shift.cpp File Reference

#include <Cabana_Core.hpp>
#include <Kokkos_Core.hpp>
#include <algorithm>
#include <iostream>
#include <vector>
#include <mpi.h>
#include "timer_macro.hpp"
#include "domain_decomposition.hpp"
#include "sort.hpp"
#include "transpose_particles.hpp"
#include "shift.hpp"

Include dependency graph for shift.cpp:

Functions
template<int PTL_N_DBL, class DataType , class MPIDevice >
void	copy_to_send_buffer (Cabana::AoSoA< DataType, DeviceType, VEC_LEN > &local_particles, int n_staying, Kokkos::View< OneParticle< PTL_N_DBL > *, MPIDevice > &buf)
template<class DataType >
void	transpose_and_shift (Kokkos::View< int *, HostType > &sendcnt, Kokkos::View< int *, HostType > &sdispls, Kokkos::View< int *, HostType > &recvcnt, Kokkos::View< int *, HostType > &rdispls, Cabana::AoSoA< DataType, DeviceType, VEC_LEN > &local_particles, int n_staying, int n_leaving, int n_arriving, const MyMPI &mpi)
void	shift (Species< DeviceType > &species, SortViews< DeviceType > &sort_views, const Grid< DeviceType > &grid, const MagneticField< DeviceType > &magnetic_field, const DomainDecomposition< DeviceType > &pol_decomp, bool send_phase0, bool plane_sort_only)
void	shift_all_species (Plasma &plasma, const Grid< DeviceType > &grid, const MagneticField< DeviceType > &magnetic_field, const DomainDecomposition< DeviceType > &pol_decomp, Shift::ShiftPh0 shift_ph0)

Function Documentation

template<int PTL_N_DBL, class DataType , class MPIDevice >

void copy_to_send_buffer	(	Cabana::AoSoA< DataType, DeviceType, VEC_LEN > &	local_particles,
		int	n_staying,
		Kokkos::View< OneParticle< PTL_N_DBL > *, MPIDevice > &	buf
	)

Here is the caller graph for this function:

void shift	(	Species< DeviceType > &	species,
		SortViews< DeviceType > &	sort_views,
		const Grid< DeviceType > &	grid,
		const MagneticField< DeviceType > &	magnetic_field,
		const DomainDecomposition< DeviceType > &	pol_decomp,
		bool	send_phase0,
		bool	plane_sort_only
	)

Shifts particles between MPI ranks. First the particles are sorted by destination rank The particles are packed into a buffer and sent with an MPI all-to-all Then they are unpacked from the receiving buffer back into the particle AoSoA The same is done for phase0, needed for ions

Parameters

[in,out]	species	contains the particles
[in]	sort_views	is needed for the sort
[in]	grid	is needed for the sort
[in]	magnetic_field	is needed for the sort
[in]	pol_decomp	is needed for the sort and shift

Returns

void

Here is the call graph for this function:

Here is the caller graph for this function:

void shift_all_species	(	Plasma &	plasma,
		const Grid< DeviceType > &	grid,
		const MagneticField< DeviceType > &	magnetic_field,
		const DomainDecomposition< DeviceType > &	pol_decomp,
		Shift::ShiftPh0	shift_ph0
	)

Shifts particles between MPI ranks. First the particles are sorted by destination rank The particles are packed into a buffer and sent with an MPI all-to-all Then they are unpacked from the receiving buffer back into the particle AoSoA The same is done for phase0, needed for ions

Parameters

[in,out]	plasma	contains all species info
[in]	grid	is needed for the sort
[in]	magnetic_field	is needed for the sort
[in]	pol_decomp	is needed for the sort

Returns

void

Here is the call graph for this function:

Here is the caller graph for this function:

template<class DataType >

void transpose_and_shift	(	Kokkos::View< int *, HostType > &	sendcnt,
		Kokkos::View< int *, HostType > &	sdispls,
		Kokkos::View< int *, HostType > &	recvcnt,
		Kokkos::View< int *, HostType > &	rdispls,
		Cabana::AoSoA< DataType, DeviceType, VEC_LEN > &	local_particles,
		int	n_staying,
		int	n_leaving,
		int	n_arriving,
		const MyMPI &	mpi
	)

Packs particles into a send buffer and shifts them between MPI ranks. If particles are on the device and GPU-aware MPI is off, the send buffer resides on the host and there is a receive buffer on the host as well. If not, then the host buffer is on device and there is no receive buffer because the arriving particles are brought straight to the paticle array. The function is templated on the size of the particle (in doubles) and the particle data type. This is done so that phase0 can be sent with the same routine

Parameters

[in]	sendcnt	is the number of particles getting sent to each rank
[in]	sdispls	is the displacement in the send buffer of each rank
[in]	recvcnt	is the number of particles getting received from each rank
[in]	rdispls	is the displacement in the recv buffer of each rank
[in,out]	local_particles	is where the particles are located on device
[in]	n_staying	is the number of particles staying on this rank
[in]	n_leaving	is the number of particles leaving this rank
[in]	n_arriving	is the number of particles arriving to this rank
[in]	mpi	contains MPI and domain decomposition parameters

Returns

void

Here is the call graph for this function: