cuda::memory::pointer_t< T > Class Template Reference

A convenience wrapper around a raw pointer "known" to the CUDA runtime and which thus has various kinds of associated information which this wrapper allows access to. More...

#include <pointer.hpp>

Public Member Functions
T *	get () const
	operator T* () const
device_t	device () const
	Returns a proxy for the device into whose global memory the pointer points.
context_t	context () const
	Returns a proxy for the context in which the memory area, into which the pointer points, was allocated.
T *	get_for_device () const
T *	get_for_host () const
pointer_t	other_side_of_region_pair () const
	pointer_t (T *ptr) noexcept
	pointer_t (const pointer_t &other) noexcept=default
	pointer_t (pointer_t &&other) noexcept=default

Detailed Description

template<typename T>
class cuda::memory::pointer_t< T >

A convenience wrapper around a raw pointer "known" to the CUDA runtime and which thus has various kinds of associated information which this wrapper allows access to.

Member Function Documentation

get()

template<typename T >

T* cuda::memory::pointer_t< T >::get ( ) const

inline

Returns

Address of the pointed-to memory, regardless of which memory space it's in and whether or not it is accessible from the host

get_for_device()

template<typename T >

T* cuda::memory::pointer_t< T >::get_for_device ( ) const

inline

Returns

A pointer into device-accessible memory (not necessary on-device memory though). CUDA ensures that, for pointers to memory not accessible on the CUDA device, nullptr is returned.

Note

With unified memory, this will typically return the same value as the original pointer.

get_for_host()

template<typename T >

T* cuda::memory::pointer_t< T >::get_for_host ( ) const

inline

Returns

A pointer into device-accessible memory (not necessary on-device memory though). CUDA ensures that, for pointers to memory not accessible on the CUDA device, nullptr is returned.

Note

With unified memory, this will typically return the same value as the original pointer.

other_side_of_region_pair()

template<typename T >

pointer_t cuda::memory::pointer_t< T >::other_side_of_region_pair ( ) const

inline

Returns

For a mapped-memory pointer, returns the other side of the mapping, i.e. if this is the device pointer, returns the host pointer, otherwise returns the device pointer. For a managed-memory pointer, or in a unified setting, returns the single pointer usable on both device and host. In other cases returns nullptr.

Note

this relies on either the device and host pointers being nullptr in the case of a non-mapped pointer; and on the device and host pointers being identical to ptr_ for managed-memory pointers.

---

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

• src/cuda/api/pointer.hpp