fjx0             = _mm_macc_pd(dx10,fscal,fjx0);
 801             fjy0             = _mm_macc_pd(dy10,fscal,fjy0);
 802             fjz0             = _mm_macc_pd(dz10,fscal,fjz0);
 803
 804             /**************************
 805              * CALCULATE INTERACTIONS *
 806              **************************/
 807
 808             /* Compute parameters for interactions between i and j atoms */
 809             qq20             = _mm_mul_pd(iq2,jq0);
 810
 811             /* REACTION-FIELD ELECTROSTATICS */
 812             felec            = _mm_mul_pd(qq20,_mm_msub_pd(rinv20,rinvsq20,krf2));
 813
 814             fscal            = felec;
 815
 816             fscal            = _mm_unpacklo_pd(fscal,_mm_setzero_pd());
 817
 818             /* Update vectorial force */
 819             fix2             = _mm_macc_pd(dx20,fscal,fix2);
 820             fiy2             = _mm_macc_pd(dy20,fscal,fiy2);
 821             fiz2             = _mm_macc_pd(dz20,fscal,fiz2);
 822
 823             fjx0             = _mm_macc_pd(dx20,fscal,fjx0);
 824             fjy0             = _mm_macc_pd(dy20,fscal,fjy0);
 825             fjz0             = _mm_macc_pd(dz20,fscal,fjz0);
 826
 827             gmx_mm_decrement_1rvec_1ptr_swizzle_pd(f+j_coord_offsetA,fjx0,fjy0,fjz0);
 828
 829             /* Inner loop uses 100 flops */
 830         }
 831
 832         /* End of innermost loop */
 833
 834         gmx_mm_update_iforce_3atom_swizzle_pd(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,
 835                                               f+i_coord_offset,fshift+i_shift_offset);
 836
 837         /* Increment number of inner iterations */
 838         inneriter                  += j_index_end - j_index_start;
 839
 840         /* Outer loop uses 18 flops */
 841     }
 842
 843     /* Increment number of outer iterations */
 844     outeriter        += nri;
 845
 846     /* Update outer/inner flops */
 847
 848     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W3_F,outeriter*18 + inneriter*100);
 849 }