Page Menu
Home
c4science
Search
Configure Global Search
Log In
Files
F78716917
angle_charmm_kokkos.cpp
No One
Temporary
Actions
Download File
Edit File
Delete File
View Transforms
Subscribe
Mute Notifications
Award Token
Subscribers
None
File Metadata
Details
File Info
Storage
Attached
Created
Thu, Aug 22, 14:09
Size
11 KB
Mime Type
text/x-c
Expires
Sat, Aug 24, 14:09 (2 d)
Engine
blob
Format
Raw Data
Handle
20057187
Attached To
rLAMMPS lammps
angle_charmm_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 "angle_charmm_kokkos.h"
#include "atom_kokkos.h"
#include "neighbor_kokkos.h"
#include "domain.h"
#include "comm.h"
#include "force.h"
#include "math_const.h"
#include "memory.h"
#include "error.h"
#include "atom_masks.h"
using namespace LAMMPS_NS;
using namespace MathConst;
#define SMALL 0.001
/* ---------------------------------------------------------------------- */
template<class DeviceType>
AngleCharmmKokkos<DeviceType>::AngleCharmmKokkos(LAMMPS *lmp) : AngleCharmm(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;
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
AngleCharmmKokkos<DeviceType>::~AngleCharmmKokkos()
{
if (!copymode) {
memory->destroy_kokkos(k_eatom,eatom);
memory->destroy_kokkos(k_vatom,vatom);
}
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
void AngleCharmmKokkos<DeviceType>::compute(int eflag_in, int vflag_in)
{
eflag = eflag_in;
vflag = vflag_in;
if (eflag || vflag) ev_setup(eflag,vflag);
else evflag = 0;
// reallocate per-atom arrays if necessary
if (eflag_atom) {
if(k_eatom.dimension_0()<maxeatom) {
memory->destroy_kokkos(k_eatom,eatom);
memory->create_kokkos(k_eatom,eatom,maxeatom,"improper:eatom");
d_eatom = k_eatom.d_view;
}
}
if (vflag_atom) {
if(k_vatom.dimension_0()<maxvatom) {
memory->destroy_kokkos(k_vatom,vatom);
memory->create_kokkos(k_vatom,vatom,maxvatom,6,"improper:vatom");
d_vatom = k_vatom.d_view;
}
}
if (eflag || vflag) atomKK->modified(execution_space,datamask_modify);
else atomKK->modified(execution_space,F_MASK);
x = atomKK->k_x.view<DeviceType>();
f = atomKK->k_f.view<DeviceType>();
neighborKK->k_anglelist.template sync<DeviceType>();
anglelist = neighborKK->k_anglelist.view<DeviceType>();
int nanglelist = neighborKK->nanglelist;
nlocal = atom->nlocal;
newton_bond = force->newton_bond;
copymode = 1;
// loop over neighbors of my atoms
EV_FLOAT ev;
if (evflag) {
if (newton_bond) {
Kokkos::parallel_reduce(Kokkos::RangePolicy<DeviceType, TagAngleCharmmCompute<1,1> >(0,nanglelist),*this,ev);
} else {
Kokkos::parallel_reduce(Kokkos::RangePolicy<DeviceType, TagAngleCharmmCompute<0,1> >(0,nanglelist),*this,ev);
}
} else {
if (newton_bond) {
Kokkos::parallel_for(Kokkos::RangePolicy<DeviceType, TagAngleCharmmCompute<1,0> >(0,nanglelist),*this);
} else {
Kokkos::parallel_for(Kokkos::RangePolicy<DeviceType, TagAngleCharmmCompute<0,0> >(0,nanglelist),*this);
}
}
DeviceType::fence();
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 AngleCharmmKokkos<DeviceType>::operator()(TagAngleCharmmCompute<NEWTON_BOND,EVFLAG>, const int &n, EV_FLOAT& ev) const {
const int i1 = anglelist(n,0);
const int i2 = anglelist(n,1);
const int i3 = anglelist(n,2);
const int type = anglelist(n,3);
// 1st bond
const F_FLOAT delx1 = x(i1,0) - x(i2,0);
const F_FLOAT dely1 = x(i1,1) - x(i2,1);
const F_FLOAT delz1 = x(i1,2) - x(i2,2);
const F_FLOAT rsq1 = delx1*delx1 + dely1*dely1 + delz1*delz1;
const F_FLOAT r1 = sqrt(rsq1);
// 2nd bond
const F_FLOAT delx2 = x(i3,0) - x(i2,0);
const F_FLOAT dely2 = x(i3,1) - x(i2,1);
const F_FLOAT delz2 = x(i3,2) - x(i2,2);
const F_FLOAT rsq2 = delx2*delx2 + dely2*dely2 + delz2*delz2;
const F_FLOAT r2 = sqrt(rsq2);
// Urey-Bradley bond
const F_FLOAT delxUB = x(i3,0) - x(i1,0);
const F_FLOAT delyUB = x(i3,1) - x(i1,1);
const F_FLOAT delzUB = x(i3,2) - x(i1,2);
const F_FLOAT rsqUB = delxUB*delxUB + delyUB*delyUB + delzUB*delzUB;
const F_FLOAT rUB = sqrt(rsqUB);
// Urey-Bradley force & energy
const F_FLOAT dr = rUB - d_r_ub[type];
const F_FLOAT rk = d_k_ub[type] * dr;
F_FLOAT forceUB = 0.0;
if (rUB > 0.0) forceUB = -2.0*rk/rUB;
E_FLOAT eangle = 0.0;
if (eflag) eangle = rk*dr;
// angle (cos and sin)
F_FLOAT c = delx1*delx2 + dely1*dely2 + delz1*delz2;
c /= r1*r2;
if (c > 1.0) c = 1.0;
if (c < -1.0) c = -1.0;
F_FLOAT s = sqrt(1.0 - c*c);
if (s < SMALL) s = SMALL;
s = 1.0/s;
// harmonic force & energy
const F_FLOAT dtheta = acos(c) - d_theta0[type];
const F_FLOAT tk = d_k[type] * dtheta;
if (eflag) eangle += tk*dtheta;
const F_FLOAT a = -2.0 * tk * s;
const F_FLOAT a11 = a*c / rsq1;
const F_FLOAT a12 = -a / (r1*r2);
const F_FLOAT a22 = a*c / rsq2;
F_FLOAT f1[3],f3[3];
f1[0] = a11*delx1 + a12*delx2 - delxUB*forceUB;
f1[1] = a11*dely1 + a12*dely2 - delyUB*forceUB;
f1[2] = a11*delz1 + a12*delz2 - delzUB*forceUB;
f3[0] = a22*delx2 + a12*delx1 + delxUB*forceUB;
f3[1] = a22*dely2 + a12*dely1 + delyUB*forceUB;
f3[2] = a22*delz2 + a12*delz1 + delzUB*forceUB;
// apply force to each of 3 atoms
if (NEWTON_BOND || i1 < nlocal) {
f(i1,0) += f1[0];
f(i1,1) += f1[1];
f(i1,2) += f1[2];
}
if (NEWTON_BOND || i2 < nlocal) {
f(i2,0) -= f1[0] + f3[0];
f(i2,1) -= f1[1] + f3[1];
f(i2,2) -= f1[2] + f3[2];
}
if (NEWTON_BOND || i3 < nlocal) {
f(i3,0) += f3[0];
f(i3,1) += f3[1];
f(i3,2) += f3[2];
}
if (EVFLAG) ev_tally(ev,i1,i2,i3,eangle,f1,f3,
delx1,dely1,delz1,delx2,dely2,delz2);
}
template<class DeviceType>
template<int NEWTON_BOND, int EVFLAG>
KOKKOS_INLINE_FUNCTION
void AngleCharmmKokkos<DeviceType>::operator()(TagAngleCharmmCompute<NEWTON_BOND,EVFLAG>, const int &n) const {
EV_FLOAT ev;
this->template operator()<NEWTON_BOND,EVFLAG>(TagAngleCharmmCompute<NEWTON_BOND,EVFLAG>(), n, ev);
}
/* ---------------------------------------------------------------------- */
template<class DeviceType>
void AngleCharmmKokkos<DeviceType>::allocate()
{
AngleCharmm::allocate();
}
/* ----------------------------------------------------------------------
set coeffs for one or more types
------------------------------------------------------------------------- */
template<class DeviceType>
void AngleCharmmKokkos<DeviceType>::coeff(int narg, char **arg)
{
AngleCharmm::coeff(narg, arg);
int n = atom->nangletypes;
Kokkos::DualView<F_FLOAT*,DeviceType> k_k("AngleCharmm::k",n+1);
Kokkos::DualView<F_FLOAT*,DeviceType> k_theta0("AngleCharmm::theta0",n+1);
Kokkos::DualView<F_FLOAT*,DeviceType> k_k_ub("AngleCharmm::k_ub",n+1);
Kokkos::DualView<F_FLOAT*,DeviceType> k_r_ub("AngleCharmm::r_ub",n+1);
d_k = k_k.d_view;
d_theta0 = k_theta0.d_view;
d_k_ub = k_k_ub.d_view;
d_r_ub = k_r_ub.d_view;
for (int i = 1; i <= n; i++) {
k_k.h_view[i] = k[i];
k_theta0.h_view[i] = theta0[i];
k_k_ub.h_view[i] = k_ub[i];
k_r_ub.h_view[i] = r_ub[i];
}
k_k.template modify<LMPHostType>();
k_theta0.template modify<LMPHostType>();
k_k_ub.template modify<LMPHostType>();
k_r_ub.template modify<LMPHostType>();
k_k.template sync<DeviceType>();
k_theta0.template sync<DeviceType>();
k_k_ub.template sync<DeviceType>();
k_r_ub.template sync<DeviceType>();
}
/* ----------------------------------------------------------------------
tally energy and virial into global and per-atom accumulators
virial = r1F1 + r2F2 + r3F3 = (r1-r2) F1 + (r3-r2) F3 = del1*f1 + del2*f3
------------------------------------------------------------------------- */
template<class DeviceType>
//template<int NEWTON_BOND>
KOKKOS_INLINE_FUNCTION
void AngleCharmmKokkos<DeviceType>::ev_tally(EV_FLOAT &ev, const int i, const int j, const int k,
F_FLOAT &eangle, F_FLOAT *f1, F_FLOAT *f3,
const F_FLOAT &delx1, const F_FLOAT &dely1, const F_FLOAT &delz1,
const F_FLOAT &delx2, const F_FLOAT &dely2, const F_FLOAT &delz2) const
{
E_FLOAT eanglethird;
F_FLOAT v[6];
if (eflag_either) {
if (eflag_global) {
if (newton_bond) ev.evdwl += eangle;
else {
eanglethird = THIRD*eangle;
if (i < nlocal) ev.evdwl += eanglethird;
if (j < nlocal) ev.evdwl += eanglethird;
if (k < nlocal) ev.evdwl += eanglethird;
}
}
if (eflag_atom) {
eanglethird = THIRD*eangle;
if (newton_bond || i < nlocal) d_eatom[i] += eanglethird;
if (newton_bond || j < nlocal) d_eatom[j] += eanglethird;
if (newton_bond || k < nlocal) d_eatom[k] += eanglethird;
}
}
if (vflag_either) {
v[0] = delx1*f1[0] + delx2*f3[0];
v[1] = dely1*f1[1] + dely2*f3[1];
v[2] = delz1*f1[2] + delz2*f3[2];
v[3] = delx1*f1[1] + delx2*f3[1];
v[4] = delx1*f1[2] + delx2*f3[2];
v[5] = dely1*f1[2] + dely2*f3[2];
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] += THIRD*v[0];
ev.v[1] += THIRD*v[1];
ev.v[2] += THIRD*v[2];
ev.v[3] += THIRD*v[3];
ev.v[4] += THIRD*v[4];
ev.v[5] += THIRD*v[5];
}
if (j < nlocal) {
ev.v[0] += THIRD*v[0];
ev.v[1] += THIRD*v[1];
ev.v[2] += THIRD*v[2];
ev.v[3] += THIRD*v[3];
ev.v[4] += THIRD*v[4];
ev.v[5] += THIRD*v[5];
}
if (k < nlocal) {
ev.v[0] += THIRD*v[0];
ev.v[1] += THIRD*v[1];
ev.v[2] += THIRD*v[2];
ev.v[3] += THIRD*v[3];
ev.v[4] += THIRD*v[4];
ev.v[5] += THIRD*v[5];
}
}
}
if (vflag_atom) {
if (newton_bond || i < nlocal) {
d_vatom(i,0) += THIRD*v[0];
d_vatom(i,1) += THIRD*v[1];
d_vatom(i,2) += THIRD*v[2];
d_vatom(i,3) += THIRD*v[3];
d_vatom(i,4) += THIRD*v[4];
d_vatom(i,5) += THIRD*v[5];
}
if (newton_bond || j < nlocal) {
d_vatom(j,0) += THIRD*v[0];
d_vatom(j,1) += THIRD*v[1];
d_vatom(j,2) += THIRD*v[2];
d_vatom(j,3) += THIRD*v[3];
d_vatom(j,4) += THIRD*v[4];
d_vatom(j,5) += THIRD*v[5];
}
if (newton_bond || k < nlocal) {
d_vatom(k,0) += THIRD*v[0];
d_vatom(k,1) += THIRD*v[1];
d_vatom(k,2) += THIRD*v[2];
d_vatom(k,3) += THIRD*v[3];
d_vatom(k,4) += THIRD*v[4];
d_vatom(k,5) += THIRD*v[5];
}
}
}
}
/* ---------------------------------------------------------------------- */
namespace LAMMPS_NS {
template class AngleCharmmKokkos<LMPDeviceType>;
#ifdef KOKKOS_HAVE_CUDA
template class AngleCharmmKokkos<LMPHostType>;
#endif
}
Event Timeline
Log In to Comment