Arch manual pages

cudaDeviceProp(3) Doxygen cudaDeviceProp(3)

NAME

cudaDeviceProp -

SYNOPSIS

Data Fields


int asyncEngineCount
 
int canMapHostMemory
 
int canUseHostPointerForRegisteredMem
 
int clockRate
 
int computeMode
 
int computePreemptionSupported
 
int concurrentKernels
 
int concurrentManagedAccess
 
int cooperativeLaunch
 
int cooperativeMultiDeviceLaunch
 
int deviceOverlap
 
int ECCEnabled
 
int globalL1CacheSupported
 
int hostNativeAtomicSupported
 
int integrated
 
int isMultiGpuBoard
 
int kernelExecTimeoutEnabled
 
int l2CacheSize
 
int localL1CacheSupported
 
int managedMemory
 
int maxGridSize [3]
 
int maxSurface1D
 
int maxSurface1DLayered [2]
 
int maxSurface2D [2]
 
int maxSurface2DLayered [3]
 
int maxSurface3D [3]
 
int maxSurfaceCubemap
 
int maxSurfaceCubemapLayered [2]
 
int maxTexture1D
 
int maxTexture1DLayered [2]
 
int maxTexture1DLinear
 
int maxTexture1DMipmap
 
int maxTexture2D [2]
 
int maxTexture2DGather [2]
 
int maxTexture2DLayered [3]
 
int maxTexture2DLinear [3]
 
int maxTexture2DMipmap [2]
 
int maxTexture3D [3]
 
int maxTexture3DAlt [3]
 
int maxTextureCubemap
 
int maxTextureCubemapLayered [2]
 
int maxThreadsDim [3]
 
int maxThreadsPerBlock
 
int maxThreadsPerMultiProcessor
 
int memoryBusWidth
 
int memoryClockRate
 
size_t memPitch
 
int multiGpuBoardGroupID
 
int multiProcessorCount
 
int pageableMemoryAccess
 
int pciBusID
 
int pciDeviceID
 
int pciDomainID
 
int regsPerBlock
 
int regsPerMultiprocessor
 
size_t sharedMemPerBlock
 
size_t sharedMemPerBlockOptin
 
size_t sharedMemPerMultiprocessor
 
int singleToDoublePrecisionPerfRatio
 
int streamPrioritiesSupported
 
size_t surfaceAlignment
 
int tccDriver
 
size_t textureAlignment
 
size_t texturePitchAlignment
 
size_t totalConstMem
 
size_t totalGlobalMem
 
int unifiedAddressing
 
int warpSize
 

Detailed Description

CUDA device properties

Field Documentation

int cudaDeviceProp::asyncEngineCount

Number of asynchronous engines

int cudaDeviceProp::canMapHostMemory

Device can map host memory with cudaHostAlloc/cudaHostGetDevicePointer

int cudaDeviceProp::canUseHostPointerForRegisteredMem

Device can access host registered memory at the same virtual address as the CPU

int cudaDeviceProp::clockRate

Clock frequency in kilohertz

int cudaDeviceProp::computeMode

Compute mode (See cudaComputeMode)

int cudaDeviceProp::computePreemptionSupported

Device supports Compute Preemption

int cudaDeviceProp::concurrentKernels

Device can possibly execute multiple kernels concurrently

int cudaDeviceProp::concurrentManagedAccess

Device can coherently access managed memory concurrently with the CPU

int cudaDeviceProp::cooperativeLaunch

Device supports launching cooperative kernels via cudaLaunchCooperativeKernel

int cudaDeviceProp::cooperativeMultiDeviceLaunch

Device can participate in cooperative kernels launched via cudaLaunchCooperativeKernelMultiDevice

int cudaDeviceProp::deviceOverlap

Device can concurrently copy memory and execute a kernel. Deprecated. Use instead asyncEngineCount.

int cudaDeviceProp::ECCEnabled

Device has ECC support enabled

int cudaDeviceProp::globalL1CacheSupported

Device supports caching globals in L1

int cudaDeviceProp::hostNativeAtomicSupported

Link between the device and the host supports native atomic operations

int cudaDeviceProp::integrated

Device is integrated as opposed to discrete

int cudaDeviceProp::isMultiGpuBoard

Device is on a multi-GPU board

int cudaDeviceProp::kernelExecTimeoutEnabled

Specified whether there is a run time limit on kernels

int cudaDeviceProp::l2CacheSize

Size of L2 cache in bytes

int cudaDeviceProp::localL1CacheSupported

Device supports caching locals in L1

int cudaDeviceProp::managedMemory

Device supports allocating managed memory on this system

int cudaDeviceProp::maxGridSize[3]

Maximum size of each dimension of a grid

int cudaDeviceProp::maxSurface1D

Maximum 1D surface size

int cudaDeviceProp::maxSurface1DLayered[2]

Maximum 1D layered surface dimensions

int cudaDeviceProp::maxSurface2D[2]

Maximum 2D surface dimensions

int cudaDeviceProp::maxSurface2DLayered[3]

Maximum 2D layered surface dimensions

int cudaDeviceProp::maxSurface3D[3]

Maximum 3D surface dimensions

int cudaDeviceProp::maxSurfaceCubemap

Maximum Cubemap surface dimensions

int cudaDeviceProp::maxSurfaceCubemapLayered[2]

Maximum Cubemap layered surface dimensions

int cudaDeviceProp::maxTexture1D

Maximum 1D texture size

int cudaDeviceProp::maxTexture1DLayered[2]

Maximum 1D layered texture dimensions

int cudaDeviceProp::maxTexture1DLinear

Maximum size for 1D textures bound to linear memory

int cudaDeviceProp::maxTexture1DMipmap

Maximum 1D mipmapped texture size

int cudaDeviceProp::maxTexture2D[2]

Maximum 2D texture dimensions

int cudaDeviceProp::maxTexture2DGather[2]

Maximum 2D texture dimensions if texture gather operations have to be performed

int cudaDeviceProp::maxTexture2DLayered[3]

Maximum 2D layered texture dimensions

int cudaDeviceProp::maxTexture2DLinear[3]

Maximum dimensions (width, height, pitch) for 2D textures bound to pitched memory

int cudaDeviceProp::maxTexture2DMipmap[2]

Maximum 2D mipmapped texture dimensions

int cudaDeviceProp::maxTexture3D[3]

Maximum 3D texture dimensions

int cudaDeviceProp::maxTexture3DAlt[3]

Maximum alternate 3D texture dimensions

int cudaDeviceProp::maxTextureCubemap

Maximum Cubemap texture dimensions

int cudaDeviceProp::maxTextureCubemapLayered[2]

Maximum Cubemap layered texture dimensions

int cudaDeviceProp::maxThreadsDim[3]

Maximum size of each dimension of a block

int cudaDeviceProp::maxThreadsPerBlock

Maximum number of threads per block

int cudaDeviceProp::maxThreadsPerMultiProcessor

Maximum resident threads per multiprocessor

int cudaDeviceProp::memoryBusWidth

Global memory bus width in bits

int cudaDeviceProp::memoryClockRate

Peak memory clock frequency in kilohertz

size_t cudaDeviceProp::memPitch

Maximum pitch in bytes allowed by memory copies

int cudaDeviceProp::multiGpuBoardGroupID

Unique identifier for a group of devices on the same multi-GPU board

int cudaDeviceProp::multiProcessorCount

Number of multiprocessors on device

int cudaDeviceProp::pageableMemoryAccess

Device supports coherently accessing pageable memory without calling cudaHostRegister on it

int cudaDeviceProp::pciBusID

PCI bus ID of the device

int cudaDeviceProp::pciDeviceID

PCI device ID of the device

int cudaDeviceProp::pciDomainID

PCI domain ID of the device

int cudaDeviceProp::regsPerBlock

32-bit registers available per block

int cudaDeviceProp::regsPerMultiprocessor

32-bit registers available per multiprocessor

size_t cudaDeviceProp::sharedMemPerBlock

Shared memory available per block in bytes

size_t cudaDeviceProp::sharedMemPerBlockOptin

Per device maximum shared memory per block usable by special opt in

size_t cudaDeviceProp::sharedMemPerMultiprocessor

Shared memory available per multiprocessor in bytes

int cudaDeviceProp::singleToDoublePrecisionPerfRatio

Ratio of single precision performance (in floating-point operations per second) to double precision performance

int cudaDeviceProp::streamPrioritiesSupported

Device supports stream priorities

size_t cudaDeviceProp::surfaceAlignment

Alignment requirements for surfaces

int cudaDeviceProp::tccDriver

1 if device is a Tesla device using TCC driver, 0 otherwise

size_t cudaDeviceProp::textureAlignment

Alignment requirement for textures

size_t cudaDeviceProp::texturePitchAlignment

Pitch alignment requirement for texture references bound to pitched memory

size_t cudaDeviceProp::totalConstMem

Constant memory available on device in bytes

size_t cudaDeviceProp::totalGlobalMem

Global memory available on device in bytes

int cudaDeviceProp::unifiedAddressing

Device shares a unified address space with the host

int cudaDeviceProp::warpSize

Warp size in threads
 

Author

Generated automatically by Doxygen from the source code.
2 Sep 2017 Version 6.0