Homec4science
Diffusion lammps (master) Browse

lammps/src/KOKKOS/bond_fene_kokkos.cpp
3e89b270fdf8master

Last Change
Raw File

bond_fene_kokkos.cpp
View Options

/* ----------------------------------------------------------------------
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: Stan Moore (SNL)
------------------------------------------------------------------------- */
#include <math.h>
#include <stdlib.h>
#include "bond_fene_kokkos.h"
#include "atom_kokkos.h"
#include "neighbor_kokkos.h"
#include "domain.h"
#include "comm.h"
#include "force.h"
#include "memory.h"
#include "error.h"
#include "atom_masks.h"
using namespace LAMMPS_NS;
/* ---------------------------------------------------------------------- */
template<class DeviceType>
BondFENEKokkos<DeviceType>::BondFENEKokkos(LAMMPS *lmp) : BondFENE(lmp)
{
atomKK = (AtomKokkos *) atom;
neighborKK = (NeighborKokkos *) neighbor;
execution_space = ExecutionSpaceFromDevice<DeviceType>::space;
datamask_read = X_MASK | F_MASK | ENERGY_MASK | VIRIAL_MASK;
datamask_modify = F_MASK | ENERGY_MASK | VIRIAL_MASK;
k_warning_flag = DAT::tdual_int_scalar("Bond:warning_flag");
d_warning_flag = k_warning_flag.view<DeviceType>();
h_warning_flag = k_warning_flag.h_view;
k_error_flag = DAT::tdual_int_scalar("Bond:error_flag");
d_error_flag = k_error_flag.view<DeviceType>();
h_error_flag = k_error_flag.h_view;
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
BondFENEKokkos<DeviceType>::~BondFENEKokkos()
{
if (!copymode) {
memory->destroy_kokkos(k_eatom,eatom);
memory->destroy_kokkos(k_vatom,vatom);
}
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
void BondFENEKokkos<DeviceType>::compute(int eflag_in, int vflag_in)
{
eflag = eflag_in;
vflag = vflag_in;
if (eflag || vflag) ev_setup(eflag,vflag,0);
else evflag = 0;
// reallocate per-atom arrays if necessary
if (eflag_atom) {
memory->destroy_kokkos(k_eatom,eatom);
memory->create_kokkos(k_eatom,eatom,maxeatom,"bond:eatom");
d_eatom = k_eatom.view<DeviceType>();
}
if (vflag_atom) {
memory->destroy_kokkos(k_vatom,vatom);
memory->create_kokkos(k_vatom,vatom,maxvatom,6,"bond:vatom");
d_vatom = k_vatom.view<DeviceType>();
}
k_k.template sync<DeviceType>();
k_r0.template sync<DeviceType>();
k_epsilon.template sync<DeviceType>();
k_sigma.template sync<DeviceType>();
x = atomKK->k_x.view<DeviceType>();
f = atomKK->k_f.view<DeviceType>();
neighborKK->k_bondlist.template sync<DeviceType>();
bondlist = neighborKK->k_bondlist.view<DeviceType>();
int nbondlist = neighborKK->nbondlist;
nlocal = atom->nlocal;
newton_bond = force->newton_bond;
h_warning_flag() = 0;
k_warning_flag.template modify<LMPHostType>();
k_warning_flag.template sync<DeviceType>();
h_error_flag() = 0;
k_error_flag.template modify<LMPHostType>();
k_error_flag.template sync<DeviceType>();
copymode = 1;
// loop over neighbors of my atoms
EV_FLOAT ev;
if (evflag) {
if (newton_bond) {
Kokkos::parallel_reduce(Kokkos::RangePolicy<DeviceType, TagBondFENECompute<1,1> >(0,nbondlist),*this,ev);
} else {
Kokkos::parallel_reduce(Kokkos::RangePolicy<DeviceType, TagBondFENECompute<0,1> >(0,nbondlist),*this,ev);
}
} else {
if (newton_bond) {
Kokkos::parallel_for(Kokkos::RangePolicy<DeviceType, TagBondFENECompute<1,0> >(0,nbondlist),*this);
} else {
Kokkos::parallel_for(Kokkos::RangePolicy<DeviceType, TagBondFENECompute<0,0> >(0,nbondlist),*this);
}
}
k_warning_flag.template modify<DeviceType>();
k_warning_flag.template sync<LMPHostType>();
if (h_warning_flag())
error->warning(FLERR,"FENE bond too long",0);
k_error_flag.template modify<DeviceType>();
k_error_flag.template sync<LMPHostType>();
if (h_error_flag())
error->one(FLERR,"Bad FENE bond");
if (eflag_global) energy += ev.evdwl;
if (vflag_global) {
virial[0] += ev.v[0];
virial[1] += ev.v[1];
virial[2] += ev.v[2];
virial[3] += ev.v[3];
virial[4] += ev.v[4];
virial[5] += ev.v[5];
}
if (eflag_atom) {
k_eatom.template modify<DeviceType>();
k_eatom.template sync<LMPHostType>();
}
if (vflag_atom) {
k_vatom.template modify<DeviceType>();
k_vatom.template sync<LMPHostType>();
}
copymode = 0;
}
template<class DeviceType>
template<int NEWTON_BOND, int EVFLAG>
KOKKOS_INLINE_FUNCTION
void BondFENEKokkos<DeviceType>::operator()(TagBondFENECompute<NEWTON_BOND,EVFLAG>, const int &n, EV_FLOAT& ev) const {
if (d_error_flag()) return;
// The f array is atomic
Kokkos::View<F_FLOAT*[3], typename DAT::t_f_array::array_layout,DeviceType,Kokkos::MemoryTraits<Kokkos::Atomic|Kokkos::Unmanaged> > a_f = f;
const int i1 = bondlist(n,0);
const int i2 = bondlist(n,1);
const int type = bondlist(n,2);
const F_FLOAT delx = x(i1,0) - x(i2,0);
const F_FLOAT dely = x(i1,1) - x(i2,1);
const F_FLOAT delz = x(i1,2) - x(i2,2);
// force from log term
const F_FLOAT rsq = delx*delx + dely*dely + delz*delz;
const F_FLOAT r0sq = d_r0[type] * d_r0[type];
F_FLOAT rlogarg = 1.0 - rsq/r0sq;
// if r -> r0, then rlogarg < 0.0 which is an error
// issue a warning and reset rlogarg = epsilon
// if r > 2*r0 something serious is wrong, abort
if (rlogarg < 0.1) {
if (!d_warning_flag())
Kokkos::atomic_fetch_add(&d_warning_flag(),1);
if (rlogarg <= -3.0 && !d_error_flag())
Kokkos::atomic_fetch_add(&d_error_flag(),1);
rlogarg = 0.1;
}
F_FLOAT fbond = -d_k[type]/rlogarg;
// force from LJ term
F_FLOAT sr6 = 0.0;
if (rsq < TWO_1_3*d_sigma[type]*d_sigma[type]) {
const F_FLOAT sr2 = d_sigma[type]*d_sigma[type]/rsq;
sr6 = sr2*sr2*sr2;
fbond += 48.0*d_epsilon[type]*sr6*(sr6-0.5)/rsq;
}
// energy
F_FLOAT ebond = 0.0;
if (eflag) {
ebond = -0.5 * d_k[type]*r0sq*log(rlogarg);
if (rsq < TWO_1_3*d_sigma[type]*d_sigma[type])
ebond += 4.0*d_epsilon[type]*sr6*(sr6-1.0) + d_epsilon[type];
}
// apply force to each of 2 atoms
if (NEWTON_BOND || i1 < nlocal) {
a_f(i1,0) += delx*fbond;
a_f(i1,1) += dely*fbond;
a_f(i1,2) += delz*fbond;
}
if (NEWTON_BOND || i2 < nlocal) {
a_f(i2,0) -= delx*fbond;
a_f(i2,1) -= dely*fbond;
a_f(i2,2) -= delz*fbond;
}
if (EVFLAG) ev_tally(ev,i1,i2,ebond,fbond,delx,dely,delz);
}
template<class DeviceType>
template<int NEWTON_BOND, int EVFLAG>
KOKKOS_INLINE_FUNCTION
void BondFENEKokkos<DeviceType>::operator()(TagBondFENECompute<NEWTON_BOND,EVFLAG>, const int &n) const {
EV_FLOAT ev;
this->template operator()<NEWTON_BOND,EVFLAG>(TagBondFENECompute<NEWTON_BOND,EVFLAG>(), n, ev);
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
void BondFENEKokkos<DeviceType>::allocate()
{
BondFENE::allocate();
int n = atom->nbondtypes;
k_k = DAT::tdual_ffloat_1d("BondFene::k",n+1);
k_r0 = DAT::tdual_ffloat_1d("BondFene::r0",n+1);
k_epsilon = DAT::tdual_ffloat_1d("BondFene::epsilon",n+1);
k_sigma = DAT::tdual_ffloat_1d("BondFene::sigma",n+1);
d_k = k_k.template view<DeviceType>();
d_r0 = k_r0.template view<DeviceType>();
d_epsilon = k_epsilon.template view<DeviceType>();
d_sigma = k_sigma.template view<DeviceType>();
}
/* ----------------------------------------------------------------------
set coeffs for one type
------------------------------------------------------------------------- */
template<class DeviceType>
void BondFENEKokkos<DeviceType>::coeff(int narg, char **arg)
{
BondFENE::coeff(narg, arg);
int n = atom->nbondtypes;
for (int i = 1; i <= n; i++) {
k_k.h_view[i] = k[i];
k_r0.h_view[i] = r0[i];
k_epsilon.h_view[i] = epsilon[i];
k_sigma.h_view[i] = sigma[i];
}
k_k.template modify<LMPHostType>();
k_r0.template modify<LMPHostType>();
k_epsilon.template modify<LMPHostType>();
k_sigma.template modify<LMPHostType>();
}
/* ----------------------------------------------------------------------
tally energy and virial into global and per-atom accumulators
------------------------------------------------------------------------- */
template<class DeviceType>
//template<int NEWTON_BOND>
KOKKOS_INLINE_FUNCTION
void BondFENEKokkos<DeviceType>::ev_tally(EV_FLOAT &ev, const int &i, const int &j,
const F_FLOAT &ebond, const F_FLOAT &fbond, const F_FLOAT &delx,
const F_FLOAT &dely, const F_FLOAT &delz) const
{
E_FLOAT ebondhalf;
F_FLOAT v[6];
// The eatom and vatom arrays are atomic
Kokkos::View<E_FLOAT*, typename DAT::t_efloat_1d::array_layout,DeviceType,Kokkos::MemoryTraits<Kokkos::Atomic|Kokkos::Unmanaged> > v_eatom = k_eatom.view<DeviceType>();
Kokkos::View<F_FLOAT*[6], typename DAT::t_virial_array::array_layout,DeviceType,Kokkos::MemoryTraits<Kokkos::Atomic|Kokkos::Unmanaged> > v_vatom = k_vatom.view<DeviceType>();
if (eflag_either) {
if (eflag_global) {
if (newton_bond) ev.evdwl += ebond;
else {
ebondhalf = 0.5*ebond;
if (i < nlocal) ev.evdwl += ebondhalf;
if (j < nlocal) ev.evdwl += ebondhalf;
}
}
if (eflag_atom) {
ebondhalf = 0.5*ebond;
if (newton_bond || i < nlocal) v_eatom[i] += ebondhalf;
if (newton_bond || j < nlocal) v_eatom[j] += ebondhalf;
}
}
if (vflag_either) {
v[0] = delx*delx*fbond;
v[1] = dely*dely*fbond;
v[2] = delz*delz*fbond;
v[3] = delx*dely*fbond;
v[4] = delx*delz*fbond;
v[5] = dely*delz*fbond;
if (vflag_global) {
if (newton_bond) {
ev.v[0] += v[0];
ev.v[1] += v[1];
ev.v[2] += v[2];
ev.v[3] += v[3];
ev.v[4] += v[4];
ev.v[5] += v[5];
} else {
if (i < nlocal) {
ev.v[0] += 0.5*v[0];
ev.v[1] += 0.5*v[1];
ev.v[2] += 0.5*v[2];
ev.v[3] += 0.5*v[3];
ev.v[4] += 0.5*v[4];
ev.v[5] += 0.5*v[5];
}
if (j < nlocal) {
ev.v[0] += 0.5*v[0];
ev.v[1] += 0.5*v[1];
ev.v[2] += 0.5*v[2];
ev.v[3] += 0.5*v[3];
ev.v[4] += 0.5*v[4];
ev.v[5] += 0.5*v[5];
}
}
}
if (vflag_atom) {
if (newton_bond || i < nlocal) {
v_vatom(i,0) += 0.5*v[0];
v_vatom(i,1) += 0.5*v[1];
v_vatom(i,2) += 0.5*v[2];
v_vatom(i,3) += 0.5*v[3];
v_vatom(i,4) += 0.5*v[4];
v_vatom(i,5) += 0.5*v[5];
}
if (newton_bond || j < nlocal) {
v_vatom(j,0) += 0.5*v[0];
v_vatom(j,1) += 0.5*v[1];
v_vatom(j,2) += 0.5*v[2];
v_vatom(j,3) += 0.5*v[3];
v_vatom(j,4) += 0.5*v[4];
v_vatom(j,5) += 0.5*v[5];
}
}
}
}
/* ---------------------------------------------------------------------- */
namespace LAMMPS_NS {
template class BondFENEKokkos<LMPDeviceType>;
#ifdef KOKKOS_HAVE_CUDA
template class BondFENEKokkos<LMPHostType>;
#endif
}
c4science · Help