cv::cuda::HostMem クラス

Class with reference counting wrapping special memory type allocation functions from CUDA. [詳解]

#include <cuda.hpp>

公開型
enum	AllocType { PAGE_LOCKED = 1 , SHARED = 2 , WRITE_COMBINED = 4 }

公開メンバ関数
CV_WRAP	HostMem (HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	HostMem (const HostMem &m)
CV_WRAP	HostMem (int rows, int cols, int type, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
CV_WRAP	HostMem (Size size, int type, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
CV_WRAP	HostMem (InputArray arr, HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)
	creates from host memory with coping data
HostMem &	operator= (const HostMem &m)
CV_WRAP void	swap (HostMem &b)
	swaps with other smart pointer
CV_WRAP HostMem	clone () const
	returns deep copy of the matrix, i.e. the data is copied
CV_WRAP void	create (int rows, int cols, int type)
	allocates new matrix data unless the matrix already has specified size and type.
void	create (Size size, int type)
CV_WRAP HostMem	reshape (int cn, int rows=0) const
void	release ()
	decrements reference counter and released memory if needed.
CV_WRAP Mat	createMatHeader () const
	returns matrix header with disabled reference counting for HostMem data.
GpuMat	createGpuMatHeader () const
	Maps CPU memory to GPU address space and creates the cuda::GpuMat header without reference counting for it. [詳解]
CV_WRAP bool	isContinuous () const
CV_WRAP size_t	elemSize () const
CV_WRAP size_t	elemSize1 () const
CV_WRAP int	type () const
CV_WRAP int	depth () const
CV_WRAP int	channels () const
CV_WRAP size_t	step1 () const
CV_WRAP Size	size () const
CV_WRAP bool	empty () const

静的公開メンバ関数
static MatAllocator *	getAllocator (HostMem::AllocType alloc_type=HostMem::AllocType::PAGE_LOCKED)

公開変数類
int	flags
int	rows
int	cols
CV_PROP size_t	step
uchar *	data
int *	refcount
uchar *	datastart
const uchar *	dataend
AllocType	alloc_type

詳解

Class with reference counting wrapping special memory type allocation functions from CUDA.

Its interface is also Mat-like but with additional memory type parameters.

• PAGE_LOCKED sets a page locked memory type used commonly for fast and asynchronous uploading/downloading data from/to GPU.
• SHARED specifies a zero copy memory allocation that enables mapping the host memory to GPU address space, if supported.
• WRITE_COMBINED sets the write combined buffer that is not cached by CPU. Such buffers are used to supply GPU with data when GPU only reads it. The advantage is a better CPU cache utilization.

覚え書き

Allocation size of such memory types is usually limited. For more details, see CUDA 2.2 Pinned Memory APIs document or CUDA C Programming Guide.

関数詳解

createGpuMatHeader()

GpuMat cv::cuda::HostMem::createGpuMatHeader

(

)

const

Maps CPU memory to GPU address space and creates the cuda::GpuMat header without reference counting for it.

This can be done only if memory was allocated with the SHARED flag and if it is supported by the hardware. Laptops often share video and CPU memory, so address spaces can be mapped, which eliminates an extra copy.

reshape()

CV_WRAP HostMem cv::cuda::HostMem::reshape	(	int	cn,
		int	rows = 0
	)		const

creates alternative HostMem header for the same data, with different number of channels and/or different number of rows

---

このクラス詳解は次のファイルから抽出されました:

• core/cuda.hpp