Page MenuHomec4science

angle_sdk_omp.cpp
No OneTemporary

File Metadata

Created
Sat, Nov 23, 23:11

angle_sdk_omp.cpp

/* ----------------------------------------------------------------------
LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
http://lammps.sandia.gov, Sandia National Laboratories
Steve Plimpton, sjplimp@sandia.gov
Copyright (2003) Sandia Corporation. Under the terms of Contract
DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
certain rights in this software. This software is distributed under
the GNU General Public License.
See the README file in the top-level LAMMPS directory.
------------------------------------------------------------------------- */
/* ----------------------------------------------------------------------
Contributing author: Axel Kohlmeyer (Temple U)
------------------------------------------------------------------------- */
#include "angle_sdk_omp.h"
#include "atom.h"
#include "neighbor.h"
#include "domain.h"
#include "comm.h"
#include "force.h"
#include "math_const.h"
#include <math.h>
#include "lj_sdk_common.h"
#include "suffix.h"
using namespace LAMMPS_NS;
using namespace MathConst;
using namespace LJSDKParms;
#define SMALL 0.001
/* ---------------------------------------------------------------------- */
AngleSDKOMP::AngleSDKOMP(class LAMMPS *lmp)
: AngleSDK(lmp), ThrOMP(lmp,THR_ANGLE)
{
suffix_flag |= Suffix::OMP;
}
/* ---------------------------------------------------------------------- */
void AngleSDKOMP::compute(int eflag, int vflag)
{
if (eflag || vflag) {
ev_setup(eflag,vflag);
} else evflag = 0;
const int nall = atom->nlocal + atom->nghost;
const int nthreads = comm->nthreads;
const int inum = neighbor->nanglelist;
#if defined(_OPENMP)
#pragma omp parallel default(none) shared(eflag,vflag)
#endif
{
int ifrom, ito, tid;
loop_setup_thr(ifrom, ito, tid, inum, nthreads);
ThrData *thr = fix->get_thr(tid);
ev_setup_thr(eflag, vflag, nall, eatom, vatom, thr);
if (inum > 0) {
if (evflag) {
if (eflag) {
if (force->newton_bond) eval<1,1,1>(ifrom, ito, thr);
else eval<1,1,0>(ifrom, ito, thr);
} else {
if (force->newton_bond) eval<1,0,1>(ifrom, ito, thr);
else eval<1,0,0>(ifrom, ito, thr);
}
} else {
if (force->newton_bond) eval<0,0,1>(ifrom, ito, thr);
else eval<0,0,0>(ifrom, ito, thr);
}
}
reduce_thr(this, eflag, vflag, thr);
} // end of omp parallel region
}
template <int EVFLAG, int EFLAG, int NEWTON_BOND>
void AngleSDKOMP::eval(int nfrom, int nto, ThrData * const thr)
{
int i1,i2,i3,n,type;
double delx1,dely1,delz1,delx2,dely2,delz2,delx3,dely3,delz3;
double eangle,f1[3],f3[3],e13,f13;
double dtheta,tk;
double rsq1,rsq2,rsq3,r1,r2,c,s,a,a11,a12,a22;
const dbl3_t * _noalias const x = (dbl3_t *) atom->x[0];
dbl3_t * _noalias const f = (dbl3_t *) thr->get_f()[0];
const int4_t * _noalias const anglelist = (int4_t *) neighbor->anglelist[0];
const int nlocal = atom->nlocal;
