3D array descriptor
Array descriptor
CUDA array memory requirements
CUDA array sparse properties
External memory buffer descriptor
External memory handle descriptor
Win32 handle referencing the semaphore object. Valid when
type is one of the following:
External memory mipmap descriptor
External semaphore handle descriptor
Win32 handle referencing the semaphore object. Valid when
type is one of the following:
External semaphore signal parameters
Parameters for fence objects
Parameters for keyed mutex objects
External semaphore wait parameters
Parameters for fence objects
Parameters for keyed mutex objects
Semaphore signal node parameters
Semaphore wait node parameters
Graph instantiation parameters
Host node parameters
GPU kernel node parameters
Kernel launch parameters
2D memory copy parameters
3D memory cross-context copy parameters
3D memory copy parameters
Memset node parameters
Memory allocation node parameters
CUDA Resource descriptor
Resource view descriptor
Texture descriptor
Specifies an access policy for a window, a contiguous extent of memory
beginning at base_ptr and ending at base_ptr + num_bytes.
num_bytes is limited by CU_DEVICE_ATTRIBUTE_MAX_ACCESS_POLICY_WINDOW_SIZE.
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.
Specifies the CUDA array or CUDA mipmapped array memory mapping information
Legacy device properties
Execution Affinity Parameters
Value for ::CU_EXEC_AFFINITY_TYPE_SM_COUNT
Result information returned by cuGraphExecUpdate
CUDA IPC event handle
CUDA IPC mem handle
Memory access descriptor
Specifies the allocation properties for a allocation.
Specifies a memory location.
Specifies the properties of allocations made from the pool.
Opaque data for exporting a pool allocation
Tensor map descriptor. Requires compiler support for aligning to 64 bytes.
Specifies performance hint with ::CUaccessPolicyWindow for hitProp and missProp members.
Texture reference addressing modes
Sparse subresource types
Array formats
Cluster scheduling policies. These may be passed to ::cuFuncSetAttribute or ::cuKernelSetAttribute
P2P Attributes
Device properties
Flags to indicate search status. For more details see ::cuGetProcAddress
Execution Affinity Types
External memory handle types
External semaphore handle types
Texture reference filtering modes
The scopes for ::cuFlushGPUDirectRDMAWrites
The targets for ::cuFlushGPUDirectRDMAWrites
Function cache configurations
Function properties
CUDA Graph Update error types
Graph instantiation results
Graph node types
Device code formats
Online compiler and linker options
Library options to be specified with ::cuLibraryLoadData() or ::cuLibraryLoadFromFile()
Limits
Specifies the memory protection flags for mapping.
Flag for requesting different optimal and required granularities for an allocation.
Flags for specifying particular handle types
Defines the allocation types available
Memory handle types
Specifies the type of location
Memory operation types
CUDA memory pool attributes
Specifies the handle type for address range
Memory advise values
Memory types
CUDA Lazy Loading status
Pointer information
Resource view format
Resource types
Shared memory configurations
Operations for ::cuStreamBatchMemOp
Possible modes for stream capture thread interactions. For more details see
::cuStreamBeginCapture and ::cuThreadExchangeStreamCaptureMode
Possible stream capture statuses returned by ::cuStreamIsCapturing
Tensor map data type
Tensor map out-of-bounds fill type
Tensor map interleave layout type
Tensor map L2 promotion type
Tensor map swizzling mode of shared memory banks
Error codes
3D array descriptor
3D array descriptor
Array descriptor
Array descriptor
CUDA array memory requirements
CUDA array memory requirements
CUDA array sparse properties
CUDA array sparse properties
External memory buffer descriptor
External memory buffer descriptor
External memory handle descriptor
External memory handle descriptor
External memory mipmap descriptor
External memory mipmap descriptor
External semaphore handle descriptor
External semaphore handle descriptor
External semaphore signal parameters
External semaphore signal parameters
External semaphore wait parameters
External semaphore wait parameters
Semaphore signal node parameters
Semaphore signal node parameters
Semaphore wait node parameters
Semaphore wait node parameters
Graph instantiation parameters
Host node parameters
Host node parameters
GPU kernel node parameters
GPU kernel node parameters
Kernel launch parameters
Kernel launch parameters
2D memory copy parameters
2D memory copy parameters
3D memory copy parameters
3D memory cross-context copy parameters
3D memory cross-context copy parameters
3D memory copy parameters
Memset node parameters
Memset node parameters
Memory allocation node parameters
CUDA Resource descriptor
CUDA Resource descriptor
Resource view descriptor
Resource view descriptor
Texture descriptor
Texture descriptor
Access policy window
Specifies an access policy for a window, a contiguous extent of memory
beginning at base_ptr and ending at base_ptr + num_bytes.
num_bytes is limited by CU_DEVICE_ATTRIBUTE_MAX_ACCESS_POLICY_WINDOW_SIZE.
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.
Specifies the CUDA array or CUDA mipmapped array memory mapping information
Specifies the CUDA array or CUDA mipmapped array memory mapping information
Legacy device properties
Legacy device properties
Execution Affinity Parameters
Execution Affinity Parameters
Value for ::CU_EXEC_AFFINITY_TYPE_SM_COUNT
Value for ::CU_EXEC_AFFINITY_TYPE_SM_COUNT
Result information returned by cuGraphExecUpdate
Result information returned by cuGraphExecUpdate
CUDA host function
\param userData Argument value passed to the function
CUDA IPC event handle
CUDA IPC event handle
CUDA IPC mem handle
CUDA IPC mem handle
Graph kernel node attributes union, used with ::cuKernelNodeSetAttribute/::cuKernelNodeGetAttribute
Memory access descriptor
Memory access descriptor
Specifies the allocation properties for a allocation.
Specifies the allocation properties for a allocation.
Specifies a memory location.
Specifies a memory location.
Specifies the properties of allocations made from the pool.
Specifies the properties of allocations made from the pool.
Opaque data for exporting a pool allocation
Opaque data for exporting a pool allocation
Block size to per-block dynamic shared memory mapping for a certain
kernel \param blockSize Block size of the kernel.
Stream attributes union, used with ::cuStreamSetAttribute/::cuStreamGetAttribute
Per-operation parameters for ::cuStreamBatchMemOp
Per-operation parameters for ::cuStreamBatchMemOp
CUDA stream callback
\param hStream The stream the callback was added to, as passed to ::cuStreamAddCallback. May be NULL.
\param status ::CUDA_SUCCESS or any persistent error on the stream.
\param userData User parameter provided at registration.
Tensor map descriptor. Requires compiler support for aligning to 64 bytes.