PCL
pcl::CUDADevice Class Reference

Access to core CUDA device services. More...

#include <CUDADevice.h>

Public Member Functions

 CUDADevice ()=delete
 
 CUDADevice (const CUDADevice &)=delete
 
CUDADeviceoperator= (CUDADevice &)=delete
 

Static Public Member Functions

static bool IsAvailable () noexcept
 
static int MaxThreadsPerBlock () noexcept
 
static IsoString Name ()
 
static size_type SharedMemoryPerBlock () noexcept
 
static size_type TotalGlobalMemory () noexcept
 

Detailed Description

TODO: Write description.

Definition at line 73 of file CUDADevice.h.

Constructor & Destructor Documentation

◆ CUDADevice() [1/2]

pcl::CUDADevice::CUDADevice ( )
delete

Default constructor - deleted, not an instantiable class.

◆ CUDADevice() [2/2]

pcl::CUDADevice::CUDADevice ( const CUDADevice )
delete

Copy constructor - deleted, not an instantiable class.

Member Function Documentation

◆ IsAvailable()

static bool pcl::CUDADevice::IsAvailable ( )
staticnoexcept

Returns true iff a valid and operational CUDA device is currently available on the running PixInsight platform.

◆ MaxThreadsPerBlock()

static int pcl::CUDADevice::MaxThreadsPerBlock ( )
staticnoexcept

Returns the maximum number of threads per block available in the active CUDA device, or zero if no valid CUDA device is available.

◆ Name()

static IsoString pcl::CUDADevice::Name ( )
static

Returns the identifying name of the active CUDA device, or an empty string if there is no valid CUDA device available on the running PixInsight platform.

◆ operator=()

CUDADevice& pcl::CUDADevice::operator= ( CUDADevice )
delete

Copy assignment operator - deleted, not an instantiable class.

◆ SharedMemoryPerBlock()

static size_type pcl::CUDADevice::SharedMemoryPerBlock ( )
staticnoexcept

Returns the total shared memory available per block on the active CUDA device in bytes, or zero if no valid CUDA device is available.

◆ TotalGlobalMemory()

static size_type pcl::CUDADevice::TotalGlobalMemory ( )
staticnoexcept

Returns the total global memory available on the active CUDA device in bytes, or zero if no valid CUDA device is available.


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