runtime ¶

cudaDeviceGetLimit, cuCtxSetLimit

See also

cuda.bindings.runtime.cudaDeviceGetLimit(limit: cudaLimit)¶

Return resource limits.

Returns in *pValue the current size of limit. The following cudaLimit values are supported.

cudaLimitStackSize is the stack size in bytes of each GPU thread.
cudaLimitPrintfFifoSize is the size in bytes of the shared FIFO used by the printf() device system call.
cudaLimitMallocHeapSize is the size in bytes of the heap used by the malloc() and free() device system calls.
cudaLimitDevRuntimeSyncDepth is the maximum grid depth at which a thread can isssue the device runtime call cudaDeviceSynchronize() to wait on child grid launches to complete. This functionality is removed for devices of compute capability >= 9.0, and hence will return error cudaErrorUnsupportedLimit on such devices.
cudaLimitDevRuntimePendingLaunchCount is the maximum number of outstanding device runtime launches.
cudaLimitMaxL2FetchGranularity is the L2 cache fetch granularity.
cudaLimitPersistingL2CacheSize is the persisting L2 cache size in bytes.

Parameters:

limit (cudaLimit) – Limit to query

Returns:

cudaError_t – cudaSuccess, cudaErrorUnsupportedLimit, cudaErrorInvalidValue
pValue (int) – Returned size of the limit

See also

cudaDeviceSetLimit, cuCtxGetLimit

cuda.bindings.runtime.cudaDeviceGetTexture1DLinearMaxWidth(cudaChannelFormatDesc fmtDesc: Optional[cudaChannelFormatDesc], int device)¶

Returns the maximum number of elements allocatable in a 1D linear texture for a given element size.

Returns in maxWidthInElements the maximum number of elements allocatable in a 1D linear texture for given format descriptor fmtDesc.

Parameters:

fmtDesc (cudaChannelFormatDesc) – Texture format description.
None (int) – None

Returns:

cudaError_t – cudaSuccess, cudaErrorUnsupportedLimit, cudaErrorInvalidValue
maxWidthInElements (int) – Returns maximum number of texture elements allocatable for given fmtDesc.

See also

cuDeviceGetTexture1DLinearMaxWidth

cuda.bindings.runtime.cudaDeviceGetCacheConfig()¶

Returns the preferred cache configuration for the current device.

On devices where the L1 cache and shared memory use the same hardware resources, this returns through pCacheConfig the preferred cache configuration for the current device. This is only a preference. The runtime will use the requested configuration if possible, but it is free to choose a different configuration if required to execute functions.

This will return a pCacheConfig of cudaFuncCachePreferNone on devices where the size of the L1 cache and shared memory are fixed.

The supported cache configurations are:

cudaFuncCachePreferNone: no preference for shared memory or L1 (default)
cudaFuncCachePreferShared: prefer larger shared memory and smaller L1 cache
cudaFuncCachePreferL1: prefer larger L1 cache and smaller shared memory
cudaFuncCachePreferEqual: prefer equal size L1 cache and shared memory

Returns:

cudaError_t – cudaSuccess
pCacheConfig (cudaFuncCache) – Returned cache configuration

See also

cudaDeviceSetCacheConfig, cudaFuncSetCacheConfig (C API), cudaFuncSetCacheConfig (C++ API), cuCtxGetCacheConfig

cuda.bindings.runtime.cudaDeviceGetStreamPriorityRange()¶

Returns numerical values that correspond to the least and greatest stream priorities.

Returns in *leastPriority and *greatestPriority the numerical values that correspond to the least and greatest stream priorities respectively. Stream priorities follow a convention where lower numbers imply greater priorities. The range of meaningful stream priorities is given by [*greatestPriority, *leastPriority]. If the user attempts to create a stream with a priority value that is outside the the meaningful range as specified by this API, the priority is automatically clamped down or up to either *leastPriority or *greatestPriority respectively. See cudaStreamCreateWithPriority for details on creating a priority stream. A NULL may be passed in for *leastPriority or *greatestPriority if the value is not desired.

This function will return ‘0’ in both *leastPriority and *greatestPriority if the current context’s device does not support stream priorities (see cudaDeviceGetAttribute).

Returns:

cudaError_t – cudaSuccess
leastPriority (int) – Pointer to an int in which the numerical value for least stream priority is returned
greatestPriority (int) – Pointer to an int in which the numerical value for greatest stream priority is returned

cuda.bindings.runtime.cudaDeviceSetCacheConfig(cacheConfig: cudaFuncCache)¶

Sets the preferred cache configuration for the current device.

On devices where the L1 cache and shared memory use the same hardware resources, this sets through cacheConfig the preferred cache configuration for the current device. This is only a preference. The runtime will use the requested configuration if possible, but it is free to choose a different configuration if required to execute the function. Any function preference set via cudaFuncSetCacheConfig (C API) or cudaFuncSetCacheConfig (C++ API) will be preferred over this device-wide setting. Setting the device-wide cache configuration to cudaFuncCachePreferNone will cause subsequent kernel launches to prefer to not change the cache configuration unless required to launch the kernel.

This setting does nothing on devices where the size of the L1 cache and shared memory are fixed.

Launching a kernel with a different preference than the most recent preference setting may insert a device-side synchronization point.

The supported cache configurations are:

cudaFuncCachePreferNone: no preference for shared memory or L1 (default)
cudaFuncCachePreferShared: prefer larger shared memory and smaller L1 cache
cudaFuncCachePreferL1: prefer larger L1 cache and smaller shared memory
cudaFuncCachePreferEqual: prefer equal size L1 cache and shared memory

Parameters:: cacheConfig (cudaFuncCache) – Requested cache configuration
Returns:: cudaSuccess
Return type:: cudaError_t

See also

cudaDeviceGetCacheConfig, cudaFuncSetCacheConfig (C API), cudaFuncSetCacheConfig (C++ API), cuCtxSetCacheConfig

cuda.bindings.runtime.cudaDeviceGetByPCIBusId(char *pciBusId)¶

Returns a handle to a compute device.

Returns in *device a device ordinal given a PCI bus ID string.

where domain, bus, device, and function are all hexadecimal values

Parameters:

pciBusId (bytes) – String in one of the following forms:

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice
device (int) – Returned device ordinal

cuda.bindings.runtime.cudaDeviceGetPCIBusId(int length, int device)¶

Returns a PCI Bus Id string for the device.

Returns an ASCII string identifying the device dev in the NULL- terminated string pointed to by pciBusId. length specifies the maximum length of the string that may be returned.

where domain, bus, device, and function are all hexadecimal values. pciBusId should be large enough to store 13 characters including the NULL-terminator.

Parameters:

length (int) – Maximum length of string to store in name
device (int) – Device to get identifier string for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice
pciBusId (bytes) – Returned identifier string for the device in the following format

cuda.bindings.runtime.cudaIpcGetEventHandle(event)¶

Gets an interprocess handle for a previously allocated event.

Takes as input a previously allocated event. This event must have been created with the cudaEventInterprocess and cudaEventDisableTiming flags set. This opaque handle may be copied into other processes and opened with cudaIpcOpenEventHandle to allow efficient hardware synchronization between GPU work in different processes.

After the event has been been opened in the importing process, cudaEventRecord, cudaEventSynchronize, cudaStreamWaitEvent and cudaEventQuery may be used in either process. Performing operations on the imported event after the exported event has been freed with cudaEventDestroy will result in undefined behavior.

IPC functionality is restricted to devices with support for unified addressing on Linux and Windows operating systems. IPC functionality on Windows is supported for compatibility purposes but not recommended as it comes with performance cost. Users can test their device for IPC functionality by calling cudaDeviceGetAttribute with cudaDevAttrIpcEventSupport

Parameters:

event (CUevent or cudaEvent_t) – Event allocated with cudaEventInterprocess and cudaEventDisableTiming flags.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorMemoryAllocation, cudaErrorMapBufferObjectFailed, cudaErrorNotSupported, cudaErrorInvalidValue
handle (cudaIpcEventHandle_t) – Pointer to a user allocated cudaIpcEventHandle in which to return the opaque event handle

cuda.bindings.runtime.cudaIpcOpenEventHandle(cudaIpcEventHandle_t handle: cudaIpcEventHandle_t)¶

Opens an interprocess event handle for use in the current process.

Opens an interprocess event handle exported from another process with cudaIpcGetEventHandle. This function returns a cudaEvent_t that behaves like a locally created event with the cudaEventDisableTiming flag specified. This event must be freed with cudaEventDestroy.

Performing operations on the imported event after the exported event has been freed with cudaEventDestroy will result in undefined behavior.

Parameters:

handle (cudaIpcEventHandle_t) – Interprocess handle to open

Returns:

cudaError_t – cudaSuccess, cudaErrorMapBufferObjectFailed, cudaErrorNotSupported, cudaErrorInvalidValue, cudaErrorDeviceUninitialized
event (cudaEvent_t) – Returns the imported event

cuda.bindings.runtime.cudaIpcGetMemHandle(devPtr)¶

Gets an interprocess memory handle for an existing device memory allocation.

Takes a pointer to the base of an existing device memory allocation created with cudaMalloc and exports it for use in another process. This is a lightweight operation and may be called multiple times on an allocation without adverse effects.

If a region of memory is freed with cudaFree and a subsequent call to cudaMalloc returns memory with the same device address, cudaIpcGetMemHandle will return a unique handle for the new memory.

Parameters:

devPtr (Any) – Base pointer to previously allocated device memory

Returns:

cudaError_t – cudaSuccess, cudaErrorMemoryAllocation, cudaErrorMapBufferObjectFailed, cudaErrorNotSupported, cudaErrorInvalidValue
handle (cudaIpcMemHandle_t) – Pointer to user allocated cudaIpcMemHandle to return the handle in.

cuda.bindings.runtime.cudaIpcOpenMemHandle(cudaIpcMemHandle_t handle: cudaIpcMemHandle_t, unsigned int flags)¶

Opens an interprocess memory handle exported from another process and returns a device pointer usable in the local process.

Maps memory exported from another process with cudaIpcGetMemHandle into the current device address space. For contexts on different devices cudaIpcOpenMemHandle can attempt to enable peer access between the devices as if the user called cudaDeviceEnablePeerAccess. This behavior is controlled by the cudaIpcMemLazyEnablePeerAccess flag. cudaDeviceCanAccessPeer can determine if a mapping is possible.

cudaIpcOpenMemHandle can open handles to devices that may not be visible in the process calling the API.

Contexts that may open cudaIpcMemHandles are restricted in the following way. cudaIpcMemHandles from each device in a given process may only be opened by one context per device per other process.

If the memory handle has already been opened by the current context, the reference count on the handle is incremented by 1 and the existing device pointer is returned.

Memory returned from cudaIpcOpenMemHandle must be freed with cudaIpcCloseMemHandle.

Calling cudaFree on an exported memory region before calling cudaIpcCloseMemHandle in the importing context will result in undefined behavior.

Parameters:

handle (cudaIpcMemHandle_t) – cudaIpcMemHandle to open
flags (unsigned int) – Flags for this operation. Must be specified as cudaIpcMemLazyEnablePeerAccess

Returns:

cudaError_t – cudaSuccess, cudaErrorMapBufferObjectFailed, cudaErrorInvalidResourceHandle, cudaErrorDeviceUninitialized, cudaErrorTooManyPeers, cudaErrorNotSupported, cudaErrorInvalidValue
devPtr (Any) – Returned device pointer

Notes

No guarantees are made about the address returned in *devPtr.: In particular, multiple processes may not receive the same address for the same handle.

cuda.bindings.runtime.cudaIpcCloseMemHandle(devPtr)¶

Attempts to close memory mapped with cudaIpcOpenMemHandle.

Decrements the reference count of the memory returnd by cudaIpcOpenMemHandle by 1. When the reference count reaches 0, this API unmaps the memory. The original allocation in the exporting process as well as imported mappings in other processes will be unaffected.

Any resources used to enable peer access will be freed if this is the last mapping using them.

Parameters:: devPtr (Any) – Device pointer returned by cudaIpcOpenMemHandle
Returns:: cudaSuccess, cudaErrorMapBufferObjectFailed, cudaErrorNotSupported, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaDeviceFlushGPUDirectRDMAWrites(target: cudaFlushGPUDirectRDMAWritesTarget, scope: cudaFlushGPUDirectRDMAWritesScope)¶

Blocks until remote writes are visible to the specified scope.

Blocks until remote writes to the target context via mappings created through GPUDirect RDMA APIs, like nvidia_p2p_get_pages (see https://docs.nvidia.com/cuda/gpudirect-rdma for more information), are visible to the specified scope.

If the scope equals or lies within the scope indicated by cudaDevAttrGPUDirectRDMAWritesOrdering, the call will be a no-op and can be safely omitted for performance. This can be determined by comparing the numerical values between the two enums, with smaller scopes having smaller values.

Users may query support for this API via cudaDevAttrGPUDirectRDMAFlushWritesOptions.

Parameters:

target (cudaFlushGPUDirectRDMAWritesTarget) – The target of the operation, see cudaFlushGPUDirectRDMAWritesTarget
scope (cudaFlushGPUDirectRDMAWritesScope) – The scope of the operation, see cudaFlushGPUDirectRDMAWritesScope

Returns:

cudaSuccess, cudaErrorNotSupported,

Return type:

cuFlushGPUDirectRDMAWrites

See also

cuda.bindings.runtime.cudaDeviceRegisterAsyncNotification(int device, callbackFunc, userData)¶

Registers a callback function to receive async notifications.

Registers callbackFunc to receive async notifications.

The userData parameter is passed to the callback function at async notification time. Likewise, callback is also passed to the callback function to distinguish between multiple registered callbacks.

The callback function being registered should be designed to return quickly (~10ms). Any long running tasks should be queued for execution on an application thread.

Callbacks may not call cudaDeviceRegisterAsyncNotification or cudaDeviceUnregisterAsyncNotification. Doing so will result in cudaErrorNotPermitted. Async notification callbacks execute in an undefined order and may be serialized.

Returns in *callback a handle representing the registered callback instance.

Parameters:

device (int) – The device on which to register the callback
callbackFunc (cudaAsyncCallback) – The function to register as a callback
userData (Any) – A generic pointer to user data. This is passed into the callback function.

Returns:

cudaError_t – cudaSuccess cudaErrorNotSupported cudaErrorInvalidDevice cudaErrorInvalidValue cudaErrorNotPermitted cudaErrorUnknown
callback (cudaAsyncCallbackHandle_t) – A handle representing the registered callback instance

See also

cudaDeviceUnregisterAsyncNotification

cuda.bindings.runtime.cudaDeviceUnregisterAsyncNotification(int device, callback)¶

Unregisters an async notification callback.

Unregisters callback so that the corresponding callback function will stop receiving async notifications.

Parameters:

device (int) – The device from which to remove callback.
callback (cudaAsyncCallbackHandle_t) – The callback instance to unregister from receiving async notifications.

Returns:

cudaSuccess cudaErrorNotSupported cudaErrorInvalidDevice cudaErrorInvalidValue cudaErrorNotPermitted cudaErrorUnknown

Return type:

cudaDeviceRegisterAsyncNotification

See also

cuda.bindings.runtime.cudaGetDeviceCount()¶

Returns the number of compute-capable devices.

Returns in *count the number of devices with compute capability greater or equal to 2.0 that are available for execution.

Returns:

cudaError_t – cudaSuccess
count (int) – Returns the number of devices with compute capability greater or equal to 2.0

cuda.bindings.runtime.cudaGetDeviceProperties(int device)¶

Returns information about the compute-device.

Returns in *prop the properties of device dev. The cudaDeviceProp structure is defined as:

View CUDA Toolkit Documentation for a C++ code example

where:

name[256] is an ASCII string identifying the device.
uuid is a 16-byte unique identifier.
totalGlobalMem is the total amount of global memory available on the device in bytes.
sharedMemPerBlock is the maximum amount of shared memory available to a thread block in bytes.
regsPerBlock is the maximum number of 32-bit registers available to a thread block.
warpSize is the warp size in threads.
memPitch is the maximum pitch in bytes allowed by the memory copy functions that involve memory regions allocated through cudaMallocPitch().
maxThreadsPerBlock is the maximum number of threads per block.
maxThreadsDim[3] contains the maximum size of each dimension of a block.
maxGridSize[3] contains the maximum size of each dimension of a grid.
clockRate is the clock frequency in kilohertz.
totalConstMem is the total amount of constant memory available on the device in bytes.
major, minor are the major and minor revision numbers defining the device’s compute capability.
textureAlignment is the alignment requirement; texture base addresses that are aligned to textureAlignment bytes do not need an offset applied to texture fetches.
texturePitchAlignment is the pitch alignment requirement for 2D texture references that are bound to pitched memory.
deviceOverlap is 1 if the device can concurrently copy memory between host and device while executing a kernel, or 0 if not. Deprecated, use instead asyncEngineCount.
multiProcessorCount is the number of multiprocessors on the device.
kernelExecTimeoutEnabled is 1 if there is a run time limit for kernels executed on the device, or 0 if not.
integrated is 1 if the device is an integrated (motherboard) GPU and 0 if it is a discrete (card) component.
canMapHostMemory is 1 if the device can map host memory into the CUDA address space for use with cudaHostAlloc()/cudaHostGetDevicePointer(), or 0 if not.
computeMode is the compute mode that the device is currently in. Available modes are as follows:
- cudaComputeModeDefault: Default mode - Device is not restricted and multiple threads can use cudaSetDevice() with this device.
- cudaComputeModeProhibited: Compute-prohibited mode - No threads can use cudaSetDevice() with this device.
- cudaComputeModeExclusiveProcess: Compute-exclusive-process mode - Many threads in one process will be able to use cudaSetDevice() with this device. When an occupied exclusive mode device is chosen with cudaSetDevice, all subsequent non-device management runtime functions will return cudaErrorDevicesUnavailable.
maxTexture1D is the maximum 1D texture size.
maxTexture1DMipmap is the maximum 1D mipmapped texture texture size.
maxTexture1DLinear is the maximum 1D texture size for textures bound to linear memory.
maxTexture2D[2] contains the maximum 2D texture dimensions.
maxTexture2DMipmap[2] contains the maximum 2D mipmapped texture dimensions.
maxTexture2DLinear[3] contains the maximum 2D texture dimensions for 2D textures bound to pitch linear memory.
maxTexture2DGather[2] contains the maximum 2D texture dimensions if texture gather operations have to be performed.
maxTexture3D[3] contains the maximum 3D texture dimensions.
maxTexture3DAlt[3] contains the maximum alternate 3D texture dimensions.
maxTextureCubemap is the maximum cubemap texture width or height.
maxTexture1DLayered[2] contains the maximum 1D layered texture dimensions.
maxTexture2DLayered[3] contains the maximum 2D layered texture dimensions.
maxTextureCubemapLayered[2] contains the maximum cubemap layered texture dimensions.
maxSurface1D is the maximum 1D surface size.
maxSurface2D[2] contains the maximum 2D surface dimensions.
maxSurface3D[3] contains the maximum 3D surface dimensions.
maxSurface1DLayered[2] contains the maximum 1D layered surface dimensions.
maxSurface2DLayered[3] contains the maximum 2D layered surface dimensions.
maxSurfaceCubemap is the maximum cubemap surface width or height.
maxSurfaceCubemapLayered[2] contains the maximum cubemap layered surface dimensions.
surfaceAlignment specifies the alignment requirements for surfaces.
concurrentKernels is 1 if the device supports executing multiple kernels within the same context simultaneously, or 0 if not. It is not guaranteed that multiple kernels will be resident on the device concurrently so this feature should not be relied upon for correctness.
ECCEnabled is 1 if the device has ECC support turned on, or 0 if not.
pciBusID is the PCI bus identifier of the device.
pciDeviceID is the PCI device (sometimes called slot) identifier of the device.
pciDomainID is the PCI domain identifier of the device.
tccDriver is 1 if the device is using a TCC driver or 0 if not.
asyncEngineCount is 1 when the device can concurrently copy memory between host and device while executing a kernel. It is 2 when the device can concurrently copy memory between host and device in both directions and execute a kernel at the same time. It is 0 if neither of these is supported.
unifiedAddressing is 1 if the device shares a unified address space with the host and 0 otherwise.
memoryClockRate is the peak memory clock frequency in kilohertz.
memoryBusWidth is the memory bus width in bits.
l2CacheSize is L2 cache size in bytes.
persistingL2CacheMaxSize is L2 cache’s maximum persisting lines size in bytes.
maxThreadsPerMultiProcessor is the number of maximum resident threads per multiprocessor.
streamPrioritiesSupported is 1 if the device supports stream priorities, or 0 if it is not supported.
globalL1CacheSupported is 1 if the device supports caching of globals in L1 cache, or 0 if it is not supported.
localL1CacheSupported is 1 if the device supports caching of locals in L1 cache, or 0 if it is not supported.
sharedMemPerMultiprocessor is the maximum amount of shared memory available to a multiprocessor in bytes; this amount is shared by all thread blocks simultaneously resident on a multiprocessor.
regsPerMultiprocessor is the maximum number of 32-bit registers available to a multiprocessor; this number is shared by all thread blocks simultaneously resident on a multiprocessor.
managedMemory is 1 if the device supports allocating managed memory on this system, or 0 if it is not supported.
isMultiGpuBoard is 1 if the device is on a multi-GPU board (e.g. Gemini cards), and 0 if not;
multiGpuBoardGroupID is a unique identifier for a group of devices associated with the same board. Devices on the same multi- GPU board will share the same identifier.
hostNativeAtomicSupported is 1 if the link between the device and the host supports native atomic operations, or 0 if it is not supported.
singleToDoublePrecisionPerfRatio is the ratio of single precision performance (in floating-point operations per second) to double precision performance.
pageableMemoryAccess is 1 if the device supports coherently accessing pageable memory without calling cudaHostRegister on it, and 0 otherwise.
concurrentManagedAccess is 1 if the device can coherently access managed memory concurrently with the CPU, and 0 otherwise.
computePreemptionSupported is 1 if the device supports Compute Preemption, and 0 otherwise.
canUseHostPointerForRegisteredMem is 1 if the device can access host registered memory at the same virtual address as the CPU, and 0 otherwise.
cooperativeLaunch is 1 if the device supports launching cooperative kernels via cudaLaunchCooperativeKernel, and 0 otherwise.
cooperativeMultiDeviceLaunch is 1 if the device supports launching cooperative kernels via cudaLaunchCooperativeKernelMultiDevice, and 0 otherwise.
sharedMemPerBlockOptin is the per device maximum shared memory per block usable by special opt in
pageableMemoryAccessUsesHostPageTables is 1 if the device accesses pageable memory via the host’s page tables, and 0 otherwise.
directManagedMemAccessFromHost is 1 if the host can directly access managed memory on the device without migration, and 0 otherwise.
maxBlocksPerMultiProcessor is the maximum number of thread blocks that can reside on a multiprocessor.
accessPolicyMaxWindowSize is the maximum value of num_bytes.
reservedSharedMemPerBlock is the shared memory reserved by CUDA driver per block in bytes
hostRegisterSupported is 1 if the device supports host memory registration via cudaHostRegister, and 0 otherwise.
sparseCudaArraySupported is 1 if the device supports sparse CUDA arrays and sparse CUDA mipmapped arrays, 0 otherwise
hostRegisterReadOnlySupported is 1 if the device supports using the cudaHostRegister flag cudaHostRegisterReadOnly to register memory that must be mapped as read-only to the GPU
timelineSemaphoreInteropSupported is 1 if external timeline semaphore interop is supported on the device, 0 otherwise
memoryPoolsSupported is 1 if the device supports using the cudaMallocAsync and cudaMemPool family of APIs, 0 otherwise
gpuDirectRDMASupported is 1 if the device supports GPUDirect RDMA APIs, 0 otherwise
gpuDirectRDMAFlushWritesOptions is a bitmask to be interpreted according to the cudaFlushGPUDirectRDMAWritesOptions enum
gpuDirectRDMAWritesOrdering See the cudaGPUDirectRDMAWritesOrdering enum for numerical values
memoryPoolSupportedHandleTypes is a bitmask of handle types supported with mempool-based IPC
deferredMappingCudaArraySupported is 1 if the device supports deferred mapping CUDA arrays and CUDA mipmapped arrays
ipcEventSupported is 1 if the device supports IPC Events, and 0 otherwise
unifiedFunctionPointers is 1 if the device support unified pointers, and 0 otherwise

Parameters:

device (int) – Device number to get properties for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice
prop (cudaDeviceProp) – Properties for the specified device

cuda.bindings.runtime.cudaDeviceGetAttribute(attr: cudaDeviceAttr, int device)¶

Returns information about the device.

Returns in *value the integer value of the attribute attr on device device. The supported attributes are:

cudaDevAttrMaxThreadsPerBlock: Maximum number of threads per block
cudaDevAttrMaxBlockDimX: Maximum x-dimension of a block
cudaDevAttrMaxBlockDimY: Maximum y-dimension of a block
cudaDevAttrMaxBlockDimZ: Maximum z-dimension of a block
cudaDevAttrMaxGridDimX: Maximum x-dimension of a grid
cudaDevAttrMaxGridDimY: Maximum y-dimension of a grid
cudaDevAttrMaxGridDimZ: Maximum z-dimension of a grid
cudaDevAttrMaxSharedMemoryPerBlock: Maximum amount of shared memory available to a thread block in bytes
cudaDevAttrTotalConstantMemory: Memory available on device for constant variables in a CUDA C kernel in bytes
cudaDevAttrWarpSize: Warp size in threads
cudaDevAttrMaxPitch: Maximum pitch in bytes allowed by the memory copy functions that involve memory regions allocated through cudaMallocPitch()
cudaDevAttrMaxTexture1DWidth: Maximum 1D texture width
cudaDevAttrMaxTexture1DLinearWidth: Maximum width for a 1D texture bound to linear memory
cudaDevAttrMaxTexture1DMipmappedWidth: Maximum mipmapped 1D texture width
cudaDevAttrMaxTexture2DWidth: Maximum 2D texture width
cudaDevAttrMaxTexture2DHeight: Maximum 2D texture height
cudaDevAttrMaxTexture2DLinearWidth: Maximum width for a 2D texture bound to linear memory
cudaDevAttrMaxTexture2DLinearHeight: Maximum height for a 2D texture bound to linear memory
cudaDevAttrMaxTexture2DLinearPitch: Maximum pitch in bytes for a 2D texture bound to linear memory
cudaDevAttrMaxTexture2DMipmappedWidth: Maximum mipmapped 2D texture width
cudaDevAttrMaxTexture2DMipmappedHeight: Maximum mipmapped 2D texture height
cudaDevAttrMaxTexture3DWidth: Maximum 3D texture width
cudaDevAttrMaxTexture3DHeight: Maximum 3D texture height
cudaDevAttrMaxTexture3DDepth: Maximum 3D texture depth
cudaDevAttrMaxTexture3DWidthAlt: Alternate maximum 3D texture width, 0 if no alternate maximum 3D texture size is supported
cudaDevAttrMaxTexture3DHeightAlt: Alternate maximum 3D texture height, 0 if no alternate maximum 3D texture size is supported
cudaDevAttrMaxTexture3DDepthAlt: Alternate maximum 3D texture depth, 0 if no alternate maximum 3D texture size is supported
cudaDevAttrMaxTextureCubemapWidth: Maximum cubemap texture width or height
cudaDevAttrMaxTexture1DLayeredWidth: Maximum 1D layered texture width
cudaDevAttrMaxTexture1DLayeredLayers: Maximum layers in a 1D layered texture
cudaDevAttrMaxTexture2DLayeredWidth: Maximum 2D layered texture width
cudaDevAttrMaxTexture2DLayeredHeight: Maximum 2D layered texture height
cudaDevAttrMaxTexture2DLayeredLayers: Maximum layers in a 2D layered texture
cudaDevAttrMaxTextureCubemapLayeredWidth: Maximum cubemap layered texture width or height
cudaDevAttrMaxTextureCubemapLayeredLayers: Maximum layers in a cubemap layered texture
cudaDevAttrMaxSurface1DWidth: Maximum 1D surface width
cudaDevAttrMaxSurface2DWidth: Maximum 2D surface width
cudaDevAttrMaxSurface2DHeight: Maximum 2D surface height
cudaDevAttrMaxSurface3DWidth: Maximum 3D surface width
cudaDevAttrMaxSurface3DHeight: Maximum 3D surface height
cudaDevAttrMaxSurface3DDepth: Maximum 3D surface depth
cudaDevAttrMaxSurface1DLayeredWidth: Maximum 1D layered surface width
cudaDevAttrMaxSurface1DLayeredLayers: Maximum layers in a 1D layered surface
cudaDevAttrMaxSurface2DLayeredWidth: Maximum 2D layered surface width
cudaDevAttrMaxSurface2DLayeredHeight: Maximum 2D layered surface height
cudaDevAttrMaxSurface2DLayeredLayers: Maximum layers in a 2D layered surface
cudaDevAttrMaxSurfaceCubemapWidth: Maximum cubemap surface width
cudaDevAttrMaxSurfaceCubemapLayeredWidth: Maximum cubemap layered surface width
cudaDevAttrMaxSurfaceCubemapLayeredLayers: Maximum layers in a cubemap layered surface
cudaDevAttrMaxRegistersPerBlock: Maximum number of 32-bit registers available to a thread block
cudaDevAttrClockRate: Peak clock frequency in kilohertz
cudaDevAttrTextureAlignment: Alignment requirement; texture base addresses aligned to textureAlign bytes do not need an offset applied to texture fetches
cudaDevAttrTexturePitchAlignment: Pitch alignment requirement for 2D texture references bound to pitched memory
cudaDevAttrGpuOverlap: 1 if the device can concurrently copy memory between host and device while executing a kernel, or 0 if not
cudaDevAttrMultiProcessorCount: Number of multiprocessors on the device
cudaDevAttrKernelExecTimeout: 1 if there is a run time limit for kernels executed on the device, or 0 if not
cudaDevAttrIntegrated: 1 if the device is integrated with the memory subsystem, or 0 if not
cudaDevAttrCanMapHostMemory: 1 if the device can map host memory into the CUDA address space, or 0 if not
cudaDevAttrComputeMode: Compute mode is the compute mode that the device is currently in. Available modes are as follows:
- cudaComputeModeDefault: Default mode - Device is not restricted and multiple threads can use cudaSetDevice() with this device.
- cudaComputeModeProhibited: Compute-prohibited mode - No threads can use cudaSetDevice() with this device.
- cudaComputeModeExclusiveProcess: Compute-exclusive- process mode - Many threads in one process will be able to use cudaSetDevice() with this device.
cudaDevAttrConcurrentKernels: 1 if the device supports executing multiple kernels within the same context simultaneously, or 0 if not. It is not guaranteed that multiple kernels will be resident on the device concurrently so this feature should not be relied upon for correctness.
cudaDevAttrEccEnabled: 1 if error correction is enabled on the device, 0 if error correction is disabled or not supported by the device
cudaDevAttrPciBusId: PCI bus identifier of the device
cudaDevAttrPciDeviceId: PCI device (also known as slot) identifier of the device
cudaDevAttrTccDriver: 1 if the device is using a TCC driver. TCC is only available on Tesla hardware running Windows Vista or later.
cudaDevAttrMemoryClockRate: Peak memory clock frequency in kilohertz
cudaDevAttrGlobalMemoryBusWidth: Global memory bus width in bits
cudaDevAttrL2CacheSize: Size of L2 cache in bytes. 0 if the device doesn’t have L2 cache.
cudaDevAttrMaxThreadsPerMultiProcessor: Maximum resident threads per multiprocessor
cudaDevAttrUnifiedAddressing: 1 if the device shares a unified address space with the host, or 0 if not
cudaDevAttrComputeCapabilityMajor: Major compute capability version number
cudaDevAttrComputeCapabilityMinor: Minor compute capability version number
cudaDevAttrStreamPrioritiesSupported: 1 if the device supports stream priorities, or 0 if not
cudaDevAttrGlobalL1CacheSupported: 1 if device supports caching globals in L1 cache, 0 if not
cudaDevAttrLocalL1CacheSupported: 1 if device supports caching locals in L1 cache, 0 if not
cudaDevAttrMaxSharedMemoryPerMultiprocessor: Maximum amount of shared memory available to a multiprocessor in bytes; this amount is shared by all thread blocks simultaneously resident on a multiprocessor
cudaDevAttrMaxRegistersPerMultiprocessor: Maximum number of 32-bit registers available to a multiprocessor; this number is shared by all thread blocks simultaneously resident on a multiprocessor
cudaDevAttrManagedMemory: 1 if device supports allocating managed memory, 0 if not
cudaDevAttrIsMultiGpuBoard: 1 if device is on a multi-GPU board, 0 if not
cudaDevAttrMultiGpuBoardGroupID: Unique identifier for a group of devices on the same multi-GPU board
cudaDevAttrHostNativeAtomicSupported: 1 if the link between the device and the host supports native atomic operations
cudaDevAttrSingleToDoublePrecisionPerfRatio: Ratio of single precision performance (in floating-point operations per second) to double precision performance
cudaDevAttrPageableMemoryAccess: 1 if the device supports coherently accessing pageable memory without calling cudaHostRegister on it, and 0 otherwise
cudaDevAttrConcurrentManagedAccess: 1 if the device can coherently access managed memory concurrently with the CPU, and 0 otherwise
cudaDevAttrComputePreemptionSupported: 1 if the device supports Compute Preemption, 0 if not
cudaDevAttrCanUseHostPointerForRegisteredMem: 1 if the device can access host registered memory at the same virtual address as the CPU, and 0 otherwise
cudaDevAttrCooperativeLaunch: 1 if the device supports launching cooperative kernels via cudaLaunchCooperativeKernel, and 0 otherwise
cudaDevAttrCooperativeMultiDeviceLaunch: 1 if the device supports launching cooperative kernels via cudaLaunchCooperativeKernelMultiDevice, and 0 otherwise
cudaDevAttrCanFlushRemoteWrites: 1 if the device supports flushing of outstanding remote writes, and 0 otherwise
cudaDevAttrHostRegisterSupported: 1 if the device supports host memory registration via cudaHostRegister, and 0 otherwise
cudaDevAttrPageableMemoryAccessUsesHostPageTables: 1 if the device accesses pageable memory via the host’s page tables, and 0 otherwise
cudaDevAttrDirectManagedMemAccessFromHost: 1 if the host can directly access managed memory on the device without migration, and 0 otherwise
cudaDevAttrMaxSharedMemoryPerBlockOptin: Maximum per block shared memory size on the device. This value can be opted into when using cudaFuncSetAttribute
cudaDevAttrMaxBlocksPerMultiprocessor: Maximum number of thread blocks that can reside on a multiprocessor
cudaDevAttrMaxPersistingL2CacheSize: Maximum L2 persisting lines capacity setting in bytes
cudaDevAttrMaxAccessPolicyWindowSize: Maximum value of num_bytes
cudaDevAttrReservedSharedMemoryPerBlock: Shared memory reserved by CUDA driver per block in bytes
cudaDevAttrSparseCudaArraySupported: 1 if the device supports sparse CUDA arrays and sparse CUDA mipmapped arrays.
cudaDevAttrHostRegisterReadOnlySupported: Device supports using the cudaHostRegister flag cudaHostRegisterReadOnly to register memory that must be mapped as read-only to the GPU
cudaDevAttrMemoryPoolsSupported: 1 if the device supports using the cudaMallocAsync and cudaMemPool family of APIs, and 0 otherwise
cudaDevAttrGPUDirectRDMASupported: 1 if the device supports GPUDirect RDMA APIs, and 0 otherwise
cudaDevAttrGPUDirectRDMAFlushWritesOptions: bitmask to be interpreted according to the cudaFlushGPUDirectRDMAWritesOptions enum
cudaDevAttrGPUDirectRDMAWritesOrdering: see the cudaGPUDirectRDMAWritesOrdering enum for numerical values
cudaDevAttrMemoryPoolSupportedHandleTypes: Bitmask of handle types supported with mempool based IPC
cudaDevAttrDeferredMappingCudaArraySupported : 1 if the device supports deferred mapping CUDA arrays and CUDA mipmapped arrays.
cudaDevAttrIpcEventSupport: 1 if the device supports IPC Events.
cudaDevAttrNumaConfig: NUMA configuration of a device: value is of type cudaDeviceNumaConfig enum
cudaDevAttrNumaId: NUMA node ID of the GPU memory

Parameters:

attr (cudaDeviceAttr) – Device attribute to query
device (int) – Device number to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue
value (int) – Returned device attribute value

cuda.bindings.runtime.cudaDeviceGetDefaultMemPool(int device)¶

Returns the default mempool of a device.

The default mempool of a device contains device memory from that device.

Parameters:

device (int) – None

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue cudaErrorNotSupported
memPool (cudaMemPool_t) – None

cuda.bindings.runtime.cudaDeviceSetMemPool(int device, memPool)¶

Sets the current memory pool of a device.

The memory pool must be local to the specified device. Unless a mempool is specified in the cudaMallocAsync call, cudaMallocAsync allocates from the current mempool of the provided stream’s device. By default, a device’s current memory pool is its default memory pool.

Parameters:

device (int) – None
memPool (CUmemoryPool or cudaMemPool_t) – None

Returns:

cudaSuccess, cudaErrorInvalidValue cudaErrorInvalidDevice cudaErrorNotSupported

Return type:

Notes

Use cudaMallocFromPoolAsync to specify asynchronous allocations from a device different than the one the stream runs on.

cuda.bindings.runtime.cudaDeviceGetMemPool(int device)¶

Gets the current mempool for a device.

Returns the last pool provided to cudaDeviceSetMemPool for this device or the device’s default memory pool if cudaDeviceSetMemPool has never been called. By default the current mempool is the default mempool for a device, otherwise the returned pool must have been set with cuDeviceSetMemPool or cudaDeviceSetMemPool.

Parameters:

device (int) – None

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue cudaErrorNotSupported
memPool (cudaMemPool_t) – None

cuda.bindings.runtime.cudaDeviceGetNvSciSyncAttributes(nvSciSyncAttrList, int device, int flags)¶

Return NvSciSync attributes that this device can support.

Returns in nvSciSyncAttrList, the properties of NvSciSync that this CUDA device, dev can support. The returned nvSciSyncAttrList can be used to create an NvSciSync that matches this device’s capabilities.

If NvSciSyncAttrKey_RequiredPerm field in nvSciSyncAttrList is already set this API will return cudaErrorInvalidValue.

The applications should set nvSciSyncAttrList to a valid NvSciSyncAttrList failing which this API will return cudaErrorInvalidHandle.

The flags controls how applications intends to use the NvSciSync created from the nvSciSyncAttrList. The valid flags are:

cudaNvSciSyncAttrSignal, specifies that the applications intends to signal an NvSciSync on this CUDA device.
cudaNvSciSyncAttrWait, specifies that the applications intends to wait on an NvSciSync on this CUDA device.

At least one of these flags must be set, failing which the API returns cudaErrorInvalidValue. Both the flags are orthogonal to one another: a developer may set both these flags that allows to set both wait and signal specific attributes in the same nvSciSyncAttrList.

Note that this API updates the input nvSciSyncAttrList with values equivalent to the following public attribute key-values: NvSciSyncAttrKey_RequiredPerm is set to

NvSciSyncAccessPerm_SignalOnly if cudaNvSciSyncAttrSignal is set in flags.
NvSciSyncAccessPerm_WaitOnly if cudaNvSciSyncAttrWait is set in flags.
NvSciSyncAccessPerm_WaitSignal if both cudaNvSciSyncAttrWait and cudaNvSciSyncAttrSignal are set in flags. NvSciSyncAttrKey_PrimitiveInfo is set to
NvSciSyncAttrValPrimitiveType_SysmemSemaphore on any valid device.
NvSciSyncAttrValPrimitiveType_Syncpoint if device is a Tegra device.
NvSciSyncAttrValPrimitiveType_SysmemSemaphorePayload64b if device is GA10X+. NvSciSyncAttrKey_GpuId is set to the same UUID that is returned in None from cudaDeviceGetProperties for this device.

cudaSuccess, cudaErrorDeviceUninitialized, cudaErrorInvalidValue, cudaErrorInvalidHandle, cudaErrorInvalidDevice, cudaErrorNotSupported, cudaErrorMemoryAllocation

Parameters:

nvSciSyncAttrList (Any) – Return NvSciSync attributes supported.
device (int) – Valid Cuda Device to get NvSciSync attributes for.
flags (int) – flags describing NvSciSync usage.

Return type:

cudaSuccess, cudaErrorInvalidDevice,

cuda.bindings.runtime.cudaDeviceGetP2PAttribute(attr: cudaDeviceP2PAttr, int srcDevice, int dstDevice)¶

Queries attributes of the link between two devices.

Returns in *value the value of the requested attribute attrib of the link between srcDevice and dstDevice. The supported attributes are:

cudaDevP2PAttrPerformanceRank: A relative value indicating the performance of the link between two devices. Lower value means better performance (0 being the value used for most performant link).
cudaDevP2PAttrAccessSupported: 1 if peer access is enabled.
cudaDevP2PAttrNativeAtomicSupported: 1 if native atomic operations over the link are supported.
cudaDevP2PAttrCudaArrayAccessSupported: 1 if accessing CUDA arrays over the link is supported.

Returns cudaErrorInvalidDevice if srcDevice or dstDevice are not valid or if they represent the same device.

Returns cudaErrorInvalidValue if attrib is not valid or if value is a null pointer.

Parameters:

attrib (cudaDeviceP2PAttr) – The requested attribute of the link between srcDevice and dstDevice.
srcDevice (int) – The source device of the target link.
dstDevice (int) – The destination device of the target link.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue
value (int) – Returned value of the requested attribute

cuda.bindings.runtime.cudaChooseDevice(cudaDeviceProp prop: Optional[cudaDeviceProp])¶

Select compute-device which best matches criteria.

Returns in *device the device which has properties that best match *prop.

Parameters:

prop (cudaDeviceProp) – Desired device properties

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
device (int) – Device with best match

cuda.bindings.runtime.cudaInitDevice(int device, unsigned int deviceFlags, unsigned int flags)¶

Initialize device to be used for GPU executions.

This function will initialize the CUDA Runtime structures and primary context on device when called, but the context will not be made current to device.

When cudaInitDeviceFlagsAreValid is set in flags, deviceFlags are applied to the requested device. The values of deviceFlags match those of the flags parameters in cudaSetDeviceFlags. The effect may be verified by cudaGetDeviceFlags.

This function will return an error if the device is in cudaComputeModeExclusiveProcess and is occupied by another process or if the device is in cudaComputeModeProhibited.

Parameters:

device (int) – Device on which the runtime will initialize itself.
deviceFlags (unsigned int) – Parameters for device operation.
flags (unsigned int) – Flags for controlling the device initialization.

Returns:

Return type:

cuda.bindings.runtime.cudaSetDevice(int device)¶

Set device to be used for GPU executions.

Sets device as the current device for the calling host thread. Valid device id’s are 0 to (cudaGetDeviceCount() - 1).

Any device memory subsequently allocated from this host thread using cudaMalloc(), cudaMallocPitch() or cudaMallocArray() will be physically resident on device. Any host memory allocated from this host thread using cudaMallocHost() or cudaHostAlloc() or cudaHostRegister() will have its lifetime associated with device. Any streams or events created from this host thread will be associated with device. Any kernels launched from this host thread using the <<<>>> operator or cudaLaunchKernel() will be executed on device.

This call may be made from any host thread, to any device, and at any time. This function will do no synchronization with the previous or new device, and should only take significant time when it initializes the runtime’s context state. This call will bind the primary context of the specified device to the calling thread and all the subsequent memory allocations, stream and event creations, and kernel launches will be associated with the primary context. This function will also immediately initialize the runtime state on the primary context, and the context will be current on device immediately. This function will return an error if the device is in cudaComputeModeExclusiveProcess and is occupied by another process or if the device is in cudaComputeModeProhibited.

It is not required to call cudaInitDevice before using this function.

Parameters:: device (int) – Device on which the active host thread should execute the device code.
Returns:: cudaSuccess, cudaErrorInvalidDevice, cudaErrorDeviceUnavailable,
Return type:: cudaError_t

cuda.bindings.runtime.cudaGetDevice()¶

Returns which device is currently being used.

Returns in *device the current device for the calling host thread.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorDeviceUnavailable,
device (int) – Returns the device on which the active host thread executes the device code.

cuda.bindings.runtime.cudaSetDeviceFlags(unsigned int flags)¶

Sets flags to be used for device executions.

Records flags as the flags for the current device. If the current device has been set and that device has already been initialized, the previous flags are overwritten. If the current device has not been initialized, it is initialized with the provided flags. If no device has been made current to the calling thread, a default device is selected and initialized with the provided flags.

The three LSBs of the flags parameter can be used to control how the CPU thread interacts with the OS scheduler when waiting for results from the device.

cudaDeviceScheduleAuto: The default value if the flags parameter is zero, uses a heuristic based on the number of active CUDA contexts in the process C and the number of logical processors in the system P. If C > P, then CUDA will yield to other OS threads when waiting for the device, otherwise CUDA will not yield while waiting for results and actively spin on the processor. Additionally, on Tegra devices, cudaDeviceScheduleAuto uses a heuristic based on the power profile of the platform and may choose cudaDeviceScheduleBlockingSync for low-powered devices.
cudaDeviceScheduleSpin: Instruct CUDA to actively spin when waiting for results from the device. This can decrease latency when waiting for the device, but may lower the performance of CPU threads if they are performing work in parallel with the CUDA thread.
cudaDeviceScheduleYield: Instruct CUDA to yield its thread when waiting for results from the device. This can increase latency when waiting for the device, but can increase the performance of CPU threads performing work in parallel with the device.
cudaDeviceScheduleBlockingSync: Instruct CUDA to block the CPU thread on a synchronization primitive when waiting for the device to finish work.
cudaDeviceBlockingSync: Instruct CUDA to block the CPU thread on a synchronization primitive when waiting for the device to finish work. Deprecated: This flag was deprecated as of CUDA 4.0 and replaced with cudaDeviceScheduleBlockingSync.
cudaDeviceMapHost: This flag enables allocating pinned host memory that is accessible to the device. It is implicit for the runtime but may be absent if a context is created using the driver API. If this flag is not set, cudaHostGetDevicePointer() will always return a failure code.
cudaDeviceLmemResizeToMax: Instruct CUDA to not reduce local memory after resizing local memory for a kernel. This can prevent thrashing by local memory allocations when launching many kernels with high local memory usage at the cost of potentially increased memory usage. Deprecated: This flag is deprecated and the behavior enabled by this flag is now the default and cannot be disabled.
cudaDeviceSyncMemops: Ensures that synchronous memory operations initiated on this context will always synchronize. See further documentation in the section titled “API Synchronization behavior” to learn more about cases when synchronous memory operations can exhibit asynchronous behavior.

Parameters:: flags (unsigned int) – Parameters for device operation
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaGetDeviceFlags()¶

Gets the flags for the current device.

Returns in flags the flags for the current device. If there is a current device for the calling thread, the flags for the device are returned. If there is no current device, the flags for the first device are returned, which may be the default flags. Compare to the behavior of cudaSetDeviceFlags.

Typically, the flags returned should match the behavior that will be seen if the calling thread uses a device after this call, without any change to the flags or current device inbetween by this or another thread. Note that if the device is not initialized, it is possible for another thread to change the flags for the current device before it is initialized. Additionally, when using exclusive mode, if this thread has not requested a specific device, it may use a device other than the first device, contrary to the assumption made by this function.

If a context has been created via the driver API and is current to the calling thread, the flags for that context are always returned.

Flags returned by this function may specifically include cudaDeviceMapHost even though it is not accepted by cudaSetDeviceFlags because it is implicit in runtime API flags. The reason for this is that the current context may have been created via the driver API in which case the flag is not implicit and may be unset.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice
flags (unsigned int) – Pointer to store the device flags

Error Handling¶

This section describes the error handling functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaGetLastError()¶

Returns the last error from a runtime call.

Returns the last error that has been produced by any of the runtime calls in the same instance of the CUDA Runtime library in the host thread and resets it to cudaSuccess.

Note: Multiple instances of the CUDA Runtime library can be present in an application when using a library that statically links the CUDA Runtime.

Returns:: cudaSuccess, cudaErrorMissingConfiguration, cudaErrorMemoryAllocation, cudaErrorInitializationError, cudaErrorLaunchFailure, cudaErrorLaunchTimeout, cudaErrorLaunchOutOfResources, cudaErrorInvalidDeviceFunction, cudaErrorInvalidConfiguration, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidSymbol, cudaErrorUnmapBufferObjectFailed, cudaErrorInvalidDevicePointer, cudaErrorInvalidTexture, cudaErrorInvalidTextureBinding, cudaErrorInvalidChannelDescriptor, cudaErrorInvalidMemcpyDirection, cudaErrorInvalidFilterSetting, cudaErrorInvalidNormSetting, cudaErrorUnknown, cudaErrorInvalidResourceHandle, cudaErrorInsufficientDriver, cudaErrorNoDevice, cudaErrorSetOnActiveProcess, cudaErrorStartupFailure, cudaErrorInvalidPtx, cudaErrorUnsupportedPtxVersion, cudaErrorNoKernelImageForDevice, cudaErrorJitCompilerNotFound, cudaErrorJitCompilationDisabled
Return type:: cudaError_t

cuda.bindings.runtime.cudaPeekAtLastError()¶

Returns the last error from a runtime call.

Returns the last error that has been produced by any of the runtime calls in the same instance of the CUDA Runtime library in the host thread. This call does not reset the error to cudaSuccess like cudaGetLastError().

Note: Multiple instances of the CUDA Runtime library can be present in an application when using a library that statically links the CUDA Runtime.

Returns:: cudaSuccess, cudaErrorMissingConfiguration, cudaErrorMemoryAllocation, cudaErrorInitializationError, cudaErrorLaunchFailure, cudaErrorLaunchTimeout, cudaErrorLaunchOutOfResources, cudaErrorInvalidDeviceFunction, cudaErrorInvalidConfiguration, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidSymbol, cudaErrorUnmapBufferObjectFailed, cudaErrorInvalidDevicePointer, cudaErrorInvalidTexture, cudaErrorInvalidTextureBinding, cudaErrorInvalidChannelDescriptor, cudaErrorInvalidMemcpyDirection, cudaErrorInvalidFilterSetting, cudaErrorInvalidNormSetting, cudaErrorUnknown, cudaErrorInvalidResourceHandle, cudaErrorInsufficientDriver, cudaErrorNoDevice, cudaErrorSetOnActiveProcess, cudaErrorStartupFailure, cudaErrorInvalidPtx, cudaErrorUnsupportedPtxVersion, cudaErrorNoKernelImageForDevice, cudaErrorJitCompilerNotFound, cudaErrorJitCompilationDisabled
Return type:: cudaError_t

cuda.bindings.runtime.cudaGetErrorName(error: cudaError_t)¶

Returns the string representation of an error code enum name.

Returns a string containing the name of an error code in the enum. If the error code is not recognized, “unrecognized error code” is returned.

Parameters:

error (cudaError_t) – Error code to convert to string

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
bytes – char* pointer to a NULL-terminated string

cuda.bindings.runtime.cudaGetErrorString(error: cudaError_t)¶

Returns the description string for an error code.

Returns the description string for an error code. If the error code is not recognized, “unrecognized error code” is returned.

Parameters:

error (cudaError_t) – Error code to convert to string

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
bytes – char* pointer to a NULL-terminated string

Stream Management¶

This section describes the stream management functions of the CUDA runtime application programming interface.

class cuda.bindings.runtime.cudaStreamCallback_t(*args, **kwargs)¶

getPtr()¶: Get memory address of class instance

cuda.bindings.runtime.cudaStreamCreate()¶

Create an asynchronous stream.

Creates a new asynchronous stream on the context that is current to the calling host thread. If no context is current to the calling host thread, then the primary context for a device is selected, made current to the calling thread, and initialized before creating a stream on it.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pStream (cudaStream_t) – Pointer to new stream identifier

cuda.bindings.runtime.cudaStreamCreateWithFlags(unsigned int flags)¶

Create an asynchronous stream.

cudaStreamDefault: Default stream creation flag.
cudaStreamNonBlocking: Specifies that work running in the created stream may run concurrently with work in stream 0 (the NULL stream), and that the created stream should perform no implicit synchronization with stream 0.

Parameters:

flags (unsigned int) – Parameters for stream creation

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pStream (cudaStream_t) – Pointer to new stream identifier

cuda.bindings.runtime.cudaStreamCreateWithPriority(unsigned int flags, int priority)¶

Create an asynchronous stream with the specified priority.

Creates a stream with the specified priority and returns a handle in pStream. The stream is created on the context that is current to the calling host thread. If no context is current to the calling host thread, then the primary context for a device is selected, made current to the calling thread, and initialized before creating a stream on it. This affects the scheduling priority of work in the stream. Priorities provide a hint to preferentially run work with higher priority when possible, but do not preempt already-running work or provide any other functional guarantee on execution order.

priority follows a convention where lower numbers represent higher priorities. ‘0’ represents default priority. The range of meaningful numerical priorities can be queried using cudaDeviceGetStreamPriorityRange. If the specified priority is outside the numerical range returned by cudaDeviceGetStreamPriorityRange, it will automatically be clamped to the lowest or the highest number in the range.

Parameters:

flags (unsigned int) – Flags for stream creation. See cudaStreamCreateWithFlags for a list of valid flags that can be passed
priority (int) – Priority of the stream. Lower numbers represent higher priorities. See cudaDeviceGetStreamPriorityRange for more information about the meaningful stream priorities that can be passed.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pStream (cudaStream_t) – Pointer to new stream identifier

Notes

Stream priorities are supported only on GPUs with compute capability 3.5 or higher.

In the current implementation, only compute kernels launched in priority streams are affected by the stream’s priority. Stream priorities have no effect on host-to-device and device-to-host memory operations.

cuda.bindings.runtime.cudaStreamGetPriority(hStream)¶

Query the priority of a stream.

Query the priority of a stream. The priority is returned in in priority. Note that if the stream was created with a priority outside the meaningful numerical range returned by cudaDeviceGetStreamPriorityRange, this function returns the clamped priority. See cudaStreamCreateWithPriority for details about priority clamping.

Parameters:

hStream (CUstream or cudaStream_t) – Handle to the stream to be queried

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
priority (int) – Pointer to a signed integer in which the stream’s priority is returned

cuda.bindings.runtime.cudaStreamGetFlags(hStream)¶

Query the flags of a stream.

Query the flags of a stream. The flags are returned in flags. See cudaStreamCreateWithFlags for a list of valid flags.

Parameters:

hStream (CUstream or cudaStream_t) – Handle to the stream to be queried

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
flags (unsigned int) – Pointer to an unsigned integer in which the stream’s flags are returned

cuda.bindings.runtime.cudaStreamGetId(hStream)¶

Query the Id of a stream.

Query the Id of a stream. The Id is returned in streamId. The Id is unique for the life of the program.

The stream handle hStream can refer to any of the following:

a stream created via any of the CUDA runtime APIs such as cudaStreamCreate, cudaStreamCreateWithFlags and cudaStreamCreateWithPriority, or their driver API equivalents such as cuStreamCreate or cuStreamCreateWithPriority. Passing an invalid handle will result in undefined behavior.
any of the special streams such as the NULL stream, cudaStreamLegacy and cudaStreamPerThread respectively. The driver API equivalents of these are also accepted which are NULL, CU_STREAM_LEGACY and CU_STREAM_PER_THREAD.

Parameters:

hStream (CUstream or cudaStream_t) – Handle to the stream to be queried

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
streamId (unsigned long long) – Pointer to an unsigned long long in which the stream Id is returned

cuda.bindings.runtime.cudaCtxResetPersistingL2Cache()¶

Resets all persisting lines in cache to normal status.

Resets all persisting lines in cache to normal status. Takes effect on function return.

Returns:: cudaSuccess,
Return type:: cudaError_t

See also

cudaSuccess, cudaErrorNotSupported

cuda.bindings.runtime.cudaStreamCopyAttributes(dst, src)¶

Copies attributes from source stream to destination stream.

Copies attributes from source stream src to destination stream dst. Both streams must have the same context.

Parameters:

dst (CUstream or cudaStream_t) – Destination stream
src (CUstream or cudaStream_t) – Source stream For attributes see cudaStreamAttrID

Returns:

Return type:

See also

cuda.bindings.runtime.cudaStreamGetAttribute(hStream, attr: cudaStreamAttrID)¶

Queries stream attribute.

Queries attribute attr from hStream and stores it in corresponding member of value_out.

Parameters:

hStream (CUstream or cudaStream_t)
attr (cudaStreamAttrID)

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
value_out (cudaStreamAttrValue)

See also

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

cuda.bindings.runtime.cudaStreamSetAttribute(hStream, attr: cudaStreamAttrID, cudaStreamAttrValue value: Optional[cudaStreamAttrValue])¶

Sets stream attribute.

Sets attribute attr on hStream from corresponding attribute of value. The updated attribute will be applied to subsequent work submitted to the stream. It will not affect previously submitted work.

Parameters:

hStream (CUstream or cudaStream_t)
attr (cudaStreamAttrID)
value (cudaStreamAttrValue)

Returns:

Return type:

See also

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

cuda.bindings.runtime.cudaStreamDestroy(stream)¶

Destroys and cleans up an asynchronous stream.

Destroys and cleans up the asynchronous stream specified by stream.

In case the device is still doing work in the stream stream when cudaStreamDestroy() is called, the function will return immediately and the resources associated with stream will be released automatically once the device has completed all work in stream.

Parameters:: stream (CUstream or cudaStream_t) – Stream identifier
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
Return type:: cudaError_t

cuda.bindings.runtime.cudaStreamWaitEvent(stream, event, unsigned int flags)¶

Make a compute stream wait on an event.

Makes all future work submitted to stream wait for all work captured in event. See cudaEventRecord() for details on what is captured by an event. The synchronization will be performed efficiently on the device when applicable. event may be from a different device than stream.

flags include:

cudaEventWaitDefault: Default event creation flag.
cudaEventWaitExternal: Event is captured in the graph as an external event node when performing stream capture.

Parameters:

stream (CUstream or cudaStream_t) – Stream to wait
event (CUevent or cudaEvent_t) – Event to wait on
flags (unsigned int) – Parameters for the operation(See above)

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorInvalidValue, cudaErrorNotSupported

cuda.bindings.runtime.cudaStreamAddCallback(stream, callback, userData, unsigned int flags)¶

Add a callback to a compute stream.

Adds a callback to be called on the host after all currently enqueued items in the stream have completed. For each cudaStreamAddCallback call, a callback will be executed exactly once. The callback will block later work in the stream until it is finished.

The callback may be passed cudaSuccess or an error code. In the event of a device error, all subsequently executed callbacks will receive an appropriate cudaError_t.

Callbacks must not make any CUDA API calls. Attempting to use CUDA APIs may result in cudaErrorNotPermitted. Callbacks must not perform any synchronization that may depend on outstanding device work or other callbacks that are not mandated to run earlier. Callbacks without a mandated order (in independent streams) execute in undefined order and may be serialized.

For the purposes of Unified Memory, callback execution makes a number of guarantees:

The callback stream is considered idle for the duration of the callback. Thus, for example, a callback may always use memory attached to the callback stream.
The start of execution of a callback has the same effect as synchronizing an event recorded in the same stream immediately prior to the callback. It thus synchronizes streams which have been “joined” prior to the callback.
Adding device work to any stream does not have the effect of making the stream active until all preceding callbacks have executed. Thus, for example, a callback might use global attached memory even if work has been added to another stream, if it has been properly ordered with an event.
Completion of a callback does not cause a stream to become active except as described above. The callback stream will remain idle if no device work follows the callback, and will remain idle across consecutive callbacks without device work in between. Thus, for example, stream synchronization can be done by signaling from a callback at the end of the stream.

Parameters:

stream (CUstream or cudaStream_t) – Stream to add callback to
callback (cudaStreamCallback_t) – The function to call once preceding stream operations are complete
userData (Any) – User specified data to be passed to the callback function
flags (unsigned int) – Reserved for future use, must be 0

Returns:

Return type:

cudaSuccess, cudaErrorNotReady, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

Notes

This function is slated for eventual deprecation and removal. If you do not require the callback to execute in case of a device error, consider using cudaLaunchHostFunc. Additionally, this function is not supported with cudaStreamBeginCapture and cudaStreamEndCapture, unlike cudaLaunchHostFunc.

cuda.bindings.runtime.cudaStreamSynchronize(stream)¶

Waits for stream tasks to complete.

Blocks until stream has completed all operations. If the cudaDeviceScheduleBlockingSync flag was set for this device, the host thread will block until the stream is finished with all of its tasks.

Parameters:: stream (CUstream or cudaStream_t) – Stream identifier
Returns:: cudaSuccess, cudaErrorInvalidResourceHandle
Return type:: cudaError_t

cuda.bindings.runtime.cudaStreamQuery(stream)¶

Queries an asynchronous stream for completion status.

Returns cudaSuccess if all operations in stream have completed, or cudaErrorNotReady if not.

For the purposes of Unified Memory, a return value of cudaSuccess is equivalent to having called cudaStreamSynchronize().

Parameters:: stream (CUstream or cudaStream_t) – Stream identifier
Returns:: cudaSuccess, cudaErrorNotReady, cudaErrorInvalidResourceHandle
Return type:: cudaError_t

cuda.bindings.runtime.cudaStreamAttachMemAsync(stream, devPtr, size_t length, unsigned int flags)¶

Attach memory to a stream asynchronously.

Enqueues an operation in stream to specify stream association of length bytes of memory starting from devPtr. This function is a stream-ordered operation, meaning that it is dependent on, and will only take effect when, previous work in stream has completed. Any previous association is automatically replaced.

devPtr must point to an one of the following types of memories:

managed memory declared using the managed keyword or allocated with cudaMallocManaged.
a valid host-accessible region of system-allocated pageable memory. This type of memory may only be specified if the device associated with the stream reports a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess.

For managed allocations, length must be either zero or the entire allocation’s size. Both indicate that the entire allocation’s stream association is being changed. Currently, it is not possible to change stream association for a portion of a managed allocation.

For pageable allocations, length must be non-zero.

The stream association is specified using flags which must be one of cudaMemAttachGlobal, cudaMemAttachHost or cudaMemAttachSingle. The default value for flags is cudaMemAttachSingle If the cudaMemAttachGlobal flag is specified, the memory can be accessed by any stream on any device. If the cudaMemAttachHost flag is specified, the program makes a guarantee that it won’t access the memory on the device from any stream on a device that has a zero value for the device attribute cudaDevAttrConcurrentManagedAccess. If the cudaMemAttachSingle flag is specified and stream is associated with a device that has a zero value for the device attribute cudaDevAttrConcurrentManagedAccess, the program makes a guarantee that it will only access the memory on the device from stream. It is illegal to attach singly to the NULL stream, because the NULL stream is a virtual global stream and not a specific stream. An error will be returned in this case.

When memory is associated with a single stream, the Unified Memory system will allow CPU access to this memory region so long as all operations in stream have completed, regardless of whether other streams are active. In effect, this constrains exclusive ownership of the managed memory region by an active GPU to per-stream activity instead of whole-GPU activity.

Accessing memory on the device from streams that are not associated with it will produce undefined results. No error checking is performed by the Unified Memory system to ensure that kernels launched into other streams do not access this region.

It is a program’s responsibility to order calls to cudaStreamAttachMemAsync via events, synchronization or other means to ensure legal access to memory at all times. Data visibility and coherency will be changed appropriately for all kernels which follow a stream-association change.

If stream is destroyed while data is associated with it, the association is removed and the association reverts to the default visibility of the allocation as specified at cudaMallocManaged. For managed variables, the default association is always cudaMemAttachGlobal. Note that destroying a stream is an asynchronous operation, and as a result, the change to default association won’t happen until all work in the stream has completed.

Parameters:

stream (CUstream or cudaStream_t) – Stream in which to enqueue the attach operation
devPtr (Any) – Pointer to memory (must be a pointer to managed memory or to a valid host-accessible region of system-allocated memory)
length (size_t) – Length of memory (defaults to zero)
flags (unsigned int) – Must be one of cudaMemAttachGlobal, cudaMemAttachHost or cudaMemAttachSingle (defaults to cudaMemAttachSingle)

Returns:

Return type:

cuda.bindings.runtime.cudaStreamBeginCapture(stream, mode: cudaStreamCaptureMode)¶

Begins graph capture on a stream.

Begin graph capture on stream. When a stream is in capture mode, all operations pushed into the stream will not be executed, but will instead be captured into a graph, which will be returned via cudaStreamEndCapture. Capture may not be initiated if stream is cudaStreamLegacy. Capture must be ended on the same stream in which it was initiated, and it may only be initiated if the stream is not already in capture mode. The capture mode may be queried via cudaStreamIsCapturing. A unique id representing the capture sequence may be queried via cudaStreamGetCaptureInfo.

If mode is not cudaStreamCaptureModeRelaxed, cudaStreamEndCapture must be called on this stream from the same thread.

Parameters:

stream (CUstream or cudaStream_t) – Stream in which to initiate capture
mode (cudaStreamCaptureMode) – Controls the interaction of this capture sequence with other API calls that are potentially unsafe. For more details see cudaThreadExchangeStreamCaptureMode.

Returns:

Return type:

Notes

Kernels captured using this API must not use texture and surface references. Reading or writing through any texture or surface reference is undefined behavior. This restriction does not apply to texture and surface objects.

cuda.bindings.runtime.cudaStreamBeginCaptureToGraph(stream, graph, dependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], dependencyData: Optional[Tuple[cudaGraphEdgeData] | List[cudaGraphEdgeData]], size_t numDependencies, mode: cudaStreamCaptureMode)¶

Begins graph capture on a stream to an existing graph.

Begin graph capture on stream. When a stream is in capture mode, all operations pushed into the stream will not be executed, but will instead be captured into graph, which will be returned via cudaStreamEndCapture.

Capture may not be initiated if stream is cudaStreamLegacy. Capture must be ended on the same stream in which it was initiated, and it may only be initiated if the stream is not already in capture mode. The capture mode may be queried via cudaStreamIsCapturing. A unique id representing the capture sequence may be queried via cudaStreamGetCaptureInfo.

If mode is not cudaStreamCaptureModeRelaxed, cudaStreamEndCapture must be called on this stream from the same thread.

Parameters:

stream (CUstream or cudaStream_t) – Stream in which to initiate capture.
graph (CUgraph or cudaGraph_t) – Graph to capture into.
dependencies (List[cudaGraphNode_t]) – Dependencies of the first node captured in the stream. Can be NULL if numDependencies is 0.
dependencyData (List[cudaGraphEdgeData]) – Optional array of data associated with each dependency.
numDependencies (size_t) – Number of dependencies.
mode (cudaStreamCaptureMode) – Controls the interaction of this capture sequence with other API calls that are potentially unsafe. For more details see cudaThreadExchangeStreamCaptureMode.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorIllegalState

Notes

cuda.bindings.runtime.cudaThreadExchangeStreamCaptureMode(mode: cudaStreamCaptureMode)¶

Swaps the stream capture interaction mode for a thread.

Sets the calling thread’s stream capture interaction mode to the value contained in *mode, and overwrites *mode with the previous mode for the thread. To facilitate deterministic behavior across function or module boundaries, callers are encouraged to use this API in a push-pop fashion:

View CUDA Toolkit Documentation for a C++ code example

During stream capture (see cudaStreamBeginCapture), some actions, such as a call to cudaMalloc, may be unsafe. In the case of cudaMalloc, the operation is not enqueued asynchronously to a stream, and is not observed by stream capture. Therefore, if the sequence of operations captured via cudaStreamBeginCapture depended on the allocation being replayed whenever the graph is launched, the captured graph would be invalid.

Therefore, stream capture places restrictions on API calls that can be made within or concurrently to a cudaStreamBeginCapture-cudaStreamEndCapture sequence. This behavior can be controlled via this API and flags to cudaStreamBeginCapture.

A thread’s mode is one of the following:

cudaStreamCaptureModeGlobal: This is the default mode. If the local thread has an ongoing capture sequence that was not initiated with cudaStreamCaptureModeRelaxed at cuStreamBeginCapture, or if any other thread has a concurrent capture sequence initiated with cudaStreamCaptureModeGlobal, this thread is prohibited from potentially unsafe API calls.
cudaStreamCaptureModeThreadLocal: If the local thread has an ongoing capture sequence not initiated with cudaStreamCaptureModeRelaxed, it is prohibited from potentially unsafe API calls. Concurrent capture sequences in other threads are ignored.
cudaStreamCaptureModeRelaxed: The local thread is not prohibited from potentially unsafe API calls. Note that the thread is still prohibited from API calls which necessarily conflict with stream capture, for example, attempting cudaEventQuery on an event that was last recorded inside a capture sequence.

Parameters:

mode (cudaStreamCaptureMode) – Pointer to mode value to swap with the current mode

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
mode (cudaStreamCaptureMode) – Pointer to mode value to swap with the current mode

See also

cudaStreamBeginCapture

cuda.bindings.runtime.cudaStreamEndCapture(stream)¶

Ends capture on a stream, returning the captured graph.

End capture on stream, returning the captured graph via pGraph. Capture must have been initiated on stream via a call to cudaStreamBeginCapture. If capture was invalidated, due to a violation of the rules of stream capture, then a NULL graph will be returned.

If the mode argument to cudaStreamBeginCapture was not cudaStreamCaptureModeRelaxed, this call must be from the same thread as cudaStreamBeginCapture.

Parameters:

stream (CUstream or cudaStream_t) – Stream to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorStreamCaptureWrongThread
pGraph (cudaGraph_t) – The captured graph

cuda.bindings.runtime.cudaStreamIsCapturing(stream)¶

Returns a stream’s capture status.

Return the capture status of stream via pCaptureStatus. After a successful call, *pCaptureStatus will contain one of the following:

cudaStreamCaptureStatusNone: The stream is not capturing.
cudaStreamCaptureStatusActive: The stream is capturing.
cudaStreamCaptureStatusInvalidated: The stream was capturing but an error has invalidated the capture sequence. The capture sequence must be terminated with cudaStreamEndCapture on the stream where it was initiated in order to continue using stream.

Note that, if this is called on cudaStreamLegacy (the “null stream”) while a blocking stream on the same device is capturing, it will return cudaErrorStreamCaptureImplicit and *pCaptureStatus is unspecified after the call. The blocking stream capture is not invalidated.

When a blocking stream is capturing, the legacy stream is in an unusable state until the blocking stream capture is terminated. The legacy stream is not supported for stream capture, but attempted use would have an implicit dependency on the capturing stream(s).

Parameters:

stream (CUstream or cudaStream_t) – Stream to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorStreamCaptureImplicit
pCaptureStatus (cudaStreamCaptureStatus) – Returns the stream’s capture status

cuda.bindings.runtime.cudaStreamGetCaptureInfo(stream)¶

Query a stream’s capture state.

Query stream state related to stream capture.

If called on cudaStreamLegacy (the “null stream”) while a stream not created with cudaStreamNonBlocking is capturing, returns cudaErrorStreamCaptureImplicit.

Valid data (other than capture status) is returned only if both of the following are true:

the call returns cudaSuccess
the returned capture status is cudaStreamCaptureStatusActive

Parameters:

stream (CUstream or cudaStream_t) – The stream to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorStreamCaptureImplicit
captureStatus_out (cudaStreamCaptureStatus) – Location to return the capture status of the stream; required
id_out (unsigned long long) – Optional location to return an id for the capture sequence, which is unique over the lifetime of the process
graph_out (cudaGraph_t) – Optional location to return the graph being captured into. All operations other than destroy and node removal are permitted on the graph while the capture sequence is in progress. This API does not transfer ownership of the graph, which is transferred or destroyed at cudaStreamEndCapture. Note that the graph handle may be invalidated before end of capture for certain errors. Nodes that are or become unreachable from the original stream at cudaStreamEndCapture due to direct actions on the graph do not trigger cudaErrorStreamCaptureUnjoined.
dependencies_out (List[cudaGraphNode_t]) – Optional location to store a pointer to an array of nodes. The next node to be captured in the stream will depend on this set of nodes, absent operations such as event wait which modify this set. The array pointer is valid until the next API call which operates on the stream or until the capture is terminated. The node handles may be copied out and are valid until they or the graph is destroyed. The driver-owned array may also be passed directly to APIs that operate on the graph (not the stream) without copying.
numDependencies_out (int) – Optional location to store the size of the array returned in dependencies_out.

cuda.bindings.runtime.cudaStreamGetCaptureInfo_v3(stream)¶

Query a stream’s capture state (12.3+)

Query stream state related to stream capture.

If called on cudaStreamLegacy (the “null stream”) while a stream not created with cudaStreamNonBlocking is capturing, returns cudaErrorStreamCaptureImplicit.

Valid data (other than capture status) is returned only if both of the following are true:

the call returns cudaSuccess
the returned capture status is cudaStreamCaptureStatusActive

If edgeData_out is non-NULL then dependencies_out must be as well. If dependencies_out is non-NULL and edgeData_out is NULL, but there is non-zero edge data for one or more of the current stream dependencies, the call will return cudaErrorLossyQuery.

Parameters:

stream (CUstream or cudaStream_t) – The stream to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorStreamCaptureImplicit, cudaErrorLossyQuery
captureStatus_out (cudaStreamCaptureStatus) – Location to return the capture status of the stream; required
id_out (unsigned long long) – Optional location to return an id for the capture sequence, which is unique over the lifetime of the process
graph_out (cudaGraph_t) – Optional location to return the graph being captured into. All operations other than destroy and node removal are permitted on the graph while the capture sequence is in progress. This API does not transfer ownership of the graph, which is transferred or destroyed at cudaStreamEndCapture. Note that the graph handle may be invalidated before end of capture for certain errors. Nodes that are or become unreachable from the original stream at cudaStreamEndCapture due to direct actions on the graph do not trigger cudaErrorStreamCaptureUnjoined.
dependencies_out (List[cudaGraphNode_t]) – Optional location to store a pointer to an array of nodes. The next node to be captured in the stream will depend on this set of nodes, absent operations such as event wait which modify this set. The array pointer is valid until the next API call which operates on the stream or until the capture is terminated. The node handles may be copied out and are valid until they or the graph is destroyed. The driver-owned array may also be passed directly to APIs that operate on the graph (not the stream) without copying.
edgeData_out (List[cudaGraphEdgeData]) – Optional location to store a pointer to an array of graph edge data. This array parallels dependencies_out; the next node to be added has an edge to dependencies_out`[i] with annotation `edgeData_out`[i] for each `i. The array pointer is valid until the next API call which operates on the stream or until the capture is terminated.
numDependencies_out (int) – Optional location to store the size of the array returned in dependencies_out.

cuda.bindings.runtime.cudaStreamUpdateCaptureDependencies(stream, dependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, unsigned int flags)¶

Update the set of dependencies in a capturing stream (11.3+)

Modifies the dependency set of a capturing stream. The dependency set is the set of nodes that the next captured node in the stream will depend on.

Valid flags are cudaStreamAddCaptureDependencies and cudaStreamSetCaptureDependencies. These control whether the set passed to the API is added to the existing set or replaces it. A flags value of 0 defaults to cudaStreamAddCaptureDependencies.

Nodes that are removed from the dependency set via this API do not result in cudaErrorStreamCaptureUnjoined if they are unreachable from the stream at cudaStreamEndCapture.

Returns cudaErrorIllegalState if the stream is not capturing.

This API is new in CUDA 11.3. Developers requiring compatibility across minor versions of the CUDA driver to 11.0 should not use this API or provide a fallback.

Parameters:

stream (CUstream or cudaStream_t) – The stream to update
dependencies (List[cudaGraphNode_t]) – The set of dependencies to add
numDependencies (size_t) – The size of the dependencies array
flags (unsigned int) – See above

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorIllegalState

cuda.bindings.runtime.cudaStreamUpdateCaptureDependencies_v2(stream, dependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], dependencyData: Optional[Tuple[cudaGraphEdgeData] | List[cudaGraphEdgeData]], size_t numDependencies, unsigned int flags)¶

Update the set of dependencies in a capturing stream (12.3+)

Modifies the dependency set of a capturing stream. The dependency set is the set of nodes that the next captured node in the stream will depend on.

Nodes that are removed from the dependency set via this API do not result in cudaErrorStreamCaptureUnjoined if they are unreachable from the stream at cudaStreamEndCapture.

Returns cudaErrorIllegalState if the stream is not capturing.

Parameters:

stream (CUstream or cudaStream_t) – The stream to update
dependencies (List[cudaGraphNode_t]) – The set of dependencies to add
dependencyData (List[cudaGraphEdgeData]) – Optional array of data associated with each dependency.
numDependencies (size_t) – The size of the dependencies array
flags (unsigned int) – See above

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure

Event Management¶

This section describes the event management functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaEventCreate()¶

Creates an event object.

Creates an event object for the current device using cudaEventDefault.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorLaunchFailure, cudaErrorMemoryAllocation
event (cudaEvent_t) – Newly created event

cuda.bindings.runtime.cudaEventCreateWithFlags(unsigned int flags)¶

Creates an event object with the specified flags.

Creates an event object for the current device with the specified flags. Valid flags include:

cudaEventDefault: Default event creation flag.
cudaEventBlockingSync: Specifies that event should use blocking synchronization. A host thread that uses cudaEventSynchronize() to wait on an event created with this flag will block until the event actually completes.
cudaEventDisableTiming: Specifies that the created event does not need to record timing data. Events created with this flag specified and the cudaEventBlockingSync flag not specified will provide the best performance when used with cudaStreamWaitEvent() and cudaEventQuery().
cudaEventInterprocess: Specifies that the created event may be used as an interprocess event by cudaIpcGetEventHandle(). cudaEventInterprocess must be specified along with cudaEventDisableTiming.

Parameters:

flags (unsigned int) – Flags for new event

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorLaunchFailure, cudaErrorMemoryAllocation
event (cudaEvent_t) – Newly created event

cuda.bindings.runtime.cudaEventRecord(event, stream)¶

Records an event.

Captures in event the contents of stream at the time of this call. event and stream must be on the same CUDA context. Calls such as cudaEventQuery() or cudaStreamWaitEvent() will then examine or wait for completion of the work that was captured. Uses of stream after this call do not modify event. See note on default stream behavior for what is captured in the default case.

cudaEventRecord() can be called multiple times on the same event and will overwrite the previously captured state. Other APIs such as cudaStreamWaitEvent() use the most recently captured state at the time of the API call, and are not affected by later calls to cudaEventRecord(). Before the first call to cudaEventRecord(), an event represents an empty set of work, so for example cudaEventQuery() would return cudaSuccess.

Parameters:

event (CUevent or cudaEvent_t) – Event to record
stream (CUstream or cudaStream_t) – Stream in which to record event

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure

cuda.bindings.runtime.cudaEventRecordWithFlags(event, stream, unsigned int flags)¶

Records an event.

cudaEventRecordWithFlags() can be called multiple times on the same event and will overwrite the previously captured state. Other APIs such as cudaStreamWaitEvent() use the most recently captured state at the time of the API call, and are not affected by later calls to cudaEventRecordWithFlags(). Before the first call to cudaEventRecordWithFlags(), an event represents an empty set of work, so for example cudaEventQuery() would return cudaSuccess.

flags include:

cudaEventRecordDefault: Default event creation flag.
cudaEventRecordExternal: Event is captured in the graph as an external event node when performing stream capture.

Parameters:

event (CUevent or cudaEvent_t) – Event to record
stream (CUstream or cudaStream_t) – Stream in which to record event
flags (unsigned int) – Parameters for the operation(See above)

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle

cuda.bindings.runtime.cudaEventQuery(event)¶

Queries an event’s status.

Queries the status of all work currently captured by event. See cudaEventRecord() for details on what is captured by an event.

Returns cudaSuccess if all captured work has been completed, or cudaErrorNotReady if any captured work is incomplete.

For the purposes of Unified Memory, a return value of cudaSuccess is equivalent to having called cudaEventSynchronize().

Parameters:: event (CUevent or cudaEvent_t) – Event to query
Returns:: cudaSuccess, cudaErrorNotReady, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure
Return type:: cudaError_t

cuda.bindings.runtime.cudaEventSynchronize(event)¶

Waits for an event to complete.

Waits until the completion of all work currently captured in event. See cudaEventRecord() for details on what is captured by an event.

Waiting for an event that was created with the cudaEventBlockingSync flag will cause the calling CPU thread to block until the event has been completed by the device. If the cudaEventBlockingSync flag has not been set, then the CPU thread will busy-wait until the event has been completed by the device.

Parameters:: event (CUevent or cudaEvent_t) – Event to wait for
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure
Return type:: cudaError_t

cuda.bindings.runtime.cudaEventDestroy(event)¶

Destroys an event object.

Destroys the event specified by event.

An event may be destroyed before it is complete (i.e., while cudaEventQuery() would return cudaErrorNotReady). In this case, the call does not block on completion of the event, and any associated resources will automatically be released asynchronously at completion.

Parameters:: event (CUevent or cudaEvent_t) – Event to destroy
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure
Return type:: cudaError_t

cuda.bindings.runtime.cudaEventElapsedTime(start, end)¶

Computes the elapsed time between events.

Computes the elapsed time between two events (in milliseconds with a resolution of around 0.5 microseconds).

If either event was last recorded in a non-NULL stream, the resulting time may be greater than expected (even if both used the same stream handle). This happens because the cudaEventRecord() operation takes place asynchronously and there is no guarantee that the measured latency is actually just between the two events. Any number of other different stream operations could execute in between the two measured events, thus altering the timing in a significant way.

If cudaEventRecord() has not been called on either event, then cudaErrorInvalidResourceHandle is returned. If cudaEventRecord() has been called on both events but one or both of them has not yet been completed (that is, cudaEventQuery() would return cudaErrorNotReady on at least one of the events), cudaErrorNotReady is returned. If either event was created with the cudaEventDisableTiming flag, then this function will return cudaErrorInvalidResourceHandle.

Parameters:

start (CUevent or cudaEvent_t) – Starting event
end (CUevent or cudaEvent_t) – Ending event

Returns:

cudaError_t – cudaSuccess, cudaErrorNotReady, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorLaunchFailure, cudaErrorUnknown
ms (float) – Time between start and end in ms

External Resource Interoperability¶

This section describes the external resource interoperability functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaImportExternalMemory(cudaExternalMemoryHandleDesc memHandleDesc: Optional[cudaExternalMemoryHandleDesc])¶

Imports an external memory object.

Imports an externally allocated memory object and returns a handle to that in extMem_out.

The properties of the handle being imported must be described in memHandleDesc. The cudaExternalMemoryHandleDesc structure is defined as follows:

View CUDA Toolkit Documentation for a C++ code example

where type specifies the type of handle being imported. cudaExternalMemoryHandleType is defined as:

View CUDA Toolkit Documentation for a C++ code example

If type is cudaExternalMemoryHandleTypeOpaqueFd, then cudaExternalMemoryHandleDesc::handle::fd must be a valid file descriptor referencing a memory object. Ownership of the file descriptor is transferred to the CUDA driver when the handle is imported successfully. Performing any operations on the file descriptor after it is imported results in undefined behavior.

If type is cudaExternalMemoryHandleTypeOpaqueWin32, then exactly one of cudaExternalMemoryHandleDesc::handle::win32::handle and cudaExternalMemoryHandleDesc::handle::win32::name must not be NULL. If cudaExternalMemoryHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that references a memory object. Ownership of this handle is not transferred to CUDA after the import operation, so the application must release the handle using the appropriate system call. If cudaExternalMemoryHandleDesc::handle::win32::name is not NULL, then it must point to a NULL-terminated array of UTF-16 characters that refers to a memory object.

If type is cudaExternalMemoryHandleTypeOpaqueWin32Kmt, then cudaExternalMemoryHandleDesc::handle::win32::handle must be non-NULL and cudaExternalMemoryHandleDesc::handle::win32::name must be NULL. The handle specified must be a globally shared KMT handle. This handle does not hold a reference to the underlying object, and thus will be invalid when all references to the memory object are destroyed.

If type is cudaExternalMemoryHandleTypeD3D12Heap, then exactly one of cudaExternalMemoryHandleDesc::handle::win32::handle and cudaExternalMemoryHandleDesc::handle::win32::name must not be NULL. If cudaExternalMemoryHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that is returned by ID3D12Device::CreateSharedHandle when referring to a ID3D12Heap object. This handle holds a reference to the underlying object. If cudaExternalMemoryHandleDesc::handle::win32::name is not NULL, then it must point to a NULL-terminated array of UTF-16 characters that refers to a ID3D12Heap object.

If type is cudaExternalMemoryHandleTypeD3D12Resource, then exactly one of cudaExternalMemoryHandleDesc::handle::win32::handle and cudaExternalMemoryHandleDesc::handle::win32::name must not be NULL. If cudaExternalMemoryHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that is returned by ID3D12Device::CreateSharedHandle when referring to a ID3D12Resource object. This handle holds a reference to the underlying object. If cudaExternalMemoryHandleDesc::handle::win32::name is not NULL, then it must point to a NULL-terminated array of UTF-16 characters that refers to a ID3D12Resource object.

If type is cudaExternalMemoryHandleTypeD3D11Resource,then exactly one of cudaExternalMemoryHandleDesc::handle::win32::handle and cudaExternalMemoryHandleDesc::handle::win32::name must not be NULL. If cudaExternalMemoryHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that is returned by IDXGIResource1::CreateSharedHandle when referring to a ID3D11Resource object. If cudaExternalMemoryHandleDesc::handle::win32::name is not NULL, then it must point to a NULL-terminated array of UTF-16 characters that refers to a ID3D11Resource object.

If type is cudaExternalMemoryHandleTypeD3D11ResourceKmt, then cudaExternalMemoryHandleDesc::handle::win32::handle must be non-NULL and cudaExternalMemoryHandleDesc::handle::win32::name must be NULL. The handle specified must be a valid shared KMT handle that is returned by IDXGIResource::GetSharedHandle when referring to a ID3D11Resource object.

If type is cudaExternalMemoryHandleTypeNvSciBuf, then cudaExternalMemoryHandleDesc::handle::nvSciBufObject must be NON-NULL and reference a valid NvSciBuf object. If the NvSciBuf object imported into CUDA is also mapped by other drivers, then the application must use cudaWaitExternalSemaphoresAsync or cudaSignalExternalSemaphoresAsync as approprriate barriers to maintain coherence between CUDA and the other drivers. See cudaExternalSemaphoreWaitSkipNvSciBufMemSync and cudaExternalSemaphoreSignalSkipNvSciBufMemSync for memory synchronization.

The size of the memory object must be specified in size.

Specifying the flag cudaExternalMemoryDedicated in flags indicates that the resource is a dedicated resource. The definition of what a dedicated resource is outside the scope of this extension. This flag must be set if type is one of the following: cudaExternalMemoryHandleTypeD3D12Resource cudaExternalMemoryHandleTypeD3D11Resource cudaExternalMemoryHandleTypeD3D11ResourceKmt

Parameters:

memHandleDesc (cudaExternalMemoryHandleDesc) – Memory import handle descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorOperatingSystem
extMem_out (cudaExternalMemory_t) – Returned handle to an external memory object

Notes

If the Vulkan memory imported into CUDA is mapped on the CPU then the application must use vkInvalidateMappedMemoryRanges/vkFlushMappedMemoryRanges as well as appropriate Vulkan pipeline barriers to maintain coherence between CPU and GPU. For more information on these APIs, please refer to “Synchronization and Cache Control” chapter from Vulkan specification.

cuda.bindings.runtime.cudaExternalMemoryGetMappedBuffer(extMem, cudaExternalMemoryBufferDesc bufferDesc: Optional[cudaExternalMemoryBufferDesc])¶

Maps a buffer onto an imported memory object.

Maps a buffer onto an imported memory object and returns a device pointer in devPtr.

The properties of the buffer being mapped must be described in bufferDesc. The cudaExternalMemoryBufferDesc structure is defined as follows:

View CUDA Toolkit Documentation for a C++ code example

where offset is the offset in the memory object where the buffer’s base address is. size is the size of the buffer. flags must be zero.

The offset and size have to be suitably aligned to match the requirements of the external API. Mapping two buffers whose ranges overlap may or may not result in the same virtual address being returned for the overlapped portion. In such cases, the application must ensure that all accesses to that region from the GPU are volatile. Otherwise writes made via one address are not guaranteed to be visible via the other address, even if they’re issued by the same thread. It is recommended that applications map the combined range instead of mapping separate buffers and then apply the appropriate offsets to the returned pointer to derive the individual buffers.

The returned pointer devPtr must be freed using cudaFree.

Parameters:

extMem (cudaExternalMemory_t) – Handle to external memory object
bufferDesc (cudaExternalMemoryBufferDesc) – Buffer descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
devPtr (Any) – Returned device pointer to buffer

cuda.bindings.runtime.cudaExternalMemoryGetMappedMipmappedArray(extMem, cudaExternalMemoryMipmappedArrayDesc mipmapDesc: Optional[cudaExternalMemoryMipmappedArrayDesc])¶

Maps a CUDA mipmapped array onto an external memory object.

Maps a CUDA mipmapped array onto an external object and returns a handle to it in mipmap.

The properties of the CUDA mipmapped array being mapped must be described in mipmapDesc. The structure cudaExternalMemoryMipmappedArrayDesc is defined as follows:

View CUDA Toolkit Documentation for a C++ code example

where offset is the offset in the memory object where the base level of the mipmap chain is. formatDesc describes the format of the data. extent specifies the dimensions of the base level of the mipmap chain. flags are flags associated with CUDA mipmapped arrays. For further details, please refer to the documentation for cudaMalloc3DArray. Note that if the mipmapped array is bound as a color target in the graphics API, then the flag cudaArrayColorAttachment must be specified in flags. numLevels specifies the total number of levels in the mipmap chain.

The returned CUDA mipmapped array must be freed using cudaFreeMipmappedArray.

Parameters:

extMem (cudaExternalMemory_t) – Handle to external memory object
mipmapDesc (cudaExternalMemoryMipmappedArrayDesc) – CUDA array descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
mipmap (cudaMipmappedArray_t) – Returned CUDA mipmapped array

Notes

If type is cudaExternalMemoryHandleTypeNvSciBuf, then numLevels must not be greater than 1.

cuda.bindings.runtime.cudaDestroyExternalMemory(extMem)¶

Destroys an external memory object.

Destroys the specified external memory object. Any existing buffers and CUDA mipmapped arrays mapped onto this object must no longer be used and must be explicitly freed using cudaFree and cudaFreeMipmappedArray respectively.

Parameters:: extMem (cudaExternalMemory_t) – External memory object to be destroyed
Returns:: cudaSuccess, cudaErrorInvalidResourceHandle
Return type:: cudaError_t

cuda.bindings.runtime.cudaImportExternalSemaphore(cudaExternalSemaphoreHandleDesc semHandleDesc: Optional[cudaExternalSemaphoreHandleDesc])¶

Imports an external semaphore.

Imports an externally allocated synchronization object and returns a handle to that in extSem_out.

The properties of the handle being imported must be described in semHandleDesc. The cudaExternalSemaphoreHandleDesc is defined as follows:

View CUDA Toolkit Documentation for a C++ code example

where type specifies the type of handle being imported. cudaExternalSemaphoreHandleType is defined as:

View CUDA Toolkit Documentation for a C++ code example

If type is cudaExternalSemaphoreHandleTypeOpaqueFd, then cudaExternalSemaphoreHandleDesc::handle::fd must be a valid file descriptor referencing a synchronization object. Ownership of the file descriptor is transferred to the CUDA driver when the handle is imported successfully. Performing any operations on the file descriptor after it is imported results in undefined behavior.

If type is cudaExternalSemaphoreHandleTypeOpaqueWin32, then exactly one of cudaExternalSemaphoreHandleDesc::handle::win32::handle and cudaExternalSemaphoreHandleDesc::handle::win32::name must not be NULL. If cudaExternalSemaphoreHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that references a synchronization object. Ownership of this handle is not transferred to CUDA after the import operation, so the application must release the handle using the appropriate system call. If cudaExternalSemaphoreHandleDesc::handle::win32::name is not NULL, then it must name a valid synchronization object.

If type is cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt, then cudaExternalSemaphoreHandleDesc::handle::win32::handle must be non-NULL and cudaExternalSemaphoreHandleDesc::handle::win32::name must be NULL. The handle specified must be a globally shared KMT handle. This handle does not hold a reference to the underlying object, and thus will be invalid when all references to the synchronization object are destroyed.

If type is cudaExternalSemaphoreHandleTypeD3D12Fence, then exactly one of cudaExternalSemaphoreHandleDesc::handle::win32::handle and cudaExternalSemaphoreHandleDesc::handle::win32::name must not be NULL. If cudaExternalSemaphoreHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that is returned by ID3D12Device::CreateSharedHandle when referring to a ID3D12Fence object. This handle holds a reference to the underlying object. If cudaExternalSemaphoreHandleDesc::handle::win32::name is not NULL, then it must name a valid synchronization object that refers to a valid ID3D12Fence object.

If type is cudaExternalSemaphoreHandleTypeD3D11Fence, then exactly one of cudaExternalSemaphoreHandleDesc::handle::win32::handle and cudaExternalSemaphoreHandleDesc::handle::win32::name must not be NULL. If cudaExternalSemaphoreHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that is returned by ID3D11Fence::CreateSharedHandle. If cudaExternalSemaphoreHandleDesc::handle::win32::name is not NULL, then it must name a valid synchronization object that refers to a valid ID3D11Fence object.

If type is cudaExternalSemaphoreHandleTypeNvSciSync, then cudaExternalSemaphoreHandleDesc::handle::nvSciSyncObj represents a valid NvSciSyncObj.

cudaExternalSemaphoreHandleTypeKeyedMutex, then exactly one of cudaExternalSemaphoreHandleDesc::handle::win32::handle and cudaExternalSemaphoreHandleDesc::handle::win32::name must not be NULL. If cudaExternalSemaphoreHandleDesc::handle::win32::handle is not NULL, then it represent a valid shared NT handle that is returned by IDXGIResource1::CreateSharedHandle when referring to a IDXGIKeyedMutex object.

If type is cudaExternalSemaphoreHandleTypeKeyedMutexKmt, then cudaExternalSemaphoreHandleDesc::handle::win32::handle must be non-NULL and cudaExternalSemaphoreHandleDesc::handle::win32::name must be NULL. The handle specified must represent a valid KMT handle that is returned by IDXGIResource::GetSharedHandle when referring to a IDXGIKeyedMutex object.

If type is cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd, then cudaExternalSemaphoreHandleDesc::handle::fd must be a valid file descriptor referencing a synchronization object. Ownership of the file descriptor is transferred to the CUDA driver when the handle is imported successfully. Performing any operations on the file descriptor after it is imported results in undefined behavior.

If type is cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32, then exactly one of cudaExternalSemaphoreHandleDesc::handle::win32::handle and cudaExternalSemaphoreHandleDesc::handle::win32::name must not be NULL. If cudaExternalSemaphoreHandleDesc::handle::win32::handle is not NULL, then it must represent a valid shared NT handle that references a synchronization object. Ownership of this handle is not transferred to CUDA after the import operation, so the application must release the handle using the appropriate system call. If cudaExternalSemaphoreHandleDesc::handle::win32::name is not NULL, then it must name a valid synchronization object.

Parameters:

semHandleDesc (cudaExternalSemaphoreHandleDesc) – Semaphore import handle descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorOperatingSystem
extSem_out (cudaExternalSemaphore_t) – Returned handle to an external semaphore

cuda.bindings.runtime.cudaSignalExternalSemaphoresAsync(extSemArray: Optional[Tuple[cudaExternalSemaphore_t] | List[cudaExternalSemaphore_t]], paramsArray: Optional[Tuple[cudaExternalSemaphoreSignalParams] | List[cudaExternalSemaphoreSignalParams]], unsigned int numExtSems, stream)¶

Signals a set of external semaphore objects.

Enqueues a signal operation on a set of externally allocated semaphore object in the specified stream. The operations will be executed when all prior operations in the stream complete.

The exact semantics of signaling a semaphore depends on the type of the object.

If the semaphore object is any one of the following types: cudaExternalSemaphoreHandleTypeD3D12Fence, cudaExternalSemaphoreHandleTypeD3D11Fence, cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd, cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32 then the semaphore will be set to the value specified in cudaExternalSemaphoreSignalParams::params::fence::value.

If the semaphore object is of the type cudaExternalSemaphoreHandleTypeNvSciSync this API sets cudaExternalSemaphoreSignalParams::params::nvSciSync::fence to a value that can be used by subsequent waiters of the same NvSciSync object to order operations with those currently submitted in stream. Such an update will overwrite previous contents of cudaExternalSemaphoreSignalParams::params::nvSciSync::fence. By default, signaling such an external semaphore object causes appropriate memory synchronization operations to be performed over all the external memory objects that are imported as cudaExternalMemoryHandleTypeNvSciBuf. This ensures that any subsequent accesses made by other importers of the same set of NvSciBuf memory object(s) are coherent. These operations can be skipped by specifying the flag cudaExternalSemaphoreSignalSkipNvSciBufMemSync, which can be used as a performance optimization when data coherency is not required. But specifying this flag in scenarios where data coherency is required results in undefined behavior. Also, for semaphore object of the type cudaExternalSemaphoreHandleTypeNvSciSync, if the NvSciSyncAttrList used to create the NvSciSyncObj had not set the flags in cudaDeviceGetNvSciSyncAttributes to cudaNvSciSyncAttrSignal, this API will return cudaErrorNotSupported.

cudaExternalSemaphoreSignalParams::params::nvSciSync::fence associated with semaphore object of the type cudaExternalSemaphoreHandleTypeNvSciSync can be deterministic. For this the NvSciSyncAttrList used to create the semaphore object must have value of NvSciSyncAttrKey_RequireDeterministicFences key set to true. Deterministic fences allow users to enqueue a wait over the semaphore object even before corresponding signal is enqueued. For such a semaphore object, CUDA guarantees that each signal operation will increment the fence value by ‘1’. Users are expected to track count of signals enqueued on the semaphore object and insert waits accordingly. When such a semaphore object is signaled from multiple streams, due to concurrent stream execution, it is possible that the order in which the semaphore gets signaled is indeterministic. This could lead to waiters of the semaphore getting unblocked incorrectly. Users are expected to handle such situations, either by not using the same semaphore object with deterministic fence support enabled in different streams or by adding explicit dependency amongst such streams so that the semaphore is signaled in order.

Parameters:

extSemArray (List[cudaExternalSemaphore_t]) – Set of external semaphores to be signaled
paramsArray (List[cudaExternalSemaphoreSignalParams]) – Array of semaphore parameters
numExtSems (unsigned int) – Number of semaphores to signal
stream (CUstream or cudaStream_t) – Stream to enqueue the signal operations in

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle cudaErrorTimeout

cuda.bindings.runtime.cudaWaitExternalSemaphoresAsync(extSemArray: Optional[Tuple[cudaExternalSemaphore_t] | List[cudaExternalSemaphore_t]], paramsArray: Optional[Tuple[cudaExternalSemaphoreWaitParams] | List[cudaExternalSemaphoreWaitParams]], unsigned int numExtSems, stream)¶

Waits on a set of external semaphore objects.

Enqueues a wait operation on a set of externally allocated semaphore object in the specified stream. The operations will be executed when all prior operations in the stream complete.

The exact semantics of waiting on a semaphore depends on the type of the object.

If the semaphore object is any one of the following types: cudaExternalSemaphoreHandleTypeOpaqueFd, cudaExternalSemaphoreHandleTypeOpaqueWin32, cudaExternalSemaphoreHandleTypeOpaqueWin32Kmt then waiting on the semaphore will wait until the semaphore reaches the signaled state. The semaphore will then be reset to the unsignaled state. Therefore for every signal operation, there can only be one wait operation.

If the semaphore object is any one of the following types: cudaExternalSemaphoreHandleTypeD3D12Fence, cudaExternalSemaphoreHandleTypeD3D11Fence, cudaExternalSemaphoreHandleTypeTimelineSemaphoreFd, cudaExternalSemaphoreHandleTypeTimelineSemaphoreWin32 then waiting on the semaphore will wait until the value of the semaphore is greater than or equal to cudaExternalSemaphoreWaitParams::params::fence::value.

If the semaphore object is of the type cudaExternalSemaphoreHandleTypeNvSciSync then, waiting on the semaphore will wait until the cudaExternalSemaphoreSignalParams::params::nvSciSync::fence is signaled by the signaler of the NvSciSyncObj that was associated with this semaphore object. By default, waiting on such an external semaphore object causes appropriate memory synchronization operations to be performed over all external memory objects that are imported as cudaExternalMemoryHandleTypeNvSciBuf. This ensures that any subsequent accesses made by other importers of the same set of NvSciBuf memory object(s) are coherent. These operations can be skipped by specifying the flag cudaExternalSemaphoreWaitSkipNvSciBufMemSync, which can be used as a performance optimization when data coherency is not required. But specifying this flag in scenarios where data coherency is required results in undefined behavior. Also, for semaphore object of the type cudaExternalSemaphoreHandleTypeNvSciSync, if the NvSciSyncAttrList used to create the NvSciSyncObj had not set the flags in cudaDeviceGetNvSciSyncAttributes to cudaNvSciSyncAttrWait, this API will return cudaErrorNotSupported.

If the semaphore object is any one of the following types: cudaExternalSemaphoreHandleTypeKeyedMutex, cudaExternalSemaphoreHandleTypeKeyedMutexKmt, then the keyed mutex will be acquired when it is released with the key specified in cudaExternalSemaphoreSignalParams::params::keyedmutex::key or until the timeout specified by cudaExternalSemaphoreSignalParams::params::keyedmutex::timeoutMs has lapsed. The timeout interval can either be a finite value specified in milliseconds or an infinite value. In case an infinite value is specified the timeout never elapses. The windows INFINITE macro must be used to specify infinite timeout

Parameters:

extSemArray (List[cudaExternalSemaphore_t]) – External semaphores to be waited on
paramsArray (List[cudaExternalSemaphoreWaitParams]) – Array of semaphore parameters
numExtSems (unsigned int) – Number of semaphores to wait on
stream (CUstream or cudaStream_t) – Stream to enqueue the wait operations in

Returns:

Return type:

cuda.bindings.runtime.cudaDestroyExternalSemaphore(extSem)¶

Destroys an external semaphore.

Destroys an external semaphore object and releases any references to the underlying resource. Any outstanding signals or waits must have completed before the semaphore is destroyed.

Parameters:: extSem (cudaExternalSemaphore_t) – External semaphore to be destroyed
Returns:: cudaSuccess, cudaErrorInvalidResourceHandle
Return type:: cudaError_t

Execution Control¶

This section describes the execution control functions of the CUDA runtime application programming interface.

Some functions have overloaded C++ API template versions documented separately in the C++ API Routines module.

cuda.bindings.runtime.cudaFuncSetCacheConfig(func, cacheConfig: cudaFuncCache)¶

Sets the preferred cache configuration for a device function.

On devices where the L1 cache and shared memory use the same hardware resources, this sets through cacheConfig the preferred cache configuration for the function specified via func. This is only a preference. The runtime will use the requested configuration if possible, but it is free to choose a different configuration if required to execute func.

func is a device function symbol and must be declared as a None function. If the specified function does not exist, then cudaErrorInvalidDeviceFunction is returned. For templated functions, pass the function symbol as follows: func_name<template_arg_0,…,template_arg_N>

This setting does nothing on devices where the size of the L1 cache and shared memory are fixed.

Launching a kernel with a different preference than the most recent preference setting may insert a device-side synchronization point.

The supported cache configurations are:

cudaFuncCachePreferNone: no preference for shared memory or L1 (default)
cudaFuncCachePreferShared: prefer larger shared memory and smaller L1 cache
cudaFuncCachePreferL1: prefer larger L1 cache and smaller shared memory
cudaFuncCachePreferEqual: prefer equal size L1 cache and shared memory

Parameters:

func (Any) – Device function symbol
cacheConfig (cudaFuncCache) – Requested cache configuration

Returns:

cudaSuccess, :py:obj:`~.cudaErrorInvalidDeviceFunction`2

Return type:

cudaSuccess, cudaErrorInvalidDeviceFunction, cudaErrorInvalidValue

See also

cudaFuncSetCacheConfig (C++ API), cudaFuncGetAttributes (C API), cudaLaunchKernel (C API), cuFuncSetCacheConfig

cuda.bindings.runtime.cudaFuncGetAttributes(func)¶

Find out attributes for a given function.

This function obtains the attributes of a function specified via func. func is a device function symbol and must be declared as a None function. The fetched attributes are placed in attr. If the specified function does not exist, then cudaErrorInvalidDeviceFunction is returned. For templated functions, pass the function symbol as follows: func_name<template_arg_0,…,template_arg_N>

Note that some function attributes such as maxThreadsPerBlock may vary based on the device that is currently being used.

Parameters:

func (Any) – Device function symbol

Returns:

cudaError_t – cudaSuccess, :py:obj:`~.cudaErrorInvalidDeviceFunction`2
attr (cudaFuncAttributes) – Return pointer to function’s attributes

See also

cudaFuncSetCacheConfig (C API), cudaFuncGetAttributes (C++ API), cudaLaunchKernel (C API), cuFuncGetAttribute

cuda.bindings.runtime.cudaFuncSetAttribute(func, attr: cudaFuncAttribute, int value)¶

Set attributes for a given function.

This function sets the attributes of a function specified via func. The parameter func must be a pointer to a function that executes on the device. The parameter specified by func must be declared as a None function. The enumeration defined by attr is set to the value defined by value. If the specified function does not exist, then cudaErrorInvalidDeviceFunction is returned. If the specified attribute cannot be written, or if the value is incorrect, then cudaErrorInvalidValue is returned.

Valid values for attr are:

cudaFuncAttributeMaxDynamicSharedMemorySize - The requested maximum size in bytes of dynamically-allocated shared memory. The sum of this value and the function attribute sharedSizeBytes cannot exceed the device attribute cudaDevAttrMaxSharedMemoryPerBlockOptin. The maximal size of requestable dynamic shared memory may differ by GPU architecture.
cudaFuncAttributePreferredSharedMemoryCarveout - On devices where the L1 cache and shared memory use the same hardware resources, this sets the shared memory carveout preference, in percent of the total shared memory. See cudaDevAttrMaxSharedMemoryPerMultiprocessor. This is only a hint, and the driver can choose a different ratio if required to execute the function.
cudaFuncAttributeRequiredClusterWidth: The required cluster width in blocks. The width, height, and depth values must either all be 0 or all be positive. The validity of the cluster dimensions is checked at launch time. If the value is set during compile time, it cannot be set at runtime. Setting it at runtime will return cudaErrorNotPermitted.
cudaFuncAttributeRequiredClusterHeight: The required cluster height in blocks. The width, height, and depth values must either all be 0 or all be positive. The validity of the cluster dimensions is checked at launch time. If the value is set during compile time, it cannot be set at runtime. Setting it at runtime will return cudaErrorNotPermitted.
cudaFuncAttributeRequiredClusterDepth: The required cluster depth in blocks. The width, height, and depth values must either all be 0 or all be positive. The validity of the cluster dimensions is checked at launch time. If the value is set during compile time, it cannot be set at runtime. Setting it at runtime will return cudaErrorNotPermitted.
cudaFuncAttributeNonPortableClusterSizeAllowed: Indicates whether the function can be launched with non-portable cluster size. 1 is allowed, 0 is disallowed.
cudaFuncAttributeClusterSchedulingPolicyPreference: The block scheduling policy of a function. The value type is cudaClusterSchedulingPolicy.

cudaLaunchKernel (C++ API), cudaFuncSetCacheConfig (C++ API), cudaFuncGetAttributes (C API),

Parameters:

func (Any) – Function to get attributes of
attr (cudaFuncAttribute) – Attribute to set
value (int) – Value to set

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorInvalidValue, cudaErrorNotSupported

cuda.bindings.runtime.cudaLaunchHostFunc(stream, fn, userData)¶

Enqueues a host function call in a stream.

Enqueues a host function to run in a stream. The function will be called after currently enqueued work and will block work added after it.

The host function must not make any CUDA API calls. Attempting to use a CUDA API may result in cudaErrorNotPermitted, but this is not required. The host function must not perform any synchronization that may depend on outstanding CUDA work not mandated to run earlier. Host functions without a mandated order (such as in independent streams) execute in undefined order and may be serialized.

For the purposes of Unified Memory, execution makes a number of guarantees:

The stream is considered idle for the duration of the function’s execution. Thus, for example, the function may always use memory attached to the stream it was enqueued in.
The start of execution of the function has the same effect as synchronizing an event recorded in the same stream immediately prior to the function. It thus synchronizes streams which have been “joined” prior to the function.
Adding device work to any stream does not have the effect of making the stream active until all preceding host functions and stream callbacks have executed. Thus, for example, a function might use global attached memory even if work has been added to another stream, if the work has been ordered behind the function call with an event.
Completion of the function does not cause a stream to become active except as described above. The stream will remain idle if no device work follows the function, and will remain idle across consecutive host functions or stream callbacks without device work in between. Thus, for example, stream synchronization can be done by signaling from a host function at the end of the stream.

Note that, in constrast to cuStreamAddCallback, the function will not be called in the event of an error in the CUDA context.

Parameters:

hStream (CUstream or cudaStream_t) – Stream to enqueue function call in
fn (cudaHostFn_t) – The function to call once preceding stream operations are complete
userData (Any) – User-specified data to be passed to the function

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation, cudaErrorHostMemoryAlreadyRegistered, cudaErrorNotSupported

Occupancy¶

This section describes the occupancy calculation functions of the CUDA runtime application programming interface.

Besides the occupancy calculator functions (cudaOccupancyMaxActiveBlocksPerMultiprocessor and cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags), there are also C++ only occupancy-based launch configuration functions documented in C++ API Routines module.

See cudaOccupancyMaxPotentialBlockSize (C++ API), cudaOccupancyMaxPotentialBlockSize (C++ API), cudaOccupancyMaxPotentialBlockSizeVariableSMem (C++ API), cudaOccupancyMaxPotentialBlockSizeVariableSMem (C++ API) cudaOccupancyAvailableDynamicSMemPerBlock (C++ API),

cuda.bindings.runtime.cudaOccupancyMaxActiveBlocksPerMultiprocessor(func, int blockSize, size_t dynamicSMemSize)¶

Returns occupancy for a device function.

Returns in *numBlocks the maximum number of active blocks per streaming multiprocessor for the device function.

Parameters:

func (Any) – Kernel function for which occupancy is calculated
blockSize (int) – Block size the kernel is intended to be launched with
dynamicSMemSize (size_t) – Per-block dynamic shared memory usage intended, in bytes

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidDeviceFunction, cudaErrorInvalidValue, cudaErrorUnknown,
numBlocks (int) – Returned occupancy

See also

cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags, cudaOccupancyMaxPotentialBlockSize, cudaOccupancyMaxPotentialBlockSizeWithFlags, cudaOccupancyMaxPotentialBlockSizeVariableSMem, cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags, cudaOccupancyAvailableDynamicSMemPerBlock, cuOccupancyMaxActiveBlocksPerMultiprocessor

cuda.bindings.runtime.cudaOccupancyAvailableDynamicSMemPerBlock(func, int numBlocks, int blockSize)¶

Returns dynamic shared memory available per block when launching numBlocks blocks on SM.

Returns in *dynamicSmemSize the maximum size of dynamic shared memory to allow numBlocks blocks per SM.

Parameters:

func (Any) – Kernel function for which occupancy is calculated
numBlocks (int) – Number of blocks to fit on SM
blockSize (int) – Size of the block

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidDeviceFunction, cudaErrorInvalidValue, cudaErrorUnknown,
dynamicSmemSize (int) – Returned maximum dynamic shared memory

See also

cuda.bindings.runtime.cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(func, int blockSize, size_t dynamicSMemSize, unsigned int flags)¶

Returns occupancy for a device function with the specified flags.

Returns in *numBlocks the maximum number of active blocks per streaming multiprocessor for the device function.

The flags parameter controls how special cases are handled. Valid flags include:

cudaOccupancyDefault: keeps the default behavior as cudaOccupancyMaxActiveBlocksPerMultiprocessor
cudaOccupancyDisableCachingOverride: This flag suppresses the default behavior on platform where global caching affects occupancy. On such platforms, if caching is enabled, but per-block SM resource usage would result in zero occupancy, the occupancy calculator will calculate the occupancy as if caching is disabled. Setting this flag makes the occupancy calculator to return 0 in such cases. More information can be found about this feature in the “Unified L1/Texture Cache” section of the Maxwell tuning guide.

Parameters:

func (Any) – Kernel function for which occupancy is calculated
blockSize (int) – Block size the kernel is intended to be launched with
dynamicSMemSize (size_t) – Per-block dynamic shared memory usage intended, in bytes
flags (unsigned int) – Requested behavior for the occupancy calculator

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidDeviceFunction, cudaErrorInvalidValue, cudaErrorUnknown,
numBlocks (int) – Returned occupancy

See also

cudaOccupancyMaxActiveBlocksPerMultiprocessor, cudaOccupancyMaxPotentialBlockSize, cudaOccupancyMaxPotentialBlockSizeWithFlags, cudaOccupancyMaxPotentialBlockSizeVariableSMem, cudaOccupancyMaxPotentialBlockSizeVariableSMemWithFlags, cudaOccupancyAvailableDynamicSMemPerBlock, cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags

Memory Management¶

This section describes the memory management functions of the CUDA runtime application programming interface.

Some functions have overloaded C++ API template versions documented separately in the C++ API Routines module.

cuda.bindings.runtime.cudaMallocManaged(size_t size, unsigned int flags)¶

Allocates memory that will be automatically managed by the Unified Memory system.

Allocates size bytes of managed memory on the device and returns in *devPtr a pointer to the allocated memory. If the device doesn’t support allocating managed memory, cudaErrorNotSupported is returned. Support for managed memory can be queried using the device attribute cudaDevAttrManagedMemory. The allocated memory is suitably aligned for any kind of variable. The memory is not cleared. If size is 0, cudaMallocManaged returns cudaErrorInvalidValue. The pointer is valid on the CPU and on all GPUs in the system that support managed memory. All accesses to this pointer must obey the Unified Memory programming model.

flags specifies the default stream association for this allocation. flags must be one of cudaMemAttachGlobal or cudaMemAttachHost. The default value for flags is cudaMemAttachGlobal. If cudaMemAttachGlobal is specified, then this memory is accessible from any stream on any device. If cudaMemAttachHost is specified, then the allocation should not be accessed from devices that have a zero value for the device attribute cudaDevAttrConcurrentManagedAccess; an explicit call to cudaStreamAttachMemAsync will be required to enable access on such devices.

If the association is later changed via cudaStreamAttachMemAsync to a single stream, the default association, as specifed during cudaMallocManaged, is restored when that stream is destroyed. For managed variables, the default association is always cudaMemAttachGlobal. Note that destroying a stream is an asynchronous operation, and as a result, the change to default association won’t happen until all work in the stream has completed.

Memory allocated with cudaMallocManaged should be released with cudaFree.

Device memory oversubscription is possible for GPUs that have a non- zero value for the device attribute cudaDevAttrConcurrentManagedAccess. Managed memory on such GPUs may be evicted from device memory to host memory at any time by the Unified Memory driver in order to make room for other allocations.

In a system where all GPUs have a non-zero value for the device attribute cudaDevAttrConcurrentManagedAccess, managed memory may not be populated when this API returns and instead may be populated on access. In such systems, managed memory can migrate to any processor’s memory at any time. The Unified Memory driver will employ heuristics to maintain data locality and prevent excessive page faults to the extent possible. The application can also guide the driver about memory usage patterns via cudaMemAdvise. The application can also explicitly migrate memory to a desired processor’s memory via cudaMemPrefetchAsync.

