Crate rcudnn

Source
Expand description

Provides a safe and convenient wrapper for the CUDA cuDNN API.

This crate (1.0.0) was developed against cuDNN v3.

§Architecture

This crate provides three levels of entrace.

FFI
The ffi module exposes the foreign function interface and cuDNN specific types. Usually, there should be no use to touch it if you only want to use cuDNN in you application. The ffi is provided by the rust-cudnn-sys crate and gets reexported here.

Low-Level
The api module exposes already a complete and safe wrapper for the cuDNN API, including proper Rust Errors. Usually there should be not need to use the API directly though, as the Cudnn module, as described in the next block, provides all the API functionality but provides a more convenient interface.

High-Level
The cudnn module exposes the Cudnn struct, which provides a very convenient, easy-to-understand interface for the cuDNN API. There should be not much need to obtain and read the cuDNN manual. Initialize the Cudnn struct and you can call the available methods wich are representing all the available cuDNN operations.

§Examples

extern crate rcudnn as cudnn;
extern crate libc;
use cudnn::{Cudnn, TensorDescriptor};
use cudnn::utils::{ScalParams, DataType};
fn main() {
//  Initialize a new cuDNN context and allocates resources.
    let cudnn = Cudnn::new().unwrap();
//  Create a cuDNN Tensor Descriptor for `src` and `dest` memory.
    let src_desc = TensorDescriptor::new(&[2, 2, 2], &[4, 2, 1], DataType::Float).unwrap();
    let dest_desc = TensorDescriptor::new(&[2, 2, 2], &[4, 2, 1], DataType::Float).unwrap();
    let acti = cudnn.init_activation().unwrap();
//  Obtain the `src` and memory pointer on the GPU.
//  NOTE: You wouldn't do it like that. You need to really allocate memory on the GPU with e.g. CUDA or Collenchyma.
    let src_data: *const ::libc::c_void = ::std::ptr::null();
    let dest_data: *mut ::libc::c_void = ::std::ptr::null_mut();
//  Now you can compute the forward sigmoid activation on your GPU.
    cudnn.sigmoid_forward::<f32>(&acti, &src_desc, src_data, &dest_desc, dest_data, ScalParams::default());
}

§Notes

rust-cudnn was developed at Autumn for the Rust Machine Intelligence Framework Leaf.

rust-cudnn is part of the High-Performance Computation Framework Collenchyma, for the Neural Network Plugin. Rust CUDNN is now maintained by Juice

Modules§

cuda
Defines Cuda Device Memory.
utils
Describes utility functionality for CUDA cuDNN.

Structs§

API
Defines the Cuda cuDNN API.
ActivationDescriptor
Describes a ActivationDescriptor.
CUevent_st
CUexternalMemory_st
CUexternalSemaphore_st
CUfunc_st
CUgraphExec_st
CUgraphNode_st
CUgraph_st
CUmemPoolHandle_st
CUstream_st
CUuserObject_st
CUuuid_st
ConvolutionDescriptor
Describes a Convolution Descriptor.
Cudnn
Provides a the high-level interface to CUDA’s cuDNN.
DropoutDescriptor
Describes a DropoutDescriptor.
FilterDescriptor
Describes a Filter Descriptor.
NormalizationDescriptor
Describes a LRN Descriptor.
PoolingDescriptor
Describes a Pooling Descriptor.
RnnDescriptor
Describes a Recurrent Descriptor.
TensorDescriptor
Describes a TensorDescriptor.
cudaAccessPolicyWindow
Specifies an access policy for a window, a contiguous extent of memory beginning at base_ptr and ending at base_ptr + num_bytes. Partition into many segments and assign segments such that. sum of “hit segments” / window == approx. ratio. sum of “miss segments” / window == approx 1-ratio. Segments and ratio specifications are fitted to the capabilities of the architecture. Accesses in a hit segment apply the hitProp access policy. Accesses in a miss segment apply the missProp access policy.
cudaArray
cudaArraySparseProperties
Sparse CUDA array and CUDA mipmapped array properties
cudaArraySparseProperties__bindgen_ty_1
cudaChannelFormatDesc
CUDA Channel format descriptor
cudaDeviceProp
CUDA device properties
cudaExtent
CUDA extent
cudaExternalMemoryBufferDesc
External memory buffer descriptor
cudaExternalMemoryHandleDesc
External memory handle descriptor
cudaExternalMemoryHandleDesc__bindgen_ty_1__bindgen_ty_1
Win32 handle referencing the semaphore object. Valid when type is one of the following:
cudaExternalMemoryMipmappedArrayDesc
External memory mipmap descriptor
cudaExternalSemaphoreHandleDesc
External semaphore handle descriptor
cudaExternalSemaphoreHandleDesc__bindgen_ty_1__bindgen_ty_1
Win32 handle referencing the semaphore object. Valid when type is one of the following:
cudaExternalSemaphoreSignalNodeParams
External semaphore signal node parameters
cudaExternalSemaphoreSignalParams
External semaphore signal parameters, compatible with driver type
cudaExternalSemaphoreSignalParams__bindgen_ty_1
cudaExternalSemaphoreSignalParams__bindgen_ty_1__bindgen_ty_1
Parameters for fence objects
cudaExternalSemaphoreSignalParams__bindgen_ty_1__bindgen_ty_3
Parameters for keyed mutex objects
cudaExternalSemaphoreWaitNodeParams
External semaphore wait node parameters
cudaExternalSemaphoreWaitParams
External semaphore wait parameters, compatible with driver type
cudaExternalSemaphoreWaitParams__bindgen_ty_1
cudaExternalSemaphoreWaitParams__bindgen_ty_1__bindgen_ty_1
Parameters for fence objects
cudaExternalSemaphoreWaitParams__bindgen_ty_1__bindgen_ty_3
Parameters for keyed mutex objects
cudaFuncAttributes
CUDA function attributes
cudaGraphicsResource
CUDA graphics interop resource
cudaHostNodeParams
CUDA host node parameters
cudaIpcEventHandle_st
CUDA IPC event handle
cudaIpcMemHandle_st
CUDA IPC memory handle
cudaKernelNodeParams
CUDA GPU kernel node parameters
cudaLaunchParams
CUDA launch parameters
cudaMemAccessDesc
Memory access descriptor
cudaMemAllocNodeParams
Memory allocation node parameters
cudaMemLocation
Specifies a memory location.
cudaMemPoolProps
Specifies the properties of allocations made from the pool.
cudaMemPoolPtrExportData
Opaque data for exporting a pool allocation
cudaMemcpy3DParms
CUDA 3D memory copying parameters
cudaMemcpy3DPeerParms
CUDA 3D cross-device memory copying parameters
cudaMemsetParams
CUDA Memset node parameters
cudaMipmappedArray
cudaPitchedPtr
CUDA Pitched memory pointer
cudaPointerAttributes
CUDA pointer attributes
cudaPos
CUDA 3D position
cudaResourceDesc
CUDA resource descriptor
cudaResourceDesc__bindgen_ty_1__bindgen_ty_1
cudaResourceDesc__bindgen_ty_1__bindgen_ty_2
cudaResourceDesc__bindgen_ty_1__bindgen_ty_3
cudaResourceDesc__bindgen_ty_1__bindgen_ty_4
cudaResourceViewDesc
CUDA resource view descriptor
cudaTextureDesc
CUDA texture descriptor
cudnnActivationStruct
cudnnAlgorithmPerformanceStruct
cudnnAlgorithmStruct
cudnnAlgorithmUnionStruct
cudnnAttnStruct
cudnnCTCLossStruct
cudnnContext
cudnnConvolutionBwdDataAlgoPerfStruct
cudnnConvolutionBwdFilterAlgoPerfStruct
cudnnConvolutionFwdAlgoPerfStruct
cudnnConvolutionStruct
cudnnDebugStruct
cudnnDropoutStruct
cudnnFilterStruct
cudnnFusedOpsConstParamStruct
cudnnFusedOpsPlanStruct
cudnnFusedOpsVariantParamStruct
cudnnLRNStruct
cudnnOpTensorStruct
cudnnPersistentRNNPlan
cudnnPoolingStruct
cudnnRNNDataStruct
cudnnRNNStruct
cudnnReduceTensorStruct
cudnnRuntimeTag_t
cudnnSeqDataStruct
cudnnSpatialTransformerStruct
cudnnTensorStruct
cudnnTensorTransformStruct
dim3
surfaceReference
CUDA Surface reference
textureReference
CUDA texture reference

Enums§

Error
Defines CUDA’s cuDNN errors.
cudaAccessProperty
Specifies performance hint with ::cudaAccessPolicyWindow for hitProp and missProp members.
cudaCGScope
CUDA cooperative group scope
cudaChannelFormatKind
Channel format kind
cudaComputeMode
CUDA device compute modes
cudaDataType
cudaDataType_t
cudaDeviceAttr
CUDA device attributes
cudaDeviceP2PAttr
CUDA device P2P attributes
cudaError
CUDA error types
cudaError_t
CUDA error types
cudaExternalMemoryHandleType
External memory handle types
cudaExternalSemaphoreHandleType
External semaphore handle types
cudaFlushGPUDirectRDMAWritesOptions
CUDA GPUDirect RDMA flush writes APIs supported on the device
cudaFlushGPUDirectRDMAWritesScope
CUDA GPUDirect RDMA flush writes scopes
cudaFlushGPUDirectRDMAWritesTarget
CUDA GPUDirect RDMA flush writes targets
cudaFuncAttribute
CUDA function attributes that can be set using ::cudaFuncSetAttribute
cudaFuncCache
CUDA function cache configurations
cudaGPUDirectRDMAWritesOrdering
CUDA GPUDirect RDMA flush writes ordering features of the device
cudaGetDriverEntryPointFlags
Flags to specify search options to be used with ::cudaGetDriverEntryPoint For more details see ::cuGetProcAddress
cudaGraphDebugDotFlags
CUDA Graph debug write options
cudaGraphExecUpdateResult
CUDA Graph Update error types
cudaGraphInstantiateFlags
Flags for instantiating a graph
cudaGraphMemAttributeType
Graph memory attributes
cudaGraphNodeType
CUDA Graph node types
cudaGraphicsCubeFace
CUDA graphics interop array indices for cube maps
cudaGraphicsMapFlags
CUDA graphics interop map flags
cudaGraphicsRegisterFlags
CUDA graphics interop register flags
cudaKernelNodeAttrID
Graph kernel node Attributes
cudaLimit
CUDA Limits
cudaMemAccessFlags
Specifies the memory protection flags for mapping.
cudaMemAllocationHandleType
Flags for specifying particular handle types
cudaMemAllocationType
Defines the allocation types available
cudaMemLocationType
Specifies the type of location
cudaMemPoolAttr
CUDA memory pool attributes
cudaMemRangeAttribute
CUDA range attributes
cudaMemcpyKind
CUDA memory copy types
cudaMemoryAdvise
CUDA Memory Advise values
cudaMemoryType
CUDA memory types
cudaOutputMode
CUDA Profiler Output modes
cudaOutputMode_t
CUDA Profiler Output modes
cudaResourceType
CUDA resource types
cudaResourceViewFormat
CUDA texture resource view formats
cudaRoundMode
cudaSharedCarveout
Shared memory carveout configurations. These may be passed to cudaFuncSetAttribute
cudaSharedMemConfig
CUDA shared memory configuration
cudaStreamAttrID
Stream Attributes
cudaStreamCaptureMode
Possible modes for stream capture thread interactions. For more details see ::cudaStreamBeginCapture and ::cudaThreadExchangeStreamCaptureMode
cudaStreamCaptureStatus
Possible stream capture statuses returned by ::cudaStreamIsCapturing
cudaStreamUpdateCaptureDependenciesFlags
Flags for ::cudaStreamUpdateCaptureDependencies
cudaSurfaceBoundaryMode
CUDA Surface boundary modes
cudaSurfaceFormatMode
CUDA Surface format modes
cudaSynchronizationPolicy
cudaTextureAddressMode
CUDA texture address modes
cudaTextureFilterMode
CUDA texture filter modes
cudaTextureReadMode
CUDA texture read modes
cudaUserObjectFlags
Flags for user objects for graphs
cudaUserObjectRetainFlags
Flags for retaining user object references for graphs
cudnnActivationMode_t
cudnnBackendAttributeName_t
cudnnBackendAttributeType_t
cudnnBackendBehaviorNote_t
cudnnBackendDescriptorType_t
cudnnBackendHeurMode_t
cudnnBackendKnobType_t
cudnnBackendLayoutType_t
cudnnBackendNumericalNote_t
cudnnBatchNormMode_t
cudnnBatchNormOps_t
cudnnBnFinalizeStatsMode_t
cudnnCTCLossAlgo_t
cudnnConvolutionBwdDataAlgo_t
cudnnConvolutionBwdFilterAlgo_t
cudnnConvolutionFwdAlgo_t
cudnnConvolutionMode_t
cudnnDataType_t
cudnnDeterminism_t
cudnnDirectionMode_t
cudnnDivNormMode_t
cudnnErrQueryMode_t
cudnnFoldingDirection_t
cudnnForwardMode_t
cudnnFusedOpsConstParamLabel_t
cudnnFusedOpsPointerPlaceHolder_t
cudnnFusedOpsVariantParamLabel_t
cudnnFusedOps_t
cudnnGenStatsMode_t
cudnnIndicesType_t
cudnnLRNMode_t
cudnnLossNormalizationMode_t
cudnnMathType_t
cudnnMultiHeadAttnWeightKind_t
cudnnNanPropagation_t
cudnnNormAlgo_t
cudnnNormMode_t
cudnnNormOps_t
cudnnOpTensorOp_t
cudnnPointwiseMode_t
cudnnPoolingMode_t
cudnnRNNAlgo_t
cudnnRNNBiasMode_t
cudnnRNNClipMode_t
cudnnRNNDataLayout_t
cudnnRNNInputMode_t
cudnnRNNMode_t
cudnnReduceTensorIndices_t
cudnnReduceTensorOp_t
cudnnReorderType_t
cudnnSamplerType_t
cudnnSeqDataAxis_t
cudnnSeverity_t
cudnnSoftmaxAlgorithm_t
cudnnSoftmaxMode_t
cudnnStatus_t
cudnnTensorFormat_t
cudnnWgradMode_t

Constants§

CUDNN_ADV_INFER_MAJOR
CUDNN_ADV_INFER_MINOR
CUDNN_ADV_INFER_PATCH
CUDNN_ADV_TRAIN_MAJOR
CUDNN_ADV_TRAIN_MINOR
CUDNN_ADV_TRAIN_PATCH
CUDNN_ATTN_DISABLE_PROJ_BIASES
CUDNN_ATTN_ENABLE_PROJ_BIASES
CUDNN_ATTN_QUERYMAP_ALL_TO_ONE
CUDNN_ATTN_QUERYMAP_ONE_TO_ONE
CUDNN_ATTN_WKIND_COUNT
CUDNN_BN_MIN_EPSILON
CUDNN_CNN_INFER_MAJOR
CUDNN_CNN_INFER_MINOR
CUDNN_CNN_INFER_PATCH
CUDNN_CNN_TRAIN_MAJOR
CUDNN_CNN_TRAIN_MINOR
CUDNN_CNN_TRAIN_PATCH
CUDNN_DIM_MAX
CUDNN_LRN_MAX_N
CUDNN_LRN_MIN_BETA
CUDNN_LRN_MIN_K
CUDNN_LRN_MIN_N
CUDNN_MAJOR
CUDNN_MINOR
CUDNN_OPS_INFER_MAJOR
CUDNN_OPS_INFER_MINOR
CUDNN_OPS_INFER_PATCH
CUDNN_OPS_TRAIN_MAJOR
CUDNN_OPS_TRAIN_MINOR
CUDNN_OPS_TRAIN_PATCH
CUDNN_PATCHLEVEL
CUDNN_RNN_PADDED_IO_DISABLED
CUDNN_RNN_PADDED_IO_ENABLED
CUDNN_SEQDATA_DIM_COUNT
CUDNN_VERSION
libraryPropertyType_t_MAJOR_VERSION
libraryPropertyType_t_MINOR_VERSION
libraryPropertyType_t_PATCH_LEVEL

Functions§

cudaArrayGetInfo
\brief Gets info about the specified cudaArray
cudaArrayGetPlane
\brief Gets a CUDA array plane from a CUDA array
cudaArrayGetSparseProperties
cudaBindSurfaceToArray
\brief Binds an array to a surface
cudaBindTexture
\brief Binds a memory area to a texture
cudaBindTexture2D
\brief Binds a 2D memory area to a texture
cudaBindTextureToArray
\brief Binds an array to a texture
cudaBindTextureToMipmappedArray
\brief Binds a mipmapped array to a texture
cudaChooseDevice
\brief Select compute-device which best matches criteria
cudaCreateChannelDesc
\brief Returns a channel descriptor using the specified format
cudaCreateSurfaceObject
\brief Creates a surface object
cudaCreateTextureObject
\brief Creates a texture object
cudaCtxResetPersistingL2Cache
\brief Resets all persisting lines in cache to normal status.
cudaDestroyExternalMemory
\brief Destroys an external memory object.
cudaDestroyExternalSemaphore
\brief Destroys an external semaphore
cudaDestroySurfaceObject
\brief Destroys a surface object
cudaDestroyTextureObject
\brief Destroys a texture object
cudaDeviceCanAccessPeer
\brief Queries if a device may directly access a peer device’s memory.
cudaDeviceDisablePeerAccess
\brief Disables direct access to memory allocations on a peer device.
cudaDeviceEnablePeerAccess
\brief Enables direct access to memory allocations on a peer device.
cudaDeviceFlushGPUDirectRDMAWrites
cudaDeviceGetAttribute
\brief Returns information about the device
cudaDeviceGetByPCIBusId
\brief Returns a handle to a compute device
cudaDeviceGetCacheConfig
\brief Returns the preferred cache configuration for the current device.
cudaDeviceGetDefaultMemPool
\brief Returns the default mempool of a device
cudaDeviceGetGraphMemAttribute
cudaDeviceGetLimit
\brief Returns resource limits
cudaDeviceGetMemPool
\brief Gets the current mempool for a device
cudaDeviceGetNvSciSyncAttributes
\brief Return NvSciSync attributes that this device can support.
cudaDeviceGetP2PAttribute
\brief Queries attributes of the link between two devices.
cudaDeviceGetPCIBusId
\brief Returns a PCI Bus Id string for the device
cudaDeviceGetSharedMemConfig
\brief Returns the shared memory configuration for the current device.
cudaDeviceGetStreamPriorityRange
\brief Returns numerical values that correspond to the least and greatest stream priorities.
cudaDeviceGetTexture1DLinearMaxWidth
cudaDeviceGraphMemTrim
cudaDeviceReset
\brief Destroy all allocations and reset all state on the current device in the current process.
cudaDeviceSetCacheConfig
\brief Sets the preferred cache configuration for the current device.
cudaDeviceSetGraphMemAttribute
cudaDeviceSetLimit
\brief Set resource limits
cudaDeviceSetMemPool
\brief Sets the current memory pool of a device
cudaDeviceSetSharedMemConfig
\brief Sets the shared memory configuration for the current device.
cudaDeviceSynchronize
\brief Wait for compute device to finish
cudaDriverGetVersion
\brief Returns the latest version of CUDA supported by the driver
cudaEventCreate
\brief Creates an event object
cudaEventCreateWithFlags
\brief Creates an event object with the specified flags
cudaEventDestroy
\brief Destroys an event object
cudaEventElapsedTime
\brief Computes the elapsed time between events
cudaEventQuery
\brief Queries an event’s status
cudaEventRecord
\brief Records an event
cudaEventRecordWithFlags
cudaEventSynchronize
\brief Waits for an event to complete
cudaExternalMemoryGetMappedBuffer
\brief Maps a buffer onto an imported memory object
cudaExternalMemoryGetMappedMipmappedArray
\brief Maps a CUDA mipmapped array onto an external memory object
cudaFree
\brief Frees memory on the device
cudaFreeArray
\brief Frees an array on the device
cudaFreeAsync
\brief Frees memory with stream ordered semantics
cudaFreeHost
\brief Frees page-locked memory
cudaFreeMipmappedArray
\brief Frees a mipmapped array on the device
cudaFuncGetAttributes
\brief Find out attributes for a given function
cudaFuncSetAttribute
\brief Set attributes for a given function
cudaFuncSetCacheConfig
\brief Sets the preferred cache configuration for a device function
cudaFuncSetSharedMemConfig
\brief Sets the shared memory configuration for a device function
cudaGetChannelDesc
\brief Get the channel descriptor of an array
cudaGetDevice
\brief Returns which device is currently being used
cudaGetDeviceCount
\brief Returns the number of compute-capable devices
cudaGetDeviceFlags
\brief Gets the flags for the current device
cudaGetDeviceProperties
\brief Returns information about the compute-device
cudaGetDriverEntryPoint
\brief Returns the requested driver API function pointer
cudaGetErrorName
\brief Returns the string representation of an error code enum name
cudaGetErrorString
\brief Returns the description string for an error code
cudaGetExportTable
\cond impl_private
cudaGetFuncBySymbol
\brief Get pointer to device entry function that matches entry function \p symbolPtr
cudaGetLastError
\brief Returns the last error from a runtime call
cudaGetMipmappedArrayLevel
\brief Gets a mipmap level of a CUDA mipmapped array
cudaGetSurfaceObjectResourceDesc
\brief Returns a surface object’s resource descriptor Returns the resource descriptor for the surface object specified by \p surfObject.
cudaGetSurfaceReference
\brief Get the surface reference associated with a symbol
cudaGetSymbolAddress
\brief Finds the address associated with a CUDA symbol
cudaGetSymbolSize
\brief Finds the size of the object associated with a CUDA symbol
cudaGetTextureAlignmentOffset
\brief Get the alignment offset of a texture
cudaGetTextureObjectResourceDesc
\brief Returns a texture object’s resource descriptor
cudaGetTextureObjectResourceViewDesc
\brief Returns a texture object’s resource view descriptor
cudaGetTextureObjectTextureDesc
\brief Returns a texture object’s texture descriptor
cudaGetTextureReference
\brief Get the texture reference associated with a symbol
cudaGraphAddChildGraphNode
\brief Creates a child graph node and adds it to a graph
cudaGraphAddDependencies
\brief Adds dependency edges to a graph.
cudaGraphAddEmptyNode
\brief Creates an empty node and adds it to a graph
cudaGraphAddEventRecordNode
cudaGraphAddEventWaitNode
cudaGraphAddExternalSemaphoresSignalNode
cudaGraphAddExternalSemaphoresWaitNode
cudaGraphAddHostNode
\brief Creates a host execution node and adds it to a graph
cudaGraphAddKernelNode
\brief Creates a kernel execution node and adds it to a graph
cudaGraphAddMemAllocNode
cudaGraphAddMemFreeNode
cudaGraphAddMemcpyNode
\brief Creates a memcpy node and adds it to a graph
cudaGraphAddMemcpyNode1D
cudaGraphAddMemcpyNodeFromSymbol
cudaGraphAddMemcpyNodeToSymbol
cudaGraphAddMemsetNode
\brief Creates a memset node and adds it to a graph
cudaGraphChildGraphNodeGetGraph
\brief Gets a handle to the embedded graph of a child graph node
cudaGraphClone
\brief Clones a graph
cudaGraphCreate
\brief Creates a graph
cudaGraphDebugDotPrint
\brief Write a DOT file describing graph structure
cudaGraphDestroy
\brief Destroys a graph
cudaGraphDestroyNode
\brief Remove a node from the graph
cudaGraphEventRecordNodeGetEvent
cudaGraphEventRecordNodeSetEvent
cudaGraphEventWaitNodeGetEvent
cudaGraphEventWaitNodeSetEvent
cudaGraphExecChildGraphNodeSetParams
cudaGraphExecDestroy
\brief Destroys an executable graph
cudaGraphExecEventRecordNodeSetEvent
cudaGraphExecEventWaitNodeSetEvent
cudaGraphExecExternalSemaphoresSignalNodeSetParams
cudaGraphExecExternalSemaphoresWaitNodeSetParams
cudaGraphExecHostNodeSetParams
\brief Sets the parameters for a host node in the given graphExec.
cudaGraphExecKernelNodeSetParams
\brief Sets the parameters for a kernel node in the given graphExec
cudaGraphExecMemcpyNodeSetParams
\brief Sets the parameters for a memcpy node in the given graphExec.
cudaGraphExecMemcpyNodeSetParams1D
cudaGraphExecMemcpyNodeSetParamsFromSymbol
cudaGraphExecMemcpyNodeSetParamsToSymbol
cudaGraphExecMemsetNodeSetParams
\brief Sets the parameters for a memset node in the given graphExec.
cudaGraphExecUpdate
\brief Check whether an executable graph can be updated with a graph and perform the update if possible
cudaGraphExternalSemaphoresSignalNodeGetParams
cudaGraphExternalSemaphoresSignalNodeSetParams
cudaGraphExternalSemaphoresWaitNodeGetParams
cudaGraphExternalSemaphoresWaitNodeSetParams
cudaGraphGetEdges
\brief Returns a graph’s dependency edges
cudaGraphGetNodes
\brief Returns a graph’s nodes
cudaGraphGetRootNodes
\brief Returns a graph’s root nodes
cudaGraphHostNodeGetParams
\brief Returns a host node’s parameters
cudaGraphHostNodeSetParams
\brief Sets a host node’s parameters
cudaGraphInstantiate
\brief Creates an executable graph from a graph
cudaGraphInstantiateWithFlags
cudaGraphKernelNodeCopyAttributes
\brief Copies attributes from source node to destination node.
cudaGraphKernelNodeGetAttribute
\brief Queries node attribute.
cudaGraphKernelNodeGetParams
\brief Returns a kernel node’s parameters
cudaGraphKernelNodeSetAttribute
\brief Sets node attribute.
cudaGraphKernelNodeSetParams
\brief Sets a kernel node’s parameters
cudaGraphLaunch
\brief Launches an executable graph in a stream
cudaGraphMemAllocNodeGetParams
cudaGraphMemFreeNodeGetParams
cudaGraphMemcpyNodeGetParams
\brief Returns a memcpy node’s parameters
cudaGraphMemcpyNodeSetParams
\brief Sets a memcpy node’s parameters
cudaGraphMemcpyNodeSetParams1D
cudaGraphMemcpyNodeSetParamsFromSymbol
cudaGraphMemcpyNodeSetParamsToSymbol
cudaGraphMemsetNodeGetParams
\brief Returns a memset node’s parameters
cudaGraphMemsetNodeSetParams
\brief Sets a memset node’s parameters
cudaGraphNodeFindInClone
\brief Finds a cloned version of a node
cudaGraphNodeGetDependencies
\brief Returns a node’s dependencies
cudaGraphNodeGetDependentNodes
\brief Returns a node’s dependent nodes
cudaGraphNodeGetType
\brief Returns a node’s type
cudaGraphReleaseUserObject
\brief Release a user object reference from a graph
cudaGraphRemoveDependencies
\brief Removes dependency edges from a graph.
cudaGraphRetainUserObject
\brief Retain a reference to a user object from a graph
cudaGraphUpload
cudaGraphicsMapResources
\brief Map graphics resources for access by CUDA
cudaGraphicsResourceGetMappedMipmappedArray
\brief Get a mipmapped array through which to access a mapped graphics resource.
cudaGraphicsResourceGetMappedPointer
\brief Get an device pointer through which to access a mapped graphics resource.
cudaGraphicsResourceSetMapFlags
\brief Set usage flags for mapping a graphics resource
cudaGraphicsSubResourceGetMappedArray
\brief Get an array through which to access a subresource of a mapped graphics resource.
cudaGraphicsUnmapResources
\brief Unmap graphics resources.
cudaGraphicsUnregisterResource
\brief Unregisters a graphics resource for access by CUDA
cudaHostAlloc
\brief Allocates page-locked memory on the host
cudaHostGetDevicePointer
\brief Passes back device pointer of mapped host memory allocated by cudaHostAlloc or registered by cudaHostRegister
cudaHostGetFlags
\brief Passes back flags used to allocate pinned host memory allocated by cudaHostAlloc
cudaHostRegister
\brief Registers an existing host memory range for use by CUDA
cudaHostUnregister
\brief Unregisters a memory range that was registered with cudaHostRegister
cudaImportExternalMemory
\brief Imports an external memory object
cudaImportExternalSemaphore
\brief Imports an external semaphore
cudaIpcCloseMemHandle
\brief Attempts to close memory mapped with cudaIpcOpenMemHandle
cudaIpcGetEventHandle
\brief Gets an interprocess handle for a previously allocated event
cudaIpcGetMemHandle
\brief Gets an interprocess memory handle for an existing device memory allocation
cudaIpcOpenEventHandle
\brief Opens an interprocess event handle for use in the current process
cudaIpcOpenMemHandle
\brief Opens an interprocess memory handle exported from another process and returns a device pointer usable in the local process.
cudaLaunchCooperativeKernel
\brief Launches a device function where thread blocks can cooperate and synchronize as they execute
cudaLaunchCooperativeKernelMultiDevice
\brief Launches device functions on multiple devices where thread blocks can cooperate and synchronize as they execute
cudaLaunchHostFunc
\brief Enqueues a host function call in a stream
cudaLaunchKernel
\brief Launches a device function
cudaMalloc
\brief Allocate memory on the device
cudaMalloc3D
\brief Allocates logical 1D, 2D, or 3D memory objects on the device
cudaMalloc3DArray
\brief Allocate an array on the device
cudaMallocArray
\brief Allocate an array on the device
cudaMallocAsync
\brief Allocates memory with stream ordered semantics
cudaMallocFromPoolAsync
\brief Allocates memory from a specified pool with stream ordered semantics.
cudaMallocHost
\brief Allocates page-locked memory on the host
cudaMallocManaged
cudaMallocMipmappedArray
\brief Allocate a mipmapped array on the device
cudaMallocPitch
\brief Allocates pitched memory on the device
cudaMemAdvise
\brief Advise about the usage of a given memory range
cudaMemGetInfo
\brief Gets free and total device memory
cudaMemPoolCreate
\brief Creates a memory pool
cudaMemPoolDestroy
\brief Destroys the specified memory pool
cudaMemPoolExportPointer
\brief Export data to share a memory pool allocation between processes.
cudaMemPoolExportToShareableHandle
\brief Exports a memory pool to the requested handle type.
cudaMemPoolGetAccess
\brief Returns the accessibility of a pool from a device
cudaMemPoolGetAttribute
\brief Gets attributes of a memory pool
cudaMemPoolImportFromShareableHandle
\brief imports a memory pool from a shared handle.
cudaMemPoolImportPointer
\brief Import a memory pool allocation from another process.
cudaMemPoolSetAccess
\brief Controls visibility of pools between devices
cudaMemPoolSetAttribute
\brief Sets attributes of a memory pool
cudaMemPoolTrimTo
\brief Tries to release memory back to the OS
cudaMemPrefetchAsync
\brief Prefetches memory to the specified destination device
cudaMemRangeGetAttribute
\brief Query an attribute of a given memory range
cudaMemRangeGetAttributes
\brief Query attributes of a given memory range.
cudaMemcpy
\brief Copies data between host and device
cudaMemcpy2D
\brief Copies data between host and device
cudaMemcpy2DArrayToArray
\brief Copies data between host and device
cudaMemcpy2DAsync
\brief Copies data between host and device
cudaMemcpy2DFromArray
\brief Copies data between host and device
cudaMemcpy2DFromArrayAsync
\brief Copies data between host and device
cudaMemcpy2DToArray
\brief Copies data between host and device
cudaMemcpy2DToArrayAsync
\brief Copies data between host and device
cudaMemcpy3D
\brief Copies data between 3D objects
cudaMemcpy3DAsync
\brief Copies data between 3D objects
cudaMemcpy3DPeer
\brief Copies memory between devices
cudaMemcpy3DPeerAsync
\brief Copies memory between devices asynchronously.
cudaMemcpyArrayToArray
\brief Copies data between host and device
cudaMemcpyAsync
\brief Copies data between host and device
cudaMemcpyFromArray
\brief Copies data between host and device
cudaMemcpyFromArrayAsync
\brief Copies data between host and device
cudaMemcpyFromSymbol
\brief Copies data from the given symbol on the device
cudaMemcpyFromSymbolAsync
\brief Copies data from the given symbol on the device
cudaMemcpyPeer
\brief Copies memory between two devices
cudaMemcpyPeerAsync
\brief Copies memory between two devices asynchronously.
cudaMemcpyToArray
\brief Copies data between host and device
cudaMemcpyToArrayAsync
\brief Copies data between host and device
cudaMemcpyToSymbol
\brief Copies data to the given symbol on the device
cudaMemcpyToSymbolAsync
\brief Copies data to the given symbol on the device
cudaMemset
\brief Initializes or sets device memory to a value
cudaMemset2D
\brief Initializes or sets device memory to a value
cudaMemset2DAsync
\brief Initializes or sets device memory to a value
cudaMemset3D
\brief Initializes or sets device memory to a value
cudaMemset3DAsync
\brief Initializes or sets device memory to a value
cudaMemsetAsync
\brief Initializes or sets device memory to a value
cudaMipmappedArrayGetSparseProperties
cudaOccupancyAvailableDynamicSMemPerBlock
\brief Returns dynamic shared memory available per block when launching \p numBlocks blocks on SM.
cudaOccupancyMaxActiveBlocksPerMultiprocessor
\brief Returns occupancy for a device function
cudaOccupancyMaxActiveBlocksPerMultiprocessorWithFlags
\brief Returns occupancy for a device function with the specified flags
cudaPeekAtLastError
\brief Returns the last error from a runtime call
cudaPointerGetAttributes
\brief Returns attributes about a specified pointer
cudaRuntimeGetVersion
\brief Returns the CUDA Runtime version
cudaSetDevice
\brief Set device to be used for GPU executions
cudaSetDeviceFlags
\brief Sets flags to be used for device executions
cudaSetDoubleForDevice
\brief Converts a double argument to be executed on a device
cudaSetDoubleForHost
\brief Converts a double argument after execution on a device
cudaSetValidDevices
\brief Set a list of devices that can be used for CUDA
cudaSignalExternalSemaphoresAsync_v2
cudaStreamAddCallback
\brief Add a callback to a compute stream
cudaStreamAttachMemAsync
cudaStreamBeginCapture
\brief Begins graph capture on a stream
cudaStreamCopyAttributes
\brief Copies attributes from source stream to destination stream.
cudaStreamCreate
\brief Create an asynchronous stream
cudaStreamCreateWithFlags
\brief Create an asynchronous stream
cudaStreamCreateWithPriority
\brief Create an asynchronous stream with the specified priority
cudaStreamDestroy
\brief Destroys and cleans up an asynchronous stream
cudaStreamEndCapture
\brief Ends capture on a stream, returning the captured graph
cudaStreamGetAttribute
\brief Queries stream attribute.
cudaStreamGetCaptureInfo
\brief Query capture status of a stream
cudaStreamGetCaptureInfo_v2
\brief Query a stream’s capture state (11.3+)
cudaStreamGetFlags
\brief Query the flags of a stream
cudaStreamGetPriority
\brief Query the priority of a stream
cudaStreamIsCapturing
\brief Returns a stream’s capture status
cudaStreamQuery
\brief Queries an asynchronous stream for completion status
cudaStreamSetAttribute
\brief Sets stream attribute.
cudaStreamSynchronize
\brief Waits for stream tasks to complete
cudaStreamUpdateCaptureDependencies
\brief Update the set of dependencies in a capturing stream (11.3+)
cudaStreamWaitEvent
\brief Make a compute stream wait on an event
cudaThreadExchangeStreamCaptureMode
\brief Swaps the stream capture interaction mode for a thread
cudaThreadExit
\brief Exit and clean up from CUDA launches
cudaThreadGetCacheConfig
\brief Returns the preferred cache configuration for the current device.
cudaThreadGetLimit
\brief Returns resource limits
cudaThreadSetCacheConfig
\brief Sets the preferred cache configuration for the current device.
cudaThreadSetLimit
\brief Set resource limits
cudaThreadSynchronize
\brief Wait for compute device to finish
cudaUnbindTexture
\brief Unbinds a texture
cudaUserObjectCreate
\brief Create a user object
cudaUserObjectRelease
\brief Release a reference to a user object
cudaUserObjectRetain
\brief Retain a reference to a user object
cudaWaitExternalSemaphoresAsync_v2
cudnnActivationBackward
cudnnActivationForward
cudnnAddTensor
cudnnAdvInferVersionCheck
cudnnAdvTrainVersionCheck
cudnnBackendCreateDescriptor
cudnnBackendDestroyDescriptor
cudnnBackendExecute
cudnnBackendFinalize
cudnnBackendGetAttribute
cudnnBackendInitialize
cudnnBackendSetAttribute
cudnnBatchNormalizationBackward
cudnnBatchNormalizationBackwardEx
cudnnBatchNormalizationForwardInference
cudnnBatchNormalizationForwardTraining
cudnnBatchNormalizationForwardTrainingEx
cudnnBuildRNNDynamic
cudnnCTCLoss
cudnnCTCLoss_v8
cudnnCnnInferVersionCheck
cudnnCnnTrainVersionCheck
cudnnConvolutionBackwardBias
cudnnConvolutionBackwardData
cudnnConvolutionBackwardFilter
cudnnConvolutionBiasActivationForward
cudnnConvolutionForward
cudnnCopyAlgorithmDescriptor
cudnnCreate
cudnnCreateActivationDescriptor
cudnnCreateAlgorithmDescriptor
cudnnCreateAlgorithmPerformance
cudnnCreateAttnDescriptor
cudnnCreateCTCLossDescriptor
cudnnCreateConvolutionDescriptor
cudnnCreateDropoutDescriptor
cudnnCreateFilterDescriptor
cudnnCreateFusedOpsConstParamPack
cudnnCreateFusedOpsPlan
cudnnCreateFusedOpsVariantParamPack
cudnnCreateLRNDescriptor
cudnnCreateOpTensorDescriptor
cudnnCreatePersistentRNNPlan
cudnnCreatePoolingDescriptor
cudnnCreateRNNDataDescriptor
cudnnCreateRNNDescriptor
cudnnCreateReduceTensorDescriptor
cudnnCreateSeqDataDescriptor
cudnnCreateSpatialTransformerDescriptor
cudnnCreateTensorDescriptor
cudnnCreateTensorTransformDescriptor
Create an empty tensor transform descriptor
cudnnDeriveBNTensorDescriptor
cudnnDeriveNormTensorDescriptor
cudnnDestroy
cudnnDestroyActivationDescriptor
cudnnDestroyAlgorithmDescriptor
cudnnDestroyAlgorithmPerformance
cudnnDestroyAttnDescriptor
cudnnDestroyCTCLossDescriptor
cudnnDestroyConvolutionDescriptor
cudnnDestroyDropoutDescriptor
cudnnDestroyFilterDescriptor
cudnnDestroyFusedOpsConstParamPack
cudnnDestroyFusedOpsPlan
cudnnDestroyFusedOpsVariantParamPack
cudnnDestroyLRNDescriptor
cudnnDestroyOpTensorDescriptor
cudnnDestroyPersistentRNNPlan
cudnnDestroyPoolingDescriptor
cudnnDestroyRNNDataDescriptor
cudnnDestroyRNNDescriptor
cudnnDestroyReduceTensorDescriptor
cudnnDestroySeqDataDescriptor
cudnnDestroySpatialTransformerDescriptor
cudnnDestroyTensorDescriptor
cudnnDestroyTensorTransformDescriptor
Destroys a previously created tensor transform descriptor.
cudnnDivisiveNormalizationBackward
cudnnDivisiveNormalizationForward
cudnnDropoutBackward
cudnnDropoutForward
cudnnDropoutGetReserveSpaceSize
cudnnDropoutGetStatesSize
cudnnFindConvolutionBackwardDataAlgorithm
cudnnFindConvolutionBackwardDataAlgorithmEx
cudnnFindConvolutionBackwardFilterAlgorithm
cudnnFindConvolutionBackwardFilterAlgorithmEx
cudnnFindConvolutionForwardAlgorithm
cudnnFindConvolutionForwardAlgorithmEx
cudnnFindRNNBackwardDataAlgorithmEx
cudnnFindRNNBackwardWeightsAlgorithmEx
cudnnFindRNNForwardInferenceAlgorithmEx
cudnnFindRNNForwardTrainingAlgorithmEx
cudnnFusedOpsExecute
cudnnGetActivationDescriptor
cudnnGetActivationDescriptorSwishBeta
cudnnGetAlgorithmDescriptor
cudnnGetAlgorithmPerformance
cudnnGetAlgorithmSpaceSize
cudnnGetAttnDescriptor
cudnnGetBatchNormalizationBackwardExWorkspaceSize
cudnnGetBatchNormalizationForwardTrainingExWorkspaceSize
cudnnGetBatchNormalizationTrainingExReserveSpaceSize
cudnnGetCTCLossDescriptor
cudnnGetCTCLossDescriptorEx
cudnnGetCTCLossDescriptor_v8
cudnnGetCTCLossWorkspaceSize
cudnnGetCTCLossWorkspaceSize_v8
cudnnGetCallback
cudnnGetConvolution2dDescriptor
cudnnGetConvolution2dForwardOutputDim
cudnnGetConvolutionBackwardDataAlgorithmMaxCount
cudnnGetConvolutionBackwardDataAlgorithm_v7
cudnnGetConvolutionBackwardDataWorkspaceSize
cudnnGetConvolutionBackwardFilterAlgorithmMaxCount
cudnnGetConvolutionBackwardFilterAlgorithm_v7
cudnnGetConvolutionBackwardFilterWorkspaceSize
cudnnGetConvolutionForwardAlgorithmMaxCount
cudnnGetConvolutionForwardAlgorithm_v7
cudnnGetConvolutionForwardWorkspaceSize
cudnnGetConvolutionGroupCount
cudnnGetConvolutionMathType
cudnnGetConvolutionNdDescriptor
cudnnGetConvolutionNdForwardOutputDim
cudnnGetConvolutionReorderType
cudnnGetCudartVersion
cudnnGetDropoutDescriptor
cudnnGetErrorString
cudnnGetFilter4dDescriptor
cudnnGetFilterNdDescriptor
cudnnGetFilterSizeInBytes
cudnnGetFoldedConvBackwardDataDescriptors
cudnnGetFusedOpsConstParamPackAttribute
cudnnGetFusedOpsVariantParamPackAttribute
cudnnGetLRNDescriptor
cudnnGetMultiHeadAttnBuffers
cudnnGetMultiHeadAttnWeights
cudnnGetNormalizationBackwardWorkspaceSize
cudnnGetNormalizationForwardTrainingWorkspaceSize
cudnnGetNormalizationTrainingReserveSpaceSize
cudnnGetOpTensorDescriptor
cudnnGetPooling2dDescriptor
cudnnGetPooling2dForwardOutputDim
cudnnGetPoolingNdDescriptor
cudnnGetPoolingNdForwardOutputDim
cudnnGetProperty
cudnnGetRNNBackwardDataAlgorithmMaxCount
cudnnGetRNNBackwardWeightsAlgorithmMaxCount
cudnnGetRNNBiasMode
cudnnGetRNNDataDescriptor
cudnnGetRNNDescriptor_v6
cudnnGetRNNDescriptor_v8
cudnnGetRNNForwardInferenceAlgorithmMaxCount
cudnnGetRNNForwardTrainingAlgorithmMaxCount
cudnnGetRNNLinLayerBiasParams
cudnnGetRNNLinLayerMatrixParams
cudnnGetRNNMatrixMathType
cudnnGetRNNPaddingMode
cudnnGetRNNParamsSize
cudnnGetRNNProjectionLayers
cudnnGetRNNTempSpaceSizes
cudnnGetRNNTrainingReserveSize
cudnnGetRNNWeightParams
cudnnGetRNNWeightSpaceSize
cudnnGetRNNWorkspaceSize
cudnnGetReduceTensorDescriptor
cudnnGetReductionIndicesSize
cudnnGetReductionWorkspaceSize
cudnnGetSeqDataDescriptor
cudnnGetStream
cudnnGetTensor4dDescriptor
cudnnGetTensorNdDescriptor
cudnnGetTensorSizeInBytes
cudnnGetTensorTransformDescriptor
Retrieves the values stored in a previously initialized tensor transform descriptor.
cudnnGetVersion
cudnnIm2Col
cudnnInitTransformDest
Create a destination descriptor for cudnnTransformTensor
cudnnLRNCrossChannelBackward
cudnnLRNCrossChannelForward
cudnnMakeFusedOpsPlan
cudnnMultiHeadAttnBackwardData
cudnnMultiHeadAttnBackwardWeights
cudnnMultiHeadAttnForward
cudnnNormalizationBackward
cudnnNormalizationForwardInference
cudnnNormalizationForwardTraining
cudnnOpTensor
cudnnOpsInferVersionCheck
cudnnOpsTrainVersionCheck
cudnnPoolingBackward
cudnnPoolingForward
cudnnQueryRuntimeError
cudnnRNNBackwardData
cudnnRNNBackwardDataEx
cudnnRNNBackwardData_v8
cudnnRNNBackwardWeights
cudnnRNNBackwardWeightsEx
cudnnRNNBackwardWeights_v8
cudnnRNNForward
cudnnRNNForwardInference
cudnnRNNForwardInferenceEx
cudnnRNNForwardTraining
cudnnRNNForwardTrainingEx
cudnnRNNGetClip
cudnnRNNGetClip_v8
cudnnRNNSetClip
cudnnRNNSetClip_v8
cudnnReduceTensor
cudnnReorderFilterAndBias
cudnnRestoreAlgorithm
cudnnRestoreDropoutDescriptor
cudnnSaveAlgorithm
cudnnScaleTensor
cudnnSetActivationDescriptor
cudnnSetActivationDescriptorSwishBeta
cudnnSetAlgorithmDescriptor
cudnnSetAlgorithmPerformance
cudnnSetAttnDescriptor
cudnnSetCTCLossDescriptor
cudnnSetCTCLossDescriptorEx
cudnnSetCTCLossDescriptor_v8
cudnnSetCallback
cudnnSetConvolution2dDescriptor
cudnnSetConvolutionGroupCount
cudnnSetConvolutionMathType
cudnnSetConvolutionNdDescriptor
cudnnSetConvolutionReorderType
cudnnSetDropoutDescriptor
cudnnSetFilter4dDescriptor
cudnnSetFilterNdDescriptor
cudnnSetFusedOpsConstParamPackAttribute
cudnnSetFusedOpsVariantParamPackAttribute
cudnnSetLRNDescriptor
cudnnSetOpTensorDescriptor
cudnnSetPersistentRNNPlan
cudnnSetPooling2dDescriptor
cudnnSetPoolingNdDescriptor
cudnnSetRNNAlgorithmDescriptor
cudnnSetRNNBiasMode
cudnnSetRNNDataDescriptor
cudnnSetRNNDescriptor_v6
cudnnSetRNNDescriptor_v8
cudnnSetRNNMatrixMathType
cudnnSetRNNPaddingMode
cudnnSetRNNProjectionLayers
cudnnSetReduceTensorDescriptor
cudnnSetSeqDataDescriptor
cudnnSetSpatialTransformerNdDescriptor
cudnnSetStream
cudnnSetTensor
cudnnSetTensor4dDescriptor
cudnnSetTensor4dDescriptorEx
cudnnSetTensorNdDescriptor
cudnnSetTensorNdDescriptorEx
cudnnSetTensorTransformDescriptor
Initialize a previously created tensor transform descriptor.
cudnnSoftmaxBackward
cudnnSoftmaxForward
cudnnSpatialTfGridGeneratorBackward
cudnnSpatialTfGridGeneratorForward
cudnnSpatialTfSamplerBackward
cudnnSpatialTfSamplerForward
cudnnTransformFilter
cudnnTransformTensor
cudnnTransformTensorEx
tensor_vec_id_c
Return C Handle for a Vector of Tensor Descriptors

Type Aliases§

CUuuid
__int32_t
__int64_t
__uint32_t
cudaArray_const_t
CUDA array (as source copy argument)
cudaArray_t
CUDA array
cudaEvent_t
CUDA event types
cudaExternalMemory_t
CUDA external memory
cudaExternalSemaphore_t
CUDA external semaphore
cudaFunction_t
CUDA function
cudaGraphExec_t
CUDA executable (launchable) graph
cudaGraphNode_t
CUDA graph node.
cudaGraph_t
CUDA graph
cudaGraphicsResource_t
CUDA graphics resource types
cudaHostFn_t
CUDA host function \param userData Argument value passed to the function
cudaIpcEventHandle_t
CUDA IPC event handle
cudaIpcMemHandle_t
CUDA IPC memory handle
cudaMemPool_t
CUDA memory pool
cudaMipmappedArray_const_t
CUDA mipmapped array (as source argument)
cudaMipmappedArray_t
CUDA mipmapped array
cudaStreamCallback_t
Type of stream callback functions. \param stream The stream as passed to ::cudaStreamAddCallback, may be NULL. \param status ::cudaSuccess or any persistent error on the stream. \param userData User parameter provided at registration.
cudaStream_t
CUDA stream
cudaSurfaceObject_t
An opaque value that represents a CUDA Surface object
cudaTextureObject_t
An opaque value that represents a CUDA texture object
cudaUUID_t
cudaUserObject_t
CUDA user object for graphs
cudnnActivationDescriptor_t
cudnnAlgorithmDescriptor_t
cudnnAlgorithmPerformance_t
cudnnAlgorithm_t
cudnnAttnDescriptor_t
cudnnAttnQueryMap_t
cudnnBackendDescriptor_t
cudnnCTCLossDescriptor_t
cudnnCallback_t
cudnnConvolutionBwdDataAlgoPerf_t
cudnnConvolutionBwdFilterAlgoPerf_t
cudnnConvolutionDescriptor_t
cudnnConvolutionFwdAlgoPerf_t
cudnnDebug_t
cudnnDropoutDescriptor_t
cudnnFilterDescriptor_t
cudnnFusedOpsConstParamPack_t
cudnnFusedOpsPlan_t
cudnnFusedOpsVariantParamPack_t
cudnnHandle_t
cudnnLRNDescriptor_t
cudnnOpTensorDescriptor_t
cudnnPersistentRNNPlan_t
cudnnPoolingDescriptor_t
cudnnRNNDataDescriptor_t
cudnnRNNDescriptor_t
cudnnRNNPaddingMode_t
cudnnReduceTensorDescriptor_t
cudnnSeqDataDescriptor_t
cudnnSpatialTransformerDescriptor_t
cudnnTensorDescriptor_t
cudnnTensorTransformDescriptor_t
libraryPropertyType
libraryPropertyType_t

Unions§

cudaExternalMemoryHandleDesc__bindgen_ty_1
cudaExternalSemaphoreHandleDesc__bindgen_ty_1
cudaExternalSemaphoreSignalParams__bindgen_ty_1__bindgen_ty_2
cudaExternalSemaphoreWaitParams__bindgen_ty_1__bindgen_ty_2
cudaKernelNodeAttrValue
Graph kernel node attributes union, used with ::cudaGraphKernelNodeSetAttribute/::cudaGraphKernelNodeGetAttribute
cudaResourceDesc__bindgen_ty_1
cudaStreamAttrValue
Stream attributes union used with ::cudaStreamSetAttribute/::cudaStreamGetAttribute
cudnnAlgorithmUnionStruct_Algorithm