Manages memory, thread locks, and task scheduling for a CUDA context. More...

#include <PxCudaContextManager.h>

Public Member Functions
virtual void	acquireContext ()=0
	Acquire the CUDA context for the current thread. More...

virtual void	releaseContext ()=0
	Release the CUDA context from the current thread. More...

virtual PxCudaMemoryManager *	getMemoryManager ()=0
	Return the PxCudaMemoryManager instance associated with this CUDA context. More...

virtual class PxGpuDispatcher *	getGpuDispatcher ()=0
	Return the PxGpuDispatcher instance associated with this CUDA context. More...

virtual bool	contextIsValid () const =0
	Context manager has a valid CUDA context. More...

virtual bool	supportsArchSM10 () const =0
	G80. More...

virtual bool	supportsArchSM11 () const =0
	G92. More...

virtual bool	supportsArchSM12 () const =0
	GT200. More...

virtual bool	supportsArchSM13 () const =0
	GT260. More...

virtual bool	supportsArchSM20 () const =0
	GF100. More...

virtual bool	supportsArchSM30 () const =0
	GK100. More...

virtual bool	supportsArchSM35 () const =0
	GK110. More...

virtual bool	supportsArchSM50 () const =0
	GM100. More...

virtual bool	supportsArchSM52 () const =0
	GM200. More...

virtual bool	isIntegrated () const =0
	true if GPU is an integrated (MCP) part More...

virtual bool	canMapHostMemory () const =0
	true if GPU map host memory to GPU (0-copy) More...

virtual int	getDriverVersion () const =0
	returns cached value of cuGetDriverVersion() More...

virtual size_t	getDeviceTotalMemBytes () const =0
	returns cached value of device memory size More...

virtual int	getMultiprocessorCount () const =0
	returns cache value of SM unit count More...

virtual unsigned int	getClockRate () const =0
	returns cached value of SM clock frequency More...

virtual int	getSharedMemPerBlock () const =0
	returns total amount of shared memory available per block in bytes More...

virtual unsigned int	getMaxThreadsPerBlock () const =0
	returns the maximum number of threads per block More...

virtual const char *	getDeviceName () const =0
	returns device name retrieved from driver More...

virtual PxCudaInteropMode::Enum	getInteropMode () const =0
	interop mode the context was created with More...

virtual void	setUsingConcurrentStreams (bool)=0
	turn on/off using concurrent streams for GPU work More...

virtual bool	getUsingConcurrentStreams () const =0
	true if GPU work can run in concurrent streams More...

virtual bool	registerResourceInCudaGL (CUgraphicsResource &resource, PxU32 buffer)=0
	Register a rendering resource with CUDA. More...

virtual bool	registerResourceInCudaD3D (CUgraphicsResource &resource, void *resourcePointer)=0
	Register a rendering resource with CUDA. More...

virtual bool	unregisterResourceInCuda (CUgraphicsResource resource)=0
	Unregister a rendering resource with CUDA. More...

virtual int	usingDedicatedPhysXGPU () const =0
	Determine if the user has configured a dedicated PhysX GPU in the NV Control Panel. More...

virtual void	release ()=0
	Release the PxCudaContextManager. More...

Protected Member Functions
virtual	~PxCudaContextManager ()
	protected destructor, use release() method More...

Detailed Description

Manages memory, thread locks, and task scheduling for a CUDA context.

A PxCudaContextManager manages access to a single CUDA context, allowing it to be shared between multiple scenes. Memory allocations are dynamic: starting with an initial heap size and growing on demand by a configurable page size. The context must be acquired from the manager before using any CUDA APIs.

The PxCudaContextManager is based on the CUDA driver API and explictly does not support the the CUDA runtime API (aka, CUDART).

To enable CUDA use by an APEX scene, a PxCudaContextManager must be created (supplying your own CUDA context, or allowing a new context to be allocated for you), the PxGpuDispatcher for that context is retrieved via the getGpuDispatcher() method, and this is assigned to the TaskManager that is given to the scene via its NxApexSceneDesc.

Constructor & Destructor Documentation

virtual PxCudaContextManager::~PxCudaContextManager ( )

inlineprotectedvirtual

protected destructor, use release() method

Member Function Documentation

virtual void PxCudaContextManager::acquireContext ( )

pure virtual

Acquire the CUDA context for the current thread.

Acquisitions are allowed to be recursive within a single thread. You can acquire the context multiple times so long as you release it the same count.

The context must be acquired before using most CUDA functions.

It is not necessary to acquire the CUDA context inside GpuTask launch functions, because the PxGpuDispatcher will have already acquired the context for its worker thread. However it is not harmfull to (re)acquire the context in code that is shared between GpuTasks and non-task functions.

virtual bool PxCudaContextManager::canMapHostMemory ( ) const

pure virtual

true if GPU map host memory to GPU (0-copy)

virtual bool PxCudaContextManager::contextIsValid ( ) const

pure virtual

Context manager has a valid CUDA context.