eangle = 0.0;
for (n = nfrom; n < nto; n++) {
i1 = anglelist[n].a;
i2 = anglelist[n].b;
i3 = anglelist[n].c;
type = anglelist[n].t;
// 1st bond
delx1 = x[i1].x - x[i2].x;
dely1 = x[i1].y - x[i2].y;
delz1 = x[i1].z - x[i2].z;
rsq1 = delx1*delx1 + dely1*dely1 + delz1*delz1;
r1 = sqrt(rsq1);
// 2nd bond
delx2 = x[i3].x - x[i2].x;
dely2 = x[i3].y - x[i2].y;
delz2 = x[i3].z - x[i2].z;
rsq2 = delx2*delx2 + dely2*dely2 + delz2*delz2;
r2 = sqrt(rsq2);
// angle (cos and sin)
c = delx1*delx2 + dely1*dely2 + delz1*delz2;
c /= r1*r2;
if (c > 1.0) c = 1.0;
if (c < -1.0) c = -1.0;
s = sqrt(1.0 - c*c);
if (s < SMALL) s = SMALL;
s = 1.0/s;
// 1-3 LJ interaction.
// we only want to use the repulsive part,
// and it can be scaled (or off).
// so this has to be done here and not in the
// general non-bonded code.
f13 = e13 = delx3 = dely3 = delz3 = 0.0;
if (repflag) {
delx3 = x[i1].x - x[i3].x;
dely3 = x[i1].y - x[i3].y;
delz3 = x[i1].z - x[i3].z;
rsq3 = delx3*delx3 + dely3*dely3 + delz3*delz3;
const int type1 = atom->type[i1];
const int type3 = atom->type[i3];
if (rsq3 < rminsq[type1][type3]) {
const int ljt = lj_type[type1][type3];
const double r2inv = 1.0/rsq3;
if (ljt == LJ12_4) {
const double r4inv=r2inv*r2inv;
f13 = r4inv*(lj1[type1][type3]*r4inv*r4inv - lj2[type1][type3]);
if (EFLAG) e13 = r4inv*(lj3[type1][type3]*r4inv*r4inv - lj4[type1][type3]);
} else if (ljt == LJ9_6) {
const double r3inv = r2inv*sqrt(r2inv);
const double r6inv = r3inv*r3inv;
f13 = r6inv*(lj1[type1][type3]*r3inv - lj2[type1][type3]);
if (EFLAG) e13 = r6inv*(lj3[type1][type3]*r3inv - lj4[type1][type3]);
} else if (ljt == LJ12_6) {
const double r6inv = r2inv*r2inv*r2inv;
f13 = r6inv*(lj1[type1][type3]*r6inv - lj2[type1][type3]);
if (EFLAG) e13 = r6inv*(lj3[type1][type3]*r6inv - lj4[type1][type3]);
}
// make sure energy is 0.0 at the cutoff.
if (EFLAG) e13 -= emin[type1][type3];
f13 *= r2inv;
}
}
// force & energy
dtheta = acos(c) - theta0[type];
tk = k[type] * dtheta;
if (EFLAG) eangle = tk*dtheta;
a = -2.0 * tk * s;
a11 = a*c / rsq1;
a12 = -a / (r1*r2);
a22 = a*c / rsq2;
f1[0] = a11*delx1 + a12*delx2;
f1[1] = a11*dely1 + a12*dely2;
f1[2] = a11*delz1 + a12*delz2;
f3[0] = a22*delx2 + a12*delx1;
f3[1] = a22*dely2 + a12*dely1;
f3[2] = a22*delz2 + a12*delz1;
// apply force to each of the 3 atoms
if (NEWTON_BOND || i1 < nlocal) {
f[i1].x += f1[0] + f13*delx3;
f[i1].y += f1[1] + f13*dely3;
f[i1].z += f1[2] + f13*delz3;
}
if (NEWTON_BOND || i2 < nlocal) {
f[i2].x -= f1[0] + f3[0];
f[i2].y -= f1[1] + f3[1];
f[i2].z -= f1[2] + f3[2];
}
if (NEWTON_BOND || i3 < nlocal) {
f[i3].x += f3[0] - f13*delx3;
f[i3].y += f3[1] - f13*dely3;
f[i3].z += f3[2] - f13*delz3;
}
if (EVFLAG) {
ev_tally_thr(this,i1,i2,i3,nlocal,NEWTON_BOND,eangle,f1,f3,
delx1,dely1,delz1,delx2,dely2,delz2,thr);
if (repflag) ev_tally13_thr(this,i1,i3,nlocal,NEWTON_BOND,
e13,f13,delx3,dely3,delz3,thr);
}
}
}

Event Timeline