In a multi-GPU system where all of the GPUs have a zero value for the device attribute cudaDevAttrConcurrentManagedAccess and all the GPUs have peer-to-peer support with each other, the physical storage for managed memory is created on the GPU which is active at the time cudaMallocManaged is called. All other GPUs will reference the data at reduced bandwidth via peer mappings over the PCIe bus. The Unified Memory driver does not migrate memory among such GPUs.

In a multi-GPU system where not all GPUs have peer-to-peer support with each other and where the value of the device attribute cudaDevAttrConcurrentManagedAccess is zero for at least one of those GPUs, the location chosen for physical storage of managed memory is system-dependent.

On Linux, the location chosen will be device memory as long as the current set of active contexts are on devices that either have peer- to-peer support with each other or have a non-zero value for the device attribute cudaDevAttrConcurrentManagedAccess. If there is an active context on a GPU that does not have a non-zero value for that device attribute and it does not have peer-to-peer support with the other devices that have active contexts on them, then the location for physical storage will be ‘zero-copy’ or host memory. Note that this means that managed memory that is located in device memory is migrated to host memory if a new context is created on a GPU that doesn’t have a non-zero value for the device attribute and does not support peer-to-peer with at least one of the other devices that has an active context. This in turn implies that context creation may fail if there is insufficient host memory to migrate all managed allocations.
On Windows, the physical storage is always created in ‘zero-copy’ or host memory. All GPUs will reference the data at reduced bandwidth over the PCIe bus. In these circumstances, use of the environment variable CUDA_VISIBLE_DEVICES is recommended to restrict CUDA to only use those GPUs that have peer-to-peer support. Alternatively, users can also set CUDA_MANAGED_FORCE_DEVICE_ALLOC to a non-zero value to force the driver to always use device memory for physical storage. When this environment variable is set to a non-zero value, all devices used in that process that support managed memory have to be peer-to-peer compatible with each other. The error cudaErrorInvalidDevice will be returned if a device that supports managed memory is used and it is not peer-to-peer compatible with any of the other managed memory supporting devices that were previously used in that process, even if cudaDeviceReset has been called on those devices. These environment variables are described in the CUDA programming guide under the “CUDA environment variables” section.

Parameters:

size (size_t) – Requested allocation size in bytes
flags (unsigned int) – Must be either cudaMemAttachGlobal or cudaMemAttachHost (defaults to cudaMemAttachGlobal)

Returns:

cudaError_t – cudaSuccess, cudaErrorMemoryAllocation, cudaErrorNotSupported, cudaErrorInvalidValue
devPtr (Any) – Pointer to allocated device memory

cuda.bindings.runtime.cudaMalloc(size_t size)¶

Allocate memory on the device.

Allocates size bytes of linear memory on the device and returns in *devPtr a pointer to the allocated memory. The allocated memory is suitably aligned for any kind of variable. The memory is not cleared. cudaMalloc() returns cudaErrorMemoryAllocation in case of failure.

The device version of cudaFree cannot be used with a *devPtr allocated using the host API, and vice versa.

Parameters:

size (size_t) – Requested allocation size in bytes

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
devPtr (Any) – Pointer to allocated device memory

cuda.bindings.runtime.cudaMallocHost(size_t size)¶

Allocates page-locked memory on the host.

On systems where pageableMemoryAccessUsesHostPageTables is true, cudaMallocHost may not page-lock the allocated memory.

Page-locking excessive amounts of memory with cudaMallocHost() may degrade system performance, since it reduces the amount of memory available to the system for paging. As a result, this function is best used sparingly to allocate staging areas for data exchange between host and device.

Parameters:

size (size_t) – Requested allocation size in bytes

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
ptr (Any) – Pointer to allocated host memory

cuda.bindings.runtime.cudaMallocPitch(size_t width, size_t height)¶

Allocates pitched memory on the device.

Allocates at least width (in bytes) * height bytes of linear memory on the device and returns in *devPtr a pointer to the allocated memory. The function may pad the allocation to ensure that corresponding pointers in any given row will continue to meet the alignment requirements for coalescing as the address is updated from row to row. The pitch returned in *pitch by cudaMallocPitch() is the width in bytes of the allocation. The intended usage of pitch is as a separate parameter of the allocation, used to compute addresses within the 2D array. Given the row and column of an array element of type T, the address is computed as:

View CUDA Toolkit Documentation for a C++ code example

For allocations of 2D arrays, it is recommended that programmers consider performing pitch allocations using cudaMallocPitch(). Due to pitch alignment restrictions in the hardware, this is especially true if the application will be performing 2D memory copies between different regions of device memory (whether linear memory or CUDA arrays).

Parameters:

width (size_t) – Requested pitched allocation width (in bytes)
height (size_t) – Requested pitched allocation height

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
devPtr (Any) – Pointer to allocated pitched device memory
pitch (int) – Pitch for allocation

cuda.bindings.runtime.cudaMallocArray(cudaChannelFormatDesc desc: Optional[cudaChannelFormatDesc], size_t width, size_t height, unsigned int flags)¶

Allocate an array on the device.

Allocates a CUDA array according to the cudaChannelFormatDesc structure desc and returns a handle to the new CUDA array in *array.

The cudaChannelFormatDesc is defined as:

View CUDA Toolkit Documentation for a C++ code example

where cudaChannelFormatKind is one of cudaChannelFormatKindSigned, cudaChannelFormatKindUnsigned, or cudaChannelFormatKindFloat.

The flags parameter enables different options to be specified that affect the allocation, as follows.

cudaArrayDefault: This flag’s value is defined to be 0 and provides default array allocation
cudaArraySurfaceLoadStore: Allocates an array that can be read from or written to using a surface reference
cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the array.
cudaArraySparse: Allocates a CUDA array without physical backing memory. The subregions within this sparse array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. The physical backing memory must be allocated via cuMemCreate.
cudaArrayDeferredMapping: Allocates a CUDA array without physical backing memory. The entire array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. The physical backing memory must be allocated via cuMemCreate.

width and height must meet certain size requirements. See cudaMalloc3DArray() for more details.

Parameters:

desc (cudaChannelFormatDesc) – Requested channel format
width (size_t) – Requested array allocation width
height (size_t) – Requested array allocation height
flags (unsigned int) – Requested properties of allocated array

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
array (cudaArray_t) – Pointer to allocated array in device memory

cuda.bindings.runtime.cudaFree(devPtr)¶

Frees memory on the device.

Frees the memory space pointed to by devPtr, which must have been returned by a previous call to one of the following memory allocation APIs - cudaMalloc(), cudaMallocPitch(), cudaMallocManaged(), cudaMallocAsync(), cudaMallocFromPoolAsync().

Note - This API will not perform any implicit synchronization when the pointer was allocated with cudaMallocAsync or cudaMallocFromPoolAsync. Callers must ensure that all accesses to these pointer have completed before invoking cudaFree. For best performance and memory reuse, users should use cudaFreeAsync to free memory allocated via the stream ordered memory allocator. For all other pointers, this API may perform implicit synchronization.

If cudaFree`(`devPtr) has already been called before, an error is returned. If devPtr is 0, no operation is performed. cudaFree() returns cudaErrorValue in case of failure.

The device version of cudaFree cannot be used with a *devPtr allocated using the host API, and vice versa.

Parameters:: devPtr (Any) – Device pointer to memory to free
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaFreeHost(ptr)¶

Frees page-locked memory.

Frees the memory space pointed to by hostPtr, which must have been returned by a previous call to cudaMallocHost() or cudaHostAlloc().

Parameters:: ptr (Any) – Pointer to memory to free
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaFreeArray(array)¶

Frees an array on the device.

Frees the CUDA array array, which must have been returned by a previous call to cudaMallocArray(). If devPtr is 0, no operation is performed.

Parameters:: array (cudaArray_t) – Pointer to array to free
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaFreeMipmappedArray(mipmappedArray)¶

Frees a mipmapped array on the device.

Frees the CUDA mipmapped array mipmappedArray, which must have been returned by a previous call to cudaMallocMipmappedArray(). If devPtr is 0, no operation is performed.

Parameters:: mipmappedArray (cudaMipmappedArray_t) – Pointer to mipmapped array to free
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaHostAlloc(size_t size, unsigned int flags)¶

Allocates page-locked memory on the host.

Allocates size bytes of host memory that is page-locked and accessible to the device. The driver tracks the virtual memory ranges allocated with this function and automatically accelerates calls to functions such as cudaMemcpy(). Since the memory can be accessed directly by the device, it can be read or written with much higher bandwidth than pageable memory obtained with functions such as malloc(). Allocating excessive amounts of pinned memory may degrade system performance, since it reduces the amount of memory available to the system for paging. As a result, this function is best used sparingly to allocate staging areas for data exchange between host and device.

The flags parameter enables different options to be specified that affect the allocation, as follows.

cudaHostAllocDefault: This flag’s value is defined to be 0 and causes cudaHostAlloc() to emulate cudaMallocHost().
cudaHostAllocPortable: The memory returned by this call will be considered as pinned memory by all CUDA contexts, not just the one that performed the allocation.
cudaHostAllocMapped: Maps the allocation into the CUDA address space. The device pointer to the memory may be obtained by calling cudaHostGetDevicePointer().
cudaHostAllocWriteCombined: Allocates the memory as write-combined (WC). WC memory can be transferred across the PCI Express bus more quickly on some system configurations, but cannot be read efficiently by most CPUs. WC memory is a good option for buffers that will be written by the CPU and read by the device via mapped pinned memory or host->device transfers.

All of these flags are orthogonal to one another: a developer may allocate memory that is portable, mapped and/or write-combined with no restrictions.

In order for the cudaHostAllocMapped flag to have any effect, the CUDA context must support the cudaDeviceMapHost flag, which can be checked via cudaGetDeviceFlags(). The cudaDeviceMapHost flag is implicitly set for contexts created via the runtime API.

The cudaHostAllocMapped flag may be specified on CUDA contexts for devices that do not support mapped pinned memory. The failure is deferred to cudaHostGetDevicePointer() because the memory may be mapped into other CUDA contexts via the cudaHostAllocPortable flag.

Memory allocated by this function must be freed with cudaFreeHost().

Parameters:

size (size_t) – Requested allocation size in bytes
flags (unsigned int) – Requested properties of allocated memory

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
pHost (Any) – Device pointer to allocated memory

See also

cudaSetDeviceFlags, cudaMallocHost (C API), cudaFreeHost, cudaGetDeviceFlags, cuMemHostAlloc

cuda.bindings.runtime.cudaHostRegister(ptr, size_t size, unsigned int flags)¶

Registers an existing host memory range for use by CUDA.

Page-locks the memory range specified by ptr and size and maps it for the device(s) as specified by flags. This memory range also is added to the same tracking mechanism as cudaHostAlloc() to automatically accelerate calls to functions such as cudaMemcpy(). Since the memory can be accessed directly by the device, it can be read or written with much higher bandwidth than pageable memory that has not been registered. Page-locking excessive amounts of memory may degrade system performance, since it reduces the amount of memory available to the system for paging. As a result, this function is best used sparingly to register staging areas for data exchange between host and device.

On systems where pageableMemoryAccessUsesHostPageTables is true, cudaHostRegister will not page-lock the memory range specified by ptr but only populate unpopulated pages.

cudaHostRegister is supported only on I/O coherent devices that have a non-zero value for the device attribute cudaDevAttrHostRegisterSupported.

The flags parameter enables different options to be specified that affect the allocation, as follows.

cudaHostRegisterDefault: On a system with unified virtual addressing, the memory will be both mapped and portable. On a system with no unified virtual addressing, the memory will be neither mapped nor portable.
cudaHostRegisterPortable: The memory returned by this call will be considered as pinned memory by all CUDA contexts, not just the one that performed the allocation.
cudaHostRegisterMapped: Maps the allocation into the CUDA address space. The device pointer to the memory may be obtained by calling cudaHostGetDevicePointer().
cudaHostRegisterIoMemory: The passed memory pointer is treated as pointing to some memory-mapped I/O space, e.g. belonging to a third-party PCIe device, and it will marked as non cache- coherent and contiguous.
cudaHostRegisterReadOnly: The passed memory pointer is treated as pointing to memory that is considered read-only by the device. On platforms without cudaDevAttrPageableMemoryAccessUsesHostPageTables, this flag is required in order to register memory mapped to the CPU as read-only. Support for the use of this flag can be queried from the device attribute cudaDeviceAttrReadOnlyHostRegisterSupported. Using this flag with a current context associated with a device that does not have this attribute set will cause cudaHostRegister to error with cudaErrorNotSupported.

All of these flags are orthogonal to one another: a developer may page- lock memory that is portable or mapped with no restrictions.

The CUDA context must have been created with the cudaMapHost flag in order for the cudaHostRegisterMapped flag to have any effect.

The cudaHostRegisterMapped flag may be specified on CUDA contexts for devices that do not support mapped pinned memory. The failure is deferred to cudaHostGetDevicePointer() because the memory may be mapped into other CUDA contexts via the cudaHostRegisterPortable flag.

For devices that have a non-zero value for the device attribute cudaDevAttrCanUseHostPointerForRegisteredMem, the memory can also be accessed from the device using the host pointer ptr. The device pointer returned by cudaHostGetDevicePointer() may or may not match the original host pointer ptr and depends on the devices visible to the application. If all devices visible to the application have a non-zero value for the device attribute, the device pointer returned by cudaHostGetDevicePointer() will match the original pointer ptr. If any device visible to the application has a zero value for the device attribute, the device pointer returned by cudaHostGetDevicePointer() will not match the original host pointer ptr, but it will be suitable for use on all devices provided Unified Virtual Addressing is enabled. In such systems, it is valid to access the memory using either pointer on devices that have a non-zero value for the device attribute. Note however that such devices should access the memory using only of the two pointers and not both.

The memory page-locked by this function must be unregistered with cudaHostUnregister().

Parameters:

ptr (Any) – Host pointer to memory to page-lock
size (size_t) – Size in bytes of the address range to page-lock in bytes
flags (unsigned int) – Flags for allocation request

Returns:

Return type:

cudaHostAlloc, cuMemHostGetFlags

cuda.bindings.runtime.cudaHostUnregister(ptr)¶

Unregisters a memory range that was registered with cudaHostRegister.

Unmaps the memory range whose base address is specified by ptr, and makes it pageable again.

The base address must be the same one specified to cudaHostRegister().

Parameters:: ptr (Any) – Host pointer to memory to unregister
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorHostMemoryNotRegistered
Return type:: cudaError_t

cuda.bindings.runtime.cudaHostGetDevicePointer(pHost, unsigned int flags)¶

Passes back device pointer of mapped host memory allocated by cudaHostAlloc or registered by cudaHostRegister.

Passes back the device pointer corresponding to the mapped, pinned host buffer allocated by cudaHostAlloc() or registered by cudaHostRegister().

cudaHostGetDevicePointer() will fail if the cudaDeviceMapHost flag was not specified before deferred context creation occurred, or if called on a device that does not support mapped, pinned memory.

For devices that have a non-zero value for the device attribute cudaDevAttrCanUseHostPointerForRegisteredMem, the memory can also be accessed from the device using the host pointer pHost. The device pointer returned by cudaHostGetDevicePointer() may or may not match the original host pointer pHost and depends on the devices visible to the application. If all devices visible to the application have a non-zero value for the device attribute, the device pointer returned by cudaHostGetDevicePointer() will match the original pointer pHost. If any device visible to the application has a zero value for the device attribute, the device pointer returned by cudaHostGetDevicePointer() will not match the original host pointer pHost, but it will be suitable for use on all devices provided Unified Virtual Addressing is enabled. In such systems, it is valid to access the memory using either pointer on devices that have a non-zero value for the device attribute. Note however that such devices should access the memory using only of the two pointers and not both.

flags provides for future releases. For now, it must be set to 0.

Parameters:

pHost (Any) – Requested host pointer mapping
flags (unsigned int) – Flags for extensions (must be 0 for now)

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
pDevice (Any) – Returned device pointer for mapped memory

cuda.bindings.runtime.cudaHostGetFlags(pHost)¶

Passes back flags used to allocate pinned host memory allocated by cudaHostAlloc.

cudaHostGetFlags() will fail if the input pointer does not reside in an address range allocated by cudaHostAlloc().

Parameters:

pHost (Any) – Host pointer

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pFlags (unsigned int) – Returned flags word

See also

cuda.bindings.runtime.cudaMalloc3D(cudaExtent extent: cudaExtent)¶

Allocates logical 1D, 2D, or 3D memory objects on the device.

Allocates at least width * height * depth bytes of linear memory on the device and returns a cudaPitchedPtr in which ptr is a pointer to the allocated memory. The function may pad the allocation to ensure hardware alignment requirements are met. The pitch returned in the pitch field of pitchedDevPtr is the width in bytes of the allocation.

The returned cudaPitchedPtr contains additional fields xsize and ysize, the logical width and height of the allocation, which are equivalent to the width and height extent parameters provided by the programmer during allocation.

For allocations of 2D and 3D objects, it is highly recommended that programmers perform allocations using cudaMalloc3D() or cudaMallocPitch(). Due to alignment restrictions in the hardware, this is especially true if the application will be performing memory copies involving 2D or 3D objects (whether linear memory or CUDA arrays).

Parameters:

extent (cudaExtent) – Requested allocation size (width field in bytes)

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
pitchedDevPtr (cudaPitchedPtr) – Pointer to allocated pitched device memory

cuda.bindings.runtime.cudaMalloc3DArray(cudaChannelFormatDesc desc: Optional[cudaChannelFormatDesc], cudaExtent extent: cudaExtent, unsigned int flags)¶

Allocate an array on the device.

Allocates a CUDA array according to the cudaChannelFormatDesc structure desc and returns a handle to the new CUDA array in *array.

The cudaChannelFormatDesc is defined as:

View CUDA Toolkit Documentation for a C++ code example

where cudaChannelFormatKind is one of cudaChannelFormatKindSigned, cudaChannelFormatKindUnsigned, or cudaChannelFormatKindFloat.

cudaMalloc3DArray() can allocate the following:

A 1D array is allocated if the height and depth extents are both zero.
A 2D array is allocated if only the depth extent is zero.
A 3D array is allocated if all three extents are non-zero.
A 1D layered CUDA array is allocated if only the height extent is zero and the cudaArrayLayered flag is set. Each layer is a 1D array. The number of layers is determined by the depth extent.
A 2D layered CUDA array is allocated if all three extents are non- zero and the cudaArrayLayered flag is set. Each layer is a 2D array. The number of layers is determined by the depth extent.
A cubemap CUDA array is allocated if all three extents are non-zero and the cudaArrayCubemap flag is set. Width must be equal to height, and depth must be six. A cubemap is a special type of 2D layered CUDA array, where the six layers represent the six faces of a cube. The order of the six layers in memory is the same as that listed in cudaGraphicsCubeFace.
A cubemap layered CUDA array is allocated if all three extents are non-zero, and both, cudaArrayCubemap and cudaArrayLayered flags are set. Width must be equal to height, and depth must be a multiple of six. A cubemap layered CUDA array is a special type of 2D layered CUDA array that consists of a collection of cubemaps. The first six layers represent the first cubemap, the next six layers form the second cubemap, and so on.

The flags parameter enables different options to be specified that affect the allocation, as follows.

cudaArrayDefault: This flag’s value is defined to be 0 and provides default array allocation
cudaArrayLayered: Allocates a layered CUDA array, with the depth extent indicating the number of layers
cudaArrayCubemap: Allocates a cubemap CUDA array. Width must be equal to height, and depth must be six. If the cudaArrayLayered flag is also set, depth must be a multiple of six.
cudaArraySurfaceLoadStore: Allocates a CUDA array that could be read from or written to using a surface reference.
cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the CUDA array. Texture gather can only be performed on 2D CUDA arrays.
cudaArraySparse: Allocates a CUDA array without physical backing memory. The subregions within this sparse array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. This flag can only be used for creating 2D, 3D or 2D layered sparse CUDA arrays. The physical backing memory must be allocated via cuMemCreate.
cudaArrayDeferredMapping: Allocates a CUDA array without physical backing memory. The entire array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. The physical backing memory must be allocated via cuMemCreate.

The width, height and depth extents must meet certain size requirements as listed in the following table. All values are specified in elements.

Note that 2D CUDA arrays have different size requirements if the cudaArrayTextureGather flag is set. In that case, the valid range for (width, height, depth) is ((1,maxTexture2DGather[0]), (1,maxTexture2DGather[1]), 0).

View CUDA Toolkit Documentation for a table example

Parameters:

desc (cudaChannelFormatDesc) – Requested channel format
extent (cudaExtent) – Requested allocation size (width field in elements)
flags (unsigned int) – Flags for extensions

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
array (cudaArray_t) – Pointer to allocated array in device memory

cuda.bindings.runtime.cudaMallocMipmappedArray(cudaChannelFormatDesc desc: Optional[cudaChannelFormatDesc], cudaExtent extent: cudaExtent, unsigned int numLevels, unsigned int flags)¶

Allocate a mipmapped array on the device.

Allocates a CUDA mipmapped array according to the cudaChannelFormatDesc structure desc and returns a handle to the new CUDA mipmapped array in *mipmappedArray. numLevels specifies the number of mipmap levels to be allocated. This value is clamped to the range [1, 1 + floor(log2(max(width, height, depth)))].

The cudaChannelFormatDesc is defined as:

View CUDA Toolkit Documentation for a C++ code example

where cudaChannelFormatKind is one of cudaChannelFormatKindSigned, cudaChannelFormatKindUnsigned, or cudaChannelFormatKindFloat.

cudaMallocMipmappedArray() can allocate the following:

A 1D mipmapped array is allocated if the height and depth extents are both zero.
A 2D mipmapped array is allocated if only the depth extent is zero.
A 3D mipmapped array is allocated if all three extents are non-zero.
A 1D layered CUDA mipmapped array is allocated if only the height extent is zero and the cudaArrayLayered flag is set. Each layer is a 1D mipmapped array. The number of layers is determined by the depth extent.
A 2D layered CUDA mipmapped array is allocated if all three extents are non-zero and the cudaArrayLayered flag is set. Each layer is a 2D mipmapped array. The number of layers is determined by the depth extent.
A cubemap CUDA mipmapped array is allocated if all three extents are non-zero and the cudaArrayCubemap flag is set. Width must be equal to height, and depth must be six. The order of the six layers in memory is the same as that listed in cudaGraphicsCubeFace.
A cubemap layered CUDA mipmapped array is allocated if all three extents are non-zero, and both, cudaArrayCubemap and cudaArrayLayered flags are set. Width must be equal to height, and depth must be a multiple of six. A cubemap layered CUDA mipmapped array is a special type of 2D layered CUDA mipmapped array that consists of a collection of cubemap mipmapped arrays. The first six layers represent the first cubemap mipmapped array, the next six layers form the second cubemap mipmapped array, and so on.

The flags parameter enables different options to be specified that affect the allocation, as follows.

cudaArrayDefault: This flag’s value is defined to be 0 and provides default mipmapped array allocation
cudaArrayLayered: Allocates a layered CUDA mipmapped array, with the depth extent indicating the number of layers
cudaArrayCubemap: Allocates a cubemap CUDA mipmapped array. Width must be equal to height, and depth must be six. If the cudaArrayLayered flag is also set, depth must be a multiple of six.
cudaArraySurfaceLoadStore: This flag indicates that individual mipmap levels of the CUDA mipmapped array will be read from or written to using a surface reference.
cudaArrayTextureGather: This flag indicates that texture gather operations will be performed on the CUDA array. Texture gather can only be performed on 2D CUDA mipmapped arrays, and the gather operations are performed only on the most detailed mipmap level.
cudaArraySparse: Allocates a CUDA mipmapped array without physical backing memory. The subregions within this sparse array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. This flag can only be used for creating 2D, 3D or 2D layered sparse CUDA mipmapped arrays. The physical backing memory must be allocated via cuMemCreate.
cudaArrayDeferredMapping: Allocates a CUDA mipmapped array without physical backing memory. The entire array can later be mapped onto a physical memory allocation by calling cuMemMapArrayAsync. The physical backing memory must be allocated via cuMemCreate.

The width, height and depth extents must meet certain size requirements as listed in the following table. All values are specified in elements.

View CUDA Toolkit Documentation for a table example

Parameters:

desc (cudaChannelFormatDesc) – Requested channel format
extent (cudaExtent) – Requested allocation size (width field in elements)
numLevels (unsigned int) – Number of mipmap levels to allocate
flags (unsigned int) – Flags for extensions

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
mipmappedArray (cudaMipmappedArray_t) – Pointer to allocated mipmapped array in device memory

cuda.bindings.runtime.cudaGetMipmappedArrayLevel(mipmappedArray, unsigned int level)¶

Gets a mipmap level of a CUDA mipmapped array.

Returns in *levelArray a CUDA array that represents a single mipmap level of the CUDA mipmapped array mipmappedArray.

If level is greater than the maximum number of levels in this mipmapped array, cudaErrorInvalidValue is returned.

If mipmappedArray is NULL, cudaErrorInvalidResourceHandle is returned.

Parameters:

mipmappedArray (cudaMipmappedArray_const_t) – CUDA mipmapped array
level (unsigned int) – Mipmap level

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue cudaErrorInvalidResourceHandle
levelArray (cudaArray_t) – Returned mipmap level CUDA array

cuda.bindings.runtime.cudaMemcpy3D(cudaMemcpy3DParms p: Optional[cudaMemcpy3DParms])¶

Copies data between 3D objects.

View CUDA Toolkit Documentation for a C++ code example

cudaMemcpy3D() copies data betwen two 3D objects. The source and destination objects may be in either host memory, device memory, or a CUDA array. The source, destination, extent, and kind of copy performed is specified by the cudaMemcpy3DParms struct which should be initialized to zero before use:

View CUDA Toolkit Documentation for a C++ code example

The struct passed to cudaMemcpy3D() must specify one of srcArray or srcPtr and one of dstArray or dstPtr. Passing more than one non-zero source or destination will cause cudaMemcpy3D() to return an error.

The srcPos and dstPos fields are optional offsets into the source and destination objects and are defined in units of each object’s elements. The element for a host or device pointer is assumed to be unsigned char.

The extent field defines the dimensions of the transferred area in elements. If a CUDA array is participating in the copy, the extent is defined in terms of that array’s elements. If no CUDA array is participating in the copy then the extents are defined in elements of unsigned char.

The kind field defines the direction of the copy. It must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. For cudaMemcpyHostToHost or cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost passed as kind and cudaArray type passed as source or destination, if the kind implies cudaArray type to be present on the host, cudaMemcpy3D() will disregard that implication and silently correct the kind based on the fact that cudaArray type can only be present on the device.

If the source and destination are both arrays, cudaMemcpy3D() will return an error if they do not have the same element size.

The source and destination object may not overlap. If overlapping source and destination objects are specified, undefined behavior will result.

The source object must entirely contain the region defined by srcPos and extent. The destination object must entirely contain the region defined by dstPos and extent.

cudaMemcpy3D() returns an error if the pitch of srcPtr or dstPtr exceeds the maximum allowed. The pitch of a cudaPitchedPtr allocated with cudaMalloc3D() will always be valid.

Parameters:: p (cudaMemcpy3DParms) – 3D memory copy parameters
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection
Return type:: cudaError_t

See also

cudaMalloc3D, cudaMalloc3DArray, cudaMemset3D, cudaMemcpy3DAsync, cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, make_cudaExtent, make_cudaPos, cuMemcpy3D

cuda.bindings.runtime.cudaMemcpy3DPeer(cudaMemcpy3DPeerParms p: Optional[cudaMemcpy3DPeerParms])¶

Copies memory between devices.

Perform a 3D memory copy according to the parameters specified in p. See the definition of the cudaMemcpy3DPeerParms structure for documentation of its parameters.

Note that this function is synchronous with respect to the host only if the source or destination of the transfer is host memory. Note also that this copy is serialized with respect to all pending and future asynchronous work in to the current device, the copy’s source device, and the copy’s destination device (use cudaMemcpy3DPeerAsync to avoid this synchronization).

Parameters:: p (cudaMemcpy3DPeerParms) – Parameters for the memory copy
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice, cudaErrorInvalidPitchValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaMemcpy3DAsync(cudaMemcpy3DParms p: Optional[cudaMemcpy3DParms], stream)¶

Copies data between 3D objects.

View CUDA Toolkit Documentation for a C++ code example

cudaMemcpy3DAsync() copies data betwen two 3D objects. The source and destination objects may be in either host memory, device memory, or a CUDA array. The source, destination, extent, and kind of copy performed is specified by the cudaMemcpy3DParms struct which should be initialized to zero before use:

View CUDA Toolkit Documentation for a C++ code example

The struct passed to cudaMemcpy3DAsync() must specify one of srcArray or srcPtr and one of dstArray or dstPtr. Passing more than one non-zero source or destination will cause cudaMemcpy3DAsync() to return an error.

The kind field defines the direction of the copy. It must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. For cudaMemcpyHostToHost or cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost passed as kind and cudaArray type passed as source or destination, if the kind implies cudaArray type to be present on the host, cudaMemcpy3DAsync() will disregard that implication and silently correct the kind based on the fact that cudaArray type can only be present on the device.

If the source and destination are both arrays, cudaMemcpy3DAsync() will return an error if they do not have the same element size.

The source and destination object may not overlap. If overlapping source and destination objects are specified, undefined behavior will result.

The source object must lie entirely within the region defined by srcPos and extent. The destination object must lie entirely within the region defined by dstPos and extent.

cudaMemcpy3DAsync() returns an error if the pitch of srcPtr or dstPtr exceeds the maximum allowed. The pitch of a cudaPitchedPtr allocated with cudaMalloc3D() will always be valid.

cudaMemcpy3DAsync() is asynchronous with respect to the host, so the call may return before the copy is complete. The copy can optionally be associated to a stream by passing a non-zero stream argument. If kind is cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost and stream is non-zero, the copy may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

p (cudaMemcpy3DParms) – 3D memory copy parameters
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice, cudaErrorInvalidPitchValue

See also

cudaMalloc3D, cudaMalloc3DArray, cudaMemset3D, cudaMemcpy3D, cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, :::cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, make_cudaExtent, make_cudaPos, cuMemcpy3DAsync

cuda.bindings.runtime.cudaMemcpy3DPeerAsync(cudaMemcpy3DPeerParms p: Optional[cudaMemcpy3DPeerParms], stream)¶

Copies memory between devices asynchronously.

Perform a 3D memory copy according to the parameters specified in p. See the definition of the cudaMemcpy3DPeerParms structure for documentation of its parameters.

Parameters:

p (cudaMemcpy3DPeerParms) – Parameters for the memory copy
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaArrayGetMemoryRequirements

cuda.bindings.runtime.cudaMemGetInfo()¶

Gets free and total device memory.

Returns in *total the total amount of memory available to the the current context. Returns in *free the amount of memory on the device that is free according to the OS. CUDA is not guaranteed to be able to allocate all of the memory that the OS reports as free. In a multi- tenet situation, free estimate returned is prone to race condition where a new allocation/free done by a different process or a different thread in the same process between the time when free memory was estimated and reported, will result in deviation in free value reported and actual free memory.

The integrated GPU on Tegra shares memory with CPU and other component of the SoC. The free and total values returned by the API excludes the SWAP memory space maintained by the OS on some platforms. The OS may move some of the memory pages into swap area as the GPU or CPU allocate or access memory. See Tegra app note on how to calculate total and free memory on Tegra.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorLaunchFailure
free (int) – Returned free memory in bytes
total (int) – Returned total memory in bytes

See also

cuMemGetInfo

cuda.bindings.runtime.cudaArrayGetInfo(array)¶

Gets info about the specified cudaArray.

Returns in *desc, *extent and *flags respectively, the type, shape and flags of array.

Any of *desc, *extent and *flags may be specified as NULL.

Parameters:

array (cudaArray_t) – The cudaArray to get info for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
desc (cudaChannelFormatDesc) – Returned array type
extent (cudaExtent) – Returned array shape. 2D arrays will have depth of zero
flags (unsigned int) – Returned array flags

cuda.bindings.runtime.cudaArrayGetPlane(hArray, unsigned int planeIdx)¶

Gets a CUDA array plane from a CUDA array.

Returns in pPlaneArray a CUDA array that represents a single format plane of the CUDA array hArray.

If planeIdx is greater than the maximum number of planes in this array or if the array does not have a multi-planar format e.g: cudaChannelFormatKindNV12, then cudaErrorInvalidValue is returned.

Note that if the hArray has format cudaChannelFormatKindNV12, then passing in 0 for planeIdx returns a CUDA array of the same size as hArray but with one 8-bit channel and cudaChannelFormatKindUnsigned as its format kind. If 1 is passed for planeIdx, then the returned CUDA array has half the height and width of hArray with two 8-bit channels and cudaChannelFormatKindUnsigned as its format kind.

Parameters:

hArray (cudaArray_t) – CUDA array
planeIdx (unsigned int) – Plane index

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue cudaErrorInvalidResourceHandle
pPlaneArray (cudaArray_t) – Returned CUDA array referenced by the planeIdx

See also

cuArrayGetPlane

cuda.bindings.runtime.cudaArrayGetMemoryRequirements(array, int device)¶

Returns the memory requirements of a CUDA array.

Returns the memory requirements of a CUDA array in memoryRequirements If the CUDA array is not allocated with flag cudaArrayDeferredMapping cudaErrorInvalidValue will be returned.

The returned value in size represents the total size of the CUDA array. The returned value in alignment represents the alignment necessary for mapping the CUDA array.

Parameters:

array (cudaArray_t) – CUDA array to get the memory requirements of
device (int) – Device to get the memory requirements for

Returns:

cudaError_t – cudaSuccess cudaErrorInvalidValue
memoryRequirements (cudaArrayMemoryRequirements) – Pointer to cudaArrayMemoryRequirements

cuda.bindings.runtime.cudaMipmappedArrayGetMemoryRequirements(mipmap, int device)¶

Returns the memory requirements of a CUDA mipmapped array.

Returns the memory requirements of a CUDA mipmapped array in memoryRequirements If the CUDA mipmapped array is not allocated with flag cudaArrayDeferredMapping cudaErrorInvalidValue will be returned.

The returned value in size represents the total size of the CUDA mipmapped array. The returned value in alignment represents the alignment necessary for mapping the CUDA mipmapped array.

Parameters:

mipmap (cudaMipmappedArray_t) – CUDA mipmapped array to get the memory requirements of
device (int) – Device to get the memory requirements for

Returns:

cudaError_t – cudaSuccess cudaErrorInvalidValue
memoryRequirements (cudaArrayMemoryRequirements) – Pointer to cudaArrayMemoryRequirements

See also

cuda.bindings.runtime.cudaArrayGetSparseProperties(array)¶

Returns the layout properties of a sparse CUDA array.

Returns the layout properties of a sparse CUDA array in sparseProperties. If the CUDA array is not allocated with flag cudaArraySparse cudaErrorInvalidValue will be returned.

If the returned value in flags contains cudaArraySparsePropertiesSingleMipTail, then miptailSize represents the total size of the array. Otherwise, it will be zero. Also, the returned value in miptailFirstLevel is always zero. Note that the array must have been allocated using cudaMallocArray or cudaMalloc3DArray. For CUDA arrays obtained using cudaMipmappedArrayGetLevel, cudaErrorInvalidValue will be returned. Instead, cudaMipmappedArrayGetSparseProperties must be used to obtain the sparse properties of the entire CUDA mipmapped array to which array belongs to.

Parameters:

array (cudaArray_t) – The CUDA array to get the sparse properties of

Returns:

cudaError_t – cudaSuccess cudaErrorInvalidValue
sparseProperties (cudaArraySparseProperties) – Pointer to return the cudaArraySparseProperties

cuda.bindings.runtime.cudaMipmappedArrayGetSparseProperties(mipmap)¶

Returns the layout properties of a sparse CUDA mipmapped array.

Returns the sparse array layout properties in sparseProperties. If the CUDA mipmapped array is not allocated with flag cudaArraySparse cudaErrorInvalidValue will be returned.

For non-layered CUDA mipmapped arrays, miptailSize returns the size of the mip tail region. The mip tail region includes all mip levels whose width, height or depth is less than that of the tile. For layered CUDA mipmapped arrays, if flags contains cudaArraySparsePropertiesSingleMipTail, then miptailSize specifies the size of the mip tail of all layers combined. Otherwise, miptailSize specifies mip tail size per layer. The returned value of miptailFirstLevel is valid only if miptailSize is non-zero.

Parameters:

mipmap (cudaMipmappedArray_t) – The CUDA mipmapped array to get the sparse properties of

Returns:

cudaError_t – cudaSuccess cudaErrorInvalidValue
sparseProperties (cudaArraySparseProperties) – Pointer to return cudaArraySparseProperties

cuda.bindings.runtime.cudaMemcpy(dst, src, size_t count, kind: cudaMemcpyKind)¶

Copies data between host and device.

Copies count bytes from the memory area pointed to by src to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. Calling cudaMemcpy() with dst and src pointers that do not match the direction of the copy results in an undefined behavior.

ote_sync

dstAny
Destination memory address

srcAny
Source memory address

countsize_t
Size in bytes to copy

kindcudaMemcpyKind
Type of transfer

cudaError_t
cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidMemcpyDirection

cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpyDtoH, cuMemcpyHtoD, cuMemcpyDtoD, cuMemcpy

cuda.bindings.runtime.cudaMemcpyPeer(dst, int dstDevice, src, int srcDevice, size_t count)¶

Copies memory between two devices.

Copies memory from one device to memory on another device. dst is the base device pointer of the destination memory and dstDevice is the destination device. src is the base device pointer of the source memory and srcDevice is the source device. count specifies the number of bytes to copy.

Note that this function is asynchronous with respect to the host, but serialized with respect all pending and future asynchronous work in to the current device, srcDevice, and dstDevice (use cudaMemcpyPeerAsync to avoid this synchronization).

Parameters:

dst (Any) – Destination device pointer
dstDevice (int) – Destination device
src (Any) – Source device pointer
srcDevice (int) – Source device
count (size_t) – Size of memory copy in bytes

Returns:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

cuda.bindings.runtime.cudaMemcpy2D(dst, size_t dpitch, src, size_t spitch, size_t width, size_t height, kind: cudaMemcpyKind)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the memory area pointed to by src to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. dpitch and spitch are the widths in memory in bytes of the 2D arrays pointed to by dst and src, including any padding added to the end of each row. The memory areas may not overlap. width must not exceed either dpitch or spitch. Calling cudaMemcpy2D() with dst and src pointers that do not match the direction of the copy results in an undefined behavior. cudaMemcpy2D() returns an error if dpitch or spitch exceeds the maximum allowed.

Parameters:

dst (Any) – Destination memory address
dpitch (size_t) – Pitch of destination memory
src (Any) – Source memory address
spitch (size_t) – Pitch of source memory
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2D, cuMemcpy2DUnaligned

cuda.bindings.runtime.cudaMemcpy2DToArray(dst, size_t wOffset, size_t hOffset, src, size_t spitch, size_t width, size_t height, kind: cudaMemcpyKind)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the memory area pointed to by src to the CUDA array dst starting at hOffset rows and wOffset bytes from the upper left corner, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. spitch is the width in memory in bytes of the 2D array pointed to by src, including any padding added to the end of each row. wOffset + width must not exceed the width of the CUDA array dst. width must not exceed spitch. cudaMemcpy2DToArray() returns an error if spitch exceeds the maximum allowed.

Parameters:

dst (cudaArray_t) – Destination memory address
wOffset (size_t) – Destination starting X offset (columns in bytes)
hOffset (size_t) – Destination starting Y offset (rows)
src (Any) – Source memory address
spitch (size_t) – Pitch of source memory
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2D, cuMemcpy2DUnaligned

cuda.bindings.runtime.cudaMemcpy2DFromArray(dst, size_t dpitch, src, size_t wOffset, size_t hOffset, size_t width, size_t height, kind: cudaMemcpyKind)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the CUDA array src starting at hOffset rows and wOffset bytes from the upper left corner to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. dpitch is the width in memory in bytes of the 2D array pointed to by dst, including any padding added to the end of each row. wOffset + width must not exceed the width of the CUDA array src. width must not exceed dpitch. cudaMemcpy2DFromArray() returns an error if dpitch exceeds the maximum allowed.

Parameters:

dst (Any) – Destination memory address
dpitch (size_t) – Pitch of destination memory
src (cudaArray_const_t) – Source memory address
wOffset (size_t) – Source starting X offset (columns in bytes)
hOffset (size_t) – Source starting Y offset (rows)
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2D, cuMemcpy2DUnaligned

cuda.bindings.runtime.cudaMemcpy2DArrayToArray(dst, size_t wOffsetDst, size_t hOffsetDst, src, size_t wOffsetSrc, size_t hOffsetSrc, size_t width, size_t height, kind: cudaMemcpyKind)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the CUDA array src starting at hOffsetSrc rows and wOffsetSrc bytes from the upper left corner to the CUDA array dst starting at hOffsetDst rows and wOffsetDst bytes from the upper left corner, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. wOffsetDst + width must not exceed the width of the CUDA array dst. wOffsetSrc + width must not exceed the width of the CUDA array src.

Parameters:

dst (cudaArray_t) – Destination memory address
wOffsetDst (size_t) – Destination starting X offset (columns in bytes)
hOffsetDst (size_t) – Destination starting Y offset (rows)
src (cudaArray_const_t) – Source memory address
wOffsetSrc (size_t) – Source starting X offset (columns in bytes)
hOffsetSrc (size_t) – Source starting Y offset (rows)
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2D, cuMemcpy2DUnaligned

cuda.bindings.runtime.cudaMemcpyAsync(dst, src, size_t count, kind: cudaMemcpyKind, stream)¶

Copies data between host and device.

Copies count bytes from the memory area pointed to by src to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing.

The memory areas may not overlap. Calling cudaMemcpyAsync() with dst and src pointers that do not match the direction of the copy results in an undefined behavior.

cudaMemcpyAsync() is asynchronous with respect to the host, so the call may return before the copy is complete. The copy can optionally be associated to a stream by passing a non-zero stream argument. If kind is cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost and the stream is non-zero, the copy may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

dst (Any) – Destination memory address
src (Any) – Source memory address
count (size_t) – Size in bytes to copy
kind (cudaMemcpyKind) – Type of transfer
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpy2DAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpyAsync, cuMemcpyDtoHAsync, cuMemcpyHtoDAsync, cuMemcpyDtoDAsync

cuda.bindings.runtime.cudaMemcpyPeerAsync(dst, int dstDevice, src, int srcDevice, size_t count, stream)¶

Copies memory between two devices asynchronously.

Note that this function is asynchronous with respect to the host and all work on other devices.

Parameters:

dst (Any) – Destination device pointer
dstDevice (int) – Destination device
src (Any) – Source device pointer
srcDevice (int) – Source device
count (size_t) – Size of memory copy in bytes
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

cuda.bindings.runtime.cudaMemcpy2DAsync(dst, size_t dpitch, src, size_t spitch, size_t width, size_t height, kind: cudaMemcpyKind, stream)¶

Copies data between host and device.

Calling cudaMemcpy2DAsync() with dst and src pointers that do not match the direction of the copy results in an undefined behavior. cudaMemcpy2DAsync() returns an error if dpitch or spitch is greater than the maximum allowed.

cudaMemcpy2DAsync() is asynchronous with respect to the host, so the call may return before the copy is complete. The copy can optionally be associated to a stream by passing a non-zero stream argument. If kind is cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost and stream is non-zero, the copy may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

dst (Any) – Destination memory address
dpitch (size_t) – Pitch of destination memory
src (Any) – Source memory address
spitch (size_t) – Pitch of source memory
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DToArrayAsync, cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2DAsync

cuda.bindings.runtime.cudaMemcpy2DToArrayAsync(dst, size_t wOffset, size_t hOffset, src, size_t spitch, size_t width, size_t height, kind: cudaMemcpyKind, stream)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the memory area pointed to by src to the CUDA array dst starting at hOffset rows and wOffset bytes from the upper left corner, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. spitch is the width in memory in bytes of the 2D array pointed to by src, including any padding added to the end of each row. wOffset + width must not exceed the width of the CUDA array dst. width must not exceed spitch. cudaMemcpy2DToArrayAsync() returns an error if spitch exceeds the maximum allowed.

cudaMemcpy2DToArrayAsync() is asynchronous with respect to the host, so the call may return before the copy is complete. The copy can optionally be associated to a stream by passing a non-zero stream argument. If kind is cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost and stream is non-zero, the copy may overlap with operations in other streams.

cudaMemcpy2DFromArrayAsync, cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2DAsync

Parameters:

dst (cudaArray_t) – Destination memory address
wOffset (size_t) – Destination starting X offset (columns in bytes)
hOffset (size_t) – Destination starting Y offset (rows)
src (Any) – Source memory address
spitch (size_t) – Pitch of source memory
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidPitchValue, cudaErrorInvalidMemcpyDirection

See also

cudaMemcpy, cudaMemcpy2D, cudaMemcpy2DToArray, cudaMemcpy2DFromArray, cudaMemcpy2DArrayToArray, cudaMemcpyToSymbol, cudaMemcpyFromSymbol, cudaMemcpyAsync, cudaMemcpy2DAsync

cuda.bindings.runtime.cudaMemcpy2DFromArrayAsync(dst, size_t dpitch, src, size_t wOffset, size_t hOffset, size_t width, size_t height, kind: cudaMemcpyKind, stream)¶

Copies data between host and device.

Copies a matrix (height rows of width bytes each) from the CUDA array src starting at hOffset rows and wOffset bytes from the upper left corner to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. dpitch is the width in memory in bytes of the 2D array pointed to by dst, including any padding added to the end of each row. wOffset + width must not exceed the width of the CUDA array src. width must not exceed dpitch. cudaMemcpy2DFromArrayAsync() returns an error if dpitch exceeds the maximum allowed.

cudaMemcpy2DFromArrayAsync() is asynchronous with respect to the host, so the call may return before the copy is complete. The copy can optionally be associated to a stream by passing a non-zero stream argument. If kind is cudaMemcpyHostToDevice or cudaMemcpyDeviceToHost and stream is non-zero, the copy may overlap with operations in other streams.

cudaMemcpyToSymbolAsync, cudaMemcpyFromSymbolAsync, cuMemcpy2DAsync

Parameters:

dst (Any) – Destination memory address
dpitch (size_t) – Pitch of destination memory
src (cudaArray_const_t) – Source memory address
wOffset (size_t) – Source starting X offset (columns in bytes)
hOffset (size_t) – Source starting Y offset (rows)
width (size_t) – Width of matrix transfer (columns in bytes)
height (size_t) – Height of matrix transfer (rows)
kind (cudaMemcpyKind) – Type of transfer
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

See also

cuda.bindings.runtime.cudaMemset(devPtr, int value, size_t count)¶

Initializes or sets device memory to a value.

Fills the first count bytes of the memory area pointed to by devPtr with the constant byte value value.

Note that this function is asynchronous with respect to the host unless devPtr refers to pinned host memory.

Parameters:

devPtr (Any) – Pointer to device memory
value (int) – Value to set for each byte of specified memory
count (size_t) – Size in bytes to set

Returns:

Return type:

cuda.bindings.runtime.cudaMemset2D(devPtr, size_t pitch, int value, size_t width, size_t height)¶

Initializes or sets device memory to a value.

Sets to the specified value value a matrix (height rows of width bytes each) pointed to by dstPtr. pitch is the width in bytes of the 2D array pointed to by dstPtr, including any padding added to the end of each row. This function performs fastest when the pitch is one that has been passed back by cudaMallocPitch().

Note that this function is asynchronous with respect to the host unless devPtr refers to pinned host memory.

Parameters:

devPtr (Any) – Pointer to 2D device memory
pitch (size_t) – Pitch in bytes of 2D device memory(Unused if height is 1)
value (int) – Value to set for each byte of specified memory
width (size_t) – Width of matrix set (columns in bytes)
height (size_t) – Height of matrix set (rows)

Returns:

Return type:

cuda.bindings.runtime.cudaMemset3D(cudaPitchedPtr pitchedDevPtr: cudaPitchedPtr, int value, cudaExtent extent: cudaExtent)¶

Initializes or sets device memory to a value.

Initializes each element of a 3D array to the specified value value. The object to initialize is defined by pitchedDevPtr. The pitch field of pitchedDevPtr is the width in memory in bytes of the 3D array pointed to by pitchedDevPtr, including any padding added to the end of each row. The xsize field specifies the logical width of each row in bytes, while the ysize field specifies the height of each 2D slice in rows. The pitch field of pitchedDevPtr is ignored when height and depth are both equal to 1.

The extents of the initialized region are specified as a width in bytes, a height in rows, and a depth in slices.

Extents with width greater than or equal to the xsize of pitchedDevPtr may perform significantly faster than extents narrower than the xsize. Secondarily, extents with height equal to the ysize of pitchedDevPtr will perform faster than when the height is shorter than the ysize.

This function performs fastest when the pitchedDevPtr has been allocated by cudaMalloc3D().

Note that this function is asynchronous with respect to the host unless pitchedDevPtr refers to pinned host memory.

Parameters:

pitchedDevPtr (cudaPitchedPtr) – Pointer to pitched device memory
value (int) – Value to set for each byte of specified memory
extent (cudaExtent) – Size parameters for where to set device memory (width field in bytes)

Returns:

Return type:

cuda.bindings.runtime.cudaMemsetAsync(devPtr, int value, size_t count, stream)¶

Initializes or sets device memory to a value.

Fills the first count bytes of the memory area pointed to by devPtr with the constant byte value value.

cudaMemsetAsync() is asynchronous with respect to the host, so the call may return before the memset is complete. The operation can optionally be associated to a stream by passing a non-zero stream argument. If stream is non-zero, the operation may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

devPtr (Any) – Pointer to device memory
value (int) – Value to set for each byte of specified memory
count (size_t) – Size in bytes to set
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cuda.bindings.runtime.cudaMemset2DAsync(devPtr, size_t pitch, int value, size_t width, size_t height, stream)¶

Initializes or sets device memory to a value.

cudaMemset2DAsync() is asynchronous with respect to the host, so the call may return before the memset is complete. The operation can optionally be associated to a stream by passing a non- zero stream argument. If stream is non-zero, the operation may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

devPtr (Any) – Pointer to 2D device memory
pitch (size_t) – Pitch in bytes of 2D device memory(Unused if height is 1)
value (int) – Value to set for each byte of specified memory
width (size_t) – Width of matrix set (columns in bytes)
height (size_t) – Height of matrix set (rows)
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cuda.bindings.runtime.cudaMemset3DAsync(cudaPitchedPtr pitchedDevPtr: cudaPitchedPtr, int value, cudaExtent extent: cudaExtent, stream)¶

Initializes or sets device memory to a value.

The extents of the initialized region are specified as a width in bytes, a height in rows, and a depth in slices.

This function performs fastest when the pitchedDevPtr has been allocated by cudaMalloc3D().

cudaMemset3DAsync() is asynchronous with respect to the host, so the call may return before the memset is complete. The operation can optionally be associated to a stream by passing a non- zero stream argument. If stream is non-zero, the operation may overlap with operations in other streams.

The device version of this function only handles device to device copies and cannot be given local or shared pointers.

Parameters:

pitchedDevPtr (cudaPitchedPtr) – Pointer to pitched device memory
value (int) – Value to set for each byte of specified memory
extent (cudaExtent) – Size parameters for where to set device memory (width field in bytes)
stream (CUstream or cudaStream_t) – Stream identifier

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

cuda.bindings.runtime.cudaMemPrefetchAsync(devPtr, size_t count, int dstDevice, stream)¶

Prefetches memory to the specified destination device.

Prefetches memory to the specified destination device. devPtr is the base device pointer of the memory to be prefetched and dstDevice is the destination device. count specifies the number of bytes to copy. stream is the stream in which the operation is enqueued. The memory range must refer to managed memory allocated via cudaMallocManaged or declared via managed variables, or it may also refer to system-allocated memory on systems with non-zero cudaDevAttrPageableMemoryAccess.

Passing in cudaCpuDeviceId for dstDevice will prefetch the data to host memory. If dstDevice is a GPU, then the device attribute cudaDevAttrConcurrentManagedAccess must be non-zero. Additionally, stream must be associated with a device that has a non- zero value for the device attribute cudaDevAttrConcurrentManagedAccess.

The start address and end address of the memory range will be rounded down and rounded up respectively to be aligned to CPU page size before the prefetch operation is enqueued in the stream.

If no physical memory has been allocated for this region, then this memory region will be populated and mapped on the destination device. If there’s insufficient memory to prefetch the desired region, the Unified Memory driver may evict pages from other cudaMallocManaged allocations to host memory in order to make room. Device memory allocated using cudaMalloc or cudaMallocArray will not be evicted.

By default, any mappings to the previous location of the migrated pages are removed and mappings for the new location are only setup on dstDevice. The exact behavior however also depends on the settings applied to this memory range via cudaMemAdvise as described below:

If cudaMemAdviseSetReadMostly was set on any subset of this memory range, then that subset will create a read-only copy of the pages on dstDevice.

If cudaMemAdviseSetPreferredLocation was called on any subset of this memory range, then the pages will be migrated to dstDevice even if dstDevice is not the preferred location of any pages in the memory range.

If cudaMemAdviseSetAccessedBy was called on any subset of this memory range, then mappings to those pages from all the appropriate processors are updated to refer to the new location if establishing such a mapping is possible. Otherwise, those mappings are cleared.

Note that this API is not required for functionality and only serves to improve performance by allowing the application to migrate data to a suitable location before it is accessed. Memory accesses to this range are always coherent and are allowed even when the data is actively being migrated.

Note that this function is asynchronous with respect to the host and all work on other devices.

Parameters:

devPtr (Any) – Pointer to be prefetched
count (size_t) – Size in bytes
dstDevice (int) – Destination device to prefetch to
stream (CUstream or cudaStream_t) – Stream to enqueue prefetch operation

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

cuda.bindings.runtime.cudaMemPrefetchAsync_v2(devPtr, size_t count, cudaMemLocation location: cudaMemLocation, unsigned int flags, stream)¶

Prefetches memory to the specified destination location.

Prefetches memory to the specified destination location. devPtr is the base device pointer of the memory to be prefetched and location specifies the destination location. count specifies the number of bytes to copy. stream is the stream in which the operation is enqueued. The memory range must refer to managed memory allocated via cudaMallocManaged or declared via managed variables, or it may also refer to system-allocated memory on systems with non-zero cudaDevAttrPageableMemoryAccess.

Specifying cudaMemLocationTypeDevice for type will prefetch memory to GPU specified by device ordinal id which must have non- zero value for the device attribute concurrentManagedAccess. Additionally, stream must be associated with a device that has a non-zero value for the device attribute concurrentManagedAccess. Specifying cudaMemLocationTypeHost as type will prefetch data to host memory. Applications can request prefetching memory to a specific host NUMA node by specifying cudaMemLocationTypeHostNuma for type and a valid host NUMA node id in id Users can also request prefetching memory to the host NUMA node closest to the current thread’s CPU by specifying cudaMemLocationTypeHostNumaCurrent for type. Note when type is etiher cudaMemLocationTypeHost OR cudaMemLocationTypeHostNumaCurrent, id will be ignored.

The start address and end address of the memory range will be rounded down and rounded up respectively to be aligned to CPU page size before the prefetch operation is enqueued in the stream.

By default, any mappings to the previous location of the migrated pages are removed and mappings for the new location are only setup on the destination location. The exact behavior however also depends on the settings applied to this memory range via cuMemAdvise as described below:

If cudaMemAdviseSetReadMostly was set on any subset of this memory range, then that subset will create a read-only copy of the pages on destination location. If however the destination location is a host NUMA node, then any pages of that subset that are already in another host NUMA node will be transferred to the destination.

If cudaMemAdviseSetPreferredLocation was called on any subset of this memory range, then the pages will be migrated to location even if location is not the preferred location of any pages in the memory range.

Note that this function is asynchronous with respect to the host and all work on other devices.

Parameters:

devPtr (Any) – Pointer to be prefetched
count (size_t) – Size in bytes
location (cudaMemLocation) – location to prefetch to
flags (unsigned int) – flags for future use, must be zero now.
stream (CUstream or cudaStream_t) – Stream to enqueue prefetch operation

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

cuda.bindings.runtime.cudaMemAdvise(devPtr, size_t count, advice: cudaMemoryAdvise, int device)¶

Advise about the usage of a given memory range.

Advise the Unified Memory subsystem about the usage pattern for the memory range starting at devPtr with a size of count bytes. The start address and end address of the memory range will be rounded down and rounded up respectively to be aligned to CPU page size before the advice is applied. The memory range must refer to managed memory allocated via cudaMallocManaged or declared via managed variables. The memory range could also refer to system-allocated pageable memory provided it represents a valid, host-accessible region of memory and all additional constraints imposed by advice as outlined below are also satisfied. Specifying an invalid system- allocated pageable memory range results in an error being returned.

The advice parameter can take the following values:

cudaMemAdviseSetReadMostly: This implies that the data is mostly going to be read from and only occasionally written to. Any read accesses from any processor to this region will create a read- only copy of at least the accessed pages in that processor’s memory. Additionally, if cudaMemPrefetchAsync is called on this region, it will create a read-only copy of the data on the destination processor. If any processor writes to this region, all copies of the corresponding page will be invalidated except for the one where the write occurred. The device argument is ignored for this advice. Note that for a page to be read-duplicated, the accessing processor must either be the CPU or a GPU that has a non- zero value for the device attribute cudaDevAttrConcurrentManagedAccess. Also, if a context is created on a device that does not have the device attribute cudaDevAttrConcurrentManagedAccess set, then read- duplication will not occur until all such contexts are destroyed. If the memory region refers to valid system-allocated pageable memory, then the accessing device must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess for a read-only copy to be created on that device. Note however that if the accessing device also has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then setting this advice will not create a read-only copy when that device accesses this memory region.
cudaMemAdviceUnsetReadMostly: Undoes the effect of cudaMemAdviceReadMostly and also prevents the Unified Memory driver from attempting heuristic read-duplication on the memory range. Any read-duplicated copies of the data will be collapsed into a single copy. The location for the collapsed copy will be the preferred location if the page has a preferred location and one of the read-duplicated copies was resident at that location. Otherwise, the location chosen is arbitrary.
cudaMemAdviseSetPreferredLocation: This advice sets the preferred location for the data to be the memory belonging to device. Passing in cudaCpuDeviceId for device sets the preferred location as host memory. If device is a GPU, then it must have a non-zero value for the device attribute cudaDevAttrConcurrentManagedAccess. Setting the preferred location does not cause data to migrate to that location immediately. Instead, it guides the migration policy when a fault occurs on that memory region. If the data is already in its preferred location and the faulting processor can establish a mapping without requiring the data to be migrated, then data migration will be avoided. On the other hand, if the data is not in its preferred location or if a direct mapping cannot be established, then it will be migrated to the processor accessing it. It is important to note that setting the preferred location does not prevent data prefetching done using cudaMemPrefetchAsync. Having a preferred location can override the page thrash detection and resolution logic in the Unified Memory driver. Normally, if a page is detected to be constantly thrashing between for example host and device memory, the page may eventually be pinned to host memory by the Unified Memory driver. But if the preferred location is set as device memory, then the page will continue to thrash indefinitely. If cudaMemAdviseSetReadMostly is also set on this memory region or any subset of it, then the policies associated with that advice will override the policies of this advice, unless read accesses from device will not result in a read-only copy being created on that device as outlined in description for the advice cudaMemAdviseSetReadMostly. If the memory region refers to valid system-allocated pageable memory, then device must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess.
cudaMemAdviseUnsetPreferredLocation: Undoes the effect of cudaMemAdviseSetPreferredLocation and changes the preferred location to none.
cudaMemAdviseSetAccessedBy: This advice implies that the data will be accessed by device. Passing in cudaCpuDeviceId for device will set the advice for the CPU. If device is a GPU, then the device attribute cudaDevAttrConcurrentManagedAccess must be non-zero. This advice does not cause data migration and has no impact on the location of the data per se. Instead, it causes the data to always be mapped in the specified processor’s page tables, as long as the location of the data permits a mapping to be established. If the data gets migrated for any reason, the mappings are updated accordingly. This advice is recommended in scenarios where data locality is not important, but avoiding faults is. Consider for example a system containing multiple GPUs with peer-to-peer access enabled, where the data located on one GPU is occasionally accessed by peer GPUs. In such scenarios, migrating data over to the other GPUs is not as important because the accesses are infrequent and the overhead of migration may be too high. But preventing faults can still help improve performance, and so having a mapping set up in advance is useful. Note that on CPU access of this data, the data may be migrated to host memory because the CPU typically cannot access device memory directly. Any GPU that had the cudaMemAdviceSetAccessedBy flag set for this data will now have its mapping updated to point to the page in host memory. If cudaMemAdviseSetReadMostly is also set on this memory region or any subset of it, then the policies associated with that advice will override the policies of this advice. Additionally, if the preferred location of this memory region or any subset of it is also device, then the policies associated with cudaMemAdviseSetPreferredLocation will override the policies of this advice. If the memory region refers to valid system- allocated pageable memory, then device must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess. Additionally, if device has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then this call has no effect.
cudaMemAdviseUnsetAccessedBy: Undoes the effect of cudaMemAdviseSetAccessedBy. Any mappings to the data from device may be removed at any time causing accesses to result in non-fatal page faults. If the memory region refers to valid system- allocated pageable memory, then device must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess. Additionally, if device has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then this call has no effect.

Parameters:

devPtr (Any) – Pointer to memory to set the advice for
count (size_t) – Size in bytes of the memory range
advice (cudaMemoryAdvise) – Advice to be applied for the specified memory range
device (int) – Device to apply the advice for

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice

cuda.bindings.runtime.cudaMemAdvise_v2(devPtr, size_t count, advice: cudaMemoryAdvise, cudaMemLocation location: cudaMemLocation)¶

Advise about the usage of a given memory range.

The advice parameter can take the following values:

cudaMemAdviseSetReadMostly: This implies that the data is mostly going to be read from and only occasionally written to. Any read accesses from any processor to this region will create a read- only copy of at least the accessed pages in that processor’s memory. Additionally, if cudaMemPrefetchAsync or cudaMemPrefetchAsync_v2 is called on this region, it will create a read-only copy of the data on the destination processor. If the target location for cudaMemPrefetchAsync_v2 is a host NUMA node and a read-only copy already exists on another host NUMA node, that copy will be migrated to the targeted host NUMA node. If any processor writes to this region, all copies of the corresponding page will be invalidated except for the one where the write occurred. If the writing processor is the CPU and the preferred location of the page is a host NUMA node, then the page will also be migrated to that host NUMA node. The location argument is ignored for this advice. Note that for a page to be read-duplicated, the accessing processor must either be the CPU or a GPU that has a non-zero value for the device attribute cudaDevAttrConcurrentManagedAccess. Also, if a context is created on a device that does not have the device attribute cudaDevAttrConcurrentManagedAccess set, then read-duplication will not occur until all such contexts are destroyed. If the memory region refers to valid system-allocated pageable memory, then the accessing device must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess for a read-only copy to be created on that device. Note however that if the accessing device also has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then setting this advice will not create a read-only copy when that device accesses this memory region.
cudaMemAdviceUnsetReadMostly: Undoes the effect of cudaMemAdviseSetReadMostly and also prevents the Unified Memory driver from attempting heuristic read-duplication on the memory range. Any read-duplicated copies of the data will be collapsed into a single copy. The location for the collapsed copy will be the preferred location if the page has a preferred location and one of the read-duplicated copies was resident at that location. Otherwise, the location chosen is arbitrary. Note: The location argument is ignored for this advice.
cudaMemAdviseSetPreferredLocation: This advice sets the preferred location for the data to be the memory belonging to location. When type is cudaMemLocationTypeHost, id is ignored and the preferred location is set to be host memory. To set the preferred location to a specific host NUMA node, applications must set type to cudaMemLocationTypeHostNuma and id must specify the NUMA ID of the host NUMA node. If type is set to cudaMemLocationTypeHostNumaCurrent, id will be ignored and the host NUMA node closest to the calling thread’s CPU will be used as the preferred location. If type is a cudaMemLocationTypeDevice, then id must be a valid device ordinal and the device must have a non-zero value for the device attribute cudaDevAttrConcurrentManagedAccess. Setting the preferred location does not cause data to migrate to that location immediately. Instead, it guides the migration policy when a fault occurs on that memory region. If the data is already in its preferred location and the faulting processor can establish a mapping without requiring the data to be migrated, then data migration will be avoided. On the other hand, if the data is not in its preferred location or if a direct mapping cannot be established, then it will be migrated to the processor accessing it. It is important to note that setting the preferred location does not prevent data prefetching done using cudaMemPrefetchAsync. Having a preferred location can override the page thrash detection and resolution logic in the Unified Memory driver. Normally, if a page is detected to be constantly thrashing between for example host and device memory, the page may eventually be pinned to host memory by the Unified Memory driver. But if the preferred location is set as device memory, then the page will continue to thrash indefinitely. If cudaMemAdviseSetReadMostly is also set on this memory region or any subset of it, then the policies associated with that advice will override the policies of this advice, unless read accesses from location will not result in a read-only copy being created on that procesor as outlined in description for the advice cudaMemAdviseSetReadMostly. If the memory region refers to valid system-allocated pageable memory, and type is cudaMemLocationTypeDevice then id must be a valid device that has a non- zero alue for the device attribute cudaDevAttrPageableMemoryAccess.
cudaMemAdviseUnsetPreferredLocation: Undoes the effect of cudaMemAdviseSetPreferredLocation and changes the preferred location to none. The location argument is ignored for this advice.
cudaMemAdviseSetAccessedBy: This advice implies that the data will be accessed by processor location. The type must be either cudaMemLocationTypeDevice with id representing a valid device ordinal or cudaMemLocationTypeHost and id will be ignored. All other location types are invalid. If id is a GPU, then the device attribute cudaDevAttrConcurrentManagedAccess must be non-zero. This advice does not cause data migration and has no impact on the location of the data per se. Instead, it causes the data to always be mapped in the specified processor’s page tables, as long as the location of the data permits a mapping to be established. If the data gets migrated for any reason, the mappings are updated accordingly. This advice is recommended in scenarios where data locality is not important, but avoiding faults is. Consider for example a system containing multiple GPUs with peer-to-peer access enabled, where the data located on one GPU is occasionally accessed by peer GPUs. In such scenarios, migrating data over to the other GPUs is not as important because the accesses are infrequent and the overhead of migration may be too high. But preventing faults can still help improve performance, and so having a mapping set up in advance is useful. Note that on CPU access of this data, the data may be migrated to host memory because the CPU typically cannot access device memory directly. Any GPU that had the cudaMemAdviseSetAccessedBy flag set for this data will now have its mapping updated to point to the page in host memory. If cudaMemAdviseSetReadMostly is also set on this memory region or any subset of it, then the policies associated with that advice will override the policies of this advice. Additionally, if the preferred location of this memory region or any subset of it is also location, then the policies associated with CU_MEM_ADVISE_SET_PREFERRED_LOCATION will override the policies of this advice. If the memory region refers to valid system- allocated pageable memory, and type is cudaMemLocationTypeDevice then device in id must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess. Additionally, if id has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then this call has no effect.
CU_MEM_ADVISE_UNSET_ACCESSED_BY: Undoes the effect of cudaMemAdviseSetAccessedBy. Any mappings to the data from location may be removed at any time causing accesses to result in non-fatal page faults. If the memory region refers to valid system- allocated pageable memory, and type is cudaMemLocationTypeDevice then device in id must have a non-zero value for the device attribute cudaDevAttrPageableMemoryAccess. Additionally, if id has a non-zero value for the device attribute cudaDevAttrPageableMemoryAccessUsesHostPageTables, then this call has no effect.

Parameters:

devPtr (Any) – Pointer to memory to set the advice for
count (size_t) – Size in bytes of the memory range
advice (cudaMemoryAdvise) – Advice to be applied for the specified memory range
location (cudaMemLocation) – location to apply the advice for

Returns:

Return type:

make_cudaExtent, make_cudaPos

cuda.bindings.runtime.cudaMemRangeGetAttribute(size_t dataSize, attribute: cudaMemRangeAttribute, devPtr, size_t count)¶

Query an attribute of a given memory range.

Query an attribute about the memory range starting at devPtr with a size of count bytes. The memory range must refer to managed memory allocated via cudaMallocManaged or declared via managed variables.

The attribute parameter can take the following values:

cudaMemRangeAttributeReadMostly: If this attribute is specified, data will be interpreted as a 32-bit integer, and dataSize must be 4. The result returned will be 1 if all pages in the given memory range have read-duplication enabled, or 0 otherwise.
cudaMemRangeAttributePreferredLocation: If this attribute is specified, data will be interpreted as a 32-bit integer, and dataSize must be 4. The result returned will be a GPU device id if all pages in the memory range have that GPU as their preferred location, or it will be cudaCpuDeviceId if all pages in the memory range have the CPU as their preferred location, or it will be cudaInvalidDeviceId if either all the pages don’t have the same preferred location or some of the pages don’t have a preferred location at all. Note that the actual location of the pages in the memory range at the time of the query may be different from the preferred location.
cudaMemRangeAttributeAccessedBy: If this attribute is specified, data will be interpreted as an array of 32-bit integers, and dataSize must be a non-zero multiple of 4. The result returned will be a list of device ids that had cudaMemAdviceSetAccessedBy set for that entire memory range. If any device does not have that advice set for the entire memory range, that device will not be included. If data is larger than the number of devices that have that advice set for that memory range, cudaInvalidDeviceId will be returned in all the extra space provided. For ex., if dataSize is 12 (i.e. data has 3 elements) and only device 0 has the advice set, then the result returned will be { 0, cudaInvalidDeviceId, cudaInvalidDeviceId }. If data is smaller than the number of devices that have that advice set, then only as many devices will be returned as can fit in the array. There is no guarantee on which specific devices will be returned, however.
cudaMemRangeAttributeLastPrefetchLocation: If this attribute is specified, data will be interpreted as a 32-bit integer, and dataSize must be 4. The result returned will be the last location to which all pages in the memory range were prefetched explicitly via cudaMemPrefetchAsync. This will either be a GPU id or cudaCpuDeviceId depending on whether the last location for prefetch was a GPU or the CPU respectively. If any page in the memory range was never explicitly prefetched or if all pages were not prefetched to the same location, cudaInvalidDeviceId will be returned. Note that this simply returns the last location that the applicaton requested to prefetch the memory range to. It gives no indication as to whether the prefetch operation to that location has completed or even begun.
- cudaMemRangeAttributePreferredLocationType: If this attribute is specified, data will be interpreted as a cudaMemLocationType, and dataSize must be sizeof(cudaMemLocationType). The cudaMemLocationType returned will be cudaMemLocationTypeDevice if all pages in the memory range have the same GPU as their preferred location, or cudaMemLocationType will be cudaMemLocationTypeHost if all pages in the memory range have the CPU as their preferred location, or or it will be cudaMemLocationTypeHostNuma if all the pages in the memory range have the same host NUMA node ID as their preferred location or it will be cudaMemLocationTypeInvalid if either all the pages don’t have the same preferred location or some of the pages don’t have a preferred location at all. Note that the actual location type of the pages in the memory range at the time of the query may be different from the preferred location type.
cudaMemRangeAttributePreferredLocationId: If this

attribute is specified, data will be interpreted as a 32-bit integer, and dataSize must be 4. If the cudaMemRangeAttributePreferredLocationType query for the same address range returns cudaMemLocationTypeDevice, it will be a valid device ordinal or if it returns cudaMemLocationTypeHostNuma, it will be a valid host NUMA node ID or if it returns any other location type, the id should be ignored.

cudaMemRangeAttributeLastPrefetchLocationType: If this attribute is specified, data will be interpreted as a cudaMemLocationType, and dataSize must be sizeof(cudaMemLocationType). The result returned will be the last location type to which all pages in the memory range were prefetched explicitly via cuMemPrefetchAsync. The cudaMemLocationType returned will be cudaMemLocationTypeDevice if the last prefetch location was the GPU or cudaMemLocationTypeHost if it was the CPU or cudaMemLocationTypeHostNuma if the last prefetch location was a specific host NUMA node. If any page in the memory range was never explicitly prefetched or if all pages were not prefetched to the same location, CUmemLocationType will be cudaMemLocationTypeInvalid. Note that this simply returns the last location type that the application requested to prefetch the memory range to. It gives no indication as to whether the prefetch operation to that location has completed or even begun.

cudaMemRangeAttributeLastPrefetchLocationId: If this

attribute is specified, data will be interpreted as a 32-bit integer, and dataSize must be 4. If the cudaMemRangeAttributeLastPrefetchLocationType query for the same address range returns cudaMemLocationTypeDevice, it will be a valid device ordinal or if it returns cudaMemLocationTypeHostNuma, it will be a valid host NUMA node ID or if it returns any other location type, the id should be ignored.

Parameters:

dataSize (size_t) – Array containing the size of data
attribute (cudaMemRangeAttribute) – The attribute to query
devPtr (Any) – Start of the range to query
count (size_t) – Size of the range to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
data (Any) – A pointers to a memory location where the result of each attribute query will be written to.

cuda.bindings.runtime.cudaMemRangeGetAttributes(dataSizes: Tuple[int] | List[int], attributes: Optional[Tuple[cudaMemRangeAttribute] | List[cudaMemRangeAttribute]], size_t numAttributes, devPtr, size_t count)¶

Query attributes of a given memory range.

Query attributes of the memory range starting at devPtr with a size of count bytes. The memory range must refer to managed memory allocated via cudaMallocManaged or declared via managed variables. The attributes array will be interpreted to have numAttributes entries. The dataSizes array will also be interpreted to have numAttributes entries. The results of the query will be stored in data.

The list of supported attributes are given below. Please refer to cudaMemRangeGetAttribute for attribute descriptions and restrictions.

cudaMemRangeAttributeReadMostly
cudaMemRangeAttributePreferredLocation
cudaMemRangeAttributeAccessedBy
cudaMemRangeAttributeLastPrefetchLocation
:: cudaMemRangeAttributePreferredLocationType
:: cudaMemRangeAttributePreferredLocationId
:: cudaMemRangeAttributeLastPrefetchLocationType
:: cudaMemRangeAttributeLastPrefetchLocationId

Parameters:

dataSizes (List[int]) – Array containing the sizes of each result
attributes (List[cudaMemRangeAttribute]) – An array of attributes to query (numAttributes and the number of attributes in this array should match)
numAttributes (size_t) – Number of attributes to query
devPtr (Any) – Start of the range to query
count (size_t) – Size of the range to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
data (List[Any]) – A two-dimensional array containing pointers to memory locations where the result of each attribute query will be written to.

cuda.bindings.runtime.make_cudaPitchedPtr(d, size_t p, size_t xsz, size_t ysz)¶

Returns a cudaPitchedPtr based on input parameters.

Returns a cudaPitchedPtr based on the specified input parameters d, p, xsz, and ysz.

Parameters:

d (Any) – Pointer to allocated memory
p (size_t) – Pitch of allocated memory in bytes
xsz (size_t) – Logical width of allocation in elements
ysz (size_t) – Logical height of allocation in elements

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
cudaPitchedPtr – cudaPitchedPtr specified by d, p, xsz, and ysz

See also

cuda.bindings.runtime.make_cudaPos(size_t x, size_t y, size_t z)¶

Returns a cudaPos based on input parameters.

Returns a cudaPos based on the specified input parameters x, y, and z.

Parameters:

x (size_t) – X position
y (size_t) – Y position
z (size_t) – Z position

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
cudaPos – cudaPos specified by x, y, and z

cuda.bindings.runtime.make_cudaExtent(size_t w, size_t h, size_t d)¶

Returns a cudaExtent based on input parameters.

Returns a cudaExtent based on the specified input parameters w, h, and d.

Parameters:

w (size_t) – Width in elements when referring to array memory, in bytes when referring to linear memory
h (size_t) – Height in elements
d (size_t) – Depth in elements

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
cudaExtent – cudaExtent specified by w, h, and d

See also

make_cudaPitchedPtr, make_cudaPos

Stream Ordered Memory Allocator¶

overview

The asynchronous allocator allows the user to allocate and free in stream order. All asynchronous accesses of the allocation must happen between the stream executions of the allocation and the free. If the memory is accessed outside of the promised stream order, a use before allocation / use after free error will cause undefined behavior.

The allocator is free to reallocate the memory as long as it can guarantee that compliant memory accesses will not overlap temporally. The allocator may refer to internal stream ordering as well as inter-stream dependencies (such as CUDA events and null stream dependencies) when establishing the temporal guarantee. The allocator may also insert inter-stream dependencies to establish the temporal guarantee.

Supported Platforms

Whether or not a device supports the integrated stream ordered memory allocator may be queried by calling cudaDeviceGetAttribute() with the device attribute cudaDevAttrMemoryPoolsSupported.

cuda.bindings.runtime.cudaMallocAsync(size_t size, hStream)¶

Allocates memory with stream ordered semantics.

Inserts an allocation operation into hStream. A pointer to the allocated memory is returned immediately in *dptr. The allocation must not be accessed until the the allocation operation completes. The allocation comes from the memory pool associated with the stream’s device.

Parameters:

size (size_t) – Number of bytes to allocate
hStream (CUstream or cudaStream_t) – The stream establishing the stream ordering contract and the memory pool to allocate from

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorNotSupported, cudaErrorOutOfMemory,
devPtr (Any) – Returned device pointer

Notes

The default memory pool of a device contains device memory from that device.

Basic stream ordering allows future work submitted into the same stream to use the allocation. Stream query, stream synchronize, and CUDA events can be used to guarantee that the allocation operation completes before work submitted in a separate stream runs.

During stream capture, this function results in the creation of an allocation node. In this case, the allocation is owned by the graph instead of the memory pool. The memory pool’s properties are used to set the node’s creation parameters.

cuda.bindings.runtime.cudaFreeAsync(devPtr, hStream)¶

Frees memory with stream ordered semantics.

Inserts a free operation into hStream. The allocation must not be accessed after stream execution reaches the free. After this API returns, accessing the memory from any subsequent work launched on the GPU or querying its pointer attributes results in undefined behavior.

Parameters:

dptr (Any) – memory to free
hStream (CUstream or cudaStream_t) – The stream establishing the stream ordering promise

Returns:

cudaSuccess, cudaErrorInvalidValue, cudaErrorNotSupported

Return type:

cuMemFreeAsync, cudaMallocAsync

See also

Notes

During stream capture, this function results in the creation of a free node and must therefore be passed the address of a graph allocation.

cuda.bindings.runtime.cudaMemPoolTrimTo(memPool, size_t minBytesToKeep)¶

Tries to release memory back to the OS.

Releases memory back to the OS until the pool contains fewer than minBytesToKeep reserved bytes, or there is no more memory that the allocator can safely release. The allocator cannot release OS allocations that back outstanding asynchronous allocations. The OS allocations may happen at different granularity from the user allocations.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – The memory pool to trim
minBytesToKeep (size_t) – If the pool has less than minBytesToKeep reserved, the TrimTo operation is a no-op. Otherwise the pool will be guaranteed to have at least minBytesToKeep bytes reserved after the operation.

Returns:

Return type:

Notes

: Allocations that have not been freed count as outstanding.

: Allocations that have been asynchronously freed but whose completion has not been observed on the host (eg. by a synchronize) can count as outstanding.

cuda.bindings.runtime.cudaMemPoolSetAttribute(memPool, attr: cudaMemPoolAttr, value)¶

Sets attributes of a memory pool.

Supported attributes are:

cudaMemPoolAttrReleaseThreshold: (value type = cuuint64_t) Amount of reserved memory in bytes to hold onto before trying to release memory back to the OS. When more than the release threshold bytes of memory are held by the memory pool, the allocator will try to release memory back to the OS on the next call to stream, event or context synchronize. (default 0)
cudaMemPoolReuseFollowEventDependencies: (value type = int) Allow cudaMallocAsync to use memory asynchronously freed in another stream as long as a stream ordering dependency of the allocating stream on the free action exists. Cuda events and null stream interactions can create the required stream ordered dependencies. (default enabled)
cudaMemPoolReuseAllowOpportunistic: (value type = int) Allow reuse of already completed frees when there is no dependency between the free and allocation. (default enabled)
cudaMemPoolReuseAllowInternalDependencies: (value type = int) Allow cudaMallocAsync to insert new stream dependencies in order to establish the stream ordering required to reuse a piece of memory released by cudaFreeAsync (default enabled).
cudaMemPoolAttrReservedMemHigh: (value type = cuuint64_t) Reset the high watermark that tracks the amount of backing memory that was allocated for the memory pool. It is illegal to set this attribute to a non-zero value.
cudaMemPoolAttrUsedMemHigh: (value type = cuuint64_t) Reset the high watermark that tracks the amount of used memory that was allocated for the memory pool. It is illegal to set this attribute to a non-zero value.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – The memory pool to modify
attr (cudaMemPoolAttr) – The attribute to modify
value (Any) – Pointer to the value to assign

Returns:

Return type:

cuda.bindings.runtime.cudaMemPoolGetAttribute(memPool, attr: cudaMemPoolAttr)¶

Gets attributes of a memory pool.

Supported attributes are:

cudaMemPoolAttrReleaseThreshold: (value type = cuuint64_t) Amount of reserved memory in bytes to hold onto before trying to release memory back to the OS. When more than the release threshold bytes of memory are held by the memory pool, the allocator will try to release memory back to the OS on the next call to stream, event or context synchronize. (default 0)
cudaMemPoolReuseFollowEventDependencies: (value type = int) Allow cudaMallocAsync to use memory asynchronously freed in another stream as long as a stream ordering dependency of the allocating stream on the free action exists. Cuda events and null stream interactions can create the required stream ordered dependencies. (default enabled)
cudaMemPoolReuseAllowOpportunistic: (value type = int) Allow reuse of already completed frees when there is no dependency between the free and allocation. (default enabled)
cudaMemPoolReuseAllowInternalDependencies: (value type = int) Allow cudaMallocAsync to insert new stream dependencies in order to establish the stream ordering required to reuse a piece of memory released by cudaFreeAsync (default enabled).
cudaMemPoolAttrReservedMemCurrent: (value type = cuuint64_t) Amount of backing memory currently allocated for the mempool.
cudaMemPoolAttrReservedMemHigh: (value type = cuuint64_t) High watermark of backing memory allocated for the mempool since the last time it was reset.
cudaMemPoolAttrUsedMemCurrent: (value type = cuuint64_t) Amount of memory from the pool that is currently in use by the application.
cudaMemPoolAttrUsedMemHigh: (value type = cuuint64_t) High watermark of the amount of memory from the pool that was in use by the application since the last time it was reset.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – The memory pool to get attributes of
attr (cudaMemPoolAttr) – The attribute to get

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
value (Any) – Retrieved value

cuda.bindings.runtime.cudaMemPoolSetAccess(memPool, descList: Optional[Tuple[cudaMemAccessDesc] | List[cudaMemAccessDesc]], size_t count)¶

Controls visibility of pools between devices.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – The pool being modified
map (List[cudaMemAccessDesc]) – Array of access descriptors. Each descriptor instructs the access to enable for a single gpu
count (size_t) – Number of descriptors in the map array.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidDevice, cudaErrorPeerAccessAlreadyEnabled, cudaErrorInvalidValue

cuda.bindings.runtime.cudaMemPoolGetAccess(memPool, cudaMemLocation location: Optional[cudaMemLocation])¶

Returns the accessibility of a pool from a device.

Returns the accessibility of the pool’s memory from the specified location.

Parameters:

memPool (CUmemoryPool or cudaMemPool_t) – the pool being queried
location (cudaMemLocation) – the location accessing the pool

Returns:

cudaError_t
flags (cudaMemAccessFlags) – the accessibility of the pool from the specified location

cuda.bindings.runtime.cudaMemPoolCreate(cudaMemPoolProps poolProps: Optional[cudaMemPoolProps])¶

Creates a memory pool.

Creates a CUDA memory pool and returns the handle in pool. The poolProps determines the properties of the pool such as the backing device and IPC capabilities.

To create a memory pool targeting a specific host NUMA node, applications must set cudaMemPoolProps::cudaMemLocation::type to cudaMemLocationTypeHostNuma and cudaMemPoolProps::cudaMemLocation::id must specify the NUMA ID of the host memory node. Specifying cudaMemLocationTypeHostNumaCurrent or cudaMemLocationTypeHost as the cudaMemPoolProps::cudaMemLocation::type will result in cudaErrorInvalidValue. By default, the pool’s memory will be accessible from the device it is allocated on. In the case of pools created with cudaMemLocationTypeHostNuma, their default accessibility will be from the host CPU. Applications can control the maximum size of the pool by specifying a non-zero value for maxSize. If set to 0, the maximum size of the pool will default to a system dependent value.

Applications can set handleTypes to cudaMemHandleTypeFabric in order to create cudaMemPool_t suitable for sharing within an IMEX domain. An IMEX domain is either an OS instance or a group of securely connected OS instances using the NVIDIA IMEX daemon. An IMEX channel is a global resource within the IMEX domain that represents a logical entity that aims to provide fine grained accessibility control for the participating processes. When exporter and importer CUDA processes have been granted access to the same IMEX channel, they can securely share memory. If the allocating process does not have access setup for an IMEX channel, attempting to export a CUmemoryPool with cudaMemHandleTypeFabric will result in cudaErrorNotPermitted. The nvidia-modprobe CLI provides more information regarding setting up of IMEX channels.

Parameters:

poolProps (cudaMemPoolProps) – None

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorNotSupported
memPool (cudaMemPool_t) – None

Notes

Specifying cudaMemHandleTypeNone creates a memory pool that will not support IPC.

cuda.bindings.runtime.cudaMemPoolDestroy(memPool)¶

Destroys the specified memory pool.

If any pointers obtained from this pool haven’t been freed or the pool has free operations that haven’t completed when cudaMemPoolDestroy is invoked, the function will return immediately and the resources associated with the pool will be released automatically once there are no more outstanding allocations.

Destroying the current mempool of a device sets the default mempool of that device as the current mempool for that device.

Parameters:: memPool (CUmemoryPool or cudaMemPool_t) – None
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

Notes

A device’s default memory pool cannot be destroyed.

cuda.bindings.runtime.cudaMallocFromPoolAsync(size_t size, memPool, stream)¶

Allocates memory from a specified pool with stream ordered semantics.

Parameters:

bytesize (size_t) – Number of bytes to allocate
memPool (CUmemoryPool or cudaMemPool_t) – The pool to allocate from
stream (CUstream or cudaStream_t) – The stream establishing the stream ordering semantic

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorNotSupported, cudaErrorOutOfMemory
ptr (Any) – Returned device pointer

Notes

cuda.bindings.runtime.cudaMemPoolExportToShareableHandle(memPool, handleType: cudaMemAllocationHandleType, unsigned int flags)¶

Exports a memory pool to the requested handle type.

Given an IPC capable mempool, create an OS handle to share the pool with another process. A recipient process can convert the shareable handle into a mempool with cudaMemPoolImportFromShareableHandle. Individual pointers can then be shared with the cudaMemPoolExportPointer and cudaMemPoolImportPointer APIs. The implementation of what the shareable handle is and how it can be transferred is defined by the requested handle type.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – pool to export
handleType (cudaMemAllocationHandleType) – the type of handle to create
flags (unsigned int) – must be 0

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorOutOfMemory
handle_out (Any) – pointer to the location in which to store the requested handle

Notes

: To create an IPC capable mempool, create a mempool with a CUmemAllocationHandleType other than cudaMemHandleTypeNone.

cuda.bindings.runtime.cudaMemPoolImportFromShareableHandle(shareableHandle, handleType: cudaMemAllocationHandleType, unsigned int flags)¶

imports a memory pool from a shared handle.

Specific allocations can be imported from the imported pool with cudaMemPoolImportPointer.

Parameters:

handle (Any) – OS handle of the pool to open
handleType (cudaMemAllocationHandleType) – The type of handle being imported
flags (unsigned int) – must be 0

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorOutOfMemory
pool_out (cudaMemPool_t) – Returned memory pool

Notes

Imported memory pools do not support creating new allocations. As such imported memory pools may not be used in cudaDeviceSetMemPool or cudaMallocFromPoolAsync calls.

cuda.bindings.runtime.cudaMemPoolExportPointer(ptr)¶

Export data to share a memory pool allocation between processes.

Constructs shareData_out for sharing a specific allocation from an already shared memory pool. The recipient process can import the allocation with the cudaMemPoolImportPointer api. The data is not a handle and may be shared through any IPC mechanism.

Parameters:

ptr (Any) – pointer to memory being exported

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorOutOfMemory
shareData_out (cudaMemPoolPtrExportData) – Returned export data

cuda.bindings.runtime.cudaMemPoolImportPointer(memPool, cudaMemPoolPtrExportData exportData: Optional[cudaMemPoolPtrExportData])¶

Import a memory pool allocation from another process.

Returns in ptr_out a pointer to the imported memory. The imported memory must not be accessed before the allocation operation completes in the exporting process. The imported memory must be freed from all importing processes before being freed in the exporting process. The pointer may be freed with cudaFree or cudaFreeAsync. If cudaFreeAsync is used, the free must be completed on the importing process before the free operation on the exporting process.

Parameters:

pool (CUmemoryPool or cudaMemPool_t) – pool from which to import
shareData (cudaMemPoolPtrExportData) – data specifying the memory to import

Returns:

cudaError_t – CUDA_SUCCESS, CUDA_ERROR_INVALID_VALUE, CUDA_ERROR_NOT_INITIALIZED, CUDA_ERROR_OUT_OF_MEMORY
ptr_out (Any) – pointer to imported memory

Notes

The cudaFreeAsync api may be used in the exporting process before the cudaFreeAsync operation completes in its stream as long as the cudaFreeAsync in the exporting process specifies a stream with a stream dependency on the importing process’s cudaFreeAsync.

Unified Addressing¶

This section describes the unified addressing functions of the CUDA runtime application programming interface.

Overview

CUDA devices can share a unified address space with the host.

For these devices there is no distinction between a device pointer and a host pointer – the same pointer value may be used to access memory from the host program and from a kernel running on the device (with exceptions enumerated below).

Supported Platforms

Whether or not a device supports unified addressing may be queried by calling cudaGetDeviceProperties() with the device property cudaDeviceProp::unifiedAddressing.

Unified addressing is automatically enabled in 64-bit processes .

Looking Up Information from Pointer Values

It is possible to look up information about the memory which backs a pointer value. For instance, one may want to know if a pointer points to host or device memory. As another example, in the case of device memory, one may want to know on which CUDA device the memory resides. These properties may be queried using the function cudaPointerGetAttributes()

Since pointers are unique, it is not necessary to specify information about the pointers specified to cudaMemcpy() and other copy functions.

The copy direction cudaMemcpyDefault may be used to specify that the CUDA runtime should infer the location of the pointer from its value.

Automatic Mapping of Host Allocated Host Memory

All host memory allocated through all devices using cudaMallocHost() and cudaHostAlloc() is always directly accessible from all devices that support unified addressing. This is the case regardless of whether or not the flags cudaHostAllocPortable and cudaHostAllocMapped are specified.

The pointer value through which allocated host memory may be accessed in kernels on all devices that support unified addressing is the same as the pointer value through which that memory is accessed on the host. It is not necessary to call cudaHostGetDevicePointer() to get the device pointer for these allocations.

Note that this is not the case for memory allocated using the flag cudaHostAllocWriteCombined, as discussed below.

Direct Access of Peer Memory

Upon enabling direct access from a device that supports unified addressing to another peer device that supports unified addressing using cudaDeviceEnablePeerAccess() all memory allocated in the peer device using cudaMalloc() and cudaMallocPitch() will immediately be accessible by the current device. The device pointer value through which any peer’s memory may be accessed in the current device is the same pointer value through which that memory may be accessed from the peer device.

Exceptions, Disjoint Addressing

Not all memory may be accessed on devices through the same pointer value through which they are accessed on the host. These exceptions are host memory registered using cudaHostRegister() and host memory allocated using the flag cudaHostAllocWriteCombined. For these exceptions, there exists a distinct host and device address for the memory. The device address is guaranteed to not overlap any valid host pointer range and is guaranteed to have the same value across all devices that support unified addressing.

This device address may be queried using cudaHostGetDevicePointer() when a device using unified addressing is current. Either the host or the unified device pointer value may be used to refer to this memory in cudaMemcpy() and similar functions using the cudaMemcpyDefault memory direction.

cuda.bindings.runtime.cudaPointerGetAttributes(ptr)¶

Returns attributes about a specified pointer.

Returns in *attributes the attributes of the pointer ptr. If pointer was not allocated in, mapped by or registered with context supporting unified addressing cudaErrorInvalidValue is returned.

The cudaPointerAttributes structure is defined as:

View CUDA Toolkit Documentation for a C++ code example

In this structure, the individual fields mean

type identifies type of memory. It can be cudaMemoryTypeUnregistered for unregistered host memory, cudaMemoryTypeHost for registered host memory, cudaMemoryTypeDevice for device memory or cudaMemoryTypeManaged for managed memory.
device is the device against which ptr was allocated. If ptr has memory type cudaMemoryTypeDevice then this identifies the device on which the memory referred to by ptr physically resides. If ptr has memory type cudaMemoryTypeHost then this identifies the device which was current when the allocation was made (and if that device is deinitialized then this allocation will vanish with that device’s state).
devicePointer is the device pointer alias through which the memory referred to by ptr may be accessed on the current device. If the memory referred to by ptr cannot be accessed directly by the current device then this is NULL.
hostPointer is the host pointer alias through which the memory referred to by ptr may be accessed on the host. If the memory referred to by ptr cannot be accessed directly by the host then this is NULL.

Parameters:

ptr (Any) – Pointer to get attributes for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue
attributes (cudaPointerAttributes) – Attributes for the specified pointer

Notes

In CUDA 11.0 forward passing host pointer will return cudaMemoryTypeUnregistered in type and call will return cudaSuccess.

Peer Device Memory Access¶

This section describes the peer device memory access functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaDeviceCanAccessPeer(int device, int peerDevice)¶

Queries if a device may directly access a peer device’s memory.

Returns in *canAccessPeer a value of 1 if device device is capable of directly accessing memory from peerDevice and 0 otherwise. If direct access of peerDevice from device is possible, then access may be enabled by calling cudaDeviceEnablePeerAccess().

Parameters:

device (int) – Device from which allocations on peerDevice are to be directly accessed.
peerDevice (int) – Device on which the allocations to be directly accessed by device reside.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice
canAccessPeer (int) – Returned access capability

cuda.bindings.runtime.cudaDeviceEnablePeerAccess(int peerDevice, unsigned int flags)¶

Enables direct access to memory allocations on a peer device.

On success, all allocations from peerDevice will immediately be accessible by the current device. They will remain accessible until access is explicitly disabled using cudaDeviceDisablePeerAccess() or either device is reset using cudaDeviceReset().

Note that access granted by this call is unidirectional and that in order to access memory on the current device from peerDevice, a separate symmetric call to cudaDeviceEnablePeerAccess() is required.

Note that there are both device-wide and system-wide limitations per system configuration, as noted in the CUDA Programming Guide under the section “Peer-to-Peer Memory Access”.

Returns cudaErrorInvalidDevice if cudaDeviceCanAccessPeer() indicates that the current device cannot directly access memory from peerDevice.

Returns cudaErrorPeerAccessAlreadyEnabled if direct access of peerDevice from the current device has already been enabled.

Returns cudaErrorInvalidValue if flags is not 0.

Parameters:

peerDevice (int) – Peer device to enable direct access to from the current device
flags (unsigned int) – Reserved for future use and must be set to 0

Returns:

Return type:

cudaSuccess, cudaErrorInvalidDevice, cudaErrorSetOnActiveProcess

cuda.bindings.runtime.cudaDeviceDisablePeerAccess(int peerDevice)¶

Disables direct access to memory allocations on a peer device.

Returns cudaErrorPeerAccessNotEnabled if direct access to memory on peerDevice has not yet been enabled from the current device.

Parameters:: peerDevice (int) – Peer device to disable direct access to
Returns:: cudaSuccess, cudaErrorPeerAccessNotEnabled, cudaErrorInvalidDevice
Return type:: cudaError_t

OpenGL Interoperability¶

impl_private

This section describes the OpenGL interoperability functions of the CUDA runtime application programming interface. Note that mapping of OpenGL resources is performed with the graphics API agnostic, resource mapping interface described in Graphics Interopability.

class cuda.bindings.runtime.cudaGLDeviceList(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)¶

CUDA devices corresponding to the current OpenGL context

cudaGLDeviceListAll = 1¶: The CUDA devices for all GPUs used by the current OpenGL context

cudaGLDeviceListCurrentFrame = 2¶: The CUDA devices for the GPUs used by the current OpenGL context in its currently rendering frame

cudaGLDeviceListNextFrame = 3¶: The CUDA devices for the GPUs to be used by the current OpenGL context in the next frame

cuda.bindings.runtime.cudaGLGetDevices(unsigned int cudaDeviceCount, deviceList: cudaGLDeviceList)¶

Gets the CUDA devices associated with the current OpenGL context.

Returns in *pCudaDeviceCount the number of CUDA-compatible devices corresponding to the current OpenGL context. Also returns in *pCudaDevices at most cudaDeviceCount of the CUDA-compatible devices corresponding to the current OpenGL context. If any of the GPUs being used by the current OpenGL context are not CUDA capable then the call will return cudaErrorNoDevice.

Parameters:

cudaDeviceCount (unsigned int) – The size of the output device array pCudaDevices
deviceList (cudaGLDeviceList) – The set of devices to return. This set may be cudaGLDeviceListAll for all devices, cudaGLDeviceListCurrentFrame for the devices used to render the current frame (in SLI), or cudaGLDeviceListNextFrame for the devices used to render the next frame (in SLI).

Returns:

cudaError_t – cudaSuccess cudaErrorNoDevice cudaErrorInvalidGraphicsContext cudaErrorUnknown
pCudaDeviceCount (unsigned int) – Returned number of CUDA devices corresponding to the current OpenGL context
pCudaDevices (List[int]) – Returned CUDA devices corresponding to the current OpenGL context

Notes

This function is not supported on Mac OS X.

cuda.bindings.runtime.cudaGraphicsGLRegisterImage(image, target, unsigned int flags)¶

Registers the texture or renderbuffer object specified by image for access by CUDA. A handle to the registered object is returned as resource.

target must match the type of the object, and must be one of GL_TEXTURE_2D, GL_TEXTURE_RECTANGLE, GL_TEXTURE_CUBE_MAP, GL_TEXTURE_3D, GL_TEXTURE_2D_ARRAY, or GL_RENDERBUFFER.

The register flags flags specify the intended usage, as follows:

cudaGraphicsRegisterFlagsNone: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA. This is the default value.
cudaGraphicsRegisterFlagsReadOnly: Specifies that CUDA will not write to this resource.
cudaGraphicsRegisterFlagsWriteDiscard: Specifies that CUDA will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved.
cudaGraphicsRegisterFlagsSurfaceLoadStore: Specifies that CUDA will bind this resource to a surface reference.
cudaGraphicsRegisterFlagsTextureGather: Specifies that CUDA will perform texture gather operations on this resource.

The following image formats are supported. For brevity’s sake, the list is abbreviated. For ex., {GL_R, GL_RG} X {8, 16} would expand to the following 4 formats {GL_R8, GL_R16, GL_RG8, GL_RG16} :

GL_RED, GL_RG, GL_RGBA, GL_LUMINANCE, GL_ALPHA, GL_LUMINANCE_ALPHA, GL_INTENSITY
{GL_R, GL_RG, GL_RGBA} X {8, 16, 16F, 32F, 8UI, 16UI, 32UI, 8I, 16I, 32I}
{GL_LUMINANCE, GL_ALPHA, GL_LUMINANCE_ALPHA, GL_INTENSITY} X {8, 16, 16F_ARB, 32F_ARB, 8UI_EXT, 16UI_EXT, 32UI_EXT, 8I_EXT, 16I_EXT, 32I_EXT}

The following image classes are currently disallowed:

Textures with borders
Multisampled renderbuffers

Parameters:

image (GLuint) – name of texture or renderbuffer object to be registered
target (GLenum) – Identifies the type of object specified by image
flags (unsigned int) – Register flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorOperatingSystem, cudaErrorUnknown
resource (cudaGraphicsResource) – Pointer to the returned object handle

cuda.bindings.runtime.cudaGraphicsGLRegisterBuffer(buffer, unsigned int flags)¶

Registers an OpenGL buffer object.

Registers the buffer object specified by buffer for access by CUDA. A handle to the registered object is returned as resource. The register flags flags specify the intended usage, as follows:

cudaGraphicsRegisterFlagsNone: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA. This is the default value.
cudaGraphicsRegisterFlagsReadOnly: Specifies that CUDA will not write to this resource.
cudaGraphicsRegisterFlagsWriteDiscard: Specifies that CUDA will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved.

Parameters:

buffer (GLuint) – name of buffer object to be registered
flags (unsigned int) – Register flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorOperatingSystem, cudaErrorUnknown
resource (cudaGraphicsResource) – Pointer to the returned object handle

Direct3D 9 Interoperability¶

Direct3D 10 Interoperability¶

Direct3D 11 Interoperability¶

VDPAU Interoperability¶

This section describes the VDPAU interoperability functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaVDPAUGetDevice(vdpDevice, vdpGetProcAddress)¶

Gets the CUDA device associated with a VdpDevice.

Returns the CUDA device associated with a VdpDevice, if applicable.

Parameters:

vdpDevice (VdpDevice) – A VdpDevice handle
vdpGetProcAddress (VdpGetProcAddress) – VDPAU’s VdpGetProcAddress function pointer

Returns:

cudaError_t – cudaSuccess
device (int) – Returns the device associated with vdpDevice, or -1 if the device associated with vdpDevice is not a compute device.

cuda.bindings.runtime.cudaVDPAUSetVDPAUDevice(int device, vdpDevice, vdpGetProcAddress)¶

Sets a CUDA device to use VDPAU interoperability.

Records vdpDevice as the VdpDevice for VDPAU interoperability with the CUDA device device and sets device as the current device for the calling host thread.

This function will immediately initialize the primary context on device if needed.

If device has already been initialized then this call will fail with the error cudaErrorSetOnActiveProcess. In this case it is necessary to reset device using cudaDeviceReset() before VDPAU interoperability on device may be enabled.

Parameters:

device (int) – Device to use for VDPAU interoperability
vdpDevice (VdpDevice) – The VdpDevice to interoperate with
vdpGetProcAddress (VdpGetProcAddress) – VDPAU’s VdpGetProcAddress function pointer

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown, cudaErrorLaunchTimeout

cuda.bindings.runtime.cudaGraphicsVDPAURegisterVideoSurface(vdpSurface, unsigned int flags)¶

Registers the VdpVideoSurface specified by vdpSurface for access by CUDA. A handle to the registered object is returned as resource. The surface’s intended usage is specified using flags, as follows:

cudaGraphicsMapFlagsNone: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA. This is the default value.
cudaGraphicsMapFlagsReadOnly: Specifies that CUDA will not write to this resource.
cudaGraphicsMapFlagsWriteDiscard: Specifies that CUDA will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved.

Parameters:

vdpSurface (VdpVideoSurface) – VDPAU object to be registered
flags (unsigned int) – Map flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorUnknown
resource (cudaGraphicsResource) – Pointer to the returned object handle

cuda.bindings.runtime.cudaGraphicsVDPAURegisterOutputSurface(vdpSurface, unsigned int flags)¶

Registers the VdpOutputSurface specified by vdpSurface for access by CUDA. A handle to the registered object is returned as resource. The surface’s intended usage is specified using flags, as follows:

cudaGraphicsMapFlagsNone: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA. This is the default value.
cudaGraphicsMapFlagsReadOnly: Specifies that CUDA will not write to this resource.
cudaGraphicsMapFlagsWriteDiscard: Specifies that CUDA will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved.

Parameters:

vdpSurface (VdpOutputSurface) – VDPAU object to be registered
flags (unsigned int) – Map flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorUnknown
resource (cudaGraphicsResource) – Pointer to the returned object handle

EGL Interoperability¶

This section describes the EGL interoperability functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaGraphicsEGLRegisterImage(image, unsigned int flags)¶

Registers an EGL image.

Registers the EGLImageKHR specified by image for access by CUDA. A handle to the registered object is returned as pCudaResource. Additional Mapping/Unmapping is not required for the registered resource and cudaGraphicsResourceGetMappedEglFrame can be directly called on the pCudaResource.

The application will be responsible for synchronizing access to shared objects. The application must ensure that any pending operation which access the objects have completed before passing control to CUDA. This may be accomplished by issuing and waiting for glFinish command on all GLcontexts (for OpenGL and likewise for other APIs). The application will be also responsible for ensuring that any pending operation on the registered CUDA resource has completed prior to executing subsequent commands in other APIs accesing the same memory objects. This can be accomplished by calling cuCtxSynchronize or cuEventSynchronize (preferably).

The surface’s intended usage is specified using flags, as follows:

cudaGraphicsRegisterFlagsNone: Specifies no hints about how this resource will be used. It is therefore assumed that this resource will be read from and written to by CUDA. This is the default value.
cudaGraphicsRegisterFlagsReadOnly: Specifies that CUDA will not write to this resource.
cudaGraphicsRegisterFlagsWriteDiscard: Specifies that CUDA will not read from this resource and will write over the entire contents of the resource, so none of the data previously stored in the resource will be preserved.

The EGLImageKHR is an object which can be used to create EGLImage target resource. It is defined as a void pointer. typedef void* EGLImageKHR

Parameters:

image (EGLImageKHR) – An EGLImageKHR image which can be used to create target resource.
flags (unsigned int) – Map flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorInvalidValue, cudaErrorUnknown
pCudaResource (cudaGraphicsResource) – Pointer to the returned object handle

cuda.bindings.runtime.cudaEGLStreamConsumerConnect(eglStream)¶

Connect CUDA to EGLStream as a consumer.

Connect CUDA as a consumer to EGLStreamKHR specified by eglStream.

The EGLStreamKHR is an EGL object that transfers a sequence of image frames from one API to another.

Parameters:

eglStream (EGLStreamKHR) – EGLStreamKHR handle

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
conn (cudaEglStreamConnection) – Pointer to the returned connection handle

cuda.bindings.runtime.cudaEGLStreamConsumerConnectWithFlags(eglStream, unsigned int flags)¶

Connect CUDA to EGLStream as a consumer with given flags.

Connect CUDA as a consumer to EGLStreamKHR specified by stream with specified flags defined by cudaEglResourceLocationFlags.

The flags specify whether the consumer wants to access frames from system memory or video memory. Default is cudaEglResourceLocationVidmem.

Parameters:

eglStream (EGLStreamKHR) – EGLStreamKHR handle
flags (unsigned int) – Flags denote intended location - system or video.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
conn (cudaEglStreamConnection) – Pointer to the returned connection handle

cuda.bindings.runtime.cudaEGLStreamConsumerDisconnect(conn)¶

Disconnect CUDA as a consumer to EGLStream .

Disconnect CUDA as a consumer to EGLStreamKHR.

Parameters:: conn (cudaEglStreamConnection) – Conection to disconnect.
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
Return type:: cudaError_t

cuda.bindings.runtime.cudaEGLStreamConsumerAcquireFrame(conn, pCudaResource, pStream, unsigned int timeout)¶

Acquire an image frame from the EGLStream with CUDA as a consumer.

Acquire an image frame from EGLStreamKHR. cudaGraphicsResourceGetMappedEglFrame can be called on pCudaResource to get cudaEglFrame.

Parameters:

conn (cudaEglStreamConnection) – Connection on which to acquire
pCudaResource (cudaGraphicsResource_t) – CUDA resource on which the EGLStream frame will be mapped for use.
pStream (cudaStream_t) – CUDA stream for synchronization and any data migrations implied by cudaEglResourceLocationFlags.
timeout (unsigned int) – Desired timeout in usec.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown

cuda.bindings.runtime.cudaEGLStreamConsumerReleaseFrame(conn, pCudaResource, pStream)¶

Releases the last frame acquired from the EGLStream.

Release the acquired image frame specified by pCudaResource to EGLStreamKHR.

Parameters:

conn (cudaEglStreamConnection) – Connection on which to release
pCudaResource (cudaGraphicsResource_t) – CUDA resource whose corresponding frame is to be released
pStream (cudaStream_t) – CUDA stream on which release will be done.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown

cuda.bindings.runtime.cudaEGLStreamProducerConnect(eglStream, width, height)¶

Connect CUDA to EGLStream as a producer.

Connect CUDA as a producer to EGLStreamKHR specified by stream.

The EGLStreamKHR is an EGL object that transfers a sequence of image frames from one API to another.

Parameters:

eglStream (EGLStreamKHR) – EGLStreamKHR handle
width (EGLint) – width of the image to be submitted to the stream
height (EGLint) – height of the image to be submitted to the stream

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
conn (cudaEglStreamConnection) – Pointer to the returned connection handle

cuda.bindings.runtime.cudaEGLStreamProducerDisconnect(conn)¶

Disconnect CUDA as a producer to EGLStream .

Disconnect CUDA as a producer to EGLStreamKHR.

Parameters:: conn (cudaEglStreamConnection) – Conection to disconnect.
Returns:: cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
Return type:: cudaError_t

cuda.bindings.runtime.cudaEGLStreamProducerPresentFrame(conn, cudaEglFrame eglframe: cudaEglFrame, pStream)¶

Present a CUDA eglFrame to the EGLStream with CUDA as a producer.

The cudaEglFrame is defined as:

View CUDA Toolkit Documentation for a C++ code example

For cudaEglFrame of type cudaEglFrameTypePitch, the application may present sub-region of a memory allocation. In that case, ptr will specify the start address of the sub-region in the allocation and cudaEglPlaneDesc will specify the dimensions of the sub-region.

Parameters:

conn (cudaEglStreamConnection) – Connection on which to present the CUDA array
eglframe (cudaEglFrame) – CUDA Eglstream Proucer Frame handle to be sent to the consumer over EglStream.
pStream (cudaStream_t) – CUDA stream on which to present the frame.

Returns:

Return type:

cudaSuccess, cudaErrorLaunchTimeout, cudaErrorInvalidValue, cudaErrorUnknown

cuda.bindings.runtime.cudaEGLStreamProducerReturnFrame(conn, cudaEglFrame eglframe: Optional[cudaEglFrame], pStream)¶

Return the CUDA eglFrame to the EGLStream last released by the consumer.

This API can potentially return cudaErrorLaunchTimeout if the consumer has not returned a frame to EGL stream. If timeout is returned the application can retry.

Parameters:

conn (cudaEglStreamConnection) – Connection on which to present the CUDA array
eglframe (cudaEglFrame) – CUDA Eglstream Proucer Frame handle returned from the consumer over EglStream.
pStream (cudaStream_t) – CUDA stream on which to return the frame.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorUnknown,

cuda.bindings.runtime.cudaGraphicsResourceGetMappedEglFrame(resource, unsigned int index, unsigned int mipLevel)¶

Get an eglFrame through which to access a registered EGL graphics resource.

Returns in *eglFrame an eglFrame pointer through which the registered graphics resource resource may be accessed. This API can only be called for EGL graphics resources.

The cudaEglFrame is defined as

View CUDA Toolkit Documentation for a C++ code example

Parameters:

resource (cudaGraphicsResource_t) – Registered resource to access.
index (unsigned int) – Index for cubemap surfaces.
mipLevel (unsigned int) – Mipmap level for the subresource to access.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorUnknown
eglFrame (cudaEglFrame) – Returned eglFrame.

Notes

Note that in case of multiplanar *eglFrame, pitch of only first plane (unsigned int pitch) is to be considered by the application.

cuda.bindings.runtime.cudaEventCreateFromEGLSync(eglSync, unsigned int flags)¶

Creates an event from EGLSync object.

Creates an event *phEvent from an EGLSyncKHR eglSync with the flages specified via flags. Valid flags include:

cudaEventDefault: Default event creation flag.
cudaEventBlockingSync: Specifies that the created event should use blocking synchronization. A CPU thread that uses cudaEventSynchronize() to wait on an event created with this flag will block until the event has actually been completed.

cudaEventRecord and TimingData are not supported for events created from EGLSync.

The EGLSyncKHR is an opaque handle to an EGL sync object. typedef void* EGLSyncKHR

Parameters:

eglSync (EGLSyncKHR) – Opaque handle to EGLSync object
flags (unsigned int) – Event creation flags

Returns:

cudaError_t – cudaSuccess, cudaErrorInitializationError, cudaErrorInvalidValue, cudaErrorLaunchFailure, cudaErrorMemoryAllocation
phEvent (cudaEvent_t) – Returns newly created event

Graphics Interoperability¶

This section describes the graphics interoperability functions of the CUDA runtime application programming interface.

cuda.bindings.runtime.cudaGraphicsUnregisterResource(resource)¶

Unregisters a graphics resource for access by CUDA.

Unregisters the graphics resource resource so it is not accessible by CUDA unless registered again.

If resource is invalid then cudaErrorInvalidResourceHandle is returned.

Parameters:: resource (cudaGraphicsResource_t) – Resource to unregister
Returns:: cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorUnknown
Return type:: cudaError_t

See also

cudaGraphicsD3D9RegisterResource, cudaGraphicsD3D10RegisterResource, cudaGraphicsD3D11RegisterResource, cudaGraphicsGLRegisterBuffer, cudaGraphicsGLRegisterImage, cuGraphicsUnregisterResource

cuda.bindings.runtime.cudaGraphicsResourceSetMapFlags(resource, unsigned int flags)¶

Set usage flags for mapping a graphics resource.

Set flags for mapping the graphics resource resource.

Changes to flags will take effect the next time resource is mapped. The flags argument may be any of the following:

cudaGraphicsMapFlagsNone: Specifies no hints about how resource will be used. It is therefore assumed that CUDA may read from or write to resource.
cudaGraphicsMapFlagsReadOnly: Specifies that CUDA will not write to resource.
cudaGraphicsMapFlagsWriteDiscard: Specifies CUDA will not read from resource and will write over the entire contents of resource, so none of the data previously stored in resource will be preserved.

If resource is presently mapped for access by CUDA then cudaErrorUnknown is returned. If flags is not one of the above values then cudaErrorInvalidValue is returned.

Parameters:

resource (cudaGraphicsResource_t) – Registered resource to set flags for
flags (unsigned int) – Parameters for resource mapping

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorUnknown

cuda.bindings.runtime.cudaGraphicsMapResources(int count, resources, stream)¶

Map graphics resources for access by CUDA.

Maps the count graphics resources in resources for access by CUDA.

The resources in resources may be accessed by CUDA until they are unmapped. The graphics API from which resources were registered should not access any resources while they are mapped by CUDA. If an application does so, the results are undefined.

This function provides the synchronization guarantee that any graphics calls issued before cudaGraphicsMapResources() will complete before any subsequent CUDA work issued in stream begins.

If resources contains any duplicate entries then cudaErrorInvalidResourceHandle is returned. If any of resources are presently mapped for access by CUDA then cudaErrorUnknown is returned.

Parameters:

count (int) – Number of resources to map
resources (cudaGraphicsResource_t) – Resources to map for CUDA
stream (CUstream or cudaStream_t) – Stream for synchronization

Returns:

Return type:

cudaSuccess, cudaErrorInvalidResourceHandle, cudaErrorUnknown

cuda.bindings.runtime.cudaGraphicsUnmapResources(int count, resources, stream)¶

Unmap graphics resources.

Unmaps the count graphics resources in resources.

Once unmapped, the resources in resources may not be accessed by CUDA until they are mapped again.

This function provides the synchronization guarantee that any CUDA work issued in stream before cudaGraphicsUnmapResources() will complete before any subsequently issued graphics work begins.

If resources contains any duplicate entries then cudaErrorInvalidResourceHandle is returned. If any of resources are not presently mapped for access by CUDA then cudaErrorUnknown is returned.

Parameters:

count (int) – Number of resources to unmap
resources (cudaGraphicsResource_t) – Resources to unmap
stream (CUstream or cudaStream_t) – Stream for synchronization

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorMemoryAllocation

cuda.bindings.runtime.cudaGraphicsResourceGetMappedPointer(resource)¶

Get an device pointer through which to access a mapped graphics resource.

Returns in *devPtr a pointer through which the mapped graphics resource resource may be accessed. Returns in *size the size of the memory in bytes which may be accessed from that pointer. The value set in devPtr may change every time that resource is mapped.

If resource is not a buffer then it cannot be accessed via a pointer and cudaErrorUnknown is returned. If resource is not mapped then cudaErrorUnknown is returned.

Parameters:

resource (cudaGraphicsResource_t) – None

Returns:

cudaError_t
devPtr (Any) – None
size (int) – None

cuda.bindings.runtime.cudaGraphicsSubResourceGetMappedArray(resource, unsigned int arrayIndex, unsigned int mipLevel)¶

Get an array through which to access a subresource of a mapped graphics resource.

Returns in *array an array through which the subresource of the mapped graphics resource resource which corresponds to array index arrayIndex and mipmap level mipLevel may be accessed. The value set in array may change every time that resource is mapped.

If resource is not a texture then it cannot be accessed via an array and cudaErrorUnknown is returned. If arrayIndex is not a valid array index for resource then cudaErrorInvalidValue is returned. If mipLevel is not a valid mipmap level for resource then cudaErrorInvalidValue is returned. If resource is not mapped then cudaErrorUnknown is returned.

Parameters:

resource (cudaGraphicsResource_t) – Mapped resource to access
arrayIndex (unsigned int) – Array index for array textures or cubemap face index as defined by cudaGraphicsCubeFace for cubemap textures for the subresource to access
mipLevel (unsigned int) – Mipmap level for the subresource to access

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorUnknown
array (cudaArray_t) – Returned array through which a subresource of resource may be accessed

cuda.bindings.runtime.cudaGraphicsResourceGetMappedMipmappedArray(resource)¶

Get a mipmapped array through which to access a mapped graphics resource.

Returns in *mipmappedArray a mipmapped array through which the mapped graphics resource resource may be accessed. The value set in mipmappedArray may change every time that resource is mapped.

If resource is not a texture then it cannot be accessed via an array and cudaErrorUnknown is returned. If resource is not mapped then cudaErrorUnknown is returned.

Parameters:

resource (cudaGraphicsResource_t) – Mapped resource to access

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle, cudaErrorUnknown
mipmappedArray (cudaMipmappedArray_t) – Returned mipmapped array through which resource may be accessed

Texture Object Management¶

This section describes the low level texture object management functions of the CUDA runtime application programming interface. The texture object API is only supported on devices of compute capability 3.0 or higher.

cuda.bindings.runtime.cudaGetChannelDesc(array)¶

Get the channel descriptor of an array.

Returns in *desc the channel descriptor of the CUDA array array.

Parameters:

array (cudaArray_const_t) – Memory array on device

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
desc (cudaChannelFormatDesc) – Channel format

See also

cudaCreateChannelDesc (C API), cudaCreateTextureObject, cudaCreateSurfaceObject

cuda.bindings.runtime.cudaCreateChannelDesc(int x, int y, int z, int w, f: cudaChannelFormatKind)¶

Returns a channel descriptor using the specified format.

Returns a channel descriptor with format f and number of bits of each component x, y, z, and w. The cudaChannelFormatDesc is defined as:

View CUDA Toolkit Documentation for a C++ code example

where cudaChannelFormatKind is one of cudaChannelFormatKindSigned, cudaChannelFormatKindUnsigned, or cudaChannelFormatKindFloat.

Parameters:

x (int) – X component
y (int) – Y component
z (int) – Z component
w (int) – W component
f (cudaChannelFormatKind) – Channel format

Returns:

cudaError_t.cudaSuccess – cudaError_t.cudaSuccess
cudaChannelFormatDesc – Channel descriptor with format f

cuda.bindings.runtime.cudaCreateTextureObject(cudaResourceDesc pResDesc: Optional[cudaResourceDesc], cudaTextureDesc pTexDesc: Optional[cudaTextureDesc], cudaResourceViewDesc pResViewDesc: Optional[cudaResourceViewDesc])¶

Creates a texture object.

Creates a texture object and returns it in pTexObject. pResDesc describes the data to texture from. pTexDesc describes how the data should be sampled. pResViewDesc is an optional argument that specifies an alternate format for the data described by pResDesc, and also describes the subresource region to restrict access to when texturing. pResViewDesc can only be specified if the type of resource is a CUDA array or a CUDA mipmapped array not in a block compressed format.

Texture objects are only supported on devices of compute capability 3.0 or higher. Additionally, a texture object is an opaque value, and, as such, should only be accessed through CUDA API calls.

The cudaResourceDesc structure is defined as:

View CUDA Toolkit Documentation for a C++ code example

where:

resType specifies the type of resource to texture from. CUresourceType is defined as:
View CUDA Toolkit Documentation for a C++ code example

If resType is set to cudaResourceTypeArray, cudaResourceDesc::res::array::array must be set to a valid CUDA array handle.

If resType is set to cudaResourceTypeMipmappedArray, cudaResourceDesc::res::mipmap::mipmap must be set to a valid CUDA mipmapped array handle and normalizedCoords must be set to true.

If resType is set to cudaResourceTypeLinear, cudaResourceDesc::res::linear::devPtr must be set to a valid device pointer, that is aligned to textureAlignment. cudaResourceDesc::res::linear::desc describes the format and the number of components per array element. cudaResourceDesc::res::linear::sizeInBytes specifies the size of the array in bytes. The total number of elements in the linear address range cannot exceed maxTexture1DLinear. The number of elements is computed as (sizeInBytes / sizeof(desc)).

If resType is set to cudaResourceTypePitch2D, cudaResourceDesc::res::pitch2D::devPtr must be set to a valid device pointer, that is aligned to textureAlignment. cudaResourceDesc::res::pitch2D::desc describes the format and the number of components per array element. cudaResourceDesc::res::pitch2D::width and cudaResourceDesc::res::pitch2D::height specify the width and height of the array in elements, and cannot exceed cudaResourceDesc::res::pitch2D::pitchInBytes specifies the pitch between two rows in bytes and has to be aligned to texturePitchAlignment. Pitch cannot exceed :py:obj:`~.cudaDeviceProp.maxTexture2DLinear`[2].

The cudaTextureDesc struct is defined as

View CUDA Toolkit Documentation for a C++ code example

where

addressMode specifies the addressing mode for each dimension of the texture data. cudaTextureAddressMode is defined as:
View CUDA Toolkit Documentation for a C++ code example
This is ignored if resType is cudaResourceTypeLinear. Also, if normalizedCoords is set to zero, cudaAddressModeWrap and cudaAddressModeMirror won’t be supported and will be switched to cudaAddressModeClamp.
filterMode specifies the filtering mode to be used when fetching from the texture. cudaTextureFilterMode is defined as:
View CUDA Toolkit Documentation for a C++ code example
This is ignored if resType is cudaResourceTypeLinear.
readMode specifies whether integer data should be converted to floating point or not. cudaTextureReadMode is defined as:
View CUDA Toolkit Documentation for a C++ code example
Note that this applies only to 8-bit and 16-bit integer formats. 32-bit integer format would not be promoted, regardless of whether or not this readMode is set cudaReadModeNormalizedFloat is specified.
sRGB specifies whether sRGB to linear conversion should be performed during texture fetch.
borderColor specifies the float values of color. where: addressMode is cudaAddressModeBorder.
normalizedCoords specifies whether the texture coordinates will be normalized or not.
maxAnisotropy specifies the maximum anistropy ratio to be used when doing anisotropic filtering. This value will be clamped to the range [1,16].
mipmapFilterMode specifies the filter mode when the calculated mipmap level lies between two defined mipmap levels.
mipmapLevelBias specifies the offset to be applied to the calculated mipmap level.
minMipmapLevelClamp specifies the lower end of the mipmap level range to clamp access to.
maxMipmapLevelClamp specifies the upper end of the mipmap level range to clamp access to.
disableTrilinearOptimization specifies whether the trilinear filtering optimizations will be disabled.
seamlessCubemap specifies whether seamless cube map filtering is enabled. This flag can only be specified if the underlying resource is a CUDA array or a CUDA mipmapped array that was created with the flag cudaArrayCubemap. When seamless cube map filtering is enabled, texture address modes specified by addressMode are ignored. Instead, if the filterMode is set to cudaFilterModePoint the address mode cudaAddressModeClamp will be applied for all dimensions. If the filterMode is set to cudaFilterModeLinear seamless cube map filtering will be performed when sampling along the cube face borders.

The cudaResourceViewDesc struct is defined as

View CUDA Toolkit Documentation for a C++ code example

where:

format specifies how the data contained in the CUDA array or CUDA mipmapped array should be interpreted. Note that this can incur a change in size of the texture data. If the resource view format is a block compressed format, then the underlying CUDA array or CUDA mipmapped array has to have a 32-bit unsigned integer format with 2 or 4 channels, depending on the block compressed format. For ex., BC1 and BC4 require the underlying CUDA array to have a 32-bit unsigned int with 2 channels. The other BC formats require the underlying resource to have the same 32-bit unsigned int format but with 4 channels.
width specifies the new width of the texture data. If the resource view format is a block compressed format, this value has to be 4 times the original width of the resource. For non block compressed formats, this value has to be equal to that of the original resource.
height specifies the new height of the texture data. If the resource view format is a block compressed format, this value has to be 4 times the original height of the resource. For non block compressed formats, this value has to be equal to that of the original resource.
depth specifies the new depth of the texture data. This value has to be equal to that of the original resource.
firstMipmapLevel specifies the most detailed mipmap level. This will be the new mipmap level zero. For non-mipmapped resources, this value has to be zero.:py:obj:~.cudaTextureDesc.minMipmapLevelClamp and maxMipmapLevelClamp will be relative to this value. For ex., if the firstMipmapLevel is set to 2, and a minMipmapLevelClamp of 1.2 is specified, then the actual minimum mipmap level clamp will be 3.2.
lastMipmapLevel specifies the least detailed mipmap level. For non-mipmapped resources, this value has to be zero.
firstLayer specifies the first layer index for layered textures. This will be the new layer zero. For non- layered resources, this value has to be zero.
lastLayer specifies the last layer index for layered textures. For non-layered resources, this value has to be zero.

Parameters:

pResDesc (cudaResourceDesc) – Resource descriptor
pTexDesc (cudaTextureDesc) – Texture descriptor
pResViewDesc (cudaResourceViewDesc) – Resource view descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pTexObject (cudaTextureObject_t) – Texture object to create

cuda.bindings.runtime.cudaDestroyTextureObject(texObject)¶

Destroys a texture object.

Destroys the texture object specified by texObject.

Parameters:: texObject (cudaTextureObject_t) – Texture object to destroy
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaGetTextureObjectResourceDesc(texObject)¶

Returns a texture object’s resource descriptor.

Returns the resource descriptor for the texture object specified by texObject.

Parameters:

texObject (cudaTextureObject_t) – Texture object

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pResDesc (cudaResourceDesc) – Resource descriptor

cuda.bindings.runtime.cudaGetTextureObjectTextureDesc(texObject)¶

Returns a texture object’s texture descriptor.

Returns the texture descriptor for the texture object specified by texObject.

Parameters:

texObject (cudaTextureObject_t) – Texture object

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pTexDesc (cudaTextureDesc) – Texture descriptor

cuda.bindings.runtime.cudaGetTextureObjectResourceViewDesc(texObject)¶

Returns a texture object’s resource view descriptor.

Returns the resource view descriptor for the texture object specified by texObject. If no resource view was specified, cudaErrorInvalidValue is returned.

Parameters:

texObject (cudaTextureObject_t) – Texture object

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pResViewDesc (cudaResourceViewDesc) – Resource view descriptor

Surface Object Management¶

This section describes the low level texture object management functions of the CUDA runtime application programming interface. The surface object API is only supported on devices of compute capability 3.0 or higher.

cuda.bindings.runtime.cudaCreateSurfaceObject(cudaResourceDesc pResDesc: Optional[cudaResourceDesc])¶

Creates a surface object.

Creates a surface object and returns it in pSurfObject. pResDesc describes the data to perform surface load/stores on. resType must be cudaResourceTypeArray and cudaResourceDesc::res::array::array must be set to a valid CUDA array handle.

Surface objects are only supported on devices of compute capability 3.0 or higher. Additionally, a surface object is an opaque value, and, as such, should only be accessed through CUDA API calls.

Parameters:

pResDesc (cudaResourceDesc) – Resource descriptor

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidChannelDescriptor, cudaErrorInvalidResourceHandle
pSurfObject (cudaSurfaceObject_t) – Surface object to create

cuda.bindings.runtime.cudaDestroySurfaceObject(surfObject)¶

Destroys a surface object.

Destroys the surface object specified by surfObject.

Parameters:: surfObject (cudaSurfaceObject_t) – Surface object to destroy
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaGetSurfaceObjectResourceDesc(surfObject)¶

Returns a surface object’s resource descriptor Returns the resource descriptor for the surface object specified by surfObject.

Parameters:

surfObject (cudaSurfaceObject_t) – Surface object

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pResDesc (cudaResourceDesc) – Resource descriptor

Version Management¶

cuda.bindings.runtime.cudaDriverGetVersion()¶

Returns the latest version of CUDA supported by the driver.

Returns in *driverVersion the latest version of CUDA supported by the driver. The version is returned as (1000 * major + 10 * minor). For example, CUDA 9.2 would be represented by 9020. If no driver is installed, then 0 is returned as the driver version.

This function automatically returns cudaErrorInvalidValue if driverVersion is NULL.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
driverVersion (int) – Returns the CUDA driver version.

cuda.bindings.runtime.cudaRuntimeGetVersion()¶

Returns the CUDA Runtime version.

Returns in *runtimeVersion the version number of the current CUDA Runtime instance. The version is returned as (1000 * major + 10 * minor). For example, CUDA 9.2 would be represented by 9020.

As of CUDA 12.0, this function no longer initializes CUDA. The purpose of this API is solely to return a compile-time constant stating the CUDA Toolkit version in the above format.

This function automatically returns cudaErrorInvalidValue if the runtimeVersion argument is NULL.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
runtimeVersion (int) – Returns the CUDA Runtime version.

cuda.bindings.runtime.getLocalRuntimeVersion()¶

Returns the CUDA Runtime version of local shared library.

Returns in *runtimeVersion the version number of the current CUDA Runtime instance. The version is returned as (1000 * major + 10 * minor). For example, CUDA 9.2 would be represented by 9020.

As of CUDA 12.0, this function no longer initializes CUDA. The purpose of this API is solely to return a compile-time constant stating the CUDA Toolkit version in the above format.

This function automatically returns cudaErrorInvalidValue if the runtimeVersion argument is NULL.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
runtimeVersion (int) – Returns the CUDA Runtime version.

Graph Management¶

This section describes the graph management functions of CUDA runtime application programming interface.

cuda.bindings.runtime.cudaGraphCreate(unsigned int flags)¶

Creates a graph.

Creates an empty graph, which is returned via pGraph.

Parameters:

flags (unsigned int) – Graph creation flags, must be 0

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
pGraph (cudaGraph_t) – Returns newly created graph

cuda.bindings.runtime.cudaGraphAddKernelNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaKernelNodeParams pNodeParams: Optional[cudaKernelNodeParams])¶

Creates a kernel execution node and adds it to a graph.

Creates a new kernel execution node and adds it to graph with numDependencies dependencies specified via pDependencies and arguments specified in pNodeParams. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

The cudaKernelNodeParams structure is defined as:

View CUDA Toolkit Documentation for a C++ code example

When the graph is launched, the node will invoke kernel func on a (gridDim.x x gridDim.y x gridDim.z) grid of blocks. Each block contains (blockDim.x x blockDim.y x blockDim.z) threads.

sharedMem sets the amount of dynamic shared memory that will be available to each thread block.

Kernel parameters to func can be specified in one of two ways:

1) Kernel parameters can be specified via kernelParams. If the kernel has N parameters, then kernelParams needs to be an array of N pointers. Each pointer, from `kernelParams`[0] to `kernelParams`[N-1], points to the region of memory from which the actual parameter will be copied. The number of kernel parameters and their offsets and sizes do not need to be specified as that information is retrieved directly from the kernel’s image.

2) Kernel parameters can also be packaged by the application into a single buffer that is passed in via extra. This places the burden on the application of knowing each kernel parameter’s size and alignment/padding within the buffer. The extra parameter exists to allow this function to take additional less commonly used arguments. extra specifies a list of names of extra settings and their corresponding values. Each extra setting name is immediately followed by the corresponding value. The list must be terminated with either NULL or CU_LAUNCH_PARAM_END.

CU_LAUNCH_PARAM_END, which indicates the end of the extra array;
CU_LAUNCH_PARAM_BUFFER_POINTER, which specifies that the next value in extra will be a pointer to a buffer containing all the kernel parameters for launching kernel func;
CU_LAUNCH_PARAM_BUFFER_SIZE, which specifies that the next value in extra will be a pointer to a size_t containing the size of the buffer specified with CU_LAUNCH_PARAM_BUFFER_POINTER;

The error cudaErrorInvalidValue will be returned if kernel parameters are specified with both kernelParams and extra (i.e. both kernelParams and extra are non-NULL).

The kernelParams or extra array, as well as the argument values it points to, are copied during this call.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
pNodeParams (cudaKernelNodeParams) – Parameters for the GPU execution node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction
pGraphNode (cudaGraphNode_t) – Returns newly created node

Notes

Kernels launched using graphs must not use texture and surface references. Reading or writing through any texture or surface reference is undefined behavior. This restriction does not apply to texture and surface objects.

cuda.bindings.runtime.cudaGraphKernelNodeGetParams(node)¶

Returns a kernel node’s parameters.

Returns the parameters of kernel node node in pNodeParams. The kernelParams or extra array returned in pNodeParams, as well as the argument values it points to, are owned by the node. This memory remains valid until the node is destroyed or its parameters are modified, and should not be modified directly. Use cudaGraphKernelNodeSetParams to update the parameters of this node.

The params will contain either kernelParams or extra, according to which of these was most recently set on the node.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction
pNodeParams (cudaKernelNodeParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphKernelNodeSetParams(node, cudaKernelNodeParams pNodeParams: Optional[cudaKernelNodeParams])¶

Sets a kernel node’s parameters.

Sets the parameters of kernel node node to pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
pNodeParams (cudaKernelNodeParams) – Parameters to copy

Returns:

Return type:

cuda.bindings.runtime.cudaGraphKernelNodeCopyAttributes(hSrc, hDst)¶

Copies attributes from source node to destination node.

Copies attributes from source node src to destination node dst. Both node must have the same context.

Parameters:

dst (CUgraphNode or cudaGraphNode_t) – Destination node
src (CUgraphNode or cudaGraphNode_t) – Source node For list of attributes see cudaKernelNodeAttrID

Returns:

cudaSuccess, cudaErrorInvalidContext

Return type:

See also

cuda.bindings.runtime.cudaGraphKernelNodeGetAttribute(hNode, attr: cudaKernelNodeAttrID)¶

Queries node attribute.

Queries attribute attr from node hNode and stores it in corresponding member of value_out.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t)
attr (cudaKernelNodeAttrID)

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle
value_out (cudaKernelNodeAttrValue)

See also

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidResourceHandle

cuda.bindings.runtime.cudaGraphKernelNodeSetAttribute(hNode, attr: cudaKernelNodeAttrID, cudaKernelNodeAttrValue value: Optional[cudaKernelNodeAttrValue])¶

Sets node attribute.

Sets attribute attr on node hNode from corresponding attribute of value.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t)
attr (cudaKernelNodeAttrID)
value (cudaKernelNodeAttrValue)

Returns:

Return type:

See also

cuda.bindings.runtime.cudaGraphAddMemcpyNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaMemcpy3DParms pCopyParams: Optional[cudaMemcpy3DParms])¶

Creates a memcpy node and adds it to a graph.

Creates a new memcpy node and adds it to graph with numDependencies dependencies specified via pDependencies. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

When the graph is launched, the node will perform the memcpy described by pCopyParams. See cudaMemcpy3D() for a description of the structure and its restrictions.

Memcpy nodes have some additional restrictions with regards to managed memory, if the system contains at least one device which has a zero value for the device attribute cudaDevAttrConcurrentManagedAccess.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
pCopyParams (cudaMemcpy3DParms) – Parameters for the memory copy

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphAddMemcpyNode1D(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, dst, src, size_t count, kind: cudaMemcpyKind)¶

Creates a 1D memcpy node and adds it to a graph.

Creates a new 1D memcpy node and adds it to graph with numDependencies dependencies specified via pDependencies. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

When the graph is launched, the node will copy count bytes from the memory area pointed to by src to the memory area pointed to by dst, where kind specifies the direction of the copy, and must be one of cudaMemcpyHostToHost, cudaMemcpyHostToDevice, cudaMemcpyDeviceToHost, cudaMemcpyDeviceToDevice, or cudaMemcpyDefault. Passing cudaMemcpyDefault is recommended, in which case the type of transfer is inferred from the pointer values. However, cudaMemcpyDefault is only allowed on systems that support unified virtual addressing. Launching a memcpy node with dst and src pointers that do not match the direction of the copy results in an undefined behavior.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
dst (Any) – Destination memory address
src (Any) – Source memory address
count (size_t) – Size in bytes to copy
kind (cudaMemcpyKind) – Type of transfer

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphMemcpyNodeGetParams(node)¶

Returns a memcpy node’s parameters.

Returns the parameters of memcpy node node in pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pNodeParams (cudaMemcpy3DParms) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphMemcpyNodeSetParams(node, cudaMemcpy3DParms pNodeParams: Optional[cudaMemcpy3DParms])¶

Sets a memcpy node’s parameters.

Sets the parameters of memcpy node node to pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
pNodeParams (cudaMemcpy3DParms) – Parameters to copy

Returns:

Return type:

See also

cudaGraphNodeSetParams, cudaMemcpy3D, cudaGraphMemcpyNodeSetParamsToSymbol, cudaGraphMemcpyNodeSetParamsFromSymbol, cudaGraphMemcpyNodeSetParams1D, cudaGraphAddMemcpyNode, cudaGraphMemcpyNodeGetParams

cuda.bindings.runtime.cudaGraphMemcpyNodeSetParams1D(node, dst, src, size_t count, kind: cudaMemcpyKind)¶

Sets a memcpy node’s parameters to perform a 1-dimensional copy.

Sets the parameters of memcpy node node to the copy described by the provided parameters.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
dst (Any) – Destination memory address
src (Any) – Source memory address
count (size_t) – Size in bytes to copy
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddMemsetNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaMemsetParams pMemsetParams: Optional[cudaMemsetParams])¶

Creates a memset node and adds it to a graph.

Creates a new memset node and adds it to graph with numDependencies dependencies specified via pDependencies. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

The element size must be 1, 2, or 4 bytes. When the graph is launched, the node will perform the memset described by pMemsetParams.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
pMemsetParams (cudaMemsetParams) – Parameters for the memory set

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDevice
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphMemsetNodeGetParams(node)¶

Returns a memset node’s parameters.

Returns the parameters of memset node node in pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pNodeParams (cudaMemsetParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphMemsetNodeSetParams(node, cudaMemsetParams pNodeParams: Optional[cudaMemsetParams])¶

Sets a memset node’s parameters.

Sets the parameters of memset node node to pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
pNodeParams (cudaMemsetParams) – Parameters to copy

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddHostNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaHostNodeParams pNodeParams: Optional[cudaHostNodeParams])¶

Creates a host execution node and adds it to a graph.

Creates a new CPU execution node and adds it to graph with numDependencies dependencies specified via pDependencies and arguments specified in pNodeParams. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

When the graph is launched, the node will invoke the specified CPU function. Host nodes are not supported under MPS with pre-Volta GPUs.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
pNodeParams (cudaHostNodeParams) – Parameters for the host node

Returns:

cudaError_t – cudaSuccess, cudaErrorNotSupported, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphHostNodeGetParams(node)¶

Returns a host node’s parameters.

Returns the parameters of host node node in pNodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pNodeParams (cudaHostNodeParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphHostNodeSetParams(node, cudaHostNodeParams pNodeParams: Optional[cudaHostNodeParams])¶

Sets a host node’s parameters.

Sets the parameters of host node node to nodeParams.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
pNodeParams (cudaHostNodeParams) – Parameters to copy

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddChildGraphNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, childGraph)¶

Creates a child graph node and adds it to a graph.

Creates a new node which executes an embedded graph, and adds it to graph with numDependencies dependencies specified via pDependencies. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

If hGraph contains allocation or free nodes, this call will return an error.

The node executes an embedded child graph. The child graph is cloned in this call.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
childGraph (CUgraph or cudaGraph_t) – The graph to clone into this node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphChildGraphNodeGetGraph(node)¶

Gets a handle to the embedded graph of a child graph node.

Gets a handle to the embedded graph in a child graph node. This call does not clone the graph. Changes to the graph will be reflected in the node, and the node retains ownership of the graph.

Allocation and free nodes cannot be added to the returned graph. Attempting to do so will return an error.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the embedded graph for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraph (cudaGraph_t) – Location to store a handle to the graph

cuda.bindings.runtime.cudaGraphAddEmptyNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies)¶

Creates an empty node and adds it to a graph.

Creates a new node which performs no operation, and adds it to graph with numDependencies dependencies specified via pDependencies. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

An empty node performs no operation during execution, but can be used for transitive ordering. For example, a phased execution graph with 2 groups of n nodes with a barrier between them can be represented using an empty node and 2*n dependency edges, rather than no empty node and n^2 dependency edges.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphAddEventRecordNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, event)¶

Creates an event record node and adds it to a graph.

Creates a new event record node and adds it to hGraph with numDependencies dependencies specified via dependencies and event specified in event. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. dependencies may not have any duplicate entries. A handle to the new node will be returned in phGraphNode.

Each launch of the graph will record event to capture execution of the node’s dependencies.

These nodes may not be used in loops or conditionals.

Parameters:

hGraph (CUgraph or cudaGraph_t) – Graph to which to add the node
dependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
event (CUevent or cudaEvent_t) – Event for the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
phGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphEventRecordNodeGetEvent(node)¶

Returns the event associated with an event record node.

Returns the event of event record node hNode in event_out.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to get the event for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
event_out (cudaEvent_t) – Pointer to return the event

cuda.bindings.runtime.cudaGraphEventRecordNodeSetEvent(node, event)¶

Sets an event record node’s event.

Sets the event of event record node hNode to event.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to set the event for
event (CUevent or cudaEvent_t) – Event to use

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddEventWaitNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, event)¶

Creates an event wait node and adds it to a graph.

Creates a new event wait node and adds it to hGraph with numDependencies dependencies specified via dependencies and event specified in event. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. dependencies may not have any duplicate entries. A handle to the new node will be returned in phGraphNode.

The graph node will wait for all work captured in event. See cuEventRecord() for details on what is captured by an event. The synchronization will be performed efficiently on the device when applicable. event may be from a different context or device than the launch stream.

These nodes may not be used in loops or conditionals.

Parameters:

hGraph (CUgraph or cudaGraph_t) – Graph to which to add the node
dependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
event (CUevent or cudaEvent_t) – Event for the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
phGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphEventWaitNodeGetEvent(node)¶

Returns the event associated with an event wait node.

Returns the event of event wait node hNode in event_out.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to get the event for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
event_out (cudaEvent_t) – Pointer to return the event

cuda.bindings.runtime.cudaGraphEventWaitNodeSetEvent(node, event)¶

Sets an event wait node’s event.

Sets the event of event wait node hNode to event.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to set the event for
event (CUevent or cudaEvent_t) – Event to use

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddExternalSemaphoresSignalNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaExternalSemaphoreSignalNodeParams nodeParams: Optional[cudaExternalSemaphoreSignalNodeParams])¶

Creates an external semaphore signal node and adds it to a graph.

Creates a new external semaphore signal node and adds it to graph with numDependencies dependencies specified via dependencies and arguments specified in nodeParams. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. dependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

Performs a signal operation on a set of externally allocated semaphore objects when the node is launched. The operation(s) will occur after all of the node’s dependencies have completed.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
nodeParams (cudaExternalSemaphoreSignalNodeParams) – Parameters for the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphExternalSemaphoresSignalNodeGetParams(hNode)¶

Returns an external semaphore signal node’s parameters.

Returns the parameters of an external semaphore signal node hNode in params_out. The extSemArray and paramsArray returned in params_out, are owned by the node. This memory remains valid until the node is destroyed or its parameters are modified, and should not be modified directly. Use cudaGraphExternalSemaphoresSignalNodeSetParams to update the parameters of this node.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
params_out (cudaExternalSemaphoreSignalNodeParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphExternalSemaphoresSignalNodeSetParams(hNode, cudaExternalSemaphoreSignalNodeParams nodeParams: Optional[cudaExternalSemaphoreSignalNodeParams])¶

Sets an external semaphore signal node’s parameters.

Sets the parameters of an external semaphore signal node hNode to nodeParams.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
nodeParams (cudaExternalSemaphoreSignalNodeParams) – Parameters to copy

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddExternalSemaphoresWaitNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaExternalSemaphoreWaitNodeParams nodeParams: Optional[cudaExternalSemaphoreWaitNodeParams])¶

Creates an external semaphore wait node and adds it to a graph.

Creates a new external semaphore wait node and adds it to graph with numDependencies dependencies specified via dependencies and arguments specified in nodeParams. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. dependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

Performs a wait operation on a set of externally allocated semaphore objects when the node is launched. The node’s dependencies will not be launched until the wait operation has completed.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
nodeParams (cudaExternalSemaphoreWaitNodeParams) – Parameters for the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphExternalSemaphoresWaitNodeGetParams(hNode)¶

Returns an external semaphore wait node’s parameters.

Returns the parameters of an external semaphore wait node hNode in params_out. The extSemArray and paramsArray returned in params_out, are owned by the node. This memory remains valid until the node is destroyed or its parameters are modified, and should not be modified directly. Use cudaGraphExternalSemaphoresSignalNodeSetParams to update the parameters of this node.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
params_out (cudaExternalSemaphoreWaitNodeParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphExternalSemaphoresWaitNodeSetParams(hNode, cudaExternalSemaphoreWaitNodeParams nodeParams: Optional[cudaExternalSemaphoreWaitNodeParams])¶

Sets an external semaphore wait node’s parameters.

Sets the parameters of an external semaphore wait node hNode to nodeParams.

Parameters:

hNode (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
nodeParams (cudaExternalSemaphoreWaitNodeParams) – Parameters to copy

Returns:

Return type:

cudaSuccess, cudaErrorInvalidDevice

cuda.bindings.runtime.cudaGraphAddMemAllocNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaMemAllocNodeParams nodeParams: Optional[cudaMemAllocNodeParams])¶

Creates an allocation node and adds it to a graph.

Creates a new allocation node and adds it to graph with numDependencies dependencies specified via pDependencies and arguments specified in nodeParams. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

When cudaGraphAddMemAllocNode creates an allocation node, it returns the address of the allocation in nodeParams.dptr. The allocation’s address remains fixed across instantiations and launches.

If the allocation is freed in the same graph, by creating a free node using cudaGraphAddMemFreeNode, the allocation can be accessed by nodes ordered after the allocation node but before the free node. These allocations cannot be freed outside the owning graph, and they can only be freed once in the owning graph.

If the allocation is not freed in the same graph, then it can be accessed not only by nodes in the graph which are ordered after the allocation node, but also by stream operations ordered after the graph’s execution but before the allocation is freed.

Allocations which are not freed in the same graph can be freed by:

passing the allocation to cudaMemFreeAsync or cudaMemFree;
launching a graph with a free node for that allocation; or
specifying cudaGraphInstantiateFlagAutoFreeOnLaunch during instantiation, which makes each launch behave as though it called cudaMemFreeAsync for every unfreed allocation.

It is not possible to free an allocation in both the owning graph and another graph. If the allocation is freed in the same graph, a free node cannot be added to another graph. If the allocation is freed in another graph, a free node can no longer be added to the owning graph.

The following restrictions apply to graphs which contain allocation and/or memory free nodes:

Nodes and edges of the graph cannot be deleted.
The graph cannot be used in a child node.
Only one instantiation of the graph may exist at any point in time.
The graph cannot be cloned.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
nodeParams (cudaMemAllocNodeParams) – Parameters for the node

Returns:

cudaError_t – cudaSuccess, cudaErrorCudartUnloading, cudaErrorInitializationError, cudaErrorNotSupported, cudaErrorInvalidValue, cudaErrorOutOfMemory
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphMemAllocNodeGetParams(node)¶

Returns a memory alloc node’s parameters.

Returns the parameters of a memory alloc node hNode in params_out. The poolProps and accessDescs returned in params_out, are owned by the node. This memory remains valid until the node is destroyed. The returned parameters must not be modified.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
params_out (cudaMemAllocNodeParams) – Pointer to return the parameters

cuda.bindings.runtime.cudaGraphAddMemFreeNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, dptr)¶

Creates a memory free node and adds it to a graph.

Creates a new memory free node and adds it to graph with numDependencies dependencies specified via pDependencies and address specified in dptr. It is possible for numDependencies to be 0, in which case the node will be placed at the root of the graph. pDependencies may not have any duplicate entries. A handle to the new node will be returned in pGraphNode.

cudaGraphAddMemFreeNode will return cudaErrorInvalidValue if the user attempts to free:

an allocation twice in the same graph.
an address that was not returned by an allocation node.
an invalid address.

The following restrictions apply to graphs which contain allocation and/or memory free nodes:

Nodes and edges of the graph cannot be deleted.
The graph cannot be used in a child node.
Only one instantiation of the graph may exist at any point in time.
The graph cannot be cloned.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
dptr (Any) – Address of memory to free

Returns:

cudaError_t – cudaSuccess, cudaErrorCudartUnloading, cudaErrorInitializationError, cudaErrorNotSupported, cudaErrorInvalidValue, cudaErrorOutOfMemory
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphMemFreeNodeGetParams(node)¶

Returns a memory free node’s parameters.

Returns the address of a memory free node hNode in dptr_out.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to get the parameters for

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
dptr_out (Any) – Pointer to return the device address

cuda.bindings.runtime.cudaDeviceGraphMemTrim(int device)¶

Free unused memory that was cached on the specified device for use with graphs back to the OS.

Blocks which are not in use by a graph that is either currently executing or scheduled to execute are freed back to the operating system.

Parameters:: device (int) – The device for which cached memory should be freed.
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaDeviceGetGraphMemAttribute(int device, attr: cudaGraphMemAttributeType)¶

Query asynchronous allocation attributes related to graphs.

Valid attributes are:

cudaGraphMemAttrUsedMemCurrent: Amount of memory, in bytes, currently associated with graphs
cudaGraphMemAttrUsedMemHigh: High watermark of memory, in bytes, associated with graphs since the last time it was reset. High watermark can only be reset to zero.
cudaGraphMemAttrReservedMemCurrent: Amount of memory, in bytes, currently allocated for use by the CUDA graphs asynchronous allocator.
cudaGraphMemAttrReservedMemHigh: High watermark of memory, in bytes, currently allocated for use by the CUDA graphs asynchronous allocator.

Parameters:

device (int) – Specifies the scope of the query
attr (cudaGraphMemAttributeType) – attribute to get

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidDevice
value (Any) – retrieved value

cuda.bindings.runtime.cudaDeviceSetGraphMemAttribute(int device, attr: cudaGraphMemAttributeType, value)¶

Set asynchronous allocation attributes related to graphs.

Valid attributes are:

cudaGraphMemAttrUsedMemHigh: High watermark of memory, in bytes, associated with graphs since the last time it was reset. High watermark can only be reset to zero.
cudaGraphMemAttrReservedMemHigh: High watermark of memory, in bytes, currently allocated for use by the CUDA graphs asynchronous allocator.

Parameters:

device (int) – Specifies the scope of the query
attr (cudaGraphMemAttributeType) – attribute to get
value (Any) – pointer to value to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphClone(originalGraph)¶

Clones a graph.

This function creates a copy of originalGraph and returns it in pGraphClone. All parameters are copied into the cloned graph. The original graph may be modified after this call without affecting the clone.

Child graph nodes in the original graph are recursively copied into the clone.

Parameters:

originalGraph (CUgraph or cudaGraph_t) – Graph to clone

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorMemoryAllocation
pGraphClone (cudaGraph_t) – Returns newly created cloned graph

cuda.bindings.runtime.cudaGraphNodeFindInClone(originalNode, clonedGraph)¶

Finds a cloned version of a node.

This function returns the node in clonedGraph corresponding to originalNode in the original graph.

clonedGraph must have been cloned from originalGraph via cudaGraphClone. originalNode must have been in originalGraph at the time of the call to cudaGraphClone, and the corresponding cloned node in clonedGraph must not have been removed. The cloned node is then returned via pClonedNode.

Parameters:

originalNode (CUgraphNode or cudaGraphNode_t) – Handle to the original node
clonedGraph (CUgraph or cudaGraph_t) – Cloned graph to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pNode (cudaGraphNode_t) – Returns handle to the cloned node

See also

cudaGraphClone

cuda.bindings.runtime.cudaGraphNodeGetType(node)¶

Returns a node’s type.

Returns the node type of node in pType.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pType (cudaGraphNodeType) – Pointer to return the node type

cuda.bindings.runtime.cudaGraphGetNodes(graph, size_t numNodes=0)¶

Returns a graph’s nodes.

Returns a list of graph’s nodes. nodes may be NULL, in which case this function will return the number of nodes in numNodes. Otherwise, numNodes entries will be filled in. If numNodes is higher than the actual number of nodes, the remaining entries in nodes will be set to NULL, and the number of nodes actually obtained will be returned in numNodes.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to query
numNodes (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
nodes (List[cudaGraphNode_t]) – Pointer to return the nodes
numNodes (int) – See description

cuda.bindings.runtime.cudaGraphGetRootNodes(graph, size_t pNumRootNodes=0)¶

Returns a graph’s root nodes.

Returns a list of graph’s root nodes. pRootNodes may be NULL, in which case this function will return the number of root nodes in pNumRootNodes. Otherwise, pNumRootNodes entries will be filled in. If pNumRootNodes is higher than the actual number of root nodes, the remaining entries in pRootNodes will be set to NULL, and the number of nodes actually obtained will be returned in pNumRootNodes.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to query
pNumRootNodes (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pRootNodes (List[cudaGraphNode_t]) – Pointer to return the root nodes
pNumRootNodes (int) – See description

cuda.bindings.runtime.cudaGraphGetEdges(graph, size_t numEdges=0)¶

Returns a graph’s dependency edges.