This method should be called after creating a PxCudaContextManager, especially if the manager was responsible for allocating its own CUDA context (desc.ctx == NULL). If it returns false, there is no point in assigning this manager's PxGpuDispatcher to a TaskManager as it will be unable to execute GpuTasks.

virtual unsigned int PxCudaContextManager::getClockRate ( ) const

pure virtual

returns cached value of SM clock frequency

virtual const char* PxCudaContextManager::getDeviceName ( ) const

pure virtual

returns device name retrieved from driver

virtual size_t PxCudaContextManager::getDeviceTotalMemBytes ( ) const

pure virtual

returns cached value of device memory size

virtual int PxCudaContextManager::getDriverVersion ( ) const

pure virtual

returns cached value of cuGetDriverVersion()

virtual class PxGpuDispatcher* PxCudaContextManager::getGpuDispatcher ( )

pure virtual

Return the PxGpuDispatcher instance associated with this CUDA context.

virtual PxCudaInteropMode::Enum PxCudaContextManager::getInteropMode ( ) const

pure virtual

interop mode the context was created with

virtual unsigned int PxCudaContextManager::getMaxThreadsPerBlock ( ) const

pure virtual

returns the maximum number of threads per block

virtual PxCudaMemoryManager* PxCudaContextManager::getMemoryManager ( )

pure virtual

Return the PxCudaMemoryManager instance associated with this CUDA context.

virtual int PxCudaContextManager::getMultiprocessorCount ( ) const

pure virtual

returns cache value of SM unit count

virtual int PxCudaContextManager::getSharedMemPerBlock ( ) const

pure virtual

returns total amount of shared memory available per block in bytes

virtual bool PxCudaContextManager::getUsingConcurrentStreams ( ) const

pure virtual

true if GPU work can run in concurrent streams

virtual bool PxCudaContextManager::isIntegrated ( ) const

pure virtual

true if GPU is an integrated (MCP) part

virtual bool PxCudaContextManager::registerResourceInCudaD3D	(	CUgraphicsResource &	resource,
		void *	resourcePointer
	)

pure virtual

Register a rendering resource with CUDA.

This function is called to register render resources (allocated from Direct3D) with CUDA so that the memory may be shared between the two systems. This is only required for render resources that are designed for interop use. In APEX, each render resource descriptor that could support interop has a 'registerInCUDA' boolean variable.

The function must be called again any time your graphics device is reset, to re-register the resource.

Returns true if the registration succeeded. A registered resource must be unregistered before it can be released.

Parameters

resource	[OUT] the handle to the resource that can be used with CUDA
resourcePointer	[IN] A pointer to either IDirect3DResource9, or ID3D10Device, or ID3D11Resource to be registered.

virtual bool PxCudaContextManager::registerResourceInCudaGL	(	CUgraphicsResource &	resource,
		PxU32	buffer
	)

pure virtual

Register a rendering resource with CUDA.

This function is called to register render resources (allocated from OpenGL) with CUDA so that the memory may be shared between the two systems. This is only required for render resources that are designed for interop use. In APEX, each render resource descriptor that could support interop has a 'registerInCUDA' boolean variable.

The function must be called again any time your graphics device is reset, to re-register the resource.

Returns true if the registration succeeded. A registered resource must be unregistered before it can be released.

Parameters

resource	[OUT] the handle to the resource that can be used with CUDA
buffer	[IN] GLuint buffer index to be mapped to cuda

virtual void PxCudaContextManager::release ( )

pure virtual

Release the PxCudaContextManager.

When the manager instance is released, it also releases its PxGpuDispatcher instance and PxCudaMemoryManager. Before the memory manager is released, it frees all allocated memory pages. If the PxCudaContextManager created the CUDA context it was responsible for, it also frees that context.

Do not release the PxCudaContextManager if there are any scenes using its PxGpuDispatcher. Those scenes must be released first since there is no safe way to remove a PxGpuDispatcher from a TaskManager once the TaskManager has been given to a scene.

virtual void PxCudaContextManager::releaseContext ( )

pure virtual

Release the CUDA context from the current thread.

The CUDA context should be released as soon as practically possible, to allow other CPU threads (including the PxGpuDispatcher) to work efficiently.

virtual void PxCudaContextManager::setUsingConcurrentStreams ( bool )

pure virtual

turn on/off using concurrent streams for GPU work

virtual bool PxCudaContextManager::supportsArchSM10 ( ) const

pure virtual

G80.

virtual bool PxCudaContextManager::supportsArchSM11 ( ) const

pure virtual

G92.

virtual bool PxCudaContextManager::supportsArchSM12 ( ) const

pure virtual

GT200.

virtual bool PxCudaContextManager::supportsArchSM13 ( ) const

pure virtual

GT260.

virtual bool PxCudaContextManager::supportsArchSM20 ( ) const

pure virtual

GF100.

virtual bool PxCudaContextManager::supportsArchSM30 ( ) const

pure virtual

GK100.

virtual bool PxCudaContextManager::supportsArchSM35 ( ) const

pure virtual

GK110.

virtual bool PxCudaContextManager::supportsArchSM50 ( ) const

pure virtual

GM100.

virtual bool PxCudaContextManager::supportsArchSM52 ( ) const

pure virtual

GM200.

virtual bool PxCudaContextManager::unregisterResourceInCuda ( CUgraphicsResource resource )

pure virtual

Unregister a rendering resource with CUDA.

If a render resource was successfully registered with CUDA using the registerResourceInCuda***() methods, this function must be called to unregister the resource before the it can be released.

virtual int PxCudaContextManager::usingDedicatedPhysXGPU ( ) const

pure virtual

Determine if the user has configured a dedicated PhysX GPU in the NV Control Panel.

Note: If using CUDA Interop, this will always return false

Returns: 1 if there is a dedicated PhysX GPU; 0 if there is NOT a dedicated PhysX GPU; -1 if the routine is not implemented

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

PxCudaContextManager.h

Public Member Functions

Protected Member Functions

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation