Detailed Description

template<typename GPUField_T>
class walberla::gpu::communication::GPUPackInfo< GPUField_T >

Data packing/unpacking for ghost layer based communication of a gpu::GPUField.

Template Parameters:

GPUField_T A fully qualified GPUField.

#include <GPUPackInfo.h>

Inheritance diagram for walberla::gpu::communication::GPUPackInfo< GPUField_T >:

Public Types
typedef GPUField_T::value_type	FieldType

Public Member Functions
	GPUPackInfo (const BlockDataID &bdId)

	GPUPackInfo (const BlockDataID &bdId, const uint_t numberOfGHostLayers)

virtual	~GPUPackInfo ()

bool	constantDataExchange () const
	Should return true if the amount of data that is packed for a given block in direction "dir" is guaranteed to remain constant over time. More...

bool	threadsafeReceiving () const
	Must return false if calling unpackData and/or communicateLocal is not thread-safe. More...

void	unpackData (IBlock *receiver, stencil::Direction dir, mpi::RecvBuffer &buffer)
	Unpacks received Data. More...

void	communicateLocal (const IBlock sender, IBlock receiver, stencil::Direction dir)
	Function to copy data from one local block to another local block. More...

void	setCommunicationStream (gpuStream_t stream)

Public Member Functions inherited from walberla::communication::UniformPackInfo
	UniformPackInfo ()=default

virtual	~UniformPackInfo ()=default

void	packData (const IBlock *sender, stencil::Direction dir, mpi::SendBuffer &buffer) const
	Packs data from a block into a send buffer. More...

virtual void	beforeStartCommunication ()
	This function is called once before the communication is started by the UniformBufferedScheme. More...

virtual void	afterStartCommunication ()
	This function is called once after the communication has been started by the UniformBufferedScheme. More...

virtual void	beforeWait ()
	This function is called once before the UniformBufferedScheme waits for the communication to finish. More...

virtual void	afterWait ()
	This function is called once after the communication has been finished by the UniformBufferedScheme. More...

Protected Member Functions
void	packDataImpl (const IBlock *sender, stencil::Direction dir, mpi::SendBuffer &outBuffer) const
	Packs data from a block into a send buffer. More...

uint_t	numberOfGhostLayersToCommunicate (const GPUField_T *const field) const

Protected Member Functions inherited from walberla::communication::UniformPackInfo

Protected Attributes
const BlockDataID	bdId_

bool	communicateAllGhostLayers_

uint_t	numberOfGhostLayers_

bool	copyAsync_

gpuStream_t	communicationStream_

std::map< stencil::Direction, PinnedMemoryBuffer >	pinnedRecvBuffers_

std::map< stencil::Direction, PinnedMemoryBuffer >	pinnedSendBuffers_

Protected Attributes inherited from walberla::communication::UniformPackInfo
std::map< const IBlock *, std::map< stencil::Direction, size_t > >	bufferSize_

Member Typedef Documentation

◆ FieldType

template<typename GPUField_T >

typedef GPUField_T::value_type walberla::gpu::communication::GPUPackInfo< GPUField_T >::FieldType

Constructor & Destructor Documentation

◆ GPUPackInfo() [1/2]

template<typename GPUField_T >

walberla::gpu::communication::GPUPackInfo< GPUField_T >::GPUPackInfo ( const BlockDataID & bdId )

inline

◆ GPUPackInfo() [2/2]

template<typename GPUField_T >

walberla::gpu::communication::GPUPackInfo< GPUField_T >::GPUPackInfo	(	const BlockDataID &	bdId,
		const uint_t	numberOfGHostLayers
	)

inline

◆ ~GPUPackInfo()

template<typename GPUField_T >

virtual walberla::gpu::communication::GPUPackInfo< GPUField_T >::~GPUPackInfo ( )

inlinevirtual

Member Function Documentation

◆ communicateLocal()

template<typename GPUField_T >

void walberla::gpu::communication::GPUPackInfo< GPUField_T >::communicateLocal	(	const IBlock *	sender,
		IBlock *	receiver,
		stencil::Direction	dir
	)

virtual

Function to copy data from one local block to another local block.

Both blocks are allocated on the current process. If NOT thread-safe, threadsafeReceiving must return false!

Parameters

sender	id of block where the data should be copied from
receiver	id of block where the data should be copied to
dir	the direction of the communication ( from sender to receiver )

Implements walberla::communication::UniformPackInfo.

◆ constantDataExchange()

template<typename GPUField_T >

bool walberla::gpu::communication::GPUPackInfo< GPUField_T >::constantDataExchange ( ) const

inlinevirtual

Should return true if the amount of data that is packed for a given block in direction "dir" is guaranteed to remain constant over time.

False otherwise. If you are not sure what to return, return false! Returning false is always save. Falsely return true will lead to errors! However, if the data can be guaranteed to remain constant over time, returning true enables performance optimizations during the communication.

Implements walberla::communication::UniformPackInfo.

◆ numberOfGhostLayersToCommunicate()

template<typename GPUField_T >

uint_t walberla::gpu::communication::GPUPackInfo< GPUField_T >::numberOfGhostLayersToCommunicate ( const GPUField_T *const field ) const

protected

◆ packDataImpl()

template<typename GPUField_T >

void walberla::gpu::communication::GPUPackInfo< GPUField_T >::packDataImpl	(	const IBlock *	sender,
		stencil::Direction	dir,
		mpi::SendBuffer &	buffer
	)		const

protectedvirtual

Packs data from a block into a send buffer.

Must be thread-safe!

Parameters

sender	the block whose data should be packed into a buffer
dir	pack data for neighbor in this direction
buffer	buffer for writing the data into

Implements walberla::communication::UniformPackInfo.

◆ setCommunicationStream()

template<typename GPUField_T >

void walberla::gpu::communication::GPUPackInfo< GPUField_T >::setCommunicationStream ( gpuStream_t stream )

inline

◆ threadsafeReceiving()

template<typename GPUField_T >

bool walberla::gpu::communication::GPUPackInfo< GPUField_T >::threadsafeReceiving ( ) const

inlinevirtual

Must return false if calling unpackData and/or communicateLocal is not thread-safe.

True otherwise. If you are not sure what to return, return false! Returning false is always save. Falsely return true will most likely lead to errors! However, if both unpackData AND communicateLocal are thread-safe, returning true can lead to performance improvements.

Implements walberla::communication::UniformPackInfo.

◆ unpackData()

template<typename GPUField_T >

void walberla::gpu::communication::GPUPackInfo< GPUField_T >::unpackData	(	IBlock *	receiver,
		stencil::Direction	dir,
		mpi::RecvBuffer &	buffer
	)

virtual

Unpacks received Data.

If NOT thread-safe, threadsafeReceiving must return false!

Parameters

receiver	the block where the unpacked data should be stored into
dir	receive data from neighbor in this direction
buffer

std::map< stencil::Direction, PinnedMemoryBuffer > walberla::gpu::communication::GPUPackInfo< GPUField_T >::pinnedSendBuffers_

mutableprotected

The documentation for this class was generated from the following file:

/builds/administration/walberla-website/walberla/src/gpu/communication/GPUPackInfo.h

Detailed Description

template<typename GPUField_T> class walberla::gpu::communication::GPUPackInfo< GPUField_T >

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Member Typedef Documentation

◆ FieldType

Constructor & Destructor Documentation

◆ GPUPackInfo() [1/2]

◆ GPUPackInfo() [2/2]

◆ ~GPUPackInfo()

Member Function Documentation

◆ communicateLocal()

◆ constantDataExchange()

◆ numberOfGhostLayersToCommunicate()

◆ packDataImpl()

◆ setCommunicationStream()

◆ threadsafeReceiving()

◆ unpackData()

Member Data Documentation

◆ bdId_

◆ communicateAllGhostLayers_

◆ communicationStream_

◆ copyAsync_

◆ numberOfGhostLayers_

◆ pinnedRecvBuffers_

◆ pinnedSendBuffers_

template<typename GPUField_T>
class walberla::gpu::communication::GPUPackInfo< GPUField_T >