Returns a list of graph’s dependency edges. Edges are returned via corresponding indices in from and to; that is, the node in to`[i] has a dependency on the node in `from`[i]. `from and to may both be NULL, in which case this function only returns the number of edges in numEdges. Otherwise, numEdges entries will be filled in. If numEdges is higher than the actual number of edges, the remaining entries in from and to will be set to NULL, and the number of edges actually returned will be written to numEdges.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to get the edges from
numEdges (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
from (List[cudaGraphNode_t]) – Location to return edge endpoints
to (List[cudaGraphNode_t]) – Location to return edge endpoints
numEdges (int) – See description

cuda.bindings.runtime.cudaGraphGetEdges_v2(graph, size_t numEdges=0)¶

Returns a graph’s dependency edges (12.3+)

Returns a list of graph’s dependency edges. Edges are returned via corresponding indices in from, to and edgeData; that is, the node in to`[i] has a dependency on the node in `from`[i] with data `edgeData`[i]. `from and to may both be NULL, in which case this function only returns the number of edges in numEdges. Otherwise, numEdges entries will be filled in. If numEdges is higher than the actual number of edges, the remaining entries in from and to will be set to NULL, and the number of edges actually returned will be written to numEdges. edgeData may alone be NULL, in which case the edges must all have default (zeroed) edge data. Attempting a losst query via NULL edgeData will result in cudaErrorLossyQuery. If edgeData is non-NULL then from and to must be as well.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to get the edges from
numEdges (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorLossyQuery, cudaErrorInvalidValue
from (List[cudaGraphNode_t]) – Location to return edge endpoints
to (List[cudaGraphNode_t]) – Location to return edge endpoints
edgeData (List[cudaGraphEdgeData]) – Optional location to return edge data
numEdges (int) – See description

cuda.bindings.runtime.cudaGraphNodeGetDependencies(node, size_t pNumDependencies=0)¶

Returns a node’s dependencies.

Returns a list of node’s dependencies. pDependencies may be NULL, in which case this function will return the number of dependencies in pNumDependencies. Otherwise, pNumDependencies entries will be filled in. If pNumDependencies is higher than the actual number of dependencies, the remaining entries in pDependencies will be set to NULL, and the number of nodes actually obtained will be returned in pNumDependencies.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to query
pNumDependencies (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pDependencies (List[cudaGraphNode_t]) – Pointer to return the dependencies
pNumDependencies (int) – See description

cuda.bindings.runtime.cudaGraphNodeGetDependencies_v2(node, size_t pNumDependencies=0)¶

Returns a node’s dependencies (12.3+)

Note that if an edge has non-zero (non-default) edge data and edgeData is NULL, this API will return cudaErrorLossyQuery. If edgeData is non-NULL, then pDependencies must be as well.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to query
pNumDependencies (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorLossyQuery, cudaErrorInvalidValue
pDependencies (List[cudaGraphNode_t]) – Pointer to return the dependencies
edgeData (List[cudaGraphEdgeData]) – Optional array to return edge data for each dependency
pNumDependencies (int) – See description

cuda.bindings.runtime.cudaGraphNodeGetDependentNodes(node, size_t pNumDependentNodes=0)¶

Returns a node’s dependent nodes.

Returns a list of node’s dependent nodes. pDependentNodes may be NULL, in which case this function will return the number of dependent nodes in pNumDependentNodes. Otherwise, pNumDependentNodes entries will be filled in. If pNumDependentNodes is higher than the actual number of dependent nodes, the remaining entries in pDependentNodes will be set to NULL, and the number of nodes actually obtained will be returned in pNumDependentNodes.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to query
pNumDependentNodes (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pDependentNodes (List[cudaGraphNode_t]) – Pointer to return the dependent nodes
pNumDependentNodes (int) – See description

cuda.bindings.runtime.cudaGraphNodeGetDependentNodes_v2(node, size_t pNumDependentNodes=0)¶

Returns a node’s dependent nodes (12.3+)

Note that if an edge has non-zero (non-default) edge data and edgeData is NULL, this API will return cudaErrorLossyQuery. If edgeData is non-NULL, then pDependentNodes must be as well.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to query
pNumDependentNodes (int) – See description

Returns:

cudaError_t – cudaSuccess, cudaErrorLossyQuery, cudaErrorInvalidValue
pDependentNodes (List[cudaGraphNode_t]) – Pointer to return the dependent nodes
edgeData (List[cudaGraphEdgeData]) – Optional pointer to return edge data for dependent nodes
pNumDependentNodes (int) – See description

cuda.bindings.runtime.cudaGraphAddDependencies(graph, from_: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], to: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies)¶

Adds dependency edges to a graph.

The number of dependencies to be added is defined by numDependencies Elements in pFrom and pTo at corresponding indices define a dependency. Each node in pFrom and pTo must belong to graph.

If numDependencies is 0, elements in pFrom and pTo will be ignored. Specifying an existing dependency will return an error.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which dependencies are added
from (List[cudaGraphNode_t]) – Array of nodes that provide the dependencies
to (List[cudaGraphNode_t]) – Array of dependent nodes
numDependencies (size_t) – Number of dependencies to be added

Returns:

Return type:

cuda.bindings.runtime.cudaGraphAddDependencies_v2(graph, from_: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], to: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], edgeData: Optional[Tuple[cudaGraphEdgeData] | List[cudaGraphEdgeData]], size_t numDependencies)¶

Adds dependency edges to a graph. (12.3+)

The number of dependencies to be added is defined by numDependencies Elements in pFrom and pTo at corresponding indices define a dependency. Each node in pFrom and pTo must belong to graph.

If numDependencies is 0, elements in pFrom and pTo will be ignored. Specifying an existing dependency will return an error.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which dependencies are added
from (List[cudaGraphNode_t]) – Array of nodes that provide the dependencies
to (List[cudaGraphNode_t]) – Array of dependent nodes
edgeData (List[cudaGraphEdgeData]) – Optional array of edge data. If NULL, default (zeroed) edge data is assumed.
numDependencies (size_t) – Number of dependencies to be added

Returns:

Return type:

cuda.bindings.runtime.cudaGraphRemoveDependencies(graph, from_: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], to: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies)¶

Removes dependency edges from a graph.

The number of pDependencies to be removed is defined by numDependencies. Elements in pFrom and pTo at corresponding indices define a dependency. Each node in pFrom and pTo must belong to graph.

If numDependencies is 0, elements in pFrom and pTo will be ignored. Specifying a non-existing dependency will return an error.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph from which to remove dependencies
from (List[cudaGraphNode_t]) – Array of nodes that provide the dependencies
to (List[cudaGraphNode_t]) – Array of dependent nodes
numDependencies (size_t) – Number of dependencies to be removed

Returns:

Return type:

cuda.bindings.runtime.cudaGraphRemoveDependencies_v2(graph, from_: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], to: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], edgeData: Optional[Tuple[cudaGraphEdgeData] | List[cudaGraphEdgeData]], size_t numDependencies)¶

Removes dependency edges from a graph. (12.3+)

The number of pDependencies to be removed is defined by numDependencies. Elements in pFrom and pTo at corresponding indices define a dependency. Each node in pFrom and pTo must belong to graph.

If numDependencies is 0, elements in pFrom and pTo will be ignored. Specifying an edge that does not exist in the graph, with data matching edgeData, results in an error. edgeData is nullable, which is equivalent to passing default (zeroed) data for each edge.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph from which to remove dependencies
from (List[cudaGraphNode_t]) – Array of nodes that provide the dependencies
to (List[cudaGraphNode_t]) – Array of dependent nodes
edgeData (List[cudaGraphEdgeData]) – Optional array of edge data. If NULL, edge data is assumed to be default (zeroed).
numDependencies (size_t) – Number of dependencies to be removed

Returns:

Return type:

cuda.bindings.runtime.cudaGraphDestroyNode(node)¶

Remove a node from the graph.

Removes node from its graph. This operation also severs any dependencies of other nodes on node and vice versa.

Dependencies cannot be removed from graphs which contain allocation or free nodes. Any attempt to do so will return an error.

Parameters:: node (CUgraphNode or cudaGraphNode_t) – Node to remove
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaGraphInstantiate(graph, unsigned long long flags)¶

Creates an executable graph from a graph.

Instantiates graph as an executable graph. The graph is validated for any structural constraints or intra-node constraints which were not previously validated. If instantiation is successful, a handle to the instantiated graph is returned in pGraphExec.

The flags parameter controls the behavior of instantiation and subsequent graph launches. Valid flags are:

cudaGraphInstantiateFlagAutoFreeOnLaunch, which configures a graph containing memory allocation nodes to automatically free any unfreed memory allocations before the graph is relaunched.
cudaGraphInstantiateFlagDeviceLaunch, which configures the graph for launch from the device. If this flag is passed, the executable graph handle returned can be used to launch the graph from both the host and device. This flag cannot be used in conjunction with cudaGraphInstantiateFlagAutoFreeOnLaunch.
cudaGraphInstantiateFlagUseNodePriority, which causes the graph to use the priorities from the per-node attributes rather than the priority of the launch stream during execution. Note that priorities are only available on kernel nodes, and are copied from stream priority during stream capture.

If graph contains any allocation or free nodes, there can be at most one executable graph in existence for that graph at a time. An attempt to instantiate a second executable graph before destroying the first with cudaGraphExecDestroy will result in an error. The same also applies if graph contains any device-updatable kernel nodes.

Graphs instantiated for launch on the device have additional restrictions which do not apply to host graphs:

The graph’s nodes must reside on a single device.
The graph can only contain kernel nodes, memcpy nodes, memset nodes, and child graph nodes.
The graph cannot be empty and must contain at least one kernel, memcpy, or memset node. Operation-specific restrictions are outlined below.
Kernel nodes:
- Use of CUDA Dynamic Parallelism is not permitted.
- Cooperative launches are permitted as long as MPS is not in use.
Memcpy nodes:
- Only copies involving device memory and/or pinned device-mapped host memory are permitted.
- Copies involving CUDA arrays are not permitted.
- Both operands must be accessible from the current device, and the current device must match the device of other nodes in the graph.

If graph is not instantiated for launch on the device but contains kernels which call device-side cudaGraphLaunch() from multiple devices, this will result in an error.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to instantiate
flags (unsigned long long) – Flags to control instantiation. See CUgraphInstantiate_flags.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphExec (cudaGraphExec_t) – Returns instantiated graph

cuda.bindings.runtime.cudaGraphInstantiateWithFlags(graph, unsigned long long flags)¶

Creates an executable graph from a graph.

The flags parameter controls the behavior of instantiation and subsequent graph launches. Valid flags are:

cudaGraphInstantiateFlagAutoFreeOnLaunch, which configures a graph containing memory allocation nodes to automatically free any unfreed memory allocations before the graph is relaunched.
cudaGraphInstantiateFlagDeviceLaunch, which configures the graph for launch from the device. If this flag is passed, the executable graph handle returned can be used to launch the graph from both the host and device. This flag can only be used on platforms which support unified addressing. This flag cannot be used in conjunction with cudaGraphInstantiateFlagAutoFreeOnLaunch.
cudaGraphInstantiateFlagUseNodePriority, which causes the graph to use the priorities from the per-node attributes rather than the priority of the launch stream during execution. Note that priorities are only available on kernel nodes, and are copied from stream priority during stream capture.

If graph contains kernels which call device-side cudaGraphLaunch() from multiple devices, this will result in an error.

Graphs instantiated for launch on the device have additional restrictions which do not apply to host graphs:

The graph’s nodes must reside on a single device.
The graph can only contain kernel nodes, memcpy nodes, memset nodes, and child graph nodes.
The graph cannot be empty and must contain at least one kernel, memcpy, or memset node. Operation-specific restrictions are outlined below.
Kernel nodes:
- Use of CUDA Dynamic Parallelism is not permitted.
- Cooperative launches are permitted as long as MPS is not in use.
Memcpy nodes:
- Only copies involving device memory and/or pinned device-mapped host memory are permitted.
- Copies involving CUDA arrays are not permitted.
- Both operands must be accessible from the current device, and the current device must match the device of other nodes in the graph.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to instantiate
flags (unsigned long long) – Flags to control instantiation. See CUgraphInstantiate_flags.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphExec (cudaGraphExec_t) – Returns instantiated graph

cuda.bindings.runtime.cudaGraphInstantiateWithParams(graph, cudaGraphInstantiateParams instantiateParams: Optional[cudaGraphInstantiateParams])¶

Creates an executable graph from a graph.

Instantiates graph as an executable graph according to the instantiateParams structure. The graph is validated for any structural constraints or intra-node constraints which were not previously validated. If instantiation is successful, a handle to the instantiated graph is returned in pGraphExec.

instantiateParams controls the behavior of instantiation and subsequent graph launches, as well as returning more detailed information in the event of an error. cudaGraphInstantiateParams is defined as:

View CUDA Toolkit Documentation for a C++ code example

The flags field controls the behavior of instantiation and subsequent graph launches. Valid flags are:

cudaGraphInstantiateFlagAutoFreeOnLaunch, which configures a graph containing memory allocation nodes to automatically free any unfreed memory allocations before the graph is relaunched.
cudaGraphInstantiateFlagUpload, which will perform an upload of the graph into uploadStream once the graph has been instantiated.
cudaGraphInstantiateFlagDeviceLaunch, which configures the graph for launch from the device. If this flag is passed, the executable graph handle returned can be used to launch the graph from both the host and device. This flag can only be used on platforms which support unified addressing. This flag cannot be used in conjunction with cudaGraphInstantiateFlagAutoFreeOnLaunch.
cudaGraphInstantiateFlagUseNodePriority, which causes the graph to use the priorities from the per-node attributes rather than the priority of the launch stream during execution. Note that priorities are only available on kernel nodes, and are copied from stream priority during stream capture.

If graph contains kernels which call device-side cudaGraphLaunch() from multiple devices, this will result in an error.

Graphs instantiated for launch on the device have additional restrictions which do not apply to host graphs:

The graph’s nodes must reside on a single device.
The graph can only contain kernel nodes, memcpy nodes, memset nodes, and child graph nodes.
The graph cannot be empty and must contain at least one kernel, memcpy, or memset node. Operation-specific restrictions are outlined below.
Kernel nodes:
- Use of CUDA Dynamic Parallelism is not permitted.
- Cooperative launches are permitted as long as MPS is not in use.
Memcpy nodes:
- Only copies involving device memory and/or pinned device-mapped host memory are permitted.
- Copies involving CUDA arrays are not permitted.
- Both operands must be accessible from the current device, and the current device must match the device of other nodes in the graph.

In the event of an error, the result_out and errNode_out fields will contain more information about the nature of the error. Possible error reporting includes:

cudaGraphInstantiateError, if passed an invalid value or if an unexpected error occurred which is described by the return value of the function. errNode_out will be set to NULL.
cudaGraphInstantiateInvalidStructure, if the graph structure is invalid. errNode_out will be set to one of the offending nodes.
cudaGraphInstantiateNodeOperationNotSupported, if the graph is instantiated for device launch but contains a node of an unsupported node type, or a node which performs unsupported operations, such as use of CUDA dynamic parallelism within a kernel node. errNode_out will be set to this node.
cudaGraphInstantiateMultipleDevicesNotSupported, if the graph is instantiated for device launch but a node’s device differs from that of another node. This error can also be returned if a graph is not instantiated for device launch and it contains kernels which call device-side cudaGraphLaunch() from multiple devices. errNode_out will be set to this node.

If instantiation is successful, result_out will be set to cudaGraphInstantiateSuccess, and hErrNode_out will be set to NULL.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to instantiate
instantiateParams (cudaGraphInstantiateParams) – Instantiation parameters

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
pGraphExec (cudaGraphExec_t) – Returns instantiated graph

cuda.bindings.runtime.cudaGraphExecGetFlags(graphExec)¶

Query the instantiation flags of an executable graph.

Returns the flags that were passed to instantiation for the given executable graph. cudaGraphInstantiateFlagUpload will not be returned by this API as it does not affect the resulting executable graph.

Parameters:

graphExec (CUgraphExec or cudaGraphExec_t) – The executable graph to query

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
flags (unsigned long long) – Returns the instantiation flags

cuda.bindings.runtime.cudaGraphExecKernelNodeSetParams(hGraphExec, node, cudaKernelNodeParams pNodeParams: Optional[cudaKernelNodeParams])¶

Sets the parameters for a kernel node in the given graphExec.

Sets the parameters of a kernel node in an executable graph hGraphExec. The node is identified by the corresponding node node in the non-executable graph, from which the executable graph was instantiated.

node must not have been removed from the original graph. All nodeParams fields may change, but the following restrictions apply to func updates:

The owning device of the function cannot change.
A node whose function originally did not use CUDA dynamic parallelism cannot be updated to a function which uses CDP
A node whose function originally did not make device-side update calls cannot be updated to a function which makes device-side update calls.
If hGraphExec was not instantiated for device launch, a node whose function originally did not use device-side cudaGraphLaunch() cannot be updated to a function which uses device-side cudaGraphLaunch() unless the node resides on the same device as nodes which contained such calls at instantiate-time. If no such calls were present at instantiation, these updates cannot be performed at all.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

If node is a device-updatable kernel node, the next upload/launch of hGraphExec will overwrite any previous device-side updates. Additionally, applying host updates to a device-updatable kernel node while it is being updated from the device will result in undefined behavior.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – kernel node from the graph from which graphExec was instantiated
pNodeParams (cudaKernelNodeParams) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecMemcpyNodeSetParams(hGraphExec, node, cudaMemcpy3DParms pNodeParams: Optional[cudaMemcpy3DParms])¶

Sets the parameters for a memcpy node in the given graphExec.

Updates the work represented by node in hGraphExec as though node had contained pNodeParams at instantiation. node must remain in the graph which was used to instantiate hGraphExec. Changed edges to and from node are ignored.

The source and destination memory in pNodeParams must be allocated from the same contexts as the original source and destination memory. Both the instantiation-time memory operands and the memory operands in pNodeParams must be 1-dimensional. Zero-length operations are not supported.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

Returns cudaErrorInvalidValue if the memory operands’ mappings changed or either the original or new memory operands are multidimensional.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – Memcpy node from the graph which was used to instantiate graphExec
pNodeParams (cudaMemcpy3DParms) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecMemcpyNodeSetParams1D(hGraphExec, node, dst, src, size_t count, kind: cudaMemcpyKind)¶

Sets the parameters for a memcpy node in the given graphExec to perform a 1-dimensional copy.

Updates the work represented by node in hGraphExec as though node had contained the given params at instantiation. node must remain in the graph which was used to instantiate hGraphExec. Changed edges to and from node are ignored.

src and dst must be allocated from the same contexts as the original source and destination memory. The instantiation-time memory operands must be 1-dimensional. Zero-length operations are not supported.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

Returns cudaErrorInvalidValue if the memory operands’ mappings changed or the original memory operands are multidimensional.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – Memcpy node from the graph which was used to instantiate graphExec
dst (Any) – Destination memory address
src (Any) – Source memory address
count (size_t) – Size in bytes to copy
kind (cudaMemcpyKind) – Type of transfer

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecMemsetNodeSetParams(hGraphExec, node, cudaMemsetParams pNodeParams: Optional[cudaMemsetParams])¶

Sets the parameters for a memset node in the given graphExec.

Zero sized operations are not supported.

The new destination pointer in pNodeParams must be to the same kind of allocation as the original destination pointer and have the same context association and device mapping as the original destination pointer.

Both the value and pointer address may be updated. Changing other aspects of the memset (width, height, element size or pitch) may cause the update to be rejected. Specifically, for 2d memsets, all dimension changes are rejected. For 1d memsets, changes in height are explicitly rejected and other changes are oportunistically allowed if the resulting work maps onto the work resources already allocated for the node.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – Memset node from the graph which was used to instantiate graphExec
pNodeParams (cudaMemsetParams) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecHostNodeSetParams(hGraphExec, node, cudaHostNodeParams pNodeParams: Optional[cudaHostNodeParams])¶

Sets the parameters for a host node in the given graphExec.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – Host node from the graph which was used to instantiate graphExec
pNodeParams (cudaHostNodeParams) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecChildGraphNodeSetParams(hGraphExec, node, childGraph)¶

Updates node parameters in the child graph node in the given graphExec.

Updates the work represented by node in hGraphExec as though the nodes contained in node’s graph had the parameters contained in childGraph’s nodes at instantiation. node must remain in the graph which was used to instantiate hGraphExec. Changed edges to and from node are ignored.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. node is also not modified by this call.

The topology of childGraph, as well as the node insertion order, must match that of the graph contained in node. See cudaGraphExecUpdate() for a list of restrictions on what can be updated in an instantiated graph. The update is recursive, so child graph nodes contained within the top level child graph will also be updated.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
node (CUgraphNode or cudaGraphNode_t) – Host node from the graph which was used to instantiate graphExec
childGraph (CUgraph or cudaGraph_t) – The graph supplying the updated parameters

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecEventRecordNodeSetEvent(hGraphExec, hNode, event)¶

Sets the event for an event record node in the given graphExec.

Sets the event of an event record node in an executable graph hGraphExec. The node is identified by the corresponding node hNode in the non-executable graph, from which the executable graph was instantiated.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. hNode is also not modified by this call.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – Event record node from the graph from which graphExec was instantiated
event (CUevent or cudaEvent_t) – Updated event to use

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecEventWaitNodeSetEvent(hGraphExec, hNode, event)¶

Sets the event for an event wait node in the given graphExec.

Sets the event of an event wait node in an executable graph hGraphExec. The node is identified by the corresponding node hNode in the non-executable graph, from which the executable graph was instantiated.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. hNode is also not modified by this call.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – Event wait node from the graph from which graphExec was instantiated
event (CUevent or cudaEvent_t) – Updated event to use

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecExternalSemaphoresSignalNodeSetParams(hGraphExec, hNode, cudaExternalSemaphoreSignalNodeParams nodeParams: Optional[cudaExternalSemaphoreSignalNodeParams])¶

Sets the parameters for an external semaphore signal node in the given graphExec.

Sets the parameters of an external semaphore signal node in an executable graph hGraphExec. The node is identified by the corresponding node hNode in the non-executable graph, from which the executable graph was instantiated.

hNode must not have been removed from the original graph.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. hNode is also not modified by this call.

Changing nodeParams->numExtSems is not supported.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – semaphore signal node from the graph from which graphExec was instantiated
nodeParams (cudaExternalSemaphoreSignalNodeParams) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphExecExternalSemaphoresWaitNodeSetParams(hGraphExec, hNode, cudaExternalSemaphoreWaitNodeParams nodeParams: Optional[cudaExternalSemaphoreWaitNodeParams])¶

Sets the parameters for an external semaphore wait node in the given graphExec.

Sets the parameters of an external semaphore wait node in an executable graph hGraphExec. The node is identified by the corresponding node hNode in the non-executable graph, from which the executable graph was instantiated.

hNode must not have been removed from the original graph.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. hNode is also not modified by this call.

Changing nodeParams->numExtSems is not supported.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – semaphore wait node from the graph from which graphExec was instantiated
nodeParams (cudaExternalSemaphoreWaitNodeParams) – Updated Parameters to set

Returns:

Return type:

cuda.bindings.runtime.cudaGraphNodeSetEnabled(hGraphExec, hNode, unsigned int isEnabled)¶

Enables or disables the specified node in the given graphExec.

Sets hNode to be either enabled or disabled. Disabled nodes are functionally equivalent to empty nodes until they are reenabled. Existing node parameters are not affected by disabling/enabling the node.

The node is identified by the corresponding node hNode in the non- executable graph, from which the executable graph was instantiated.

hNode must not have been removed from the original graph.

The modifications only affect future launches of hGraphExec. Already enqueued or running launches of hGraphExec are not affected by this call. hNode is also not modified by this call.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – Node from the graph from which graphExec was instantiated
isEnabled (unsigned int) – Node is enabled if != 0, otherwise the node is disabled

Returns:

Return type:

Notes

Currently only kernel, memset and memcpy nodes are supported.

cuda.bindings.runtime.cudaGraphNodeGetEnabled(hGraphExec, hNode)¶

Query whether a node in the given graphExec is enabled.

Sets isEnabled to 1 if hNode is enabled, or 0 if hNode is disabled.

The node is identified by the corresponding node hNode in the non- executable graph, from which the executable graph was instantiated.

hNode must not have been removed from the original graph.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to set the specified node
hNode (CUgraphNode or cudaGraphNode_t) – Node from the graph from which graphExec was instantiated

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue,
isEnabled (unsigned int) – Location to return the enabled status of the node

Notes

Currently only kernel, memset and memcpy nodes are supported.

cuda.bindings.runtime.cudaGraphExecUpdate(hGraphExec, hGraph)¶

Check whether an executable graph can be updated with a graph and perform the update if possible.

Updates the node parameters in the instantiated graph specified by hGraphExec with the node parameters in a topologically identical graph specified by hGraph.

Limitations:

Kernel nodes:
- The owning context of the function cannot change.
- A node whose function originally did not use CUDA dynamic parallelism cannot be updated to a function which uses CDP.
- A node whose function originally did not make device-side update calls cannot be updated to a function which makes device-side update calls.
- A cooperative node cannot be updated to a non-cooperative node, and vice-versa.
- If the graph was instantiated with cudaGraphInstantiateFlagUseNodePriority, the priority attribute cannot change. Equality is checked on the originally requested priority values, before they are clamped to the device’s supported range.
- If hGraphExec was not instantiated for device launch, a node whose function originally did not use device-side cudaGraphLaunch() cannot be updated to a function which uses device-side cudaGraphLaunch() unless the node resides on the same device as nodes which contained such calls at instantiate-time. If no such calls were present at instantiation, these updates cannot be performed at all.
- Neither hGraph nor hGraphExec may contain device-updatable kernel nodes.
Memset and memcpy nodes:
- The CUDA device(s) to which the operand(s) was allocated/mapped cannot change.
- The source/destination memory must be allocated from the same contexts as the original source/destination memory.
- For 2d memsets, only address and assinged value may be updated.
- For 1d memsets, updating dimensions is also allowed, but may fail if the resulting operation doesn’t map onto the work resources already allocated for the node.
Additional memcpy node restrictions:
- Changing either the source or destination memory type(i.e. CU_MEMORYTYPE_DEVICE, CU_MEMORYTYPE_ARRAY, etc.) is not supported.
Conditional nodes:
- Changing node parameters is not supported.
- Changeing parameters of nodes within the conditional body graph is subject to the rules above.
- Conditional handle flags and default values are updated as part of the graph update.

Note: The API may add further restrictions in future releases. The return code should always be checked.

cudaGraphExecUpdate sets the result member of resultInfo to cudaGraphExecUpdateErrorTopologyChanged under the following conditions:

The count of nodes directly in hGraphExec and hGraph differ, in which case resultInfo->errorNode is set to NULL.
hGraph has more exit nodes than hGraph, in which case resultInfo->errorNode is set to one of the exit nodes in hGraph.
A node in hGraph has a different number of dependencies than the node from hGraphExec it is paired with, in which case resultInfo->errorNode is set to the node from hGraph.
A node in hGraph has a dependency that does not match with the corresponding dependency of the paired node from hGraphExec. resultInfo->errorNode will be set to the node from hGraph. resultInfo->errorFromNode will be set to the mismatched dependency. The dependencies are paired based on edge order and a dependency does not match when the nodes are already paired based on other edges examined in the graph.

cudaGraphExecUpdate sets the result member of resultInfo to:

cudaGraphExecUpdateError if passed an invalid value.
cudaGraphExecUpdateErrorTopologyChanged if the graph topology changed
cudaGraphExecUpdateErrorNodeTypeChanged if the type of a node changed, in which case hErrorNode_out is set to the node from hGraph.
cudaGraphExecUpdateErrorFunctionChanged if the function of a kernel node changed (CUDA driver < 11.2)
cudaGraphExecUpdateErrorUnsupportedFunctionChange if the func field of a kernel changed in an unsupported way(see note above), in which case hErrorNode_out is set to the node from hGraph
cudaGraphExecUpdateErrorParametersChanged if any parameters to a node changed in a way that is not supported, in which case hErrorNode_out is set to the node from hGraph
cudaGraphExecUpdateErrorAttributesChanged if any attributes of a node changed in a way that is not supported, in which case hErrorNode_out is set to the node from hGraph
cudaGraphExecUpdateErrorNotSupported if something about a node is unsupported, like the node’s type or configuration, in which case hErrorNode_out is set to the node from hGraph

If the update fails for a reason not listed above, the result member of resultInfo will be set to cudaGraphExecUpdateError. If the update succeeds, the result member will be set to cudaGraphExecUpdateSuccess.

cudaGraphExecUpdate returns cudaSuccess when the updated was performed successfully. It returns cudaErrorGraphExecUpdateFailure if the graph update was not performed because it included changes which violated constraints specific to instantiated graph update.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – The instantiated graph to be updated
hGraph (CUgraph or cudaGraph_t) – The graph containing the updated parameters

Returns:

cudaError_t – cudaSuccess, cudaErrorGraphExecUpdateFailure,
resultInfo (cudaGraphExecUpdateResultInfo) – the error info structure

See also

cudaGraphInstantiate

cuda.bindings.runtime.cudaGraphUpload(graphExec, stream)¶

Uploads an executable graph in a stream.

Uploads hGraphExec to the device in hStream without executing it. Uploads of the same hGraphExec will be serialized. Each upload is ordered behind both any previous work in hStream and any previous launches of hGraphExec. Uses memory cached by stream to back the allocations owned by graphExec.

Parameters:

hGraphExec (CUgraphExec or cudaGraphExec_t) – Executable graph to upload
hStream (CUstream or cudaStream_t) – Stream in which to upload the graph

Returns:

Return type:

cuda.bindings.runtime.cudaGraphLaunch(graphExec, stream)¶

Launches an executable graph in a stream.

Executes graphExec in stream. Only one instance of graphExec may be executing at a time. Each launch is ordered behind both any previous work in stream and any previous launches of graphExec. To execute a graph concurrently, it must be instantiated multiple times into multiple executable graphs.

If any allocations created by graphExec remain unfreed (from a previous launch) and graphExec was not instantiated with cudaGraphInstantiateFlagAutoFreeOnLaunch, the launch will fail with cudaErrorInvalidValue.

Parameters:

graphExec (CUgraphExec or cudaGraphExec_t) – Executable graph to launch
stream (CUstream or cudaStream_t) – Stream in which to launch the graph

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorOperatingSystem

cuda.bindings.runtime.cudaGraphExecDestroy(graphExec)¶

Destroys an executable graph.

Destroys the executable graph specified by graphExec.

Parameters:: graphExec (CUgraphExec or cudaGraphExec_t) – Executable graph to destroy
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

cuda.bindings.runtime.cudaGraphDestroy(graph)¶

Destroys a graph.

Destroys the graph specified by graph, as well as all of its nodes.

Parameters:: graph (CUgraph or cudaGraph_t) – Graph to destroy
Returns:: cudaSuccess, cudaErrorInvalidValue
Return type:: cudaError_t

See also

cudaGraphCreate

cuda.bindings.runtime.cudaGraphDebugDotPrint(graph, char *path, unsigned int flags)¶

Write a DOT file describing graph structure.

Using the provided graph, write to path a DOT formatted description of the graph. By default this includes the graph topology, node types, node id, kernel names and memcpy direction. flags can be specified to write more detailed information about each node type such as parameter values, kernel attributes, node and function handles.

Parameters:

graph (CUgraph or cudaGraph_t) – The graph to create a DOT file from
path (bytes) – The path to write the DOT file to
flags (unsigned int) – Flags from cudaGraphDebugDotFlags for specifying which additional node information to write

Returns:

Return type:

cuda.bindings.runtime.cudaUserObjectCreate(ptr, destroy, unsigned int initialRefcount, unsigned int flags)¶

Create a user object.

Create a user object with the specified destructor callback and initial reference count. The initial references are owned by the caller.

Destructor callbacks cannot make CUDA API calls and should avoid blocking behavior, as they are executed by a shared internal thread. Another thread may be signaled to perform such actions, if it does not block forward progress of tasks scheduled through CUDA.

See CUDA User Objects in the CUDA C++ Programming Guide for more information on user objects.

Parameters:

ptr (Any) – The pointer to pass to the destroy function
destroy (cudaHostFn_t) – Callback to free the user object when it is no longer in use
initialRefcount (unsigned int) – The initial refcount to create the object with, typically 1. The initial references are owned by the calling thread.
flags (unsigned int) – Currently it is required to pass cudaUserObjectNoDestructorSync, which is the only defined flag. This indicates that the destroy callback cannot be waited on by any CUDA API. Users requiring synchronization of the callback should signal its completion manually.

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue
object_out (cudaUserObject_t) – Location to return the user object handle

cuda.bindings.runtime.cudaUserObjectRetain(object, unsigned int count)¶

Retain a reference to a user object.

Retains new references to a user object. The new references are owned by the caller.

See CUDA User Objects in the CUDA C++ Programming Guide for more information on user objects.

Parameters:

object (cudaUserObject_t) – The object to retain
count (unsigned int) – The number of references to retain, typically 1. Must be nonzero and not larger than INT_MAX.

Returns:

Return type:

cuda.bindings.runtime.cudaUserObjectRelease(object, unsigned int count)¶

Release a reference to a user object.

Releases user object references owned by the caller. The object’s destructor is invoked if the reference count reaches zero.

It is undefined behavior to release references not owned by the caller, or to use a user object handle after all references are released.

See CUDA User Objects in the CUDA C++ Programming Guide for more information on user objects.

Parameters:

object (cudaUserObject_t) – The object to release
count (unsigned int) – The number of references to release, typically 1. Must be nonzero and not larger than INT_MAX.

Returns:

Return type:

cuda.bindings.runtime.cudaGraphRetainUserObject(graph, object, unsigned int count, unsigned int flags)¶

Retain a reference to a user object from a graph.

Creates or moves user object references that will be owned by a CUDA graph.

See CUDA User Objects in the CUDA C++ Programming Guide for more information on user objects.

Parameters:

graph (CUgraph or cudaGraph_t) – The graph to associate the reference with
object (cudaUserObject_t) – The user object to retain a reference for
count (unsigned int) – The number of references to add to the graph, typically 1. Must be nonzero and not larger than INT_MAX.
flags (unsigned int) – The optional flag cudaGraphUserObjectMove transfers references from the calling thread, rather than create new references. Pass 0 to create new references.

Returns:

Return type:

See also

cudaUserObjectCreate: py:obj:~.cudaUserObjectRetain, cudaUserObjectRelease, cudaGraphReleaseUserObject, cudaGraphCreate

cuda.bindings.runtime.cudaGraphReleaseUserObject(graph, object, unsigned int count)¶

Release a user object reference from a graph.

Releases user object references owned by a graph.

See CUDA User Objects in the CUDA C++ Programming Guide for more information on user objects.

Parameters:

graph (CUgraph or cudaGraph_t) – The graph that will release the reference
object (cudaUserObject_t) – The user object to release a reference for
count (unsigned int) – The number of references to release, typically 1. Must be nonzero and not larger than INT_MAX.

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction, cudaErrorNotSupported

See also

cudaUserObjectCreate: py:obj:~.cudaUserObjectRetain, cudaUserObjectRelease, cudaGraphRetainUserObject, cudaGraphCreate

cuda.bindings.runtime.cudaGraphAddNode(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], size_t numDependencies, cudaGraphNodeParams nodeParams: Optional[cudaGraphNodeParams])¶

Adds a node of arbitrary type to a graph.

Creates a new node in graph described by nodeParams with numDependencies dependencies specified via pDependencies. numDependencies may be 0. pDependencies may be null if numDependencies is 0. pDependencies may not have any duplicate entries.

nodeParams is a tagged union. The node type should be specified in the typename field, and type-specific parameters in the corresponding union member. All unused bytes - that is, reserved0 and all bytes past the utilized union member - must be set to zero. It is recommended to use brace initialization or memset to ensure all bytes are initialized.

Note that for some node types, nodeParams may contain “out parameters” which are modified during the call, such as nodeParams->alloc.dptr.

A handle to the new node will be returned in phGraphNode.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
numDependencies (size_t) – Number of dependencies
nodeParams (cudaGraphNodeParams) – Specification of the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction, cudaErrorNotSupported
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphAddNode_v2(graph, pDependencies: Optional[Tuple[cudaGraphNode_t] | List[cudaGraphNode_t]], dependencyData: Optional[Tuple[cudaGraphEdgeData] | List[cudaGraphEdgeData]], size_t numDependencies, cudaGraphNodeParams nodeParams: Optional[cudaGraphNodeParams])¶

Adds a node of arbitrary type to a graph (12.3+)

Note that for some node types, nodeParams may contain “out parameters” which are modified during the call, such as nodeParams->alloc.dptr.

A handle to the new node will be returned in phGraphNode.

Parameters:

graph (CUgraph or cudaGraph_t) – Graph to which to add the node
pDependencies (List[cudaGraphNode_t]) – Dependencies of the node
dependencyData (List[cudaGraphEdgeData]) – Optional edge data for the dependencies. If NULL, the data is assumed to be default (zeroed) for all dependencies.
numDependencies (size_t) – Number of dependencies
nodeParams (cudaGraphNodeParams) – Specification of the node

Returns:

cudaError_t – cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction, cudaErrorNotSupported
pGraphNode (cudaGraphNode_t) – Returns newly created node

cuda.bindings.runtime.cudaGraphNodeSetParams(node, cudaGraphNodeParams nodeParams: Optional[cudaGraphNodeParams])¶

Update’s a graph node’s parameters.

Sets the parameters of graph node node to nodeParams. The node type specified by nodeParams->type must match the type of node. nodeParams must be fully initialized and all unused bytes (reserved, padding) zeroed.

Modifying parameters is not supported for node types cudaGraphNodeTypeMemAlloc and cudaGraphNodeTypeMemFree.

Parameters:

node (CUgraphNode or cudaGraphNode_t) – Node to set the parameters for
nodeParams (cudaGraphNodeParams) – Parameters to copy

Returns:

Return type:

cudaSuccess, cudaErrorInvalidValue, cudaErrorInvalidDeviceFunction, cudaErrorNotSupported

cuda.bindings.runtime.cudaGraphExecNodeSetParams(graphExec, node, cudaGraphNodeParams nodeParams: Optional[cudaGraphNodeParams])¶

Update’s a graph node’s parameters in an instantiated graph.

Sets the parameters of a node in an executable graph graphExec. The node is identified by the corresponding node node in the non- executable graph from which the executable graph was instantiated. node must not have been removed from the original graph.

The modifications only affect future launches of graphExec. Already enqueued or running launches of graphExec are not affected by this call. node is also not modified by this call.

Allowed changes to parameters on executable graphs are as follows:

View CUDA Toolkit Documentation for a table example

Parameters:

graphExec (CUgraphExec or cudaGraphExec_t) – The executable graph in which to update the specified node
node (CUgraphNode or cudaGraphNode_t) – Corresponding node from the graph from which graphExec was instantiated
nodeParams (cudaGraphNodeParams) – Updated Parameters to set

Returns:

Return type: