src/gromacs/gmxlib/nonbonded/nb_kernel_c/nb_kernel_ElecCoul_VdwNone_GeomW4P1_c.c

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2012,2013,2014, 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 c kernel generator.
  37  */
  38 #include "config.h"
  39
  40 #include <math.h>
  41
  42 #include "../nb_kernel.h"
  43 #include "types/simple.h"
  44 #include "gromacs/math/vec.h"
  45 #include "nrnb.h"
  46
  47 /*
  48  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwNone_GeomW4P1_VF_c
  49  * Electrostatics interaction: Coulomb
  50  * VdW interaction:            None
  51  * Geometry:                   Water4-Particle
  52  * Calculate force/pot:        PotentialAndForce
  53  */
  54 void
  55 nb_kernel_ElecCoul_VdwNone_GeomW4P1_VF_c
  56                     (t_nblist                    * gmx_restrict       nlist,
  57                      rvec                        * gmx_restrict          xx,
  58                      rvec                        * gmx_restrict          ff,
  59                      t_forcerec                  * gmx_restrict          fr,
  60                      t_mdatoms                   * gmx_restrict     mdatoms,
  61                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
  62                      t_nrnb                      * gmx_restrict        nrnb)
  63 {
  64     int              i_shift_offset,i_coord_offset,j_coord_offset;
  65     int              j_index_start,j_index_end;
  66     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
  67     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
  68     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
  69     real             *shiftvec,*fshift,*x,*f;
  70     int              vdwioffset1;
  71     real             ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
  72     int              vdwioffset2;
  73     real             ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
  74     int              vdwioffset3;
  75     real             ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
  76     int              vdwjidx0;
  77     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
  78     real             dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
  79     real             dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
  80     real             dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
  81     real             velec,felec,velecsum,facel,crf,krf,krf2;
  82     real             *charge;
  83
  84     x                = xx[0];
  85     f                = ff[0];
  86
  87     nri              = nlist->nri;
  88     iinr             = nlist->iinr;
  89     jindex           = nlist->jindex;
  90     jjnr             = nlist->jjnr;
  91     shiftidx         = nlist->shift;
  92     gid              = nlist->gid;
  93     shiftvec         = fr->shift_vec[0];
  94     fshift           = fr->fshift[0];
  95     facel            = fr->epsfac;
  96     charge           = mdatoms->chargeA;
  97
  98     /* Setup water-specific parameters */
  99     inr              = nlist->iinr[0];
 100     iq1              = facel*charge[inr+1];
 101     iq2              = facel*charge[inr+2];
 102     iq3              = facel*charge[inr+3];
 103
 104     outeriter        = 0;
 105     inneriter        = 0;
 106
 107     /* Start outer loop over neighborlists */
 108     for(iidx=0; iidx<nri; iidx++)
 109     {
 110         /* Load shift vector for this list */
 111         i_shift_offset   = DIM*shiftidx[iidx];
 112         shX              = shiftvec[i_shift_offset+XX];
 113         shY              = shiftvec[i_shift_offset+YY];
 114         shZ              = shiftvec[i_shift_offset+ZZ];
 115
 116         /* Load limits for loop over neighbors */
 117         j_index_start    = jindex[iidx];
 118         j_index_end      = jindex[iidx+1];
 119
 120         /* Get outer coordinate index */
 121         inr              = iinr[iidx];
 122         i_coord_offset   = DIM*inr;
 123
 124         /* Load i particle coords and add shift vector */
 125         ix1              = shX + x[i_coord_offset+DIM*1+XX];
 126         iy1              = shY + x[i_coord_offset+DIM*1+YY];
 127         iz1              = shZ + x[i_coord_offset+DIM*1+ZZ];
 128         ix2              = shX + x[i_coord_offset+DIM*2+XX];
 129         iy2              = shY + x[i_coord_offset+DIM*2+YY];
 130         iz2              = shZ + x[i_coord_offset+DIM*2+ZZ];
 131         ix3              = shX + x[i_coord_offset+DIM*3+XX];
 132         iy3              = shY + x[i_coord_offset+DIM*3+YY];
 133         iz3              = shZ + x[i_coord_offset+DIM*3+ZZ];
 134
 135         fix1             = 0.0;
 136         fiy1             = 0.0;
 137         fiz1             = 0.0;
 138         fix2             = 0.0;
 139         fiy2             = 0.0;
 140         fiz2             = 0.0;
 141         fix3             = 0.0;
 142         fiy3             = 0.0;
 143         fiz3             = 0.0;
 144
 145         /* Reset potential sums */
 146         velecsum         = 0.0;
 147
 148         /* Start inner kernel loop */
 149         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 150         {
 151             /* Get j neighbor index, and coordinate index */
 152             jnr              = jjnr[jidx];
 153             j_coord_offset   = DIM*jnr;
 154
 155             /* load j atom coordinates */
 156             jx0              = x[j_coord_offset+DIM*0+XX];
 157             jy0              = x[j_coord_offset+DIM*0+YY];
 158             jz0              = x[j_coord_offset+DIM*0+ZZ];
 159
 160             /* Calculate displacement vector */
 161             dx10             = ix1 - jx0;
 162             dy10             = iy1 - jy0;
 163             dz10             = iz1 - jz0;
 164             dx20             = ix2 - jx0;
 165             dy20             = iy2 - jy0;
 166             dz20             = iz2 - jz0;
 167             dx30             = ix3 - jx0;
 168             dy30             = iy3 - jy0;
 169             dz30             = iz3 - jz0;
 170
 171             /* Calculate squared distance and things based on it */
 172             rsq10            = dx10*dx10+dy10*dy10+dz10*dz10;
 173             rsq20            = dx20*dx20+dy20*dy20+dz20*dz20;
 174             rsq30            = dx30*dx30+dy30*dy30+dz30*dz30;
 175
 176             rinv10           = gmx_invsqrt(rsq10);
 177             rinv20           = gmx_invsqrt(rsq20);
 178             rinv30           = gmx_invsqrt(rsq30);
 179
 180             rinvsq10         = rinv10*rinv10;
 181             rinvsq20         = rinv20*rinv20;
 182             rinvsq30         = rinv30*rinv30;
 183
 184             /* Load parameters for j particles */
 185             jq0              = charge[jnr+0];
 186
 187             /**************************
 188              * CALCULATE INTERACTIONS *
 189              **************************/
 190
 191             qq10             = iq1*jq0;
 192
 193             /* COULOMB ELECTROSTATICS */
 194             velec            = qq10*rinv10;
 195             felec            = velec*rinvsq10;
 196
 197             /* Update potential sums from outer loop */
 198             velecsum        += velec;
 199
 200             fscal            = felec;
 201
 202             /* Calculate temporary vectorial force */
 203             tx               = fscal*dx10;
 204             ty               = fscal*dy10;
 205             tz               = fscal*dz10;
 206
 207             /* Update vectorial force */
 208             fix1            += tx;
 209             fiy1            += ty;
 210             fiz1            += tz;
 211             f[j_coord_offset+DIM*0+XX] -= tx;
 212             f[j_coord_offset+DIM*0+YY] -= ty;
 213             f[j_coord_offset+DIM*0+ZZ] -= tz;
 214
 215             /**************************
 216              * CALCULATE INTERACTIONS *
 217              **************************/
 218
 219             qq20             = iq2*jq0;
 220
 221             /* COULOMB ELECTROSTATICS */
 222             velec            = qq20*rinv20;
 223             felec            = velec*rinvsq20;
 224
 225             /* Update potential sums from outer loop */
 226             velecsum        += velec;
 227
 228             fscal            = felec;
 229
 230             /* Calculate temporary vectorial force */
 231             tx               = fscal*dx20;
 232             ty               = fscal*dy20;
 233             tz               = fscal*dz20;
 234
 235             /* Update vectorial force */
 236             fix2            += tx;
 237             fiy2            += ty;
 238             fiz2            += tz;
 239             f[j_coord_offset+DIM*0+XX] -= tx;
 240             f[j_coord_offset+DIM*0+YY] -= ty;
 241             f[j_coord_offset+DIM*0+ZZ] -= tz;
 242
 243             /**************************
 244              * CALCULATE INTERACTIONS *
 245              **************************/
 246
 247             qq30             = iq3*jq0;
 248
 249             /* COULOMB ELECTROSTATICS */
 250             velec            = qq30*rinv30;
 251             felec            = velec*rinvsq30;
 252
 253             /* Update potential sums from outer loop */
 254             velecsum        += velec;
 255
 256             fscal            = felec;
 257
 258             /* Calculate temporary vectorial force */
 259             tx               = fscal*dx30;
 260             ty               = fscal*dy30;
 261             tz               = fscal*dz30;
 262
 263             /* Update vectorial force */
 264             fix3            += tx;
 265             fiy3            += ty;
 266             fiz3            += tz;
 267             f[j_coord_offset+DIM*0+XX] -= tx;
 268             f[j_coord_offset+DIM*0+YY] -= ty;
 269             f[j_coord_offset+DIM*0+ZZ] -= tz;
 270
 271             /* Inner loop uses 84 flops */
 272         }
 273         /* End of innermost loop */
 274
 275         tx = ty = tz = 0;
 276         f[i_coord_offset+DIM*1+XX] += fix1;
 277         f[i_coord_offset+DIM*1+YY] += fiy1;
 278         f[i_coord_offset+DIM*1+ZZ] += fiz1;
 279         tx                         += fix1;
 280         ty                         += fiy1;
 281         tz                         += fiz1;
 282         f[i_coord_offset+DIM*2+XX] += fix2;
 283         f[i_coord_offset+DIM*2+YY] += fiy2;
 284         f[i_coord_offset+DIM*2+ZZ] += fiz2;
 285         tx                         += fix2;
 286         ty                         += fiy2;
 287         tz                         += fiz2;
 288         f[i_coord_offset+DIM*3+XX] += fix3;
 289         f[i_coord_offset+DIM*3+YY] += fiy3;
 290         f[i_coord_offset+DIM*3+ZZ] += fiz3;
 291         tx                         += fix3;
 292         ty                         += fiy3;
 293         tz                         += fiz3;
 294         fshift[i_shift_offset+XX]  += tx;
 295         fshift[i_shift_offset+YY]  += ty;
 296         fshift[i_shift_offset+ZZ]  += tz;
 297
 298         ggid                        = gid[iidx];
 299         /* Update potential energies */
 300         kernel_data->energygrp_elec[ggid] += velecsum;
 301
 302         /* Increment number of inner iterations */
 303         inneriter                  += j_index_end - j_index_start;
 304
 305         /* Outer loop uses 31 flops */
 306     }
 307
 308     /* Increment number of outer iterations */
 309     outeriter        += nri;
 310
 311     /* Update outer/inner flops */
 312
 313     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_VF,outeriter*31 + inneriter*84);
 314 }
 315 /*
 316  * Gromacs nonbonded kernel:   nb_kernel_ElecCoul_VdwNone_GeomW4P1_F_c
 317  * Electrostatics interaction: Coulomb
 318  * VdW interaction:            None
 319  * Geometry:                   Water4-Particle
 320  * Calculate force/pot:        Force
 321  */
 322 void
 323 nb_kernel_ElecCoul_VdwNone_GeomW4P1_F_c
 324                     (t_nblist                    * gmx_restrict       nlist,
 325                      rvec                        * gmx_restrict          xx,
 326                      rvec                        * gmx_restrict          ff,
 327                      t_forcerec                  * gmx_restrict          fr,
 328                      t_mdatoms                   * gmx_restrict     mdatoms,
 329                      nb_kernel_data_t gmx_unused * gmx_restrict kernel_data,
 330                      t_nrnb                      * gmx_restrict        nrnb)
 331 {
 332     int              i_shift_offset,i_coord_offset,j_coord_offset;
 333     int              j_index_start,j_index_end;
 334     int              nri,inr,ggid,iidx,jidx,jnr,outeriter,inneriter;
 335     real             shX,shY,shZ,tx,ty,tz,fscal,rcutoff,rcutoff2;
 336     int              *iinr,*jindex,*jjnr,*shiftidx,*gid;
 337     real             *shiftvec,*fshift,*x,*f;
 338     int              vdwioffset1;
 339     real             ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1;
 340     int              vdwioffset2;
 341     real             ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2;
 342     int              vdwioffset3;
 343     real             ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3;
 344     int              vdwjidx0;
 345     real             jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0;
 346     real             dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10,cexp1_10,cexp2_10;
 347     real             dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20,cexp1_20,cexp2_20;
 348     real             dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30,cexp1_30,cexp2_30;
 349     real             velec,felec,velecsum,facel,crf,krf,krf2;
 350     real             *charge;
 351
 352     x                = xx[0];
 353     f                = ff[0];
 354
 355     nri              = nlist->nri;
 356     iinr             = nlist->iinr;
 357     jindex           = nlist->jindex;
 358     jjnr             = nlist->jjnr;
 359     shiftidx         = nlist->shift;
 360     gid              = nlist->gid;
 361     shiftvec         = fr->shift_vec[0];
 362     fshift           = fr->fshift[0];
 363     facel            = fr->epsfac;
 364     charge           = mdatoms->chargeA;
 365
 366     /* Setup water-specific parameters */
 367     inr              = nlist->iinr[0];
 368     iq1              = facel*charge[inr+1];
 369     iq2              = facel*charge[inr+2];
 370     iq3              = facel*charge[inr+3];
 371
 372     outeriter        = 0;
 373     inneriter        = 0;
 374
 375     /* Start outer loop over neighborlists */
 376     for(iidx=0; iidx<nri; iidx++)
 377     {
 378         /* Load shift vector for this list */
 379         i_shift_offset   = DIM*shiftidx[iidx];
 380         shX              = shiftvec[i_shift_offset+XX];
 381         shY              = shiftvec[i_shift_offset+YY];
 382         shZ              = shiftvec[i_shift_offset+ZZ];
 383
 384         /* Load limits for loop over neighbors */
 385         j_index_start    = jindex[iidx];
 386         j_index_end      = jindex[iidx+1];
 387
 388         /* Get outer coordinate index */
 389         inr              = iinr[iidx];
 390         i_coord_offset   = DIM*inr;
 391
 392         /* Load i particle coords and add shift vector */
 393         ix1              = shX + x[i_coord_offset+DIM*1+XX];
 394         iy1              = shY + x[i_coord_offset+DIM*1+YY];
 395         iz1              = shZ + x[i_coord_offset+DIM*1+ZZ];
 396         ix2              = shX + x[i_coord_offset+DIM*2+XX];
 397         iy2              = shY + x[i_coord_offset+DIM*2+YY];
 398         iz2              = shZ + x[i_coord_offset+DIM*2+ZZ];
 399         ix3              = shX + x[i_coord_offset+DIM*3+XX];
 400         iy3              = shY + x[i_coord_offset+DIM*3+YY];
 401         iz3              = shZ + x[i_coord_offset+DIM*3+ZZ];
 402
 403         fix1             = 0.0;
 404         fiy1             = 0.0;
 405         fiz1             = 0.0;
 406         fix2             = 0.0;
 407         fiy2             = 0.0;
 408         fiz2             = 0.0;
 409         fix3             = 0.0;
 410         fiy3             = 0.0;
 411         fiz3             = 0.0;
 412
 413         /* Start inner kernel loop */
 414         for(jidx=j_index_start; jidx<j_index_end; jidx++)
 415         {
 416             /* Get j neighbor index, and coordinate index */
 417             jnr              = jjnr[jidx];
 418             j_coord_offset   = DIM*jnr;
 419
 420             /* load j atom coordinates */
 421             jx0              = x[j_coord_offset+DIM*0+XX];
 422             jy0              = x[j_coord_offset+DIM*0+YY];
 423             jz0              = x[j_coord_offset+DIM*0+ZZ];
 424
 425             /* Calculate displacement vector */
 426             dx10             = ix1 - jx0;
 427             dy10             = iy1 - jy0;
 428             dz10             = iz1 - jz0;
 429             dx20             = ix2 - jx0;
 430             dy20             = iy2 - jy0;
 431             dz20             = iz2 - jz0;
 432             dx30             = ix3 - jx0;
 433             dy30             = iy3 - jy0;
 434             dz30             = iz3 - jz0;
 435
 436             /* Calculate squared distance and things based on it */
 437             rsq10            = dx10*dx10+dy10*dy10+dz10*dz10;
 438             rsq20            = dx20*dx20+dy20*dy20+dz20*dz20;
 439             rsq30            = dx30*dx30+dy30*dy30+dz30*dz30;
 440
 441             rinv10           = gmx_invsqrt(rsq10);
 442             rinv20           = gmx_invsqrt(rsq20);
 443             rinv30           = gmx_invsqrt(rsq30);
 444
 445             rinvsq10         = rinv10*rinv10;
 446             rinvsq20         = rinv20*rinv20;
 447             rinvsq30         = rinv30*rinv30;
 448
 449             /* Load parameters for j particles */
 450             jq0              = charge[jnr+0];
 451
 452             /**************************
 453              * CALCULATE INTERACTIONS *
 454              **************************/
 455
 456             qq10             = iq1*jq0;
 457
 458             /* COULOMB ELECTROSTATICS */
 459             velec            = qq10*rinv10;
 460             felec            = velec*rinvsq10;
 461
 462             fscal            = felec;
 463
 464             /* Calculate temporary vectorial force */
 465             tx               = fscal*dx10;
 466             ty               = fscal*dy10;
 467             tz               = fscal*dz10;
 468
 469             /* Update vectorial force */
 470             fix1            += tx;
 471             fiy1            += ty;
 472             fiz1            += tz;
 473             f[j_coord_offset+DIM*0+XX] -= tx;
 474             f[j_coord_offset+DIM*0+YY] -= ty;
 475             f[j_coord_offset+DIM*0+ZZ] -= tz;
 476
 477             /**************************
 478              * CALCULATE INTERACTIONS *
 479              **************************/
 480
 481             qq20             = iq2*jq0;
 482
 483             /* COULOMB ELECTROSTATICS */
 484             velec            = qq20*rinv20;
 485             felec            = velec*rinvsq20;
 486
 487             fscal            = felec;
 488
 489             /* Calculate temporary vectorial force */
 490             tx               = fscal*dx20;
 491             ty               = fscal*dy20;
 492             tz               = fscal*dz20;
 493
 494             /* Update vectorial force */
 495             fix2            += tx;
 496             fiy2            += ty;
 497             fiz2            += tz;
 498             f[j_coord_offset+DIM*0+XX] -= tx;
 499             f[j_coord_offset+DIM*0+YY] -= ty;
 500             f[j_coord_offset+DIM*0+ZZ] -= tz;
 501
 502             /**************************
 503              * CALCULATE INTERACTIONS *
 504              **************************/
 505
 506             qq30             = iq3*jq0;
 507
 508             /* COULOMB ELECTROSTATICS */
 509             velec            = qq30*rinv30;
 510             felec            = velec*rinvsq30;
 511
 512             fscal            = felec;
 513
 514             /* Calculate temporary vectorial force */
 515             tx               = fscal*dx30;
 516             ty               = fscal*dy30;
 517             tz               = fscal*dz30;
 518
 519             /* Update vectorial force */
 520             fix3            += tx;
 521             fiy3            += ty;
 522             fiz3            += tz;
 523             f[j_coord_offset+DIM*0+XX] -= tx;
 524             f[j_coord_offset+DIM*0+YY] -= ty;
 525             f[j_coord_offset+DIM*0+ZZ] -= tz;
 526
 527             /* Inner loop uses 81 flops */
 528         }
 529         /* End of innermost loop */
 530
 531         tx = ty = tz = 0;
 532         f[i_coord_offset+DIM*1+XX] += fix1;
 533         f[i_coord_offset+DIM*1+YY] += fiy1;
 534         f[i_coord_offset+DIM*1+ZZ] += fiz1;
 535         tx                         += fix1;
 536         ty                         += fiy1;
 537         tz                         += fiz1;
 538         f[i_coord_offset+DIM*2+XX] += fix2;
 539         f[i_coord_offset+DIM*2+YY] += fiy2;
 540         f[i_coord_offset+DIM*2+ZZ] += fiz2;
 541         tx                         += fix2;
 542         ty                         += fiy2;
 543         tz                         += fiz2;
 544         f[i_coord_offset+DIM*3+XX] += fix3;
 545         f[i_coord_offset+DIM*3+YY] += fiy3;
 546         f[i_coord_offset+DIM*3+ZZ] += fiz3;
 547         tx                         += fix3;
 548         ty                         += fiy3;
 549         tz                         += fiz3;
 550         fshift[i_shift_offset+XX]  += tx;
 551         fshift[i_shift_offset+YY]  += ty;
 552         fshift[i_shift_offset+ZZ]  += tz;
 553
 554         /* Increment number of inner iterations */
 555         inneriter                  += j_index_end - j_index_start;
 556
 557         /* Outer loop uses 30 flops */
 558     }
 559
 560     /* Increment number of outer iterations */
 561     outeriter        += nri;
 562
 563     /* Update outer/inner flops */
 564
 565     inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_W4_F,outeriter*30 + inneriter*81);
 566